2021 National Elder Abuse Prevalence Study Final Report 0
2021 National Elder Abuse Prevalence Study Final Report 0
2021 National Elder Abuse Prevalence Study Final Report 0
aifs.gov.au
The Australian Institute of Family Studies is committed to the creation and dissemination of research-based information
on family functioning and wellbeing. Views expressed in its publications are those of individual authors and may not
reflect those of the Australian Institute of Family Studies or the Australian Government.
Australian Institute of Family Studies
Level 4, 40 City Road, Southbank VIC 3006 Australia
Phone: (03) 9214 7888 Internet: aifs.gov.au
Suggested citation:
Qu, L., Kaspiew, R., Carson, R., Roopani, D., De Maio, J., Harvey, J., Horsfall, B. (2021). National Elder Abuse Prevalence
Study: Final Report. (Research Report). Melbourne: Australian Institute of Family Studies.
ISBN 978-1-76016-217-7 (Online)
ISBN 978-1-76016-218-4 (PDF)
Edited by Katharine Day
Typeset by Lisa Carroll
Cover images:
Image 1 © Gettyimages/halfpoint
Image 2 © Gettyimages/SilviaJansen
Images 3 and 4 © Sally Cashmore
Warm Safe Home project 16 Days of Activism installation, Port Campbell Foreshore, November 2020
Timboon Men’s Shed and Everybody’s Business Elder Abuse Prevention Network (South West Victoria)
About the Warm Safe Home project
What does a warm safe home mean to you? In many cultures the home is the key to a sense of safety. Homes shelter us
from the outside world and provide space to carry out our lives. In family violence situations, however, the home can be
a place of danger, loss of control, and/or neglect. The home is often a central asset in a family estate over which financial
abuse may be perpetrated. Housing insecurity can increase direct risk of homelessness for older people and can also
lead them to provide accommodation to loved ones who may perpetrate elder abuse. The Warm Safe Home project
spreads a simple message with a big meaning: everyone has the right to a warm safe home at every age.
2021_National Elder Abuse Prevalence Study Final Report
iii
Contents
Acknowledgements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
Acronyms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii
Glossary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii
Executive summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Policy context. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
National Elder Abuse Prevalence Study. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Prevalence of elder abuse in Australia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Socio-demographic characteristics associated with abuse. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Health status and social connection and elder abuse. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Implications for policy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Background. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2. Evidence on elder abuse. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Empirical evidence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3. Definition and concepts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Conceptual definition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Conceptual and theoretical frameworks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4. Methodology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Study outline. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Sample characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
5. Prevalence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Deriving prevalence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Prevalence findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Culturally and linguistically diverse participants. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Concerns about elder abuse in the Survey of the General Community. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
6. Who experiences elder abuse?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Experience of elder abuse by gender and age. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Characteristics most likely to be associated with abuse. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Health status, social connection and elder abuse. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Psychological health . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Regression analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Demographic characteristics and the CALD subsample. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
7. Who commits elder abuse?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Relationship between perpetrators and older persons. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Perpetrator characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
Problems associated with perpetrators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Overview of perpetrators: CALD subgroup. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
8. Seeking help. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
Responses to abuse: Survey of Older People. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Responses to concerns about abuse: Survey of the General Community . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
iv National Elder Abuse Prevalence Study: Final Report
9. Self-assessed seriousness. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
Self-assessed seriousness. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
Self-assessed seriousness and selected characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Self-assessed seriousness and help seeking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
10. Legal arrangements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Advance planning among older people. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .99
Advance planning and socio-demographic characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Power of attorney: Who was appointed?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
With whom are family agreements made?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
11. Giving and receiving support. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Assistance with financial matters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
12. Attitudes, awareness and understanding. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
Attitudes towards elder abuse: findings (SOP and SGC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
Ageism, intergenerational support and sense of entitlement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
Intergenerational support. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
Attitudes towards older people, intergenerational support and elder abuse. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
13. Elder abuse and CALD groups: synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
Prevalence of elder abuse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
Characteristics of those who experience abuse in the CALD subsample. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
CALD subsample and perpetrator profiles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
CALD subsample and legal arrangements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
CALD subsample and attitudes, awareness and understanding. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
14. Summary and implications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Health, social connection and elder abuse. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
Definition of elder abuse. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
Implications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
Appendix A: Additional tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
Appendix B: Methodology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
Appendix C: Survey questionnaires. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
v
List of figures
Figure 5.1: Survey of Older People: Participants who were classified as experiencing financial abuse –
proportion reporting each financial abuse item and number of items reported. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Figure 5.2: Survey of Older People: Participants who were classified as experiencing physical abuse –
proportion reporting each physical abuse item and number of items reported. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Figure 5.3: Survey of Older People: Participants who were classified as experiencing sexual abuse –
proportion reporting each sexual abuse item and number of items reported. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Figure 5.4: Survey of Older People: Participants who were classified as experiencing psychological
abuse – proportion reporting each psychological abuse item, by psychological abuse score (band). . . . . . . . . . . . 43
Figure 5.5: Survey of Older People: Participants who were classified as experiencing neglect –
proportion reporting each neglect item, by neglect score (band). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Figure 5.6: Survey of the General Community: Concerns about older family members and friends. . . . . . . . . . . . . 52
Figure 5.7: Survey of the General Community: Number of concerns reported about older family
members and friends. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Figure 6.1: Prevalence of elder abuse by gender. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Figure 6.2: Prevalence of elder abuse by age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
Figure 6.3: Prevalence of elder abuse by age and gender. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
Figure 7.1: Survey of Older People: Relationship of perpetrators to the participant, by subtypes and
overall (as % of perpetrators). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Figure 7.2: Survey of Older People: Relationship of perpetrators to older people who experienced at
least one type of abuse and are from a CALD background. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Figure 7.3: Survey of the General Community: Perpetrators identified for reported concerns by subtype. . . . . . . 79
Figure 8.1: Survey of Older People: Whether help or advice sought by persons who experienced each
subtype of elder abuse in the previous 12 months. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Figure 8.2: Survey of Older People: Proportion of participants reported action taken was effective by
type of abuse. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Figure 8.3: Survey of the General Community: Where an action was taken in response to concerns, what
action was taken, by gender . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
Figure 9.1: Survey of Older People: Reports of seriousness of main perpetrator’s behaviour, persons who
experience each subtype of elder abuse in the previous 12 months . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
Figure 10.1: Survey of Older People: Older persons with whom the family agreement arrangements were
made, by gender. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
Figure 11.1: Survey of Older People: Proportion of participants who received each type of assistance with
financial matters a. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
Figure 11.2: Survey of Older People: Proportion of participants who received each type of assistance with
financial matters, by gender. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
Figure 11.3: Survey of Older People: Proportion of participants who received each type of assistance with
financial matters, by age. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Figure 11.4: Survey of Older People: Older persons who received assistance in financial matters, who
provided assistance with financial matters, by gender. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
Figure 11.5: Survey of Older People: Older persons who received assistance in financial matters, who
provided assistance with financial matters, by age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
Figure 11.6: Survey of Older People: Older persons whose PIN was accessible by someone else, who had
access to PIN, by gender of participant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
Figure 11.7: Survey of Older People: Older persons whose PIN was accessible by someone else, who had
access to PIN, by age of participant. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Figure 11.8: Survey of the General Community: Proportion of SGC participants who provided various
types of assistance with financial matters in the past 12 months. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
vi National Elder Abuse Prevalence Study: Final Report
Figure 11.9: Survey of the General Community: Participants who provided assistance with financial
matters, who received the assistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
Figure 11.10: Survey of the General Community: Type of financial matter, by age of oldest person
receiving assistance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
Figure 11.11: Survey of Older People: Proportion of participants who reported experience of financial
abuse by whether receiving assistance on financial matters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
Figure 12.1: SOP and SGC: Attitudes towards elder abuse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
Figure 12.2: SOP and SGC: Mean scores of scale of acceptance attitudes, by gender. . . . . . . . . . . . . . . . . . . . . . . . . . 133
Figure 12.3: SOP and SGC: Mean scores of scale of acceptance attitudes, by age. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
Figure 12.4: SOP and SGC: Views on various abusive behaviours. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
Figure 12.5: SOP and SGC: Mean scores of scale of recognising abusive behaviours, by gender. . . . . . . . . . . . . . . . 137
Figure 12.6: SOP and SGC: Mean scores of scale of recognising abusive behaviours, by age. . . . . . . . . . . . . . . . . . . 138
Figure 12.7: Survey of the General Community: Bivariate analysis of the Ageism scale by gender – Mean score. . 141
Figure 12.8: Survey of the General Community: Ageism scale by age – Mean score. . . . . . . . . . . . . . . . . . . . . . . . . . . 142
Figure 12.9: Survey of the General Community: Bivariate analysis of the Intergenerational Support scale
by gender – Mean score . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
Figure 12.10: Survey of the General Community: Bivariate analysis of the Intergenerational Support scale
by age – Mean score. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
Figure 12.11: Survey of the General Community: Intergenerational Support Scale by country of birth –
Mean score. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
Figure 14.1: Findings and implications in a socio-ecological framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
vii
List of tables
Table 4.1: Survey of Older People: Characteristics of the SOP sample. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Table 4.2: Survey of the General Community: Characteristics of the SGC sample. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Table 4.3: Survey of Older People: Whether a language other than English spoken at home and
country/region of birth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Table 5.1: Survey of Older People: Overall prevalence of elder abuse by type. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Table 5.2: Australia in the context of international studies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Table 5.3: Survey of Older People: Prevalence (%) of psychological abuse in the previous 12 months by
psychological score band. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Table 5.4: Survey of Older People: Experience of neglect in the previous 12 months by band (frequency). . . . . . . 46
Table 5.5: Survey of Older People: Number of types of elder abuse in the previous 12 months, persons
who experienced any form of abuse. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Table 5.6: Survey of Older People: Co-occurrence of abuse in the previous 12 months for participants
who reported two or more abuse types and participants who reported one type of abuse only . . . . . . . . . . . . . . . 48
Table 5.7: Survey of Older People: Prevalence (%) of elder abuse in previous 12 months experienced by
CALD participants compared with non-CALD sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Table 5.8: Survey of Older People CALD subsample: Prevalence in previous 12 months of abuse relating
to culture by frequency score band. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
Table 5.9: Survey of the General Community: Co-occurring concerns about older family members and friends. . 53
Table 6.1: Survey of Older People: Socio-demographic characteristics associated with higher likelihood of
elder abuse a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
Table 6.2: Survey of Older People: Proportion of participants who report experience of any form of elder
abuse by socio-demographic characteristics, males, females and overall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Table 6.3: Survey of Older People: Mean scores of psychological distress (Kessler 6 scale) by whether
experiencing elder abuse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Table 6.4: Survey of Older People: Prevalence of elder abuse, by self-reported general physical health. . . . . . . . . 62
Table 6.5: Survey of Older People: Experience of elder abuse, by self-reported medical conditions or disability. 63
Table 6.6: Survey of Older People: Prevalence of elder abuse, by frequency of seeing family members or
friends living elsewhere and gender. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Table 6.7: Survey of Older People: Mean scores of sense of social support by whether experienced elder abuse. . . 64
Table 6.8: Survey of Older People: Relative importance (%) of explanatory variables in the regression
analysis for financial, physical and sexual abuse types. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Table 6.9: Survey of Older People: Relative importance (%) of explanatory variables in the regression
analysis for psychological abuse, neglect and overall elder abuse (any form) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Table 6.10: Survey of Older People: Prevalence of elder abuse (any form) by selected characteristics,
CALD and non-CALD participants. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Table 6.11: Survey of Older People: Prevalence of abuse relating to language and culture by selected
characteristics, CALD participants. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
Table 7.1: Survey of Older People: Demographic characteristics of main perpetrators, overall and by each
subtype of abuse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Table 7.2: Survey of Older People: problems that main perpetrators had, as reported by participants who
experienced elder abuse. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
Table 7.3: Characteristics of older people who experienced elder abuse, by relationship with perpetrator
and abuse subtype. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
Table 7.4: Perpetrator problems, by relationship with perpetrator and abuse subtype. . . . . . . . . . . . . . . . . . . . . . . . . 77
Table 7.5: Survey of Older People: Relationship of perpetrators to older people who are from a CALD
background and who experienced abuse relating to language and culture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
viii National Elder Abuse Prevalence Study: Final Report
Table 7.6: Survey of the General Community: Concern related to a professional carer – where the
mistreatment took place. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Table 8.1: Survey of Older People: Sources of help or advice as reported by older persons who had
sought help for their experience of elder abuse, for each form of abuse and overall . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Table 8.2: Survey of Older People: Older persons who experienced elder abuse and actions taken to stop
abusive behaviours from happening again, for each type of abuse and overall. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
Table 9.1: Survey of Older People: Reports of seriousness of main perpetrator’s behaviour, persons who
experience each subtype of elder abuse in the previous 12 months . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Table 9.2: Survey of Older People: Proportions of older persons who reported their elder abuse
experience as serious by selected characteristics, by type of elder abuse. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Table 9.3: Survey of Older People: Proportions of older persons who reported having sought help by
self‑assessed seriousness, by type of elder abuse. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Table 10.1: Survey of Older People: Whether participant currently has a will, by gender . . . . . . . . . . . . . . . . . . . . . . . 99
Table 10.2: Survey of Older People: Whether have ever given an enduring power of attorney to someone,
by gender. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
Table 10.3: Survey of Older People: Whether have family agreements, by gender. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Table 10.4: Survey of Older People: Advance planning, by whether person speaks a language other than
English at home. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Table 10.5: Survey of Older People: Proportion of people who have a will, by age. . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
Table 10.6: Survey of Older People: Older persons who had a valid/active power of attorney, who was
granted the power of attorney, by gender. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Table 10.7: Survey of Older People: Older persons who had an active power of attorney, who was
granted the power of attorney, by age. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Table 10.8: Survey of Older People: Prevalence of elder abuse, by each type of advance planning. . . . . . . . . . . . . 107
Table 11.1: Survey of Older People: proportion of participants where someone else (not account holder)
has access to bank account or credit card by age and gender. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
Table 11.2: Survey of the General Community: Main reason the person needed help, by age of recipient assisted. . . 122
Table 11.3: Survey of the General Community: Type of records kept and reporting arrangements where
assistance required with financial matters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
Table 11.4: Survey of the General Community: Records and reporting arrangements, by age of person
receiving assistance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
Table 11.5: Survey of the General Community: Records and arrangements, by whether appointed under a
power of attorney. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
Table 11.6: Survey of Older People: Proportion of participants who reported experience of financial abuse
by whether receiving assistance on specific financial activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
Table 11.7: Survey of the General Community: Provision of care and who receives their care. . . . . . . . . . . . . . . . . . . 127
Table 11.8: Survey of the General Community: Types of care provided, by age of person receiving care,
reported by participants who provided care. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
Table 11.9: Survey of the General Community: Proportion of participants who had concerns by providing
assistance with financial matters and providing care. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
Table 12.1: SOP and SGC: Mean scores of scale of acceptance attitudes towards elder abuse, by country
of birth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
Table 12.2: SOP and SGC: Mean scores for recognition of abusive behaviours, by country of birth. . . . . . . . . . . . . . 138
Table 12.3: Survey of the General Community: Ageism scale mean ratings and distribution of item scores . . . . . 140
Table 12.4: Survey of the General Community: Ageism scale by country of birth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
Table 12.5: Survey of the General Community: Distribution and mean ratings of intergenerational support
and sense of entitlement items. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
Table 12.6: Survey of the General Community: Correlations between the Ageism scale, Intergenerational
Support scale and agreement/disagreement with entitlement of older persons’ assets for provision of
regular assistance a. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
ix
Table 12.7: Survey of the General Community: Correlations between attitudes toward elder abuse and
the Ageism scale, Intergenerational Support scale and agreement/disagreement with entitlement of
older persons’ assets for provision of regular assistance a. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
Table A4.1: Survey of Older People: Characteristics of the SOP CALD sample. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
Table A4.2: Survey of Older People: Country of birth, by sample characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
Table A6.1: Survey of Older People: Proportion of participants who reported experience of elder abuse
by socio‑demographic characteristics, men, women and overall. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
Table A6.2: Survey of Older People: Proportion of participants who report experience of financial abuse
by socio‑demographic characteristics, men, women and overall. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
Table A6.3: Survey of Older People: Proportion of participants who report experience of physical abuse
by socio‑demographic characteristics, males, females and overall. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
Table A6.4: Survey of Older People: Proportion of participants who report experience of sexual abuse by
socio‑demographic characteristics, males, females and overall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
Table A6.5: Survey of Older People: Proportion of participants who report experience of psychological
abuse by socio‑demographic characteristics, males, females and overall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
Table A6.6: Survey of Older People: Proportion of participants who report experience of neglect by
socio‑demographic characteristics, males, females and overall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
Table A6.7: Odds ratios of logistic regression of elder abuse, overall (any form) and subtypes. . . . . . . . . . . . . . . . 198
Table A6.8: Odds of logistic regression of elder abuse, overall (any form) and subtypes, with sense of
social support and psychological distress scales as part of correlates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
Table A7.1: Survey of Older People: Relationship of perpetrators to the persons who experienced at least
one type of abuse. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
Table A7.2: Survey of Older People: Number of perpetrators reported, by all types and subtype of abuse . . . . . 201
Table A7.3: Survey of Older People: Relationship of perpetrators and main perpetrators to the
participant (people who experienced financial, physical or sexual abuse). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
Table A7.4: Survey of Older People: Relationship of all perpetrators and main perpetrator to participant
(person who experienced neglect). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
Table A7.5: Survey of Older People: Relationship of carers to participants. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
Table A7.6: Survey of Older People: Relationship of all perpetrators and main perpetrators to participant
(person who experienced psychological abuse). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
Table A7.7: Survey of Older People: Relationship of perpetrators to older people who experienced at
least one type of abuse and are from a CALD background. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
Table A7.8: Survey of Older People: Relationship of perpetrators to older people who experienced abuse
relating to language and culture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
Table A7.9: Survey of the General Community: Where each concern reported, who mistreated the older
person. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
Table A10.1: Survey of Older People: Whether participant has had discussion with someone in the family
about current will, by gender. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
Table A10.2: Survey of Older People: Advance planning, by whether person speaks a language other
than English at home. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
Table A10.3: Survey of Older People: Older persons whose PIN was accessible by someone else, who had
access to PIN, by gender and age. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
Table A10.4: Survey of Older People: Proportion of people who have a will, by socio‑demographic
characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
Table A10.5: Survey of Older People: Proportion of people who have a valid/active power of attorney, by
socio‑demographic characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
Table A10.6: Survey of Older People: Older persons who had an active power of attorney, who was
granted the power of attorney, by gender and age. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
Table A10.7: Survey of Older People: Older persons who had an active power of attorney, who was
granted the power of attorney, by whether speak a language other than English at home. . . . . . . . . . . . . . . . . . . . . 211
x National Elder Abuse Prevalence Study: Final Report
Table A10.8: Survey of Older People: Proportion of people who have a family agreement, by
socio‑demographic characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
Table A10.9: Survey of Older People: Prevalence of elder abuse, by each type of advance planning. . . . . . . . . . . . 214
Table A10.10: Survey of the General Community: Power of attorney and type of power of attorney . . . . . . . . . . . . 214
Table A10.11: Survey of the General Community: Participants who were appointed under a power of
attorney, proportion of how persons who granted the power of attorney were related to the participants. . . . . . 215
Table A10.12: Survey of the General Community: Socio-demographic characteristics of people appointed
pursuant to a power of attorney. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
Table A10.13: Survey of the General Community: Prevalence of family agreements. . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
Table A10.14: Survey of the General Community: Participants with a family agreement, how parties to the
family agreement were related . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
Table A10.15: Survey of the General Community: Characteristics of participants who had entered a family
agreement compared with power of attorney. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
Table A11.1: Survey of Older People: Number of types of assistance received with financial matters, by age. . . . . 219
Table A11.2: Survey of Older People: Older persons who received assistance in financial matters, who
provided assistance with financial matters, by gender. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
Table A11.3: Survey of Older People: Older persons whose PIN was accessible by someone else, who had
access to PIN, by gender of participant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
Table A11.4: Survey of Older People: Older persons whose PIN was accessible by someone else, who had
access to PIN, by age of participant. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
Table A11.5: Survey of the General Community: Older person to whom SGC participants provided
assistance with financial matters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
Table A12.1: SOP and SGC: Mean scores of scale of acceptance attitudes, by socio‑demographic characteristics . . 221
Table A12.2: SOP and SGC: Mean scores of scale of recognising abusive behaviours, by
socio‑demographic characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
Table A12.3: Survey of the General Community: Bivariate analysis of the Ageism scale by
socio‑demographic characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
Table A12.4: Survey of the General Community: Bivariate analysis of the Intergenerational Support scale
by socio‑demographic characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
Table B1: Sources of relevant questions in the Survey of Older People. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
Table B2: Sources of relevant questions in the Survey of the General Community . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
Table B3: Summary fieldwork statistics of SOP sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
Table B4: Summary of efforts by SRC to achieve the target sample. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
Table B5: Summary fieldwork statistics of SGC sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
Table B6: Summary of efforts by SRC to achieve the target sample. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
xi
Acknowledgements
This report was commissioned and funded by the Australian Attorney-General’s Department (AGD). The NSW
Department of Communities and Justice co-funded the Culturally and Linguistically Diverse groups substudy.
The authors would like to acknowledge the support of the Family Safety Branch in the AGD.
We thank the Social Research Centre for the implementation of the fieldwork. Sincere thanks also to
Professor Ilan Katz, Dr Trish Hill, Wendy Heywood, Laura Rimington and Nikki Honey (Social Policy Research
Centre), as well as Professor Briony Dow (Director, National Ageing Research Institute) and Associate Professor
Bianca Brijnath (National Ageing Research Institute) for their important advisory contributions to this project.
We extend our gratitude to members of the Expert Panel for their advice on this project: Christine Coumarelos
(Australia’s National Research Organisation for Women’s Safety), Cathy Etherington (Australian Bureau of Statistics),
David Firman (Queensland Treasury), Anna Gillbard (Elder Abuse Prevention Unit) and Melinda Leake (Australian
Institute of Health and Welfare).
We would like to thank our colleagues at the Australian Institute of Family Studies; in particular,
Andrew Whitecross, Acting Director, Anne Hollonds, former Director, Kelly Hand, Deputy Director (Research),
Dr Michael Alexander, Deputy Director (Corporate and Strategy), Dr Jennifer Prattley (Statistician) and
Jatender Mohal (Senior Manager, Data Management & Linkage) for their advice and support throughout
this research.
Sincere appreciation to the AIFS library team and, in particular, Gillian Lord for her research assistance, together
with the AIFS communications team; in particular, Katharine Day for editing this report and Lisa Carroll for
preparing the report for publication.
Finally, we are very grateful to all the people who participated in the surveys and took the time to talk to us
about their views and experiences.
Views expressed in this publication are those of individual authors and may not reflect those of the Australian
Government or the Australian Institute of Family Studies.
xii National Elder Abuse Prevalence Study: Final Report
Acronyms
Term Description
ABS Australian Bureau of Statistics
SD Standard deviation
Glossary
Term Definition
Ageism Attitudes that are prejudicial and discriminatory towards people because of their
age (Swift et al., 2018). Measured in this research through a scale that comprised
attitudes relating to benevolent and hostile ageism (see Box 12.3).
Back-coding Where the verbatim responses to questions with an ‘Other’ (specify) response
option were coded to the original response options or additional categories.
Bi-variate analysis Analysis involving two variables.
CALD In this report, CALD (Culturally and Linguistically Diverse) refers specifically to
participants in the Survey of Older People who reported speaking a language other
than English at home.
Co-occurrence The presence of two or more forms of abuse in the previous 12 months.
Community dwelling Residing in a private dwelling or self-care retirement village as opposed to residing
in institutional care including hospitals, nursing homes or other homes.
Confidence interval Confidence intervals relate to the level of confidence we have that the estimates
calculated based on the sample are a true reflection of the Australian population
of interest for this study. That is, the Australian population aged 65 years and older
living in the community. An interval is calculated using a mathematical formula that
produces upper and lower bounds. The value for the population is expected to
fall between these bounds with a certain degree of confidence. A 95% confidence
interval means that we can be 95% confident that the true population value is
between the upper and lower bounds of the interval.
In reporting the prevalence of elder abuse overall and the six subtypes, 95% confidence
intervals are provided.
The ABS online Basic Survey Design (2020a) explains confidence intervals in this way:
‘Assuming that the target population is distributed normally for the characteristic
being measured (or, if estimating the mean, the sample is sufficient to assume the
sample mean is distributed normally), the interval which contains the true value is
usually calculated as being one, two, or three standard errors above and below the
survey estimate. This interval is usually referred to as a confidence interval … There is
a 95% chance that the confidence interval which extends to two standard errors on
either side of the estimate contains the “true value”. This interval is called the 95%
confidence interval and is the most commonly used confidence interval.’
xiii
Term Definition
Correlation A statistical measure (expressed as a number) that describes the size and direction
of a relationship between two or more variables (ABS, 2013).
Enduring power of attorney A legal document that enables a person to nominate a person or organisation to make
decisions about financial and/or personal matters, if the person is unable to make these
decisions for themselves. The person appointed to make the decision is the ‘attorney’
and their power to make decisions is ‘enduring’ because it endures or continues
when the person granting the power is no longer able to make their own decisions.
Family agreement A family agreement is defined as an agreement to provide care to an older person in
return for financial support or benefit (e.g. transfer of assets, property or finances)
or a bequest in a will.
Intimate partner Defined in this report as a current partner or spouse.
Logistic regression Logistic regression is a statistical method used to assess the association between
a dependent variable and one or more independent (or explanatory) variables. The
dependent variable is a binary variable (with two values). Logistic regression can
be bivariate (one independent variable and one dependent variable) or multivariate
(many independent variables and one dependent variable). In this report, logistic
regression analysis is used to assess whether a specific variable is correlated with a
specific abuse type, taking into account any other variables in the analysis. While the
results can provide an indication of correlation (the extent to which these variables
have a linear relationship) between a specific variable and an abuse type, the results
do not indicate a causal relationship.
Mean The sum of the value of each observation in a dataset divided by the number of
observations (ABS, 2013).
Older person As defined for this research, a person aged 65 years or over.
Perpetrator Defined in this report as a person who has committed the behaviour(s) against the
older person.
Prevalence In this report, prevalence refers to the proportion of participants who were classified
as having an experience of elder abuse (a specific subtype or overall, depending on
the context). The terms ‘prevalence’, ‘rate’ and ‘proportion’ are used interchangeably
in describing the occurrence of elder abuse for the whole sample or specific
subgroups. For the technical analytic definition applied in deriving prevalence
estimates, see chapter 5.
Prevalence of abuse The proportion of people in the population who have experienced abuse (as defined
in this report) within a specified time period. The prevalence period for this report is
12 months prior to the survey.
Sample weights As surveys involve a sample (subset of the population) rather than the whole
population, sample weights are used to reduce any biases arising from some groups
in the population not being adequately represented in a sample. In this report, it is
specified where data are reported with sample weights and without sample weights
(unweighted).
Service provider Service provider refers to a professional or paid worker that provides a service.
For the analysis of perpetrators of abuse, service provider is a new category created
following back-coding (refer to footnote 14 in chapter 7). For this report, service
provider excludes professional carers (included as a separate category).
SEIFA (Socio‑economic ABS product that ranks areas in Australia according to relative socio‑economic
Indexes for Areas) advantage and disadvantage, and is based on information from the Census of
Population and Housing (see ABS, 2018a).
Standard deviation Measures the spread of data around the mean (ABS, 2013).
Standard error Measure of the variation between any estimated population value that is based on a
sample rather than true value for the population (ABS, 2013).
Statistical significance Statistical significance relates to how likely the observed effect is due to chance
or the specific sample. Results of a statistical test are statistically significant if
the p-value of the result is lower than a specified significance level. The p-value
represents the probability of obtaining results at least as extreme as the ones
observed, assuming that there is no effect or association. In this report, statistical
significance is reported for multiple levels in the relevant tables, at significance levels
of 5%, 1% and 0.1% (or expressed as *p < .05; **p < .01; ***p < .001).
xiv National Elder Abuse Prevalence Study: Final Report
Term Definition
Definitions of abuse used in this report
Elder abuse The conceptual definition of elder abuse applied in this research was developed for
the Australian context. The definition is: a single or repeated act or failure to act,
including threats, that results in harm or distress to an older person. This occurs
where there is an expectation of trust and/or where there is a power imbalance
between the party responsible and the older person.
Neglect Neglect is determined based on a score that takes into account the level of need
for support and the extent to which these needs were not met by a defined carer or
carers (who participant reported was/were responsible for a specific daily activity
that he/she needed assistance with for some or all of the time) in the 12 months
prior to the survey. Scores above three were counted as neglect. (See the section
‘Technical definitions and measures’ in chapter 5 for how scores were generated.)
Financial abuse Financial abuse is defined as the experience of one or more specified incidents, by
family member, professional carer, other professional or other known person in the
12 months prior to the survey. Frequency was not taken into account in the definition.
Psychological abuse Participants were asked about their experience of specific behaviour(s) by family
member, professional carer, other professional or other known person in the 12 months
prior to the survey, and the frequency with which they had experienced those
behaviours. Scores were derived based on the frequency of experiencing specific
behaviours. Scores above three were counted as psychological abuse. (See the section
‘Technical definitions and measures’ in chapter 5 for how scores were generated.)
Physical abuse Physical abuse is defined as the experience of one or more specified incidents, by
family member, professional carer, other professional or other known person in the
12 months prior to the survey. Frequency was not taken into account in the definition.
Sexual abuse Sexual abuse is defined as the experience of one or more specified incidents, by family
member, professional carer, other professional or other known person in the 12 months
prior to the survey. Frequency was not taken into account in the definition.
Abuse relating to language Participants who spoke a language other than English were asked if they had
or culture experienced specific behaviours (e.g. mis-translated between English and your
preferred language on purpose, denied you access to important information in your
preferred language, etc.), by family member, carer, carer worker, other professional,
other known person in the 12 months prior to the survey and the frequency with
which they had experienced those behaviours. Scores above three were counted
as abuse relating to language or culture. (See section ‘Technical definitions and
measures’ in chapter 5 for how scores were generated.)
Executive summary 1
Executive summary
Policy context
Elder abuse has gained significant attention in Australia in recent years as a serious problem requiring increased
policy focus. Five abuse subtypes are commonly recognised: financial abuse, physical abuse, sexual abuse,
psychological abuse (otherwise known as emotional abuse), and neglect.
The increasingly older age profile of the Australian population makes it particularly important to address elder
abuse effectively. The 65 and over age group is expected to more than double from 3.8 million to 8.8 million in
the next 25 years.
In Australia, research on elder abuse has been limited to studies looking at particular types of elder abuse
(e.g. financial abuse), qualitative studies and those based on administrative data from services who provide
support to older people. Such studies are unable to shed light on the proportion of older people aged 65 and
over who experience elder abuse or which subtypes are most common. Nor are they able to assess other
important issues, such as the extent to which elder abuse is under-reported.
As part of the National Plan to Respond to the Abuse of Older Australians, the Attorney-General’s Department
commissioned the most extensive empirical examination of elder abuse in Australia to date, the National Elder
Abuse Prevalence Study. This report presents the findings of that research program.
An additional survey of 3,400 people aged between 18 and 64 examined views about older people and levels of
awareness of elder abuse in the community (the Survey of the General Community (SGC)). It also looked at the
extent to which people in the community provide support to older people in the context of assisting with care,
financial arrangements and power of attorney arrangements.
The research program also included a focus on the experiences of culturally and linguistically diverse (CALD)
groups in both the SOP and the SGC. CALD participants in the SOP were asked about their experiences of abuse
relating to language and culture, in addition to their experience of the five core subtypes. This report examines
similarities and differences in elder abuse experiences between the non-CALD and CALD subsamples in both the
SOP and the SGC.
It is important to note that the SOP focused on people who live in the community and did not cover people
who live in aged care or could not participate in the survey due to cognitive decline. The Royal Commission into
Aged Care Quality and Safety has recently estimated that the prevalence of physical abuse, emotional abuse and
neglect in aged care settings is 39.2%.
2 National Elder Abuse Prevalence Study: Final Report
For the other subtypes, prevalence rates are 2.1% for financial abuse, 1.8% for physical abuse and 1% for sexual abuse.
At 15.3%, prevalence rates for the CALD subsample do not differ greatly from the overall sample. Abuse relating
to language and culture is reported by 4% of the CALD subsample.
Experiences of multiple types of abuse were reported by 3.5% of SOP participants, with the most common
abuse combinations being psychological abuse and neglect.
Overall prevalence rates were similar for men and women; however, men were slightly more likely to experience
physical abuse and more women experienced sexual abuse and neglect.
The estimates in this report are based on the experiences of people living in the community who had capacity
to participate in the study. These figures likely under-estimate the total prevalence of elder abuse because
they exclude people who lack cognitive capacity to participate or perhaps were unable to participate for other
reasons such as frailty, as well as those living in aged care settings.
Dynamics for each of the abuse subtypes vary in terms of who is susceptible, who the main perpetrator groups
are and whether victims seek help. This indicates that elder abuse is complex and different subtypes may require
tailored policy and program responses.
Low socio‑economic status is associated with a greater risk of abuse overall, and especially financial, sexual and
psychological abuse. Owning a home with debt and being in rented accommodation (including public housing)
are associated with a higher likelihood of experiencing psychological abuse and neglect. Conversely, owning a
home without debt is associated with a lower likelihood of experiencing abuse.
Marital status is a consistent influence across all abuse types, with those who are either separated or divorced
being more likely to experience abuse.
Adult children were most likely to commit financial, physical, and psychological abuse. Sons were almost
twice as likely as daughters to commit financial abuse. Adult children were on par with intimate partners as
perpetrators of neglect.
In relation to sexual abuse, friends and acquaintances were the largest perpetrator groups. Neighbours and
friends were also significant perpetrator groups for physical abuse and psychological abuse (close to one in 10
of all perpetrators for these two subtypes).
Intimate partners also featured commonly as perpetrators of physical, psychological, and sexual abuse.
Professionals and service providers were the smallest perpetrator groups. Neglect was the abuse type most likely
to be associated with these two groups as perpetrators, behind adult children and intimate partners.
Men outweighed women as perpetrators of abuse by 10 percentage points overall, especially in relation to
physical, sexual, and financial abuse.
Perpetrators were reported to have a range of problems (according to the SOP participants who reported
experiencing abuse). Most commonly, perpetrators were reported to have mental health problems (almost
one third) and financial problems (nearly one in five). The most common problems associated with financial
Executive summary 3
abuse were financial problems. Mental health issues were the most commonly reported problems for physical and
psychological abuse. For sexual abuse perpetrators, problems with alcohol predominated.
Neglect is different from the other abuse subtypes, with physical health problems being the most common
issue associated with perpetrators.
Physical abuse (39%) and psychological abuse (32%) were most likely to be described as very serious.
Seriousness ratings for financial abuse were distributed relatively evenly across the three seriousness bands
(very serious, somewhat/moderately serious, not serious).
Neglect was the least likely to be described as serious with almost one half (48%) of participants rating their
experience as not serious.
In relation to psychological wellbeing, participants who experienced abuse were more than three times as likely
to fall into a score range on the Kessler 6 psychological distress scale indicating probable serious mental
illness as those who did not (10% cf. 2%). The patterns were consistent across abuse types.
Similarly, people with poorer health were more likely than those with better health to report experiencing elder
abuse. Having a disability was associated with a higher likelihood of experiencing elder abuse.
Low social support and lack of social contact were associated with a higher likelihood of experiencing elder abuse.
Help was most likely to be sought for physical abuse, financial abuse and psychological abuse. Help was least
likely to be sought for neglect and sexual abuse. Notably, reliance on criminal justice responses for sexual abuse
was almost absent.
Most commonly, avenues for help and advice were informal and mainly involved family and friends.
A common means of trying to stop the abuse was breaking contact with or avoiding the perpetrator. In some
cases, a more significant withdrawal from social life was reported (over one-tenth). These responses raise
particular concerns, as they may have adverse consequences for the older person’s contact with other family,
friends and support networks. They also do not address perpetrator accountability.
Where older people sought professional help, they were more likely to turn to the helping professions, medical
professionals such as GPs and nurses, than legal mechanisms (e.g. lawyers and police). Actions to stop abuse
involving legal and quasi-legal services (mediation or counselling, legal advice or personal protection orders)
were not common.
Notably, of those older people who reported taking action, substantial minorities considered these actions
were ineffective. Responses indicating actions were ineffective were highest for financial abuse (over one third)
and lowest for sexual abuse (over one-quarter).
4 National Elder Abuse Prevalence Study: Final Report
Legal arrangements
Nearly nine in 10 older people participating in the SOP had a will. However, lower levels of will making were
evident among people from lower socio‑economic and CALD backgrounds. Having a will is associated with lower
levels of abuse, with differences between reports of those with and without a will in relation to financial abuse
(1.7% cf. 5%) and physical abuse (1.5% cf. 3.6%). These and other SOP data together suggest this association is
likely to be as related to socio‑economic status as it is to existence of the will itself.
Just over half of the participants in the SOP had appointed a substitute decision maker under an enduring
power of attorney. Again, lower proportions of people from lower socio‑economic and CALD backgrounds had
enduring powers of attorney. Having an enduring power of attorney is associated with lower levels of abuse,
especially financial abuse, physical abuse, and psychological abuse.
Only a small proportion of participants in the SOP (3%) reported that they had a family agreement (the SGC
findings were consistent on the prevalence of family agreements). Family agreements were more prevalent
among older people from lower socio‑economic status groups and those from a CALD background. Family
agreements were also more likely to be associated with the experience of elder abuse.
Support
There is a significant amount of support in relation to financial matters and other care needs provided to people
aged over 65. This has both negative and positive aspects. Assistance with financial management provides
opportunity for financial abuse to occur, particularly in the context of gendered financial management patterns
that see women relying on children for assistance with managing financial matters to a significantly greater
extent than men. However, the provision of care and support also provides opportunities for the disclosure and
detection of elder abuse.
Nearly half the SOP participants reported receiving help with financial matters (mostly tax, superannuation,
pensions and investments) in the past 12 months. The most common sources of help were financial planners
(65%) and partners or spouses. Women were more than twice as likely as men to report receiving assistance
with financial matters from children (22% cf. 10%).
More than a quarter of all SOP participants reported that a third party had access to bank accounts or credit
cards, mostly a partner or spouse. About half reported that records were kept in relation to transactions in this
context. Compared with men, women were substantially more likely to report their children had access to their
personal identification number (PIN).
In the SGC, a third of participants reported providing care, often help with day-to-day activities, to someone
aged 65 or over, most frequently a parent or grandparent. These SGC participants were more likely to report
holding concerns about elder abuse than SGC participants who were not involved in care activities. Direct
financial support to an older person was provided by 26% of SGC participants.
Older age groups in the SOP sample were more likely to have attitudes accepting of elder abuse. In the SGC
sample, the youngest age group (24 years or less) showed greater acceptance of elder abuse.
For both samples, particular demographic and socio‑economic characteristics were associated with a higher
degree of acceptance of elder abuse. These included lower education levels, unemployment (or retirement) and
low income.
Sexual abuse (specifically measured by questions including whether ‘talking to an older person in a sexual way
when they do not want to’) had higher levels of recognition among SGC participants than SOP participants.
SOP and SGC participants born in English speaking countries overseas showed greater recognition of elder
abuse compared to those born in Australia and non-English speaking countries.
Executive summary 5
Ageism
Most people in the SGC did not hold ageist attitudes, with a mean score of 37.4 on a scale with a 0–100 score
range (with a higher score indicating more ageist attitudes towards older people) used to measure ageist attitudes.
Some socio-demographic characteristics were associated with more ageist attitudes. Males showed a greater
tendency towards ageist attitudes than females and participants born in non-English speaking countries
showed greater ageist attitudes compared to those born in English speaking countries overseas and in Australia.
Particular demographic and socio‑economic characteristics were associated with ageist attitudes and these
included younger age, lower education levels, unemployment and low income.
Intergenerational support
SGC participants generally showed positive attitudes towards intergenerational support with regard to
supporting parents financially and allowing parents to live with adult children.
Positive attitudes to intergenerational support were also associated with a stronger belief in family members
being entitled to an older person’s assets in return for their provision of regular assistance.
The findings suggest that psychological abuse, sexual abuse and neglect warrant more attention. This is not
to say that focus should be diverted from financial abuse and physical abuse. However, with financial abuse
commonly perceived to be the most common form (e.g. Australian Law Reform Commission, 2017; Dow &
Brijnath, 2019), it is evident that other types of abuse may not have received sufficient attention to date.
With the proportion of the population aged 65 and over set to increase substantially, the development of an
evidence-based prevention framework is necessary. Such a framework would be informed by the findings in
this report and other research and would identify strategies that could be applied to prevent elder abuse from
happening. Key areas of focus would include increasing recognition and awareness of elder abuse behaviours.
In addition, the socio-demographic characteristics associated with elder abuse, such as financial strain, housing
stress and individual-level characteristics such as social isolation, mental ill health, poor physical health and
disability indicate some directions for such a framework. The fact that elder abuse mostly takes place in family
relationships, and in social contexts (though to a lesser extent), means that measures to address it need to
manage potential adverse consequences for the victim – such as isolation from family and friends.
Awareness raising should not only be part of a prevention framework but also part of a set of strategies
intended to improve identification of, and responses to, elder abuse in community and service settings. In light
of findings that elder abuse largely remains a hidden problem, proactive mechanisms to identify people who
are experiencing elder abuse or are at risk of experiencing elder abuse are particularly important. Such
mechanisms should not only focus on supporting identification of the risk of elder abuse or elder abuse itself but
also awareness of the services that are available to address it. Given the important role that family and friends
play as confidants for those experiencing elder abuse such strategies should target the general public as well as
people who work in areas where they come into regular contact with older people, such as health professionals.
These strategies could include systematic screening; for example, in health settings.
The research revealed limited reliance on formal services and significant reliance on avoidant strategies among
those who experienced elder abuse. This indicates a need for further examination of the barriers to help seeking
and whether the services presently available to address elder abuse are sufficient. In particular, given the
complex relationship dynamics involved in elder abuse, it is necessary to assess whether existing services are
appropriately designed for this context. For example, in financial abuse, the predominance of adult children
as perpetrators, the lower levels of help seeking, and the higher levels of negative effectiveness assessments of
actions, taken together, suggest that further consideration of responses to financial abuse is needed.
Given the lower profile but higher prevalence of psychological abuse and neglect, particular attention should be
paid to the assessment and development of responses in these areas. Responses to sexual abuse also warrant
6 National Elder Abuse Prevalence Study: Final Report
attention, given evidence of low levels of service engagement and almost no reliance on justice responses
in this context. Assessment of responses to financial abuse should focus on how well they manage complex
family dynamics.
Further research is also necessary in several areas. The experiences of groups not adequately covered in this
methodology, including Aboriginal and Torres Strait Islander groups, LGBTIQ groups, people with cognitive
impairment and those who reside in aged care should be examined. In addition, more in-depth insight into the
experience of CALD subgroups would assist in developing deeper understanding and tailored responses for
these groups. Importantly, further research of a longitudinal nature would provide important insights into a range
of issues, including the intensity, duration and consequences of elder abuse over time, as well as supporting a
deeper understanding of the causal links with issues such as difficult relationship dynamics, financial strain and
intergenerational wealth management practices.
Chapter 1: Introduction 7
1 Introduction
Elder abuse has gained significant attention in Australia in recent years as a serious problem requiring increased
policy focus. Five abuse subtypes are commonly recognised within the overarching term elder abuse: financial
abuse, physical abuse, sexual abuse, psychological abuse (otherwise known as emotional abuse), and neglect.
This report presents findings from the most extensive study on elder abuse in Australia to date.
The core of the three-part research program involved assessing prevalence – that is, the proportion of the
population affected in a nationally representative sample of 7,000 community dwelling individuals – of elder
abuse (the Survey of Older People (SOP)). A supporting part of the research program – a survey involving
3,400 Australians aged between 18 and 64 years – sheds light on attitudes to older people and awareness of
and concerns about elder abuse in the general community (the Survey of the General Community (SGC)). The third
part of the research program focuses on examining prevalence and attitudes among culturally and linguistically
diverse (CALD) groups through a CALD-focused analysis of the SOP and the SGC (the CALD substudy). The
research has been commissioned and funded by the Australian Attorney-General’s Department (AGD).
The 65 and older age group is a significant and growing proportion of Australia’s population. It is expected to
more than double in the next 25 or so years.
In response to recognition of the need to strengthen responses to elder abuse, the then Council of
Attorneys‑General (CAG) committed to the National Plan to Respond to the Abuse of Older Australians
(Elder Abuse) 2019–2023 (the National Plan; [CAG], 2019). A significant component of this plan is recognition
of the need to improve the evidence base on elder abuse in Australia to pave the way for more targeted and
effective practice and policy responses.
The development of a national response to elder abuse in Australia aligns with the emergence of a focus on
national approaches to other complex problems in the past decade, most notably child abuse (the National
Framework for Protecting Australia’s Children) and family violence (the National Plan to Reduce Violence Against
Women and their Children).
Background
Population dynamics
According to the most recent population estimates by the Australian Bureau of Statistics (ABS, 2020b), about
4.1 million Australians are aged 65 and over, representing 16% of the population. Projections show that the size
of the age group will continue to increase, both in total numbers and as a proportion of the population. The
population projections by the ABS (2018b) indicate that by 2066, this group will increase to 8.6–10.2 million
people and represent 21–23% of the population.
These population dynamics underline the significance of understanding the prevalence of elder abuse. With
greater numbers of people in the relevant age group, the incidence (the number of people affected) of elder
abuse will also increase. Unless effective prevention measures are implemented, this will necessitate an expansion
in the agencies and services needed to identify, address and respond to elder abuse. Developing the evidence
base on prevalence, dynamics and risk and protective factors will improve the ability of policy makers and
practitioners to assess ‘what works’ in preventing and addressing elder abuse and to tackle the challenges posed
by these population dynamics.
8 National Elder Abuse Prevalence Study: Final Report
From the perspective of culturally and linguistically diverse (CALD) communities, the make-up of the population
aged 65 and over also underscores the necessity to understand elder abuse through a CALD lens. As of 2016, more
than one third (37%) of the population aged 65 and over were born overseas (ABS, 2017a, 2017b). The majority of
this group (20% of the total population aged 65 and over) were born in countries where English is not the primary
language, most commonly Italy (3%), Greece (2%) and Germany (1%). People born in the United Kingdom
and Ireland account for a further 10% of the population. International evidence indicates that the nature and
prevalence of elder abuse is influenced by culture (e.g. Williams, Davis, & Acierno, 2017), with one meta-analysis
showing higher prevalence in non-Western countries compared to Western countries (Ho, Wong, & Ho, 2017).
The prevalence of elder abuse needs to be considered in different contexts, including whether the older person
is residing in a community or non-community dwelling. This study focuses on community dwelling people. The
ABS (2019a) found that the majority of older people aged 65 and over in Australia were residing in a household,
including a private dwelling or self-care retirement village (95%), with 5% (or 181,200 people) living in care
accommodation, which included hospitals, nursing homes, aged care hostels and other homes. Further data
on aged care services from the Australian Institute of Health and Welfare ([AIHW], 2018) indicated that over
1.2 million people received aged care services in 2016–17, with the majority (77%) accessing some form of support
or care at their home or a community-based setting. The AIHW also identified that the proportion of older
people receiving care at home has steadily increased over the last 10 years.
The risk of dementia poses a significant challenge to the health and welfare of older Australians. Although
the exact proportion of people with dementia is not known; in 2018, the AIHW estimated that approximately
9% of the population aged 65 and over in Australia have dementia. The number of people with dementia is
also expected to increase over time, from 400,000–459,000 to 550,000–590,000 in 2030 (AIHW, 2018). The
increased dependence that may follow from a decline in cognitive functioning can be a significant risk factor for
the experience of elder abuse.
Policy context
In Australia, elder abuse has received increasing attention in the past 10 years, with a range of inquiries at state,
territory and Commonwealth levels highlighting the complex nature of elder abuse as a social, legal and human
rights problem. In 2016, the NSW Parliament concluded that ‘a great deal more needs to be done to prevent
the complex, hidden and pernicious problem of elder abuse’ (New South Wales Parliament, Legislative Council.
General Purpose Standing Committee No 2., 2016). In 2017, the Australian Law Reform Commission ([ALRC],
2017) observed that the fragmentation of responsibility for elder abuse across all three levels of government
meant ‘responding to elder abuse is a complex issue – both from the perspective of laws, but also in terms of
practical responsibility’ (p. 48). In 2018, a report by the Western Australian Parliament observed that ‘[a]geism
is widely acknowledged to be at the heart of the problem of elder abuse’ (p. 13). Common themes in these
and other reports include a need for a better evidence base, a national, co-ordinated approach, and a greater
emphasis on prevention, in addition to a strong recognition of elder abuse as a breach of human rights.
The National Plan (CAG, 2019) sets out strategic objectives in five key areas. Priority One under the National Plan
is Enhancing Understanding, with a national prevalence study the primary strategy under this priority. The plan
notes that the expanding range of responses to elder abuse in Australia has been ‘developed without a robust
body of evidence to make the case that one response works better than any other’ (p. 13). The National Plan
endorses the need for an evidence base to inform better and more targeted efforts to address elder abuse.
The other four priority areas under the National Plan are:
y Improving community awareness and access to information
y Strengthening service responses
y Planning for future decision making (improving frameworks governing the use of powers of attorney and
advanced health or care directives)
y Strengthening safeguards for older adults.
The development of a National Plan was a recommendation of the Australian Law Reform Commission report,
Elder Abuse – A National Legal Response (Rec 3-1, ALRC, 2017). In addition to establishing a national policy
framework, a further rationale for this recommendation was to support ‘future planning and policy development
[for elder abuse to take place] in an integrated way’. As with other areas concerned with complex problems
such as child abuse and family violence, policy and legislative responsibility in relation to elder abuse is spread
between state, territory and Commonwealth governments and among portfolio areas including those relating to
justice, policing, human rights, human services and health (Kaspiew, Carson, & Rhoades, 2016).
Chapter 1: Introduction 9
Alongside (or in some cases ahead of) the National Plan, states and territories have also developed their own
policy frameworks and strategies. The AGD’s (2019) Stocktake of Elder Abuse Awareness, Prevention and
Response Activities in Australia provides an overview of key policy and program responses across the country
up to that date.
Elder abuse is one of a number of areas where the age profile of the Australian population creates policy
challenges (e.g. Piggott, 2016). Another such area is aged care, with the Royal Commission into Aged
Care Quality and Safety delivering its final report in February 2021. This report placed significant focus on
shortcomings in the aged care system, including in relation to the quality of care provided to residents in aged
care and their exposure to abuse and neglect (Royal Commission into Aged Care Quality and Safety, 2021a,
2021b). It outlined a need for wide-ranging reforms to the aged care system and made 148 recommendations
for reform. A research paper by the Royal Commission estimated that 39.2% of aged care residents
experience emotional abuse, physical abuse and/or neglect (Royal Commission into Aged Care, 2020).
Commonwealth‑funded home care services are also included in the Royal Commission’s Terms of Reference.
With a focus on the community dwelling population aged 65 and over, this research has very limited overlap
with the Royal Commission’s area of concern. However, concerns about family and friends in aged care settings
examined in the SGC are set out in chapter 6. The findings also consider service providers both as possible
perpetrators of elder abuse (chapter 7) and as supports for those experiencing elder abuse (chapter 11).
Alongside these pieces of work, two further studies were commissioned to build relevant insights from existing
datasets. One study assessed elder abuse in existing ABS datasets (Hill & Katz, 2019). The study concluded that
existing datasets were not able to provide a robust estimation of abuse of older people. Its analysis yielded some
relevant insights, including some from the Personal Safety Survey (PSS), the Australian prevalence study on
interpersonal violence. Relevant findings from the PSS were that 2% of people aged 55 and over had experienced
physical violence in the preceding 12 months and between 0.1 and 0.4% of women aged 55 and over reported
sexual violence. Analysis of the National Aboriginal and Torres Strait Islander Survey showed 16% of survey
participants aged 45 and over had experienced physical violence in the past 12 months.
The second commissioned study assessed the vulnerabilities of Aboriginal and Torres Strait Islander people aged
over 50 (AIHW, 2019a), in the context of acknowledging that elder abuse dynamics in the Aboriginal and Torres
Strait Islander population are ‘even more complex than in non-Aboriginal Australia’ (Dow et al., 2020 p. 567), for
a range of reasons. In addition to cultural, demographic, health and economic issues, these reasons include the
legacy of ‘separation from land and culture and the forced removal of children from families’ (Dow et al., 2020,
p. 567). The AIHW (2019a) report highlighted the proportionately lower number of Aboriginal and Torres Strait
Islander people aged 50 and over (2% of the total population) and their proportionately greater representation in
hospitalisations for non-fatal assault (17%), as victims of family and domestic violence assault (11%) and as victims
of homicide (5%) (AIHW, 2019a, Table 9.1).
The SOP examines the prevalence of the five elder abuse subtypes and the extent to which different subtypes
occur together. It also provides a socio-demographic profile of the people who report experiencing elder
abuse and those who are reported to perpetrate it. Additionally, it examines the impact of elder abuse, risk and
protective factors, and what people who experience elder abuse do about it.
10 National Elder Abuse Prevalence Study: Final Report
The aims of the SOP as defined by the commissioning agency (AGD) were to:
1. Measure the national prevalence of elder abuse, including the five main subtypes of abuse and the
co‑occurrence of subtypes.
2. Describe the characteristics and contexts of abuse, including the characteristics of perpetrators and how
people who experience elder abuse respond to it.
3. Assess the extent to which older people have adopted advance planning behaviours that may protect them
against elder abuse.
4. Identify risk and protective factors associated with the experience of elder abuse overall, and for each
individual form of abuse, if the data permits.
In the context of international approaches to elder abuse prevalence research, the National Elder Abuse
Prevalence Study is unique in including a companion study designed to examine the community context for
the SOP. This second study, the SGC, surveyed 3,400 Australians, aged between 18 and 64, about attitudes and
concerns relevant to elder abuse. It covered attitudes to older people, perceptions of elder abuse, concerns about
elder abuse in connection with people known to survey participants and whether action was taken in response to
such concerns. The SGC provides important insights into the social context for the findings of the SOP, including
awareness of elder abuse in the community and the extent to which elder abuse prevalence may be linked with
negative attitudes to older people.
The aims of the SGC, as defined by the commissioning agency, were to:
1. Describe knowledge and awareness about elder abuse and attitudes towards older people and examine how
they may contribute to social and environmental norms that allow elder abuse to occur.
2. Measure the proportion of people who provide assistance to older family members or friends and the types
of assistance they provide that could potentially enable abusive practices to occur, particularly in relation to
financial transactions and decision making.
3. Indirectly estimate the prevalence of elder abuse by asking participants about whether they have any
concerns about the abuse of their older family members or friends.
The third element of the National Elder Abuse Prevalence Study focuses on CALD groups. Given the substantial
representation of people born overseas in the Australian population, in combination with research and analysis
suggesting that the experience of elder abuse in these groups may differ from the experience of non-immigrant
populations, the CALD substudy assesses the experience and attitudes of CALD participants in the SOP and the
SGC separately from the experiences of the population not born overseas. In addition, the SOP included four
questions aimed at measuring specific types of abuse that may be related to language and culture.
The aims of the CALD substudy as defined by the commissioning agency were to:
1. Describe the characteristics and contexts of elder abuse, including the prevalence and frequencies of specific
abusive behaviours, characteristics of the people who have engaged in abuse, and actions taken by people
who have been impacted by abuse in response to their experiences.
2. Assess the extent to which older CALD Australians have adopted advance planning behaviours that may
protect them against elder abuse.
3. Identify risk and protective factors associated with the experience of CALD elder abuse overall, and for each
individual type of abuse, if the data permits.
4. Identify the prevalence of elder abuse among people with a CALD background compared to non-CALD people.
Summary
The policy context for this research is an intention on the part of governments at Commonwealth and state
and territory levels to improve policy and practice responses to elder abuse (CAG, 2019). A critical aspect of
this policy agenda is a recognition that better evidence on elder abuse – including who experiences it, who
perpetrates it, what its impacts are and how people seek help – is needed.
An important part of the impetus for this policy agenda is the increasing age profile of the Australian population,
with the 65 and over age group expected to double from 3.8 million to 8.8 million in the next 25 years (AIHW, 2018).
With very limited systematic empirical evidence on elder abuse in Australia, the National Elder Abuse Prevalence
Study represents a substantial contribution to the evidence base. The SOP based on a representative sample of
7,000 people aged 65 and older, together with the SGC based on a sample of 3,400 and the CALD substudy,
provides extensive insight into elder abuse in Australia.
Chapter 2: Evidence on elder abuse 11
This chapter provides an overview of key insights from international and Australian research on elder abuse.
Prevalence studies have been conducted in many countries in the past 20 years, with findings from Norway,
Korea, Sweden and Turkey being published in the past five years. In Australia, the evidence to date has been
limited, largely consisting of analyses of data from helplines and qualitative studies.
Empirical evidence
International prevalence studies
In the past 20 years, international research on elder abuse has grown exponentially. Many countries have
implemented prevalence studies based on representative population level samples. Among the larger earlier
studies, were those from the UK (O’Keeffe et al., 2007), Ireland (Naughton et al., 2010), the United States of
America (e.g. Acierno, Hernandez, & Kilpatrick, 2010) and Canada (National Initiative for the Care of the Elderly,
2016). In the past five years, prevalence studies from Norway (Sandmoe, Wentzel-Larsen, & Hjemdal, 2017), Korea
(Jeon, Cho, Choi, & Jang, 2019), Sweden (Ahnlund, Andersson, Snellman, Sundström, & Heimer, 2020), Japan (Koga,
Hanazato, Tsuji, Suzuki, & Kondo, 2019) and Turkey (Kulakçı Altıntas & Korkmaz Aslan, 2020) have been published.
Methodological differences, including sample age ranges, measures and analytic definitions, mean that
comparability among these studies is very limited (e.g. Dong, 2015; Williams et al., 2017). However, meta-analytic
studies can provide some indications of global prevalence benchmarks. Yon, Mikton, Gassoumis, and Wilber’s
(2017) meta-analysis of 52 prevalence studies across 28 countries demonstrated an overall prevalence rate of
15.7%, including 11.6% for psychological abuse, 6.8% for financial abuse, 4.2% for neglect, 2.6% for physical abuse
and 0.9% for sexual abuse.
Another meta-analysis, by Ho and colleagues (2017), was based on 34 population-based studies from around
the world published in English. It found a pooled elder abuse (the age group was 60 and over) prevalence rate
of 10% (Ho et al., 2017). The most common subtype of abuse was emotional abuse (47.5%), followed by financial
abuse (34%), neglect (32%), physical abuse (19.1%) and sexual abuse (3.1%). There was a greater likelihood of
women being abused (15%) compared with men (10.6%) and non-Western countries had a higher prevalence
than Western countries (17.4% cf. 7.3%).
The meta-analysis by Ho and colleagues also examined the prevalence of elder abuse according to reports
of caregivers or other third parties, based on 17 studies from different countries. The studies that the analysis
is based on largely involved groups with specific roles in relation to older people, including paid and unpaid
caregivers, aged care workers and medical professionals including nurses and doctors. Rather than being based
on representative, population level samples, these studies reflect the reports of particular groups of informants
in relation to particular groups of older people. Their analysis produced a prevalence rate of 34% as reported by
study informants in these groups.
Several explanations may be relevant for the higher rate of prevalence indicated by these study informants
compared to prevalence rates reported directly by older people. They include the possibility that older people
under-report their own experiences. Further, the studies this aspect of the analysis was based on included older
people with cognitive decline and residents in care settings. The population level studies exclude older people
with cognitive decline and mostly focus on community dwelling samples, and thus did not capture prevalence
among a particularly vulnerable subgroup in the older population.
12 National Elder Abuse Prevalence Study: Final Report
Another meta-analysis specifically examined women’s experiences of elder abuse and was based on 50
prevalence studies and focused on community-dwelling women aged 60 and over (Yon, Mikton, Gassoumis, &
Wilber, 2019). It found a pooled prevalence rate of 14.1%, with a reference period of one year prior to the survey
being conducted. The most common abuse subtype was psychological abuse (11.8%), followed by neglect (4.1%),
financial abuse (3.8%), sexual abuse (2.3%) and physical abuse (1.9%).
Supported by prevalence studies, the international literature on elder abuse sheds light on a range of other
issues, including perpetrator profiles, risk and protective factors, the impact of elder abuse and whether or not
those who experience it seek help or report it.
This body of literature also establishes that the phenomenon of elder abuse is context and culture specific, as Ho
and colleagues’ finding of higher prevalence in non-Western compared to Western countries indicates. An analysis
of the evidence on prevalence by Williams and colleagues (2017, p. 62) concluded that ‘wild’ variations in estimates
‘across region, culture, socio-political, and domestic situations’ were due not only to methodological differences in the
approaches taken to measure elder abuse in different studies but to variations in social and cultural conditions: ‘that
which is considered rudeness in one culture may well be considered severe emotional abusiveness in another culture’.
Risk factors
The body of international literature that has been developed in the past decades has supported the development
of insight into risk factors for elder abuse. A recent synthesis of this evidence by Storey (2020) has highlighted
eight victim-related factors that are associated with greater susceptibility to experiencing elder abuse and eight
factors related to perpetrators.
According to Storey’s analysis of the evidence base, the eight victim-related factors are:
y Problems with physical health are associated with greater vulnerability to not only experiencing elder abuse
but also more severe experiences.
y Similarly, mental health problems, particularly depression and cognitive decline, are risk factors not only for
elder abuse but elder abuse of greater severity.
y Problems with substance misuse, including alcohol.
y Dependence is associated with elder abuse experiences but is not a predominant cause of elder abuse.
y Problems with stress and coping can be both precursors to and consequences of elder abuse. The stress may
be caused by the perpetrator or come from other sources. It can also inhibit help seeking and lead to more
sustained and possibly severe experiences of elder abuse.
y Attitudes such as self-blame, excusing the abusive behaviour of family members, protecting perpetrators,
self‑depreciation, stoicism and apathy are risk factors for sustained experiences of elder abuse.
y Previous experiences of abuse, including abuse in childhood and neglect and intimate partner violence as an adult.
y Problems with relationships, including with adult children, conflictual relationships with family and friends and
social isolation.
A range of personal and systemic barriers to help seeking for elder abuse have been identified. At a personal
level, these include fear of the consequence of disclosure. Such feared consequences include retaliation,
abandonment, institutionalisation or ostracisation (Burnes, Lachs, Burnette, & Pillemer, 2019a; Dominguez, Storey,
& Glorney, 2019). A qualitative study based on interviews with 28 people who had been assisted by Seniors Right
Victoria demonstrated that for some participants, the consequence of help seeking did involve a change in living
arrangements in order for the abuse to stop (Vrantsidis, Dow, Joosten, Walmsley, & Blakey, 2016).
Further personal barriers include shame and embarrassment, self-blame and low self-esteem (Dominguez et al.,
2019; Storey, 2020). A desire to keep the abuse secret, particularly where it involved friends or family members
was nominated as the reason for not disclosing by more than a quarter of participants who experienced
financial abuse and emotional mistreatment in a study focusing on perpetrator identity and disclosure patterns
(Acierno et al., 2020).
Fear of the consequences of disclosure for the perpetrator is also among the personal drivers of non-disclosure
(Acierno et al., 2020; Dominguez et al., 2019; Storey, 2020; Vrantsidis et al., 2016). In Acierno and colleagues’
(2020) study, even larger proportions nominated not wanting to get the perpetrator into trouble as reasons for
non-disclosure (52% financial and 39% emotional) than wanting to keep the abuse hidden.
Victims have revealed concern about disclosure worsening their relationship with the perpetrator, fear of
exposing the perpetrator to punishment and being worried about worsening the perpetrator’s circumstances
that might be associated with the abuse (Dominguez et al., 2019; Storey, 2020; Vrantsidis et al., 2016).
Some research has also shown that social isolation on the part of the victim, and a broad social network on the
part of the perpetrator, are associated with lower help seeking (Burnes, Acierno, & Hernandez-Tajada, 2019b;
Vrantsidis et al., 2016).
Among the broader level barriers to disclosure are lack of knowledge about services and concern about whether
services have the capacity to provide adequate and appropriate help, particularly in the context of the personal
concerns outlined above (Dominguez et al., 2019).
For some groups, including those from Aboriginal and Torres Strait Islander and CALD backgrounds, a fear
of authority and a reluctance to engage with criminal justice processes have been identified as barriers to
help seeking in other contexts and are also likely to be relevant in relation to elder abuse (Family Law Council,
2012a & b).
There is limited research that provides insight into the factors that facilitate help seeking. The circumstances
in which help seeking appears more likely to occur are those where the victim has a good social network,
emotional and familial ties with the perpetrator are not strong and the victim has a sense of being betrayed by
the trusted person (Acierno et al., 2020, Dominguez et al., 2019). Situations in which the abuse is particularly
severe, including where it involves physical abuse or multiple types of abuse, are associated with higher levels of
help seeking (Burnes et al., 2019a, 2019b). Burnes’ (2019b) study also demonstrated higher levels of help seeking
where the perpetrator had previous engagement with police.
Australia
In Australia, the empirical evidence base on the scope and nature of elder abuse has to date been limited. In large
part, it has consisted of analyses of data to elder abuse hotlines, assessments of the capacity of administrative
datasets to provide insight into the extent and nature of elder abuse, and qualitative research focusing on
particular subtypes of abuse or particular populations.
14 National Elder Abuse Prevalence Study: Final Report
Helpline data
A recent analysis of data from a helpline operated by Seniors Rights Victoria (SRV) (Joosten, Gartoulla, Feldman,
Brijnath, & Dow, 2020) demonstrates that the numbers of calls to the helpline increased by 63% over seven
years, with a distinct rise in 2016, which the report attributes to the awareness raised following the release of
the reports by the Royal Commission into Family Violence in Victoria. The analysis shows that most calls were
from women (72% compared with 28% from men) and mostly concerned psychological abuse (63%) or financial
abuse (62%). The majority (91%) of abuse discussed in the calls involved family members, mostly sons (39%) or
daughters (28%). Most calls were in relation to people aged 70 or over (72%).
A similar analysis of data from calls to the Queensland Elder Abuse Prevention Unit (EAPU) published in 2015,
and covering the preceding five years, shows similar patterns in relation to gender and abuse types (Spike, 2015).
Most calls were in relation to women experiencing abuse (68%, cf. men: 31%). The most common form of abuse
across the period covered was psychological abuse (65%), with financial abuse becoming increasingly common
and reaching parity with psychological abuse. The majority of abuse was committed by adult children (31% sons
and 29% daughters). The age profile was older than the SRV analysis, with 80–84 years the most common victim
age bracket represented in the analysis.
Another analysis of EAPU data comparing the characteristics of clients with and without cognitive impairment
found they differed in their characteristics, types of elder abuse, perpetrators and relationship factors (Gillbard,
2019). In particular, those with a cognitive impairment had more complex care and support needs, were older, were
more likely to be dependent on perpetrators, and were more likely to experience neglect and social abuse and less
likely to experience physical or psychological abuse than older people in the non-cognitive impairment group.
Perpetrator characteristics differed between the two groups, with perpetrators more likely to have a history
of interpersonal issues, mental illness and substance abuse in the non-cognitive impairment group, whereas
perpetrators in the cognitive impairment group were more likely to exhibit inheritance impatience and taking
responsibility for their abuse-related behaviours. Barriers to change for victims also differed as victims without a
cognitive impairment most frequently identified fear of further abuse as a barrier whereas a lack of capacity was
most frequently identified for those with a cognitive impairment.
A 2017 study focusing on Queensland included interviews and focus groups with professionals working with
agencies concerned with older people generally and elder abuse in particular. In addition to considering the
scope of elder abuse on the basis of administrative data, the study concluded that existing data sources
were unable to support prevalence estimates (Blundell, Clare, Moir, Clare, & Webb, 2017). On the basis of the
information they were able to obtain, the researchers concluded that elder abuse was under-reported and costly.
A similar exercise was undertaken by researchers in South Australia (Lacey, Middleton, Bryant, & Garnham, 2017).
This study focused on assessing administrative datasets through interviews with key agencies and included
six interviews with older people who had experienced abuse. This study concluded that improvements to
consistency and collection of administrative data could support better understanding of trends in abuse cases
and build a better understanding of the extent and depth of the problem.
Qualitative research
In the past 15 years, qualitative research in Australia has examined a range of issues including financial abuse
among CALD and non-CALD populations (e.g. Bagshaw, Wendt, Zannettino, & Adams, 2013; Wainer, Owada,
Lowndes, & Darzins, 2011; Wainer, Darzins, & Owada, 2010) and sexual abuse among older women (Mann, Horsley,
Barrett, & Tinney, 2014).
One of the more recent studies examined experiences of older people (sample size 28) who sought assistance
in relation to elder abuse from Seniors Rights Victoria (Vrantsidis et al., 2016). The interviews revealed that all
perpetrators were family members, mostly adult sons and daughters living with the older person when the
Chapter 2: Evidence on elder abuse 15
abuse occurred, and that abuse was most frequently in the form of psychological (n = 21) and financial (n = 19)
abuse, with physical abuse also experienced by some participants (n = 7). Most older people reported the abuse
had stopped or resolved following a range of interventions and support to varying extents provided by Senior’s
Rights Victoria. Following the interventions, perpetrators had often left the older person’s home or the older
person had moved homes but stopping contact was also a frequent outcome. The interviews did not directly ask
older people about cultural experiences; however, some themes described by participants from culturally and
linguistically diverse backgrounds included experiencing negative views about women associated with particular
cultures and strong views about parental and family obligations (including caring for grandchildren).
The size and scope of a problem such as elder abuse, including the relative prevalence of abuse subtypes, can only
be measured through a prevalence study based on a sample representative of the population that is of concern.
Understanding population level prevalence is critical for a number of reasons (see e.g. Dow & Brijnath, 2019),
including the development of:
y targeted and proportionate response and prevention efforts based on a rigorous understanding of the scope
and nature of elder abuse
y screening and assessment tools for use in community and clinical settings to better identify elder abuse
y estimates of the future size of the problem in the context of an expansion in the proportion of the population
aged 65 and over.
Further, detailed systematic insights into important questions such as perpetrator profiles, risk and protective
factors and patterns in reporting are vital to informing better policy and practice strategies. These insights include:
y how proportionate and effective responses to the different abuse subtypes may be developed, including
understanding the co-occurrence of different subtypes and the extent to which some subtypes may lead to
the occurrence of other subtypes
y how risk factors for experiencing elder abuse may be decreased and protective factors increased
y how safe and effective interventions may be designed, given the evidence that elder abuse is frequently
committed by family members and that this can lead to complex emotional and physical dynamics,
particularly if the victim and perpetrator live together (e.g. Burnes et al., 2019b)
y how perpetrator-related risk factors can be better managed.
The findings from the SGC, together with the findings from the SOP, will be particularly important in designing
efforts to prevent elder abuse and raise community awareness of how to identify and respond to elder abuse in
Australia. In particular, the findings will allow the:
y examination of the extent to which the occurrence of elder abuse may be linked to views and attitudes that
condone elder abuse
y development of strategies to raise awareness of elder abuse and assist the community to identify it
y development of measures to increase knowledge in the community of how to respond to elder abuse,
including which authorities to report it to.
The focus on CALD groups in the research program is also critical to developing evidence-based responses.
Given the significant representation of people from CALD backgrounds in the Australian population, the findings
of the CALD substudy will support:
y understanding whether elder abuse is different in form and extent among CALD groups compared with
non‑CALD groups
y whether perpetrator profiles and risk and protective factors and responses are different for CALD compared
to non-CALD groups
y identifying whether awareness of and responses to elder abuse differ between CALD and non-CALD groups.
Findings in these areas will be important in assessing the extent to which tailored responses to elder abuse are
necessary for CALD individuals and communities.
16 National Elder Abuse Prevalence Study: Final Report
Summary
Elder abuse prevalence studies have been conducted in more than 50 countries around the world in the past
two decades, as governments attempt to develop better policy and practice responses to the issue. Variations
in methodology between studies limit international comparisons but the largest meta-analytic study available
indicates a global prevalence rate of 15.7%, with psychological abuse being most common at 11.6% (Yon et al., 2017).
The international evidence establishes that elder abuse is a complex phenomenon that is influenced by social,
cultural and socio‑economic factors.
In Australia, the empirical evidence base on elder abuse has been limited to studies looking at particular types of
elder abuse (e.g. financial abuse), qualitative studies and those based on administrative data from services who
provide support to older people. Such studies are unable to shed light on the proportion of older people aged
65 and over who experience elder abuse and the relative prevalence of the different subtypes of abuse. Nor are
they able to assess other important issues, such as the extent to which elder abuse is under-reported.
The evidence provided by the National Elder Abuse Prevalence Study will support better policy and practice
responses to address elder abuse across Australia, in line with the National Plan.
Chapter 3: Definition and concepts 17
This chapter explains the conceptual definition of elder abuse applied in the study and introduces the
theoretical frameworks that have informed the development of the methodology and the analysis and
interpretation of the findings. The measures used and the technical, operational definitions applied to the
six subtypes of elder abuse are set out in chapter 5.
Conceptual definition
The preparatory work for this study (see chapter 1) included the development of a working conceptual
definition informed by the international literature, existing conceptual frameworks and stakeholder consultations
(Kaspiew et al., 2019). This definition is:
a single or repeated act or failure to act, including threats, that results in harm or distress to an older
person. These occur where there is an expectation of trust and/or where there is a power imbalance
between the party responsible and the older person.
The definition involves five elements that informed the development of measures in this study:
y A single or repeated act or failure to act. The measures applied in this study captured varied forms of physical,
sexual, financial, psychological abuse, and neglect. (An additional concept of abuse relating to language and
culture was developed for the CALD substudy.) The measures are described in detail in chapter 5.
y An older person (defined for this study as a person aged 65 and older).
y A person who is responsible for the act or omission. The operational definition included family members,
friends, neighbours, acquaintances and professionals but excluded strangers.
y The consequence of harm or distress. This was measured by asking the older person how serious the abuse was.
Further discussion of the influence of these measures on the technical analytic definition is set out in chapter 5.
y A situation where the connection between the older person and the perpetrator of the act or omission
involves an expectation of trust and/or a power imbalance. For the purpose of this study, this was inferred
from the relationships between the older person and the person who was the source of the act/omission.
Further discussion is in chapter 5.
The working definition adopts an approach that is broader than the World Health Organization definition but
is still widely accepted in referring to a power imbalance as well as an expectation of trust (Bagshaw, Wendt, &
Zannettino, 2009; Clare, Clare, Blundell, & Clare, 2014; EAPU, 2019; Hightower, Smith, & Hightower, 2006; Joosten,
Vrantsidis, & Dow, 2017; Mysyuk, Gerardus, Westendorp, & Lindenberg, 2016).
In the development of this working definition, it was recognised that two aspects of it – the consequence of
harm or distress and the relationship of trust and/or power imbalance – may be assessed subjectively (from the
perspective of the older person) or objectively (from the perspective of a reasonable bystander). The implications
of this for the technical analytic definition applied in deriving prevalence estimates is set out in chapter 5.
As the working definition demonstrates, elder abuse involves a varied range of acts or omissions and a significant
range of relationship types, including intimate partners, intragenerational relationships (siblings), intergenerational
family relationships (parents and children, grandparents and grandchildren) and non-family relationships such
as those involving professionals, carers, neighbours and acquaintances. This establishes that elder abuse is a
complex phenomenon and underlines the importance of explanatory theoretical and conceptual frameworks.
18 National Elder Abuse Prevalence Study: Final Report
Theoretical frameworks have been slow to develop in the area of elder abuse but it is increasingly being
recognised that theoretical sophistication is important in developing understanding of the diverse forms and
dynamics of elder abuse and that different theories may be required to explain different subtypes of abuse
(e.g. Jackson & Hafemeister, 2013; Naughton et al., 2010). For example, dynamics may be quite different in these
three situations: financial abuse committed against a father by a son; sexual abuse committed by a friend against
an older woman; psychological abuse committed against a female partner by a male partner.
For this reason, three main perspectives have informed this research: ecological approaches, theories relevant
to family violence, and a human rights perspective. To some extent, these perspectives are complementary and
have some common elements.
With elder abuse prevalence research in Australia in its early stages, limitations in the scope of data collection
possible in the context of a 28-minute survey (see further Appendix B) have placed significant constraints on
the capacity to extensively examine some issues of theoretical relevance. For this reason, an additional theory,
outlined below (the life course perspective), has had limited influence on the methodology but is considered to
be important for future research.
This section outlines relevant theories and their influence on the methodology and interpretation of findings.
Ecological approaches
Increasingly important approaches in both the elder abuse and family violence fields are multi-systemic theories.
These models understand an individual’s experience in the context of several levels of intersecting influence,
the narrowest being the family and the broadest being social norms and values. The ecological theory was first
developed by Bronfenbrenner (1986) and has since been adapted for various contexts including elder abuse
(e.g. Acierno, Hernandez-Tejada, Muzzy, & Steve, 2009) and family violence (e.g. Our Watch, ANROWS & VicHealth,
2015). Elder abuse and family violence are recognised as ‘adjacent areas of inquiry’ (Burnight & Mosqueda, 2011).
Adapting an applied ecological model to elder abuse in Australia, the National Ageing Research Institute (NARI)
has developed a model that focuses on the relationship between the older person and the perpetrator of the
abuse (the person who is in a ‘relationship of trust’) in the context of relevant community dynamics and, more
widely, relevant cultural norms and beliefs (Joosten et al., 2017). The model reflects relevant individual-level
factors for the person who experiences the abuse (e.g. dependency based on a requirement for care, frailty,
physical or mental health problems, cognitive impairment) and the perpetrator of the abuse (e.g. dependency
on the older person, financial problems, substance abuse problems, social isolation) that have been identified as
important in research on elder abuse. Community-level issues include the care setting (e.g. hospital, residential
care, community-based care). At the broadest level, the model refers to cultural norms and beliefs concerning
ageism, intergenerational conflict and wealth distribution and family violence.
The application of multi-systemic approaches in research entails collecting data and conducting analyses that will
enable examination of the different levels of influence and their associations with the experiences of elder abuse.
In practical terms, this approach involves a focus on the older person and their social, economic and cultural
circumstances, as well as consideration of the nature and circumstances of abuse, including the characteristics
of the perpetrator. More broadly, it also requires a focus on social factors – such as attitudes to ageing and older
people – that shed light on societal factors that may influence the occurrence of, and responses to, elder abuse.
Socio-ecological approaches explicitly informed the Canadian research on attitudes and awareness (EKOS
Research Associates Inc., 2011) and the Portuguese prevalence study (Gil et al., 2015). This theoretical approach
is also compatible with the prevalence studies conducted in New York (Lifespan of Greater Rochester Inc., 2011)
and the UK (O’Keeffe et al., 2007).
In applying a three-part methodology based on a focused prevalence study (the SOP), a study examining social
attitudes and awareness in a population sample (the SGC) and a CALD substudy, this research represents
the most expansive application of a socio-ecological approach to date. This methodology means that the
experiences of abuse reported in the SOP can be considered not only in terms of the characteristics of the
Chapter 3: Definition and concepts 19
victims (chapter 6) and perpetrators (chapter 7) but also in their wider social context on the basis of the findings
of the SGC (chapters 11 and 12). Further, the findings of the CALD substudy, combining findings of the SOP
and SGC and measuring an additional subtype of elder abuse based on abuse relating to language and culture,
provide insight into the cultural dimensions of elder abuse in the context of Australia’s immigrant population
(chapters 5, 6 and 7).
Family violence
The question of whether elder abuse should be regarded as a form of family violence has been the subject
of some debate, locally (e.g. Joosten, Dow, & Gally, 2019) and internationally (e.g. Roberto & Teaster, 2017).
When there is elder abuse between spouses in later life, the overlap is clear, even though questions arise as to
whether the behaviour has been sustained over time or whether it has newly arisen. However, elder abuse that
is intergenerational (e.g. perpetrated by a child against a parent) potentially involves ‘unique dynamics’ (Joosten
et al., 2019) not adequately addressed by the emphasis on gendered dynamics of power in family violence theory.
Australia’s family violence prevention strategies are premised on a socio-ecological model that emphasises the
role of gender inequality in creating the conditions for family violence (Our Watch et al., 2015). In this model,
four levels of influence interact to create family violence. At the broadest level, societal-level norms support rigid
gender roles and condone violence against women. The second level is comprised of systemic barriers to the
achievement of gender equality. The third level involves community practices that perpetuate gender inequality
and discrimination. At the fourth, individual level, rigid gender roles, male dominance and controlling behaviours
create the personal environment for violence.
The emphasis on gender inequality in the socio-ecological model aligns with the notion of a power imbalance
in long-standing conceptualisations of family violence. Unequal power relationships between men and women
are seen as underpinning family violence as a phenomenon committed by men against women (e.g. Pence &
Paymar, 1993). Power imbalance has also been influential in some thinking about elder abuse (see e.g. National
Clearinghouse on Abuse in Later Life [NCALL], 2011), with dependency arising from age-related needs creating
a ‘power advantage’ that can be exercised to the detriment of the dependent (or older person) (Burnight &
Mosqueda, 2011).
For this reason, the working definition applied in this research included the notion of a power imbalance, in
addition to a ‘relationship of trust’ (EAPR). The analyses of the relationship contexts in which elder abuse occurs
(chapters 6 and 7) presented in this report explore dynamics of power to the extent that the data allow.
can contribute to an environment in which individuals who abuse older people fail to recognise that their
behaviour constitutes abuse; other members of society fail to notice these negative behaviours or take
action to stop them; and older people experiencing elder abuse blame themselves and are too ashamed to
seek assistance. (page 5)
A human rights model suggests that abuse of older people, rather than being exclusively ‘located within personal
relationships’, (Biggs & Haapala, 2013) also arises out of relationships between the state and individuals as a
consequence of the way that the state does or does not safeguard the human rights of older people.
A human rights approach has influenced this methodology in several important ways, including through
examining attitudes to older people in the SGC. These findings are set out in chapter 12.
Research evidence on the experience of different forms of violence and abuse at different life stages – childhood,
adolescence, young adulthood, middle age and later life – suggests some links between exposure in childhood
and adolescence and experiences (potentially as a victim or perpetrator) at later life stages. A recent review of
research evidence examining ‘interconnections among different forms of violence that span the developmental
life course’ noted that the research evidence on elder abuse in this context, although not as well-developed
as the literature on links between childhood, adolescence and mid-life experiences, indicated the existence of
connections between earlier forms of violence and elder mistreatment (Herrenkohl et al., 2020, p. 9).
Australian evidence demonstrates a link between experiences of abuse in childhood and violence in adulthood.
ABS data published in 2019 show that the 13% of adults who experienced physical or sexual abuse in childhood
had increased risks of experiencing violence in adulthood (ABS, 2019b). Compared with adults who did not
experience childhood abuse, those who did were twice as likely to experience violence as an adult (71%, cf. 31%)
and three times as likely to experience partner violence (28%, cf. 8.9%).
Both the Canadian elder abuse prevalence study and the US one examined some life course influences on the
experience of elder abuse. An experience of abuse as an adult, a youth or child was the second most important
of nine factors correlated with an experience of elder abuse in the Canadian study (National Initiative for the Care of
the Elderly, 2015). In the US study, participants who reported a prior traumatic event (defined as a situation involving a
natural disaster, a serious accident or another situation, all involving fear of being killed or seriously injured) were at
increased risk of emotional and sexual abuse, and financial abuse perpetrated by a stranger (Acierno et al., 2010).
In this current study, the capacity to implement a life course perspective was limited for two main reasons.
First, the need to limit participant burden and not have an average interview time exceeding 28 minutes placed
significant restrictions on the number of questions that could be asked. As the first prevalence study in Australia,
the survey instrument prioritised core question areas needed to assess and understand prevalence and dynamics.
Including questions on the previous experience of violence and abuse was not feasible in this context. Second,
the cross-sectional (one survey conducted at a particular point in time) nature of the study does limit capacity
to reliably examine causal influences, which would instead require a longitudinal research design (a survey that
re-interviews the same participants over a number of years) (Herrenkohl et al., 2020).
Nonetheless, the findings set out in chapter 6, particularly those that highlight higher levels of elder abuse
among those who are divorced or separated compared to those who are partnered, indicate that a life course
perspective is important in understanding elder abuse and should be considered in future research (chapter 12).
Summary
The conceptual definition applied in this research was developed for the Australian context, following a literature
review, an analysis of conceptual frameworks and consultations with stakeholders. This definition accommodates
the measurement of five core subtypes of abuse: financial, physical, sexual and psychological abuse, and neglect.
To be operationalised in research, the definition requires five elements to be examined: the victim, the
perpetrator, the relationship between them, the acts or omissions that constitute the abuse and the consequence
of harm or distress to the older person. Characteristics of the relationship identified in the definition are a
relationship of trust or a power imbalance.
Three theoretical frameworks have influenced the design of the research. An ecological approach is reflected
in the aspects of the design that examine personal, family and community dynamics with the SGC focus
on attitudes and awareness providing the broader social context for the SOP findings on the dynamics and
prevalence of elder abuse.
The second theoretical framework overlaps to some extent with the first, in that family violence theory directs
attention to power relationships at the individual and social level as an explanatory device for family violence.
With international literature establishing that vulnerability can create a susceptibility to elder abuse, power
imbalance is recognised alongside a relationship of trust as an element in the conceptual definition.
Finally, a human rights perspective, with an emphasis on ageism as an underlying driver of elder abuse and
manifesting in a lack of care and respect for older people, is also relevant to the research design and compatible
with an ecological approach.
A focus on another theoretical perspective – a life course perspective – was beyond the scope of this
methodology. However, international evidence establishing links between adverse experiences and elder abuse,
together with Australian evidence demonstrating heightened susceptibility to violence or intimate partner
violence for people who experience abuse in childhood – indicates that this is an area that warrants attention.
Chapter 4: Methodology 21
4 Methodology
This chapter provides an overview of the methodology of the National Elder Abuse Prevalence Study, including
a description of the sample characteristics for the SOP and the SGC.
Study outline
The National Elder Abuse Prevalence Study is comprised of two surveys: the Survey of Older People (SOP) and
the Survey of General Community (SGC). The design of the SOP supports the estimates of the prevalence of
elder abuse in the community-dwelling population aged 65 years and older. The SGC supports insight into the
context for elder abuse in Australia by examining attitudes, knowledge and awareness relevant to elder abuse, as
well as care-giving practices, among people aged between 18 and 64 years. The design of the two surveys was
guided by the preparatory work, which included: the review of international prevalence studies with a particular
focus on studies conducted in the last decade and on a large scale; examination of conceptual frameworks; and
consultation with subject matter experts.
The key elements of the approach adopted for the two surveys include:
y A working definition was adopted that takes a broader approach than the World Health Organization
definition (discussed in chapter 3).
y The SOP involved a probability sample of 7,000 older people aged 65 years and older who are living in private
dwellings and have the cognitive capacity to successfully engage in a telephone interview of approximately
28 minutes on average. The survey was conducted from 12 February to 1 May 2020. This survey excludes older
people without access to a telephone (landline or mobile), residents of institutional premises (e.g. prisons,
nursing homes, military bases), people incapable of undertaking an interview due to a physical or health
condition and people appearing to be under the influence of drugs or alcohol.
y The SGC comprised a probability sample involving 3,400 people from the general community who are
living in private dwellings and have the cognitive capacity to successfully engage in a telephone interview of
approximately 17 minutes on average. The SGC was conducted from 12 November to 15 December 2019.
y Both the surveys recruited participants through the random digit dialling sampling approach with a landline and
mobile phone blend, and the survey data were collected through computer-assisted telephone interviews (CATI).
y Both the surveys were cross-sectional.
y A CALD subsample drawn from the SOP and SGC samples was included to provide insight into the
experiences of certain abusive behaviours related to issues of language or culture.
Limitations
This section briefly discusses the main limitations of the methodology, including those arising from the survey
response rate and the sample characteristics. It is important to appreciate that the prevalence findings apply to
people aged 65 and older who live in the community and had capacity to complete the survey. As such, they
likely represent an under-estimate of elder abuse because they do not include people with reduced capacity due,
for example, to cognitive decline or those who live in residential care settings.
22 National Elder Abuse Prevalence Study: Final Report
Response rate
The random digit dialling (RDD) approach produced a low response rate for the SOP at 15.5% and for the SGC at
11.6%. This was expected (see AIFS & SRC, 2019, Strengthening the Evidence Base, Component 2 – Final Report)
given that the RDD approach screens the wider population in order to recruit an eligible sample by calling
phone/mobile numbers generated at random.
Several complex issues arise in relation to considering the feasibility of including and excluding specific groups
of older people. First, in relation to those older people who rely on the formal or informal support of carers
(including family members) to continue to live in private dwellings, securing interviews with these older people
would require the cooperation and assistance of carers, including to be engaged on the phone. Older people
experiencing physical frailty may be unwilling to acknowledge their experiences of abuse while in the presence
of a third party.1 Indeed, there is potential that some of the carers of older people in private dwellings who are
approached to participate may be the perpetrators of the relevant abuse.
Some mitigating strategies were employed to address these issues. The survey introduction was broadly framed,
focusing on older people’s wellbeing and the role of the research in supporting government policy development,
so as to lower the likelihood of gatekeeping from other household members and limit the scope for awareness
among perpetrators of the possibility of abuse disclosure. The fieldwork company employed experienced
interviewers for this project, both to support greater potential for participant engagement and ensure the survey
was administered sensitively.
Second, older people who rely on the support of carers due, at least in part, to their cognitive impairment, may
not have the capacity to provide informed consent to participate. Assessing this capacity raises practical and
ethical challenges in the context of a CATI data collection approach.
Third, for older people living in residential aged care, there would be challenges associated with their recruitment
and with the conduct of the research, including addressing any concerns of the care organisations and making
the necessary administrative arrangements to facilitate participation (e.g. the process of selecting participants,
conducting private interviews with minimal interruption to carers’ work and so on). Once again, assessing the
capacity of older people living in residential aged care to provide informed consent to participate raises both
practical and ethical challenges, and substantial numbers may be deemed to not have the capacity to consent or
to participate. These exclusions mean that the estimates from this study likely represent under-estimates, even
for the community dwelling population and especially so for those in residential care settings.
This means that differences among different subsamples are not amenable to in-depth exploration (see further
chapter 13).
1 This may be less of an issue if the discussion takes place over the phone rather than face-to-face at the older person’s place.
Chapter 4: Methodology 23
Several disadvantages are associated with the dual-frame random digit dialling approach and include call screen
out of unrecognised telephone numbers, likely lower response rates and greater non-response bias compared
to the approach of using a database as the sampling frame. To address these limitations several methods were
employed to improve sample precision for the Prevalence Study:
y Stratified random sampling was used to ensure an adequate representation of subgroups of the population.
y Pre-contact SMS messages were sent to potential mobile participants to alert them about the study prior to
receiving a call.
y Data weighting was used to improve the representation of the study population.
Ethics
The AIFS Human Research Ethics Committee provided ethical review and clearance for this research. As the
surveys involved engaging with older people and people from the community, it was acknowledged that during
the course of their participation in a telephone interview and, in particular, when considering the potentially
sensitive questions included in the instruments, some participants may reflect on experiences of violence/
abuse and/or on complex family and other relationship dynamics, and that this may present risk issues. These
circumstances raised significant ethical complexities for the research team to address, including the need for a
sensitive approach to data collection and the development of appropriate interview protocols to support those
participants who may become distressed as a result of participating. To address these complexities, consideration
was also given to interviewer experience, pre-training and preparation and mandatory reporting obligations.
Data weighting
Data weighting was conducted to reduce the extent of any biases arising from potential non-coverage of
population subgroups and consequent constraints on the ability to make inferences about the target population
from the sample. This process takes into account that the sample may not be sufficiently representative of the
population as some participants may have a lower likelihood of participating in the survey (e.g. men are less likely
than women to participate in a survey).
Sample weights were generated for both the SOP and the SGC according to the current population, as estimated
by the ABS, to make adjustment for differential representation of subgroups in the data.2 The variables used for
weighting were: gender, state, country of birth, language spoken at home, highest education, home ownership
status and telephone status (whether owns a landline phone). The sample weights are applied to all the analyses
presented in this report except multiple regression analyses. Unweighted sample base numbers on which specific
statistics are derived are also provided in relevant tables and/or table and figure notes.
2 Generalised regression (GREG) weighting method was used to generate the data weights. This method involves non-linear
optimisation to minimise the distance between the design and adjusted weights subject to the weights meeting the benchmarks.
24 National Elder Abuse Prevalence Study: Final Report
Analysis reporting
When presenting overall prevalence of elder abuse (any subtype and overall), for the entire sample, 95% confidence
intervals are provided in the main report, and are shown in Appendix A for prevalence figures relating to subgroups.
For the bivariate analyses involving two categorical variables, a design-based F test is used to assess whether the two
categorical variables (e.g. the prevalence of elder abuse by gender) were correlated. In relation to bivariate analysis
involving the comparison of a scale between subgroups (e.g. ageism scale by gender), weighted regression (bivariate)
is used to test whether the difference between a specific group and the reference group is statistically significant.
Statistical testing was not carried out in relation to the following areas:
y the relationship between older persons who experienced abuse and the perpetrators, or relationship between
parties on other matters (e.g. financial assistance, power of attorney)
y sources of help that older participants sought for their experience of elder abuse (chapter 8)
y actions that older participants took to stop abusive behaviours (chapter 8)
y where some of the responses were back-coded based on verbatim responses provided by the participants
(Note that a significant number of responses were back-coded. Statistical significance would be influenced by
how detailed categories were presented and how back-coding categories were developed).
Unless otherwise indicated in the report, when reporting statistics by cross-tabulation, the general approach was
to include only valid responses; that is, non-responses such as ‘Do not know’ or ‘Refused’ were excluded from the
bivariate analysis. For the majority of variables analysed, the proportion of such responses was small (less than
5% of all responses). When reporting analysis results at a high level, non-responses were included in the table for
deriving percentages, and this treatment of non-responses is indicated in table notes.
In general, results are not provided when a base sample involves fewer than 50 participants. However, in some
instances, where it is important to provide insights on specific issues (e.g. the relationship between CALD
participants who reported experiencing abuse relating to their language and culture and the perpetrators),
results are provided with appropriate caveats even when the base sample is fewer than 50.
Sample characteristics
SOP
Table 4.1 describes the profile of the total adult interviews achieved for the survey of older people. The sample
appeared to have a slight over-representation of those aged in their 70s. The data show over-representation of
older people who are female, married or widowed, born in Australia, unemployed or retired, educated beyond
Year 11 and who are homeowners. As expected, the sample includes an over-representation of older people with
a non-Indigenous background, who were born in Australia and who speak English at home. In relation to family
arrangements, older people lived as part of a couple or on their own. The sample also has a higher representation
of older people who have an education level with certificates/diplomas, degrees or higher, as well as older people
with an education level below Year 12. The weighted sample characteristics were largely similar to the 2016 Census.
Notes: 2016 Census data include population over 65 residing in a private dwelling. Refused/Don’t know responses included
in calculation of proportions for unweighted/weighted and 2016 Census data but not shown in the table above.
This explains why the total proportions for some characteristics do not sum to 100.0%. Due to a high proportion
of Don’t know/Refused in 2016 Census for education, all information for this characteristic is reported (not stated/
inadequately described). a The main English speaking countries comprise of the United Kingdom, Republic of Ireland,
South Africa, Canada, the United States of America and New Zealand. b Proportion in 2016 census for categories: With
grandchildren/Other relatives with/without unrelated persons/Unrelated persons/Other combined. c Proportion in 2016
Census for categories: ‘Employed’ includes ‘employed but away from work’. d Proportion in 2016 census for categories:
Certificate/diploma and trade/other combined. Other includes no educational attainment (Census 2016).
SGC
Table 4.2 describes the profile of adult interviews achieved for people from the general community. The results
indicate a greater representation of people with the following characteristics: aged 55–64 years, being married,
having a degree or higher qualification, and being employed. The weighted data adjusted the characteristics
similar to the 2016 Census.
Table 4.2: Survey of the General Community: Characteristics of the SGC sample
Note: Refused/Don’t know responses included in calculation of proportions for unweighted/weighted and 2016 Census data
but not shown in the table above. This explains why the total proportions for some characteristics do not sum to 100.0%.
Due to high proportion of Don’t know/Refused in 2016 Census for education, all information for this characteristic is
reported (not stated/inadequately described). a The main English speaking countries comprise of United Kingdom,
Republic of Ireland, South Africa, Canada, the United States of America and New Zealand. b Proportion in 2016 Census
for categories: With grandchildren/Other relatives with/without unrelated persons/Unrelated persons/Other combined.
c Proportion in 2016 Census for categories: ‘Employed’ includes ‘employed but away from work’. d Proportion in 2016
Census for categories: not employed and retired combined. e Proportion in 2016 Census for categories: Certificate/
Diploma and trade/other combined. Other includes no educational attainment (Census 2016).
CALD substudy
The findings of the CALD substudy presented in this report, and summarised in chapter 13, are based on the
CALD subsamples from the SOP and SGC. The analytic approach to identifying CALD groups was different in
the SOP from the SGC. In the SOP, the CALD group analysis is based on all participants who reported speaking
a language other than English at home (n = 608). This was to ensure that the questions concerning the construct
of abuse relating to language and culture were fully relevant to this group. For this subgroup, 85% were
born outside of Australia as well as speaking a language other than English at home. This approach meant
that people who were born in English speaking countries outside of Australia were not included in the SOP
CALD subgroup.
For the SGC CALD subgroup (n = 660), the analysis was based on participants who reported being born in
a country where English is not the primary language (countries other than main English speaking countries).
Additional analysis based on country of region of birth is also carried out where it is feasible.
The CALD substudy is unable to shed light on the experiences of elder abuse and attitudes to elder abuse and
older people within particular CALD groups, except to a very limited extent (see chapter 13). The methodology
applied in this study supports some broad insights into similarities and differences between the non-CALD
sample and the CALD sample; however, further research (likely qualitative in nature) would be required to
understand experiences and dynamics among different groups in greater depth.
For each CALD sample, the sample characteristics were compared with their subpopulation in the 2016 Census.
The analysis indicated that the CALD sample in the SOP had a greater proportion of people with a higher
socio‑economic status: having a degree or higher qualification, owning their home outright, from areas with
highest socio‑economic status.
Compared to the population of older people aged 65 years and older who spoke a language other than English
at home in the 2016 Census, the SOP CALD subsample had a greater representation of those who are a higher
socio‑economic status: having a degree or higher qualification, owning their home outright, from areas with
highest socio‑economic status (see Appendix A, Table A4.1 for the detailed comparison). The proportion living alone
is higher in the CALD sample, compared to the census data. The weighted data largely reduced the differences.
Chapter 4: Methodology 29
Table 4.3 shows the number of participants in the SOP according to country/region of birth. The majority of
participants were born in English speaking countries (Australia n = 4,991 and United Kingdom/Ireland n = 944)
followed by Southern and Eastern Europe (n = 323) and North-West Europe (n = 238).
Table 4.3: Survey of Older People: Whether a language other than English spoken at home and
country/region of birth
Total
Characteristic N
Speak a language other than English at home 7,000
Speak English only at home 6,392
Speak a language other than English at home 608
Country/region of birth 6,956
Australia 4,991
New Zealand 135
UK/Ireland 944
North-West Europe 238
Southern and Eastern Europe 323
Middle East, North Africa 36
South-East Asia 92
Chinese Asia 32
Southern and Central Asia 57
North America 46
Africa 62
30 National Elder Abuse Prevalence Study: Final Report
5 Prevalence
Key messages
f The findings from the SOP indicate that Australia’s overall elder abuse prevalence rate is 14.8%.
f Prevalence rates are slightly higher for women than for men.
f Prevalence rates in the CALD population do not differ greatly from the general population sample.
Introduction
This chapter sets out findings on the prevalence of elder abuse – the proportion of the community dwelling
population aged 65 and over who reported experiencing elder abuse in the past year. Prevalence for three
subtypes of abuse – physical, financial and sexual – is expressed as a single prevalence figure. For two subtypes
of abuse (psychological abuse and neglect), the prevalence rates are expressed as a score across three bands:
low, medium and high. This approach is also applied to abuse relating to language and culture.
This chapter addresses three elements of the research aims (chapter 1):
y It describes the prevalence of elder abuse in the community dwelling population aged 65 and over.
y It also describes the prevalence of elder abuse, including specific subtypes, for the CALD subsample.
y It sets out the extent to which participants in the SGC reported having concerns that someone they know
aged 65 and over had experienced elder abuse.
This chapter has four further parts. The first part explains how the prevalence estimates are derived. The second
part sets out the measures (i.e. the questions that were asked in the survey) used and the technical definition
applied in the analysis along with the findings on overall prevalence and the prevalence of the five core subtypes
of abuse. The third part sets out findings of prevalence among the CALD subsample, considering overall
prevalence, prevalence of the five core subtypes and the prevalence of the CALD-specific subtype of abuse
relating to language and culture. The extent to which participants report experiencing multiple types of abuse
is considered in sections two and three. The final part considers the extent to which participants in the SGC
expressed concerns about elder abuse in relation to someone known to them.
Chapter 5: Prevalence 31
Deriving prevalence
This section sets out the technical analytic approach used to derive prevalence estimates For each type of
abuse, the development of a technical analytic definition involves determining how the following issues influence
whether a particular experience is or is not included in the prevalence estimate:
y the perpetrator of the experience (i.e. was there a power imbalance or a trust relationship?)
y the number of times an experience happened (frequency)
y self-assessed seriousness.3
Decisions on these aspects of the technical definition were informed by the conceptual definition (chapter 3).
The following sections explain the approaches taken and their rationales.
Some international literature demonstrates that in comparison with a binary present/not present approach,
an approach involving ‘a severity framework’ (Burnes et al., 2018), such as a scoring approach, is particularly
useful for two reasons (Burnes et al., 2018). First, it is a more sensitive measure for assessing whether change
has occurred, as it can assess whether abuse has become more or less serious through changes in the range of
scores. For example, if the proportion of the population that reports abuse in the most serious range declines and
the proportion reporting abuse in the lower ranges increases, then that would suggest measures to address elder
abuse are having an impact. Such an impact would not be detectable if a binary present/not present approach
was applied. A binary approach can only measure an absolute reduction (or increase) in the proportion of the
population reporting an experience of elder abuse.
Second, a scoring approach provides a more useful set of indications for policy, program development and
practice (e.g. Burnes, Lachs, & Pillemer, 2018). In these contexts, different approaches may be required for
experiences that fall into different score ranges. Experiences at the high end of the spectrum will require
responses that are different in urgency and intensity from experiences in the lower score ranges.
In relation to psychological abuse, there is significant variation in international studies as to the measures used
and the technical definitions applied. Further, the behaviours involved in psychological abuse are varied, and in
some circumstances where they occur on an isolated or infrequent basis, it may be inappropriate to assess them
as abuse (e.g. Williams et al., 2017). For these reasons, as set out below, a unique approach was developed for
psychological abuse to reflect the context of this study – as the first Australian effort to assess prevalence in an
environment where there is growing evidence of psychological abuse as a form of violence (e.g. AIHW, 2018).
Frequency is relevant in this technical definition but impact is not.
Similarly, neglect, being based on omissions rather than actions, is particularly complex with very limited
consistency in international approaches (e.g. Stodolska, Parnicka, Tobiasz-Adamczyk, & Grodzicki, 2020). As with
psychological abuse, the behaviours involved may not of themselves be abusive if they are isolated or infrequent.
Again, the approach adopted reflects the context of this study as the first Australian attempt to measure the
prevalence of neglect. Frequency is relevant in this technical definition but impact is not.
Abuse relating to language and culture is a construct measured uniquely in this study. The technical analytic
approach is the same as for psychological abuse.
3 This refers to the older person’s perceived seriousness for them of experiencing the treatment.
32 National Elder Abuse Prevalence Study: Final Report
Impact
In some international approaches, one element of the technical analytic definition for some abuse subtypes
includes self-assessed impact in determining whether some experiences should or should not be considered
elder abuse, particularly where the behaviours involved may be considered not serious or where they have
occurred infrequently. This provides a way of narrowing the scope of behaviour that counts as elder abuse.
For this study, the research team made a decision not to include self-assessed impact in the calculation of the
prevalence estimate. This is because international research demonstrates that in some circumstances, it is evident
that seriousness assessments are lower (i.e. participants are more inclined to rate some behaviours as less serious),
potentially because of complex psychological dynamics that can arise in relation to elder abuse (Burnes et al., 2019a).
These circumstances include situations where the victim and perpetrator live together, where the victim needs care
from the perpetrator and where the perpetrator is a child or grandchild of the victim. The analysis of Burnes and
colleagues suggests that in these circumstances, lower subjective assessments reflect emotions such as a sense
of internalised burden for requiring care, or embarrassment and shame that a child or grandchild could perpetrate
abuse. This means that an emphasis on self-assessed impact could produce unreliable prevalence estimates.
In Australia, similar points have been raised in commentary about whether the definition of elder abuse should
have a subjective element. The ALRC Elder Abuse Inquiry Discussion Paper (ALRC, 2016) noted that submissions
cautioned against over-emphasising the presence of distress in assessing whether elder abuse had occurred.
Similarly, the Elder Abuse: Research Definition Background Paper acknowledged the relevance of both subjective
and objective assessments of harm and distress (Kaspiew at al., 2019).
Given that this is the first systematic attempt to assess the prevalence of elder abuse in Australia, the research
team decided to exclude self-assessed impact from the technical approach to assessing elder abuse. In the
absence of systematic evidence on self-assessed seriousness in Australia, and the risk that prevalence findings
could be skewed by the inclusion of self-assessed seriousness, the research team decided this would be the
most appropriate approach. Chapter 9 sets out findings on self-assessed seriousness, which remain important in
understanding the dynamics of elder abuse, including in the context of whether action is taken in relation to elder
abuse experiences.
Prevalence findings
This section sets out estimates on the overall prevalence of elder abuse and the prevalence of the five abuse
subtypes (financial abuse, physical abuse, sexual abuse, neglect, psychological abuse). The reference period is
the 12 months prior to the survey. Consistent with the discussion above, the prevalence estimates for financial
abuse, physical abuse and sexual abuse are based on a single figure, reflecting the proportion of the sample that
reported abusive experiences. The prevalence estimates for neglect and psychological abuse are based on a
score with the overall prevalence assessment distributed across three bands: low, medium and high.
Overview
In the 12 months prior to being surveyed, 14.8% of the sample reported experiencing at least one form of
elder abuse (Table 5.1). Psychological abuse was most commonly reported (11.7%), followed by neglect (2.9%),
financial abuse (2.1%), physical abuse (1.8%) and sexual abuse (1%).
Women were more likely than men to report an experience of elder abuse (15.9% cf. 13.6%) (see Figure 6.1), but
consistent with Yon and colleagues’ (2017) international meta-analysis, which revealed no gender differences, these
overall figures do not suggest particularly strong gender dimensions overall. However, gendered patterns are evident
in some subtypes of abuse and in the profile of perpetrators (see further chapters 6, 8 and 9 in relation to gender).
Chapter 5: Prevalence 33
Table 5.1: Survey of Older People: Overall prevalence of elder abuse by type
Notes: Based on weighted data. Participants who did not respond to relevant questions were included in the total in deriving
estimates (1–2% across subtypes).
Assessing confidence intervals involves applying a mathematical formula that produces upper and
lower bound estimates of reliability. A 95% confidence interval means that we can be 95% sure that our
prevalence estimates are in the range of the upper and lower bound estimates. On this basis, we estimate
the numbers of people affected by elder abuse in Australia in the past 12 months are:
Table 5.2 sets out the prevalence findings in this study alongside those of Yon’s meta-analysis and the findings
from the prevalence studies from the UK, Ireland, Canada, New York and the USA. Notably, the measures and
34 National Elder Abuse Prevalence Study: Final Report
analytic approaches for financial, physical and sexual abuse are consistent between the latter studies and this
study. However, sample age ranges differ, as do sample sizes and data collection techniques. Measurement and
analysis of psychological abuse and neglect vary significantly among studies. Caution should be exercised when
making comparisons between these findings due to the different methodologies used. It is worth noting that the
estimates on financial abuse and neglect were lower than the international estimate by Yon and colleagues (2017)
and this could the subject of further research.
Canada
(National NY, USA
UK Ireland Initiative for (Lifespan USA Global
Australia (O’Keeffe (Naughton the Care of the of Greater (Acierno (Yon
(2021) et al., 2007) et al., 2010) Elderly, 2015) Rochester, 2011) et al., 2010) et al., 2017)
(%) (%) (%) (%) (%) (%) (%)
AGGREGATE PREVALENCE (over past year/12 months)
14.8 2.6 2.2 8.2 7.6 11.4 15.7
(4% when (2.9% if (the authors
increased mistreatment noted that lower
to incidents based on rates may be a
involving including result of using
neighbours and neighbours more restricted
acquaintances) and people frequency
and perceived
known to the
seriousness
older person)
threshold criteria
on subtypes)
FINANCIAL ABUSE
2.1 0.7 1.3 2.6 4.2 5.2 6.8
(current
exploitation
by family
members)
PHYSICAL ABUSE
1.8 0.4 0.5 2.2 2.2 1.6 2.6
(including
sexual abuse)
SEXUAL ABUSE
1.0 0.2 0.05 1.6 2.2 0.6 0.9
(including
physical abuse)
PSYCHOLOGICAL ABUSE
11.7 0.4 1.2 2.7 1.6 4.6 11.6
(unadjusted if
one or more:
33%)
NEGLECT
2.9 1.1 0.3 1.2 1.8 5.1 4.2
(unadjusted (potential
if at least one neglect;
incident used: caregiver
5.7%) neglect was
0.5%)
Chapter 5: Prevalence 35
Participants were asked the question ‘In the past 12 months, has someone you know …’ followed by
the following items of financial abuse:
1. Taken your money, possessions or property without your permission
2. Deliberately prevented you from accessing or using your money, possessions or property
7. Not contributed to household expenses such as rent or food, or aged care/home service fees
where this was previously agreed
Financial abuse
In relation to financial abuse, the most common experience involved one financial abuse item, reported by
two‑thirds of the financial abuse cohort, while just under one-third reported experiencing multiple types of
financial abuse (Figure 5.1).
Gender differences were negligible in reports of financial abuse overall: 2% of women compared with 2.1% of men
(see Figure 6.1 and see chapter 6 for detailed discussion).
The most common form of financial abuse was being ‘pressured into giving or loaning money, possessions or
property, (41.5%) (Figure 5.1). Behaviour amounting to theft (taking money or possessions without permission)
was reported by just over a third (34%) of those who reported financial abuse. Almost as common (31.4%)
was failing to provide financial contributions or assistance (rent, food, aged care/home service fees) as
previously agreed. A catchall question – covering other actions causing financial harm – was affirmed by
28.1% of the sample.
Less common forms of financial abuse were ‘pressure to make or change a will’ (9.5%) and ‘making financial
decisions without permission’ (6.6%). There was no participant reporting experience of misusing a power of
attorney. This may reflect the limitations of the study as outlined in chapter 4; that is, the sample focused on
the older population in the community who had the capacity to undertake an interview and excluded the older
persons who were incapable of undertaking an interview due to physical or health conditions. The latter segment
of the population would be more likely to use their power of attorney but was not covered in the study.
Figure 5.1: Survey of Older People: Participants who were classified as experiencing financial abuse –
proportion reporting each financial abuse item and number of items reported
1 item 67.7
0 25 50 75
Percentage
Notes: Weighted data. Unweighted sample size, n = 158. Item ‘misused a power of attorney’ is not shown because
no participant reported experiencing it.
Chapter 5: Prevalence 37
The questions asked to assess physical abuse were preceded by ‘In the past 12 months, has someone
you know …’ followed by the following items:
1. Tied or held you down, or restrained you in any other way, or locked you up
4. Threatened you with a weapon (e.g. knife, gun or any other objects)
5. Injured you with a weapon (e.g. knife, gun or any other objects)
7. Given you drugs or too much medicine in order to control you/make you docile
Physical abuse
Like financial abuse, one type of physical abuse was most commonly reported by older persons who experienced
physical abuse (57.1%), with 42.9% reporting multiple types (Figure 5.2).
Men were slightly more likely than women to report physical abuse (2% cf. 1.6%) (see Figure 6.1).
The most common form of physical abuse reported was threats to harm (60.7%). Almost half (46.6%) of those
who reported physical abuse reported being grabbed, pushed or shoved. Just over one in five (22.3%) reported
being hit, punched, kicked or slapped. Threats with a weapon were reported by 15.3%. Any other physical hurt
was the next most common (14.3%) form of physical abuse.
Figure 5.2: Survey of Older People: Participants who were classified as experiencing physical abuse –
proportion reporting each physical abuse item and number of items reported
1 item 57.1
0 25 50 75
Percentage
Notes: Weighted data. Unweighted sample size, n = 127.
Chapter 5: Prevalence 39
Participants were asked ‘In the past 12 months, has someone you know …’ followed by the following items:
1. Touched you in a sexual way against your will
Sexual abuse
Among the 1% of the sample that reported sexual abuse, one type of sexual abuse was most commonly reported
(65%), and about one-third experienced multiple sexually abusive behaviours (Figure 5.3).
Women (1.2%) were more likely to report sexual abuse than men (0.7%) (see Figure 6.1).
The most frequently reported form of sexual abuse was being spoken to in a sexual way when this was
unwelcome (76.5%) (Figure 5.3). Nearly a third of the sexual abuse reported consisted of being touched in a
sexual way against the participant’s will (32.2%). Just over a third of sexual abuse experiences involved some
‘other unwanted sexual experience/s’. Forced sexual acts accounted for 15.9% of sexual abuse experiences.
No participants reported being made to watch pornography against their will.
Figure 5.3: Survey of Older People: Participants who were classified as experiencing sexual abuse – proportion
reporting each sexual abuse item and number of items reported
1 item 65.3
0 25 50 75 100
Percentage
Notes: Weighted data. Unweighted sample size, n = 87. Item ‘Made you watch pornography against your will’ is not shown
because no participant reported experiencing it.
Chapter 5: Prevalence 41
This approach is similar to the scoring for neglect (see below). First, for each item (question about
specific acts), a score was assigned: 0 if never experienced, 1 if experienced once, 3 if experienced a
few times, 7 if experienced many times, and 10 if experienced every day or almost every day. Then, the
sum of the scores across all the items is the overall score of psychological abuse. A higher score reflects
experiences of a higher level of psychological abuse overall. The theoretical score range is 0–70, with a
higher score indicating a higher level of experience of psychological abusive behaviours.
Experiences that are scored at three or above are counted as psychological abuse. The overall prevalence
rate is distributed across three severity bands:
y Band 1: scores between 3 and 6: experiences of an act of psychological abuse ‘a few times’ in the
last 12 months (low)
The experience of psychological abuse was assessed with reference to seven response options to the
following question: ‘In the past 12 months, has someone you know …’
1. Insulted you, called you names or swore at you in a way you found offensive or aggressive
4. Threatened to harm others that you care about (e.g. pets, children, relatives, friends)
6. Prevented you from seeing or contacting family members (e.g. grandchildren) or your doctor/nurse
For each item, the participant was asked: a) whether they experienced it in the last 12 months, b) if yes,
how often it happened to them with the response options: Once; Few times; Many times; Every day or
almost every day. Participants who reported at least one incident of psychologically abusive behaviour
were asked who did this/these to them.
42 National Elder Abuse Prevalence Study: Final Report
Psychological abuse
For the 11.7% of the sample who reported psychological abuse, the distribution was concentrated in the low and
high score ranges. Nearly half of older persons who were identified as having experienced psychological abuse
in the previous 12 months were in the low score band, and 37% were in the high score band, while 14% were in
the medium score range (Table 5.3). Table 5.3 also shows what these bands mean for the prevalence estimates
of psychological abuse. Specifically, with 4.3% in the high score band and 1.7% in the medium score band, 6% of
participants experienced psychological abuse in the medium and high score bands. Another 5.7% were in the low
score band (i.e. low frequency).
Women were more likely (12.6%) to report psychological abuse than men (10.7%) (see data shown in Figure 6.1
and accompanying explanation about significance). The distribution of score ranges for women and men was
similar, with a concentration of scores in the low and high ranges for both women and men.
Table 5.3: Survey of Older People: Prevalence (%) of psychological abuse in the previous 12 months by
psychological score band
As % of the participants
who experienced Prevalence (as % of the
Psychological abuse psychological abuse full sample) 95% Confidence interval
Band 1 (low) 48.8 5.7 5.1–6.4
Band 2 (medium) 14.2 1.7 1.3–2.1
Band 3 (high) 37.1 4.3 3.8–5.0
Total (all bands) 100.0 11.7 10.8–12.6
Number of participants (base) 843 7,000
Note: Participants who did not respond to relevant questions were included in the total in deriving prevalence estimates
(1.7% across subtypes).
In relation to the psychological abuse items, analysis of individual items is based on the medium and high groups
being considered together because of the small sample size in the medium range. In relation to the individual
abuse items, the findings for the types of psychological abuse reported (Figure 5.4) are as follows:
y Half (49.0%) the sample reported being insulted, called names or sworn at in a way they found offensive or
aggressive. This behaviour was reported by 34.6% of the sample in the low range and 62.7% of the sample in
the medium/high range.
y Just under half the sample (46.4%) reported being excluded or repeatedly ignored. This behaviour was
reported by 26.7% in the low range and 65.2% in the medium/high range.
y Behaviour that was undermining or belittling was reported by the same proportion (46.4%). This was reported
by 33.1% of the low band and 59.1% of the medium/high band.
y Actions that involved preventing the participant from seeing or contacting family members or medical
professionals were reported by 14.5% of the participants who reported psychological abuse. This was much
more common in the medium/high band (25%) than in the low band (3.5%).
y Threats to harm people that the participants cared about accounted for 10.2% of the psychological abuse
types, again with a concentration in the medium/high band (15.5%) rather than the low band (2.1%).
y A catchall question concerning any other behaviours causing emotional distress was reported by 45.5% of
the psychological abuse subsample. These participants fell into the low band in 37.4% of cases and into the
medium/high band in 53.1% of cases.
Chapter 5: Prevalence 43
Figure 5.4: Survey of Older People: Participants who were classified as experiencing psychological abuse –
proportion reporting each psychological abuse item, by psychological abuse score (band)
26.7
Excluded you or repeatedly
65.2
ignored you
46.4
33.1
Undermined or belittled 59.1
what you do
46.4
3.5
Prevented you from seeing or
contacting family members 25.0
or your doctor/nurse 14.5
4.7
Threatened to harm others
15.5
that you care about
10.2
2.1
Threatened to harm themselves 6.5
if you don’t do what they ask
4.4
37.4
Done anything else to cause
you emotional distress 53.1
45.5
0 10 20 30 40 50 60 70
Percentage
Band
Band 1 (mild) Bands 2 and 3 (moderate–severe) Total (all bands)
Notes: Weighted data for the statistics and unweighted sample sizes (in order of legend, n = 425, 418, 843).
44 National Elder Abuse Prevalence Study: Final Report
In the SOP, participants were asked whether they needed assistance for each of 10 Activities of Daily
Living (ADL) and Instrumental Activities of Daily Living (IADL)4 in the past 12 months:
1. Preparing meals
4. Travel or transport
5. Personal care such as washing or bathing (including getting in and out of the bath or shower),
dressing or undressing
For each item, the survey measured how often help was needed, and the participant’s relationship to the
person responsible for providing that help:
y Whether participant required help with the activity: 1 None of the time; 2 Rarely; 3 Some of the time; 4
All of the time.
y [If any help required, including ‘rarely’] Who provided help with this activity?5
y For each person nominated who provided help: whether participant always got the help when they
needed in the last 12 months? If not, how many times participant did not receive help from this
person? (Once; A few times; Many times; Every day or almost every day).
4 The ADLs refer to basic physical and self-care needs whereas IADLs include more complex activities that relate to a person’s ability to
live independently in the community (Lawton & Brody, 1969). Items 5–7 and 9 are classed as ADLs with items 1–4 and 8 classed as IADLs.
5 Participant could nominate two people.
Chapter 5: Prevalence 45
The prevalence assessment of neglect applied in this study is based on a score that takes into account
the level of need for support and the extent to which these needs were not met by a defined carer or
carers. The scoring is intended to assess the overall experience of an older person with support needs
that should be met in order to function every day: it considers the experience of neglect across all the
activities associated with support needs. Essentially, the neglect score range for a single activity is on a
0–10 scale, and the sum of neglect score would theoretically range from 0 to 100 (see Box 5.1 for how
neglect score is scored).
This approach results in the allocation of a score between 0 and 100. Across the possible range of scores,
a score above three is counted as neglect. For experiences that are counted as neglect, there are three
severity bands:
y Band 1: scores between 3 and 6 (low)
Neglect
Of participants who were considered as having experienced neglect in the previous 12 months, the majority
(60%) were in the low score range, just under a quarter were in the medium score range, while 16% were in the
high range (Table 5.4). Table 5.4 also shows these bands in the context of prevalence, with 1.2% being in the
medium and high score bands (i.e. higher level of frequency of not having received assistance needed with
specific daily activity/activities) and 1.7% in the low score band.
More women than men reported experiences of neglect (3.5% cf. 2.2%) (see Figure 6.1).6 The distribution of
women’s and men’s experiences across the score range were different. More women than men were in the low
(2.2% cf. 1.3%) and medium band (1% cf. 0.3%).
Table 5.4: Survey of Older People: Experience of neglect in the previous 12 months by band (frequency)
Prevalence
As % of the participants (as percentage of the
Neglect who experienced neglect full sample) % 95% Confidence interval
Band 1 (low) 60.4 1.7 1.4–2.1
Band 2 (medium) 23.7 0.7 0.5–1.0
Band 3 (high) 15.9 0.5 0.3–0.7
Total (all bands) 100.0 2.9 2.4–3.4
Number of participants (base) 208 7,000
Notes: Weighted data for the statistics and unweighted sample sizes. Proportion of partial or all missing responses for
neglect 1.8%.
In relation to neglect, the analysis of individual neglect questions was based on the low band, with the medium/
high range being combined due to small sample sizes in the medium range (Figure 5.5). The most common form
of neglect involved a failure to do routine housework (79.8% of the neglect group), with limited differences in
occurrence between the low band (77.8%) and the medium/high band (83%). The next most common form of
omission was a failure to assist with transport. This was reported by 69% of the neglect group, including 66%
in the low band and 74% in the medium/high band. A failure to assist with ‘any other day-to-day activity’ was
reported by 58% of the neglect group, spanning 59% in the low band and 56% in the medium/high band.
Figure 5.5: Survey of Older People: Participants who were classified as experiencing neglect – proportion
reporting each neglect item, by neglect score (band)
77.8
Doing routine housework 82.9
79.8
66.3
Travel or transport 74.1
69.4
57.2
Shopping for groceries or clothes 55.5
56.6
50.6
Preparing meals 54.8
52.3
27.6
Taking the right amount of
medicine at the right times 20.9
25.0
17.4
Getting in and out of bed 19.2
18.1
18.8
Eating, including cutting up food 13.8
16.9
14.9
Getting to and using the toilet 18.0
16.1
59.2
Any other day-to-day activity 55.9
57.9
0 20 40 60 80 100
Percentage
Band
Band 1 (mild) Bands 2 and 3 (moderate–severe) Total (all bands)
Note: Weighted data for the statistics and unweighted sample sizes (in order of legend, n = 138, 70, 208).
Table 5.5: Survey of Older People: Number of types of elder abuse in the previous 12 months, persons who
experienced any form of abuse
Note: Weighted data for the statistics and unweighted sample sizes.
Of participants who experienced one form of abuse, psychological abuse was the most common single form of
abuse (74%, or 56% of those who experienced any form of abuse), followed by neglect (14%, or 11% of those who
experienced any form of abuse). For the participants who experienced two or more subtypes of abuse, the most
commonly co-occurring types were psychological abuse with:
y neglect (23% of those who experienced multiple subtypes)
y physical abuse (20.9% of those who experienced multiple subtypes)
y financial abuse (20.5% of those who experienced multiple subtypes).
Table 5.6: Survey of Older People: Co-occurrence of abuse in the previous 12 months for participants who
reported two or more abuse types and participants who reported one type of abuse only
As % of participants As % of participants
who experienced who experienced any
Abuse types 2+ types form of abuse
Psychological and Neglect 23.3 5.6
Psychological and Physical 20.9 5.0
Psychological and Financial 20.5 4.9
Psychological and Sexual 7.6 1.8
Psychological, Physical and Financial 6.1 1.5
Psychological, Physical and Sexual 4.6 1.1
Psychological, Physical and Neglect 3.6 0.9
Psychological, Financial and Neglect 2.6 0.6
Psychological, Sexual and Neglect 1.9 0.5
Financial and Neglect 1.5 0.4
Psychological, Physical, Sexual, Financial and 1.5 0.4
Neglect
Psychological, Physical, Sexual and Financial 1.4 0.3
Sexual and Neglect 1.0 0.2
Psychological, Sexual and Financial 0.9 0.2
Physical and Neglect 0.7 0.2
Sexual and Financial 0.5 0.1
Physical and Sexual 0.4 0.1
Physical and Financial 0.4 0.1
Physical, Sexual and Neglect 0.3 0.1
Psychological, Physical, Financial and Neglect 0.3 0.1
Psychological, Physical, Sexual and Neglect 0.2 0.0
Chapter 5: Prevalence 49
As % of participants As % of participants
who experienced who experienced any
Abuse types 2+ types form of abuse
One form of abuse
Psychological .. 55.9
Financial .. 5.5
Neglect .. 10.5
Physical .. 2.3
Sexual .. 1.8
Total 100.0 100.0
Number of participants (base) 271 1,081
Note: Weighted data for the statistics and unweighted sample sizes.
Overall, the findings presented in this section indicate that elder abuse prevalence among CALD subgroups
is largely consistent with findings for the general population of those aged 65 and over. In total, 15.3% of the
CALD participants reported experiencing elder abuse, including abuse relating to language and culture. Among
CALD participants, 14% reported experiencing the five core subtypes and 4% reported abuse relating to language
and culture.
In relation to psychological abuse, most of the CALD subsample who reported psychological abuse reported
experiences in the low score range (6.9% of the CALD sample), with only 1.4% of the sample reporting
experiences in the medium score range. The high score range was reported by 3.5% (data not shown). This
concentration of participants in either the low score range or the high score range is consistent with the pattern
in the non‑CALD sample.
Table 5.7: Survey of Older People: Prevalence (%) of elder abuse in previous 12 months experienced by
CALD participants compared with non-CALD sample
Notes: Weighted data for the statistics and unweighted sample sizes. None of the difference in the prevalence between
the two groups is statistically significant.
50 National Elder Abuse Prevalence Study: Final Report
The development of the elder abuse questions relating to people from CALD backgrounds in the SOP
was guided by a desktop literature review as well as advice and feedback from the National Ageing
Research Institute (NARI).
Participants who spoke a language other than English at home were asked whether they had experienced
any of the following specific behaviours by someone they know in the last 12 months:
1. Not respected you when talking to you because of your culture, race or ethnicity.
2. Mistranslated between English and your preferred language on purpose, when talking about financial
or legal documents.
4. Made you feel that you are just free labour (e.g. doing most of the household work or providing
child care, etc.).
5. Limited or restricted your contact with your friends or others from the same cultural background,
including friends and relatives overseas.
6. Limited or restricted your access to culturally familiar activities (e.g. attending certain events,
watching or listening to shows in your preferred languages).
For each item, if the participant reported that it had happened to them (in the previous 12 months), they
were then asked how many times it had happened to them.
The response options were: Once; Few times; Many times; Every day or almost every day.
Following all the items, participants were asked about the person(s) who carried out those behaviour(s). If
more than one person was selected, participants were asked to select the one that affected them the most.
Table 5.8: Survey of Older People CALD subsample: Prevalence in previous 12 months of abuse relating to
culture by frequency score band
As % of the
participants who
experienced abuse
relating to their 95% Confidence
background Prevalence (%) interval
CALD: Abuse relating to their background 4.0 2.5–6.2
Band 1 (low) 57.6 2.3 1.3–4.0
Band 2 (medium) 23.1 0.9 0.3–2.6
Band 3 (high) 19.3 0.8 0.3–2.2
Number of participants (base) 27 608
Note: Weighted data for the statistics and unweighted sample sizes.
Chapter 5: Prevalence 51
The most common form of abuse relating to language and culture was ‘not respecting you when talking to you
because of your culture, race or ethnicity’ (67%) (data not shown, results based on small sample sizes and should
be treated with caution). The next most common subtype was being ‘made to feel like you are just free labour’,
reported by 45% of the language and culture abuse group.7
In relation to the other subtypes, the findings are (data not shown):
y 21% reported limitations or restrictions on access to culturally familiar activities
y 21% reported limitations or restrictions on contact with friends and relatives (including those overseas)
from the same culture
y 7% reported denial of access to important information in a preferred language
y 2% reported mistranslation between English and a preferred language in connection with financial or
legal documents.
Consistent with the co-occurrence of different types for the non-CALD sample, among the CALD participants
who reported experiencing more than one type of abuse, the most commonly co-occurring subtypes involved
psychological abuse, with close to one-half of co-occurrences entailing the following:
y psychological abuse and abuse relating to language and culture
y psychological abuse, abuse relating to language and culture, and neglect
y psychological and physical abuse
y psychological, physical, and sexual abuse and abuse relating to language and culture (data not shown in
any table).
The questions covered the five subtypes of abuse but psychological abuse was referred to as emotional abuse
(e.g. intimidating, threatening them, preventing access to family and others). The questions were framed
differently to the SOP as the SOP involved questions about the experience of specific behaviours constituting
abuse, whereas the SGC asked about the experience of each subtype of abuse. The responses to relevant
questions in the SGC and SOP were from different perspectives (i.e. a concerned community member for the
former and the direct victim for the latter) and are not intended to be used for direct comparisons.
The data presented in Figure 5.6 demonstrates that 17.4% of the SGC sample indicated they held concerns about
elder abuse. This finding should be seen in the context of the discussion in chapter 11, which establishes that
people who provide care and support to older people are more likely to become aware of elder abuse.
The most common concerns related to emotional abuse (10.5%) and financial abuse (10.1%), followed by neglect
(7.4%). Physical abuse was a concern for 3.6% and a very small proportion (0.3%) had concerns about sexual abuse.
Of the people who had concerns, around half of those who had any concerns identified one type of abuse (or
8.4% of the whole sample, Figure 5.7) and half had concerns about two or more types of abuse (8.9% of the
whole sample). Concerns about one type of abuse were almost as common (8.4%) as co-occurring types of
abuse (8.9%).
7 Sample sizes were too small to sustain analysis of specific behaviours using the score band relating to language and culture.
52 National Elder Abuse Prevalence Study: Final Report
Figure 5.6: Survey of the General Community: Concerns about older family members and friends
0 5 10 15 20
Percentage
Figure 5.7: Survey of the General Community: Number of concerns reported about older family members
and friends
8.9%
8.4%
No concerns
1 concern
2 or more concerns
82.6%
Notes: Weighted data. Unweighted sample size, n = 3,400. ‘No concerns’ includes a small number of participants who did not
state (i.e. don’t know or refusal).
Of participants who reported multiple concerns, the most commonly co-occurring concerns involved financial
and psychological abuse:
y psychological and financial (26% of those with multiple concerns)
y psychological, financial and neglect (18%)
y psychological, physical, financial and neglect (12%).
These patterns of co-occurrence (Table 5.9) differ somewhat from the co-occurrence patterns reported by older
people themselves, supporting the point that these reports are likely to reflect higher experiences of abuse that
become discernible to people other than the victim and perpetrator. Here, co-occurrence of concerns refers to
reports of having multiple concerns that may not take place at the same time or to the same older persons.
Chapter 5: Prevalence 53
Table 5.9: Survey of the General Community: Co-occurring concerns about older family members and friends
Summary
This chapter has presented findings from the SOP on the prevalence of elder abuse covering the time frame
of the preceding 12 months. Overall, the prevalence rate of elder abuse reported in this chapter is 14.8%. The
most common form of abuse is psychological abuse (11.7%), across three bands (low: 5.7%, medium: 1.7% and
high: 4.3%). Neglect is the next most common abuse subtype at 2.9% (1.7% low, 0.7% medium and 0.5% high).
For the other subtypes, prevalence rates are 2.1% for financial abuse, 1.8% for physical abuse and 1% for sexual abuse.
At 15.3%, prevalence rates for the CALD subsample do not differ greatly from the overall sample. Abuse relating
to language and culture is reported by 4% of the CALD subsample.
Experiences of multiple types of abuse were reported by a minority of the sample (3.5%) (data not shown), with
the most common abuse combinations being psychological abuse and neglect.
Overall prevalence rates were higher for women than men; the distribution of different subtypes of abuse showed
some variation according to gender. Women were more likely than men to experience psychological abuse,
sexual abuse and neglect. These findings, and patterns of significance, are examined in more detail in chapter 6.
In the SGC, 17.4% of the sample indicated they had concerns that a person aged over 65 who they know is
experiencing elder abuse. The subtypes most commonly referred to were emotional abuse and financial abuse.
Aspects of these findings suggest that some subtypes of elder abuse – such as financial abuse – may be more
obvious than others – such as sexual abuse. Further, given the rate of SGC participants reporting concerns about
multiple types of abuse (more than half of those who had concerns), it is likely that higher level cases of abuse
come to the attention of people other than those directly involved.
54 National Elder Abuse Prevalence Study: Final Report
Key messages
f People who were living with a partner were less likely to experience abuse.
f People in lower SEIFA brackets and people who own a home with debt or are in rental housing
were more likely to experience financial, sexual and psychological abuse.
f Older people with poorer health or a disability were more likely to experience elder abuse.
f Older people with poor mental health had a higher likelihood of experiencing elder abuse.
f Less frequent contact with family members and friends and a lower sense of social support were
associated with an elevated likelihood of experiencing elder abuse.
Introduction
This chapter sets out findings on the socio-demographic and economic characteristics of people who reported
experiencing abuse. It also examines whether health status and social connection are different for those who
experience abuse. The discussion in this chapter relates to the following aspects of the research aims:
y identify the characteristics and contexts of elder abuse, including in relation to specific subtypes of abuse8
y identify risk and protective factors associated with the experience of abuse.
The analysis examines the extent to which the experience of elder abuse (any form as well as each subtype)
varied according to the following socio-demographic characteristics: gender, age, Indigenous status, country of
birth, religion, whether the participant lives with their children (of any age), household composition, education,
employment, household income, socio‑economic index for areas (SEIFA), whether the participant lives in a
major city, inner regional area or in outer regions, remote, very remote areas, and whether they have children
(biological, adopted or step). Both bivariate analysis and multivariate analysis were carried out. Logistic
regression was applied to each abuse subtype and overall (experience of any form of elder abuse). This chapter
also examines the extent to which older people’s health status and social connection was linked with elder abuse.
8 The characteristics of the participants in the SGC who had concerns about abuse are presented in the appendix and not discussed in
the chapter for the sake of simplicity.
Chapter 6: Who experiences elder abuse? 55
The chapter commences with an overview of the reports of elder abuse of any type and overall by gender and
age. This is followed by findings for other demographic correlates with each abuse type. In order to simplify the
reporting, the following discussion of the results focuses on those socio-demographic characteristics that emerged
as significant in the regression analysis (except for age and gender). Findings in relation to the CALD subsample,
overall and by gender, are set out after the findings for the sample overall. Finally, the chapter presents the
findings in relation to people’s health and their social connection and their experience of elder abuse.
There were two abuse subtypes where women were more likely to experience abuse to a small but statistically
significant extent: psychological abuse (10.7% cf. 12.6%); and neglect (3.5% cf. 2.2%). Reports of financial abuse
were similar between men and women. Women were more likely to report sexual abuse (1.2% cf. 0.7%) and men
were more likely to report physical abuse but these differences were not significant (2% cf. 1.6%).
20
Gender
Men Women
15.9
15
13.6
12.6
Percentage
10.7
10
5
3.5
2.1 2.0 2.2 2.0
1.6
1.2
0.7
0
Any form Financial Neglect* Physical Psychological Sexual
of abuse* abuse abuse abuse* abuse
Notes: Weighted data for the statistics and unweighted sample sizes: men n = 2,716, women n = 4,146. The asterisks indicate
that the difference between men and women is statistically significant at a 5% significance level.
Age
Figure 6.2 presents analysis of the experience of elder abuse by age group. The overall prevalence of elder abuse
(at least one form of abuse) varied according to age, with the rate of elder abuse declining with increasing age –
from 20.3% for the youngest group (65–69 years) to 9.4% for the older age group (85+ years). This age-related
pattern was also evident for psychological abuse: the proportion of older people reporting this form of abuse fell
from 16.8% for the age group of 65–69 years to 4.7% for the age group of 85 years and over. Age-related patterns
for physical and sexual abuse were less pronounced but still significant. The prevalence of these two subtypes
was higher among the age group of 65–69 years than the older age groups. However, financial abuse was not
significantly related to age.
The patterns described may arise from the under-representation of people with cognitive decline and other
conditions, such as frailty, in the sample (see further below), and this issue is likely more marked for older age
groups given general decline in health with increasing age. See further below for more discussions on this issue.
Age-related patterns were different for neglect. Rather than decreasing with age, the prevalence of neglect
is u-shaped, falling from 4.1% in the age range 65–69 years to 1.8% in the age range of 75–79 years and then
increasing to 2.9 for those in the top age range. These age-related patterns generally applied to women, with
men showing similar patterns in raw terms that were not statistically significant (Figure 6.3). These findings are
particularly likely to reflect the limitations of the sample and should be regarded with caution.
56 National Elder Abuse Prevalence Study: Final Report
14.9
15
Percentage
11.5 11.1
10 9.4
5 4.1
2.7 2.3 2.7 2.9
1.8 1.9 1.7 1.8 1.8
0
12.1
10 9.0 8.2
5 4.7
2.8
1.7 1.1 1.4 1.5 1.4
0.5 0.3 0.4 0.4
0
Age (years)
65–69 70–74 75–79 80–84 85+
Notes: Weighted data for the statistics and unweighted sample sizes: in order of age n = 1,697, 1,927, 1,525, 1,105, 946.
The asterisks indicate that the differences across the age groups are statistically significant based on design‑based
F test (**p < .01; ***p < .001).
MEN
Any form of abuse** Financial abuse Neglect Physical abuse Psychological abuse***
20 18.7
15.4
15
Percentage
12.1 11.8
10.6 11.1 10.3
10 8.2 8.0
6.1
5
3.1 2.9 2.7 3.0 3.0
1.4 1.9 1.5 2.3 1.2 1.6 1.6 1.6 1.9
0.5
0
WOMEN
Any form of abuse*** Financial abuse Neglect** Physical abuse** Psychological abuse***
21.7
20
17.9
17.3
15 13.7
Percentage
12.4
11.1
9.9
10 8.5
7.3
5.1
5 3.2 3.4
2.9
2.3 2.2 1.9 1.9 1.3 2.0 2.7 2.7
1.7
0.7 1.0 0.6
0
Age (years)
65–69 70–74 75–79 80–84 85+
Notes: Weighted data for the statistics and unweighted sample sizes: men, in order of age n = 725, 763, 589, 404, 266;
women, in order of age n = 959, 1,161, 934, 699, 478. The asterisks indicate that the differences across the age groups
are statistically significant based on design-based F test (**p < .01; ***p < .001).
Chapter 6: Who experiences elder abuse? 57
These findings provide an assessment of the important socio-demographic characteristics linked with elder
abuse. The main points are:
y Compared to men, women were at a higher risk of experiencing any form of abuse, sexual abuse,11
psychological abuse and neglect. The patterns were consistent with the discussion above on the bivariate
analysis results.
y The likelihood of experiencing physical, sexual and psychological abuse falls with increasing age. This pattern
also applied to the overall experience of elder abuse (any form).
y A higher level of education was associated with a higher likelihood of experiencing sexual and psychological
abuse but not the three other subtypes. This association suggests that older people with a higher level of
education were more likely to identify behaviours relating to these two subtypes. This is consistent with
findings on attitudes towards elder abuse in chapter 12 that demonstrate that a higher level of education
was associated with less condoning attitudes towards elder abuse and a greater tendency to identify elder
abuse behaviours. In combination, these findings suggest that the correlation between higher education levels
and the greater prevalence of sexual and psychological abuse are related to greater recognition of certain
behaviours as abuse among this group.
y Being divorced is associated with a higher likelihood of experiencing abuse overall (any form) and all
subtypes. Those who are widowed and never married are also at an elevated risk of experiencing some
subtypes (sexual and physical for never married people). The bivariate analysis shows that participants who
were divorced or separated were twice as likely as married participants to report experiencing at least one
form of abuse (26.4% cf. 13.1%).
y Having a step-child is associated with a higher likelihood of experiencing abuse overall and physical and
psychological abuse in particular. This finding does not necessarily imply that step-children perpetrate abuse
but it is consistent with complex family dynamics and histories of conflictual relationships being a risk factor
for elder abuse (chapter 2).
y Socio‑economic status is associated with abuse overall, and some subtypes of elder abuse. Specifically:
– Low socio‑economic status (as indicated by the SEIFA relative disadvantage index) is associated with
a higher likelihood of experiencing abuse overall and the subtypes of financial abuse, sexual abuse and
psychological abuse.
– Similarly, owning a home with debt and renting accommodation (including public housing) are associated
with a higher likelihood of experiencing abuse overall and the subtypes of physical abuse, psychological
abuse, and neglect.
9 It is worth noting that sexual abuse involved a small number of occurrences. Although logistic regression would not normally be
applied to such small sample sizes, research by Vittinghoff and McCulloch (2007) concluded that logistic regression can be applied
where the number of events per predictor variable is fewer than 10 events per predictor variable; in particular, with adequate control
of confounding. The ratio of events per predictor variable in this analysis is within the acceptable range of 5–9.
10 Some variables that were presented in the bivariate analysis are excluded from the regression analysis due to either high levels of
missing data (e.g. household income) or small numbers (e.g. Indigenous status). The employment variable was excluded.
11 The gendered difference in the prevalence of sexual abuse was not statistically significant. The pattern was consistent with the
Canadian study (National Initiative for the Care of the Elderly, Canada, 2015).
58 National Elder Abuse Prevalence Study: Final Report
Table 6.1: Survey of Older People: Socio-demographic characteristics associated with higher likelihood of
elder abuse a
Second, the findings may reflect age-related differences in attitudes, awareness and recognition. The findings on
attitudes towards elder abuse outlined in chapter 12 indicated that older people in the top age ranges were more
likely to condone elder abuse and less likely to recognise elder abuse behaviours compared to older people in the
lower age ranges. Some international research is consistent with this. Using data from the New York Mistreatment
Prevalence Study conducted in 2009, Burnes and colleagues (2019a) found that the older emotional abuse
victims were, the more likely they were to downplay the experience as serious for them. This suggests that any
under-reporting of elder abuse may be disproportionally higher for the oldest age group.
Finally, with increasing age, general health status declines (AIHW, 2018), and the need for support increases. This
dependency on family members and others may have discouraged the disclosure of their experience of elder
abuse in the survey.
Chapter 6: Who experiences elder abuse? 59
Table 6.2: Survey of Older People: Proportion of participants who report experience of any form of elder abuse
by socio-demographic characteristics, males, females and overall
Notes: Weighted data for the statistics and unweighted sample sizes. The asterisks indicate that the differences in the
prevalence across categories in the variable are statistically significant based on design-based F test (*p < .05;
**p < .01; ***p <.001)
Although this research cannot shed light on the reasons why participants had no one to help, describing the
extent of this unmet need and the characteristics of this group can provide important insights into the needs of
older Australians. Further research would be required to understand why the needs of this group were unmet.
From the survey data, it is not possible to draw conclusions as to the extent that systemic (e.g. gaps in service
availability) or personal issues (such as a lack of knowledge or assistance in accessing support) are relevant.
Type of activity
Participants in the unmet needs subgroup were most likely to report an unmet need for help with doing routine
housework (1.3% of all participants), followed by taking the right amount of medicine at the right times (1.1% of
all participants) and travel or transport (1% of all participants). More generally, 1.6% of participants indicated that
there was no-one to help with ‘any other day-to-day activity’.
Participants aged 80 years and over were more likely to report needing help with at least one activity and having
no-one to help as compared with participants aged 65–79 years.
Of participants aged 85 years, 11.3% reported needing help with at least one activity and having no-one to help
(one activity: 9.1%, two or more activities: 2.2%).
Chapter 6: Who experiences elder abuse? 61
As discussed in chapter 2, Storey’s (2020) recent synthesis of evidence on the risk factors for experiencing elder
abuse identifies problems with physical and mental health as predisposing factors.
Psychological health
Psychological health was assessed by applying the Kessler 6 Psychological Distress Scale. This scale includes six
questions about anxiety and depressive symptoms experienced during the past four weeks. The scale has a score
range of 0–30, with a higher score indicting a higher level of distress. This scale screens for general mental health
in an adult population (Kessler et al., 2003). Based on the scores, participants were classified into two groups: no
probable serious mental illness and probable serious mental illness (Kessler et al., 2010).
Even where elder abuse experiences were reported, most participants (91%) did not fall into the probable serious
mental illness score range. However, it is notable that participants who did experience elder abuse were more
than three times as likely to fall into this score range compared with those who did not (9% cf. 2%).
Consistent with this, the mean scores depicted in Table 6.3, show elevated levels of psychological distress for
participants who experienced elder abuse, across all abuse types.
Mean scores for those who had experienced elder abuse were 11.8 compared with 8.9 for those who did not
experience abuse.
Across all of the five subtypes of elder abuse, mean scores for those who had experienced a specific subtype of
abuse were higher compared to those who did not experience the specific subtype (mean scores: 12.7–13.6 cf.
9.0–9.3), and the proportions of scores falling into probable serious mental illness were higher (14–17% cf. 2–3%).
In other words, older people who experienced elder abuse were significantly more likely to also experience poor
mental health.
Table 6.3: Survey of Older People: Mean scores of psychological distress (Kessler 6 scale) by whether
experiencing elder abuse
Kessler 6 scale
Scores in the range
of probable serious
mental illness (19–30) Number of participants
Abuse type Mean score (%) (base)
Financial
No financial abuse 9.3 3.0 6,830
Financial abuse 12.7*** 16.5 157
Physical
No physical abuse 9.3 3 6,861
Physical abuse 12.8*** 14.5 126
Sexual
No sexual abuse 9.3 3.1 6,900
Sexual abuse 12.7*** 14.5 87
Psychological
None 9.0 2.4 6,146
Psychological abuse 11.9*** 9.6 294
62 National Elder Abuse Prevalence Study: Final Report
Kessler 6 scale
Scores in the range
of probable serious
mental illness (19–30) Number of participants
Abuse type Mean score (%) (base)
Neglect
None 9.2 2.9 6,779
Neglect 13.6*** 15.2 35
Any form of elder abuse reported
No 8.9 2.3 5,909
Yes 11.8*** 8.8 1,078
Notes: Weighted statistics and unweighted sample bases. The asterisks indicate that the difference in the mean scores between
the two groups (whether experienced any specific type of abuse) was statistically significant based on bivariate
regression analysis (***p < .001).
There was an apparent association between older people’s health and reports of experiencing elder abuse,
with poorer health being associated with a higher likelihood of experiencing elder abuse. This applied to all the
subtypes and overall abuse. For example, older people who rated their health as fair or poor were nearly twice
as likely as those who considered their health as excellent or very good to report experiencing any form of elder
abuse (20.8% cf. 11.3%). Likewise, older people with disability or long-term medical conditions were twice as likely
as others without such health problems to report experiencing any form of elder abuse (20.6% cf. 9.8%).
Neglect was the abuse subtype most associated with poorer health conditions, reflecting the conditions that
create the need for help involved when neglect occurs. The association between health and experiences of elder
abuse was more marked for neglect than other subtypes. Specifically, the prevalence of neglect was three and
half times as high for older people with a disability or long-term health conditions as for those without such
health problems (4.7% cf. 1.3%). Neglect prevalence was 6.2% for older people who rated their health as fair/poor,
compared to 2.6% for those who rated their health as good and 0.9% for those whose ratings were excellent/very
good. (These patterns are consistent for men and women and results are shown in Appendix A.)
Table 6.4: Survey of Older People: Prevalence of elder abuse, by self-reported general physical health
General health
Excellent/
very good Good Fair/poor
(%) (%) (%)
Financial abuse 1.6 1.7 3.2 **
Physical abuse 1.6 1.6 2.2
Sexual abuse 0.8 0.7 1.5 *
Psychological abuse 9.2 11.8 15.4 ***
Neglect 0.9 2.6 6.2 ***
Any form 11.3 14.3 20.8 ***
Number of participants (base) 3,066 2,104 1,806
Notes: Weighted statistics and unweighted sample bases. The asterisks indicate that the differences in the prevalence of the
specific subtype across the three health groups were statistically significant (*p < .05; **p < .01; ***p < .001).
Chapter 6: Who experiences elder abuse? 63
Table 6.5: Survey of Older People: Experience of elder abuse, by self-reported medical conditions or disability
Notes: Weighted statistics. The asterisks indicate that the difference in the prevalence of a specific subtype between the two
groups of disability status was statistically significant (*p < .05; **p < .01; ***p < .001).
Table 6.6 shows a consistent pattern across all subtypes of abuse. Lower levels of social contact with family
and friends were associated with a higher prevalence of elder abuse. Overall, 13.7% of participants who had
face‑to‑face contact with their family and friends a few times a week or more frequently reported any type of
abuse, compared with 17.8% who saw their friends and family less often than once a week.
Across each of the abuse subtypes, the proportion who reported each abuse type was higher for those who
reported less frequent contact with family and friends compared with those who had more frequent social
contact. Although the results were not significant for some subtypes, the patterns were consistent.
Table 6.6: Survey of Older People: Prevalence of elder abuse, by frequency of seeing family members or
friends living elsewhere and gender
Notes: Weighted statistics and unweighted sample bases. The asterisks indicate that the differences in the prevalence of a
specific subtype across the three groups were statistically significant based on design-based F test (*p < .05).
64 National Elder Abuse Prevalence Study: Final Report
Social support
Social support was assessed with a series of statements concerning participants’ feelings about the availability
of people to provide support in a range of ways (four items), as well as feelings of loneliness. The items formed
a scale with a score range of 0–100, with a higher score indicating a greater sense of support (see Box 6.2 for
how this scale is scored). The responses were analysed by developing scores across three ranges, indicating low,
medium and high levels of social support.
International research literature identifies level of social support as significant to the risk of experiencing elder
abuse, with lower levels of social support associated with a higher likelihood of experiencing elder abuse
(e.g. Wong et al., 2020). However, given that members of family and social networks are prominent among
perpetrators of abuse, it is clear that there is a need to distinguish between positive social support and negative
social support (Zheng, Li, Kong, & Dong, 2019).
The response options include: strongly agree, agree, neither, disagree, strongly disagree.
Two steps are involved in generating scores for the social support scales. First, each item in a scale will be
recoded so that the item score will be 0–100 (strongly agree = 0, agree = 25, neither = 50, disagree = 75
and strongly disagree = 100). Then the mean of recoded item scores forms the scale score. The scores
range from 0–100, with higher scores indicating a greater sense of social support.
Table 6.7 demonstrates a correlation between all abuse subtypes and low social support.
Across all the five subtypes of elder abuse, the mean score of social support was lower for those who reported
experiencing a specific subtype compared to those who did not have such an experience (mean scores: 56.5–63.9
cf. 77.1–78.4). The mean score was also lower for those who experienced any form of abuse compared to those
who did not experience any form of abuse (66 cf. 78.9). In other words, people with a lower sense of social
support were more likely to report experiencing elder abuse and this pattern was consistent across all abuse
subtypes and overall abuse. Notably, the difference in mean scores between those who experienced abuse and
those who did not have such an experience was larger for neglect and physical abuse (20–21) than financial,
sexual and psychological abuse (13–17).
Table 6.7: Survey of Older People: Mean scores of sense of social support by whether experienced elder abuse
Number of participants
Abuse type Mean score (base)
Financial abuse
No financial abuse 77.2 6,757
Financial abuse 63.9*** 155
Physical
No physical abuse 77.3 6,786
Physical abuse 57.4*** 126
Sexual
No sexual abuse 77.1 6,825
Sexual abuse 61.3** 87
Psychological
No psychological abuse 78.4 6,076
Psychological abuse 61.8*** 836
Chapter 6: Who experiences elder abuse? 65
Number of participants
Abuse type Mean score (base)
Neglect
None 77.5 6,706
Neglect 56.5*** 206
Any form of elder abuse reported?
No 78.9 5,840
Yes 66.0 1,072
Notes: Weighted statistics and unweighted sample bases. The asterisks indicate that the difference between the two groups
(whether experienced any specific type of abuse) was statistically significant based on bivariate regression analysis
(*p < .05; **p < .01; ***p < .001).
Regression analysis
Regression analysis was applied to each of the abuse subtypes and overall (any form of abuse) to assess the
strength of associations described above. Tables 6.8 and 6.9 rank the relative importance of 15 variables to the
experience of elder abuse. The first table sets out findings in relation to financial, physical and sexual abuse. Table 6.9
sets out findings in relation to psychological abuse and neglect, as well as the experience of elder abuse overall.
The logistic regression analysis result provides an indication of correlations between a specific explanatory
variable and an abuse type being considered and does not indicate any direction of causality. Relative
importance analysis is an indicator of strength for such associations relative to other explanatory variables
being considered in the modelling and does not refer to any absolute correlation.
For example, the psychological distress variable is correlated with the experience of any form of abuse
(the last set of columns in the table); however, the direction of this correlation cannot be established
from this analysis. This correlation can go in either direction – a higher level of psychological distress can
make the older person more vulnerable and increase the risk of experiencing elder abuse. Conversely, the
experience of abuse can lead to a higher level of psychological distress. The relative importance (as set
out in the table) indicates that this association is stronger than other explanatory variables (rank 1); the
importance indicator is 33.7%, which means that the contribution of this to the overall logistic regression
model is 35% among the 15 sets of variables being considered in the analysis.
The results confirm the associations described above continue to hold, taking into account demographic
characteristics (regression results are in Appendix A, Table A6.7). More importantly, the further analyses
reveal that social support and psychological distress are strong correlates across abuse subtypes and overall
experiences (as shown in Table 6.8), though the pattern is somewhat weaker for sexual abuse.
For financial abuse, the top four correlates (in order) are psychological distress, social support, marital status,
socio‑economic status. For physical abuse, the top four correlates are social support, psychological distress,
marital status and age. For sexual abuse, the top four correlates are marital status, age, psychological distress
and social support.
In relation to psychological abuse, psychological distress, social support, age and disability status are the top
four correlates. For neglect, the top four correlates are social support, psychological distress, disability status and
marital status.
66 National Elder Abuse Prevalence Study: Final Report
Table 6.8: Survey of Older People: Relative importance (%) of explanatory variables in the regression analysis
for financial, physical and sexual abuse types
Notes: aImportance refers to standardised weight (contribution to the fitstat, pseudo R2 of the model estimates). b Social
contact refers to frequency of seeing family members or friends living outside the households. The analysis of relevant
importance of each variable in the regression was based on the module on dominance analysis for Stata.
Table 6.9: Survey of Older People: Relative importance (%) of explanatory variables in the regression analysis
for psychological abuse, neglect and overall elder abuse (any form)
Notes: aImportance refers to standardised weight (contribution to the fitstat, pseudo R2 of the model estimates). b Social
contact refers to frequency of seeing family members or friends living outside the households. The analysis of relevant
importance of each variable in the regression was based on the module on dominance analysis for Stata.
Chapter 6: Who experiences elder abuse? 67
The prevalence rate of elder abuse (any form, excluding abuse relating to language and culture) is lower for
the CALD sample than the non-CALD sample (14% cf. 15%). For the CALD sample, none of the differences in
the prevalence of elder abuse (at least one of the five subtypes) according to the selected characteristics was
significant due to the small sample size. The following discussion focuses on whether specific patterns were
similar to the non-CALD sample.
y The overall prevalence rate was slightly higher for women with a CALD background than their male
counterparts (14.2% cf. 13.8%). The pattern was consistent with the non-CALD sample.
y Similar to the non-CALD sample, reports of elder abuse fall with age among the CALD sample, with 16.3% for
the younger age range of 65–74 and 10% for the older age ranges of 75 years and older.
y Living alone is also associated with a higher likelihood of elder abuse for both CALD and non-CALD samples
compared to living with a partner (CALD: 14.7% cf. 13.7%; non-CALD: 16.7% cf. 13.5%). While living with others is
also associated with an elevated risk of experiencing elder abuse for the non-CALD sample (22.8%), this is not
evident for the CALD sample (14.1%).
y For both the CALD and non-CALD samples, living with children is associated with a higher likelihood of
experiencing elder abuse (CALD: 16.1% cf. 13.5%; non-CALD: 17.6% cf. 14.7%).
Table 6.10: Survey of Older People: Prevalence of elder abuse (any form) by selected characteristics, CALD and
non-CALD participants
Notes: Based on weighted data. Participants who could not be classified regarding experience of abuse due to item
non‑response were included in the total when deriving the prevalence figures. For the CALD and non‑CALD groups
(separately), the differences in the prevalence of elder abuse across categories of a specific characteristic variable
were tested for statistical significance using design-based F test (*p < .05; **p < .01; ***p < .001).
68 National Elder Abuse Prevalence Study: Final Report
Table 6.11: Survey of Older People: Prevalence of abuse relating to language and culture by selected
characteristics, CALD participants
Number of participants
Characteristic CALD abuse (base)
Gender
Male 3.9 272
Female 3.9 332
Age
65–74 5.4 334
75+ 2.1 274
Living arrangement
Couple 3.5 338
Living alone 4.1 188
Living with others 6.5 82
Whether living with children
No 3.9 487
Yes 4.4 119
Summary
The analysis set out in this section provides insight into the socio-demographic characteristics of the
participants in the sample that are associated with an experience of abuse. In addition to gender, some
characteristics are salient across abuse types. These findings indicate elder abuse is tied to age, socio‑economic
status and family form.
Participants in the 65–69 years age group were more vulnerable to abuse compared with those in older age
ranges and this pattern applied to overall elder abuse and all subtypes. However, the likelihood of neglect rises
after age 80. The pattern of lower prevalences of elder abuse associated with older age groups is likely to
reflect the sample for this study, in that older people without capacity to consent and those living in aged care
settings were excluded. Further relevant factors in these patterns are the findings of an association among the
older age groups of more condoning attitudes toward elder abuse and lower levels of recognition of elder abuse
(chapter 12).
It also appears that low socio‑economic status is associated with a greater risk of abuse overall, and especially
financial, sexual and psychological abuse. Owning a home with debt and being in rented accommodation
(including public housing) are associated with a higher likelihood of experiencing psychological abuse.
Marital status is a consistent influence across all abuse types, with those who are either separated or divorced
being more vulnerable. Conversely, people with a partner are less likely to experience abuse.
Although the vast majority of SOP participants have good psychological and physical health, the findings
indicate that elder abuse experiences are associated with poorer health ratings. Although causal direction of
these outcomes and the experience of elder abuse is uncertain, the findings in relation to elevated levels of poor
psychological and physical health and social isolation among those who experienced elder abuse are consistent
with some risk factors identified in Storey’s (2020) recent analysis (chapter 2).
Chapter 6: Who experiences elder abuse? 69
In relation to psychological wellbeing, a higher level of psychological distress was associated with a higher
likelihood of elder abuse. Participants who did experience elder abuse were more likely to fall into the score
range for probable serious mental illness compared with those who did not. The patterns were consistent across
abuse types.
A similar association between lower ratings of physical health and abuse experiences is evident, with ‘poor’
health ratings being associated with a high likelihood of experiencing elder abuse. Older people with disability or
long-term medical conditions were twice as likely as others without such health problems to report experiencing
any form of elder abuse.
Less frequent contact with family members and friends was associated with an elevated likelihood of experiencing
elder abuse. A lower sense of social support was associated with a higher likelihood of experiencing elder abuse.
70 National Elder Abuse Prevalence Study: Final Report
Key messages
f The predominant relationship dynamic in elder abuse is intergenerational and familial, with
children (18%), partners of children (7%) and grandchildren (4%) together accounting for a
significant proportion of all perpetrators.
f Familial perpetrators are particularly evident for financial abuse, with sons being 24% of
perpetrators and daughters being 12%.
f Friends (12%), acquaintances (9%) and neighbours (7%) are also significant among perpetrator
groups, especially for financial, physical and sexual abuse. Collectively, this group is only a little
smaller than the familial intergenerational group.
f The overlap between elder abuse and family violence is underscored with 10% of perpetrators
being intimate partners.
Introduction
This chapter sets out findings on who perpetrates elder abuse. The findings shed light on who the most common
perpetrators of elder abuse are and expands the analysis on the diverse dynamics associated with the different
subtypes of elder abuse. The discussion identifies who among three potential groups of perpetrators are the
most common: family members; friends, acquaintances and neighbours; and professionals who come into
contact with older people.
A limitation of these findings is that professional carers are likely to be under-represented as a potential
perpetrator group given the requirements to participate in the survey, as previously explained in chapter 4.
The discussion in this chapter starts with a profile of who the abuse perpetrators were according to the reports
of the people who experienced elder abuse in the SOP. It then provides detailed insight into perpetrators
according to the five subtypes of abuse, including an examination of the characteristics of perpetrators.
These findings are then set out in relation to the CALD subsample. Finally, findings from the SGC, concerning
perpetrators of abuse where participants had concerns about someone they know being abused, are outlined.
Chapter 7: Who commits elder abuse? 71
Overview of perpetrators
Figure 7.1 sets out findings on the relationship of the person who perpetrated abuse with the participant who
reported experiencing abuse and displays these perpetrator groups as a proportion of all identified perpetrators.12
Figure 7.1: Survey of Older People: Relationship of perpetrators to the participant, by subtypes and overall
(as % of perpetrators)
0 10 20 30 40 0 10 20 30 40 0 10 20 30 40
Percentage Percentage Percentage
Notes: Weighted data and unweighted sample sizes in order of chart, n = 1,382, 168, 265, 132, 915, 89. Multiple perpetrators
could be reported for each type of abuse. Values may not equal 100% due to rounding. a Includes biological/adopted
children.
Where participants experienced abuse from more than one person (a quarter of participants who experienced
abuse, data not shown), they were asked to identify who the perpetrators were, and which one had affected
12 Verbatim responses were back-coded to categories and combined in the following way: Other family member includes: parent,
step‑parent, parent-in-law, aunt/uncle; Professional carer includes: professional carer, person from care/community organisation;
Service provider includes: medical professional, financial professional, other professional, home help worker, person from other agency
or paid service, person from government/council service; Acquaintance includes: colleague/employer/manager, acquaintance, client,
person from sporting/social club; Other includes: other, another person living with you.
72 National Elder Abuse Prevalence Study: Final Report
them the most (labelled as the ‘main perpetrator’). Participants were then asked further questions about the
characteristics of the person identified as the ‘main perpetrator’. (The reports on the ‘main perpetrator’ are in
Appendix A, Table A7.1).
The findings demonstrate that elder abuse has strong familial dynamics, with intergenerational and intimate
partner relationships involved in victim–perpetrator dynamics to a very significant extent.
Intergenerational family relationships account for a substantial proportion of elder abuse experiences. The
largest perpetrator group consists of children (including biological and adopted children), accounting for 18% of
perpetrators. Based on the reports on main perpetrators, sons are slightly more likely than daughters to be
the main perpetrator (9% cf. 7% of perpetrators). Partners of children (sons- and daughters-in-law) are also a
significant perpetrator group at 7%. Intergenerational abuse involving grandchildren also occurs, with grandsons
and granddaughters accounting for 4% of perpetrators.
Intimate partners are also one of the larger perpetrator groups at 10%. Abuse involving siblings and abuse
involving in-laws (3% and 1% respectively) also reinforces the familial dynamic of elder abuse, as does ‘other
family members’ accounting for 5%.
Friends and acquaintances account for a significant proportion of perpetrators at 12% and 9% respectively.
Neighbours are also significant in the non-family, non-professional perpetrator group at 7%.
Abuse perpetrated by professionals, including service providers (6%) and professional carers (3%) is not as
common as abuse perpetrated by family members and those known to an older person. However, this is likely to
reflect the under-representation of people with a need for a higher level of care in this sample.
Perpetrator characteristics
Table 7.1 sheds further light on the dynamics of elder abuse through an analysis of perpetrator characteristics,
with the first column providing overall findings and the subsequent columns providing a breakdown by abuse
type. These data were collected in relation to the ‘main perpetrator’ identified by the participant. The following
aspects of the analysis are noteworthy:
y Men are more likely to commit abuse than women (55% cf. 45%), with widening disparity between men and
women as perpetrators of financial (65% cf. 35%), physical (77% cf. 24%) and sexual (79% cf. 21%) abuse.
y People in the age groups most likely to be committing abuse are 45–54 years (20%) and 65–74 (23%).
y People committing abuse are more likely to be not employed than employed (53% cf. 47%), especially for
physical abuse (75% cf. 25%).
y Overall, perpetrators are unlikely to be living with the older person (23% cf. 77%) but this is especially true
of psychological abuse (18% cf. 82%) and sexual abuse (7% and 93%). In this respect, the pattern varies for
neglect (42% live with cf. 58% do not) and financial abuse (32% and 68%).
Chapter 7: Who commits elder abuse? 73
Table 7.1: Survey of Older People: Demographic characteristics of main perpetrators, overall and by each
subtype of abuse
Notes: Weighted data, unweighted sample sizes. Each characteristic excludes don’t know and refused responses from the
analysis. These proportions may include instances where the main perpetrator is ‘unclear’. Data are not included
separately for the CALD subsample owing to small sample sizes. The analysis does not include abuse relating to
language or culture as data were not collected on perpetrator characteristics for this form of abuse.
Table 7.2: Survey of Older People: problems that main perpetrators had, as reported by participants who
experienced elder abuse
Notes: Weighted data, unweighted sample sizes. Responses that indicated ‘None of these’, don’t know and refused are
excluded from number of problems. These proportions may include instances where the main perpetrator is ‘unclear’.
Problems relating to the categories of ‘Work issues’, ‘Family issues’ and ‘Emotional/personal problems’ were back-coded
from verbatim responses. a This total for all forms of abuse does not include abuse related to language or culture as data
were not collected in relation to this form of abuse.
Overall, the majority of elder abuse victims indicated they were aware that the perpetrator had one or more
problems (72%). The most commonly reported problem was a mental health issue (32%), followed by financial
problems (21%) and physical health problems (20%). The analysis shows that different problems are associated
with different forms of abuse to varied extents, underlining again the complex dynamics involved.
y For financial abuse, physical abuse and psychological abuse: financial, mental health and physical issues
were the most common problems the perpetrators had. Financial problems were more prominent for the
perpetrators of financial abuse (55%), while mental health problems were more marked for the perpetrators of
physical abuse (63%). Mental health problems were the leading problem for the perpetrators of psychological
abuse (38%), and it was similarly common for the perpetrators of financial abuse (37%).
y A different pattern emerged for sexual abuse, with this being the only type of abuse associated with a
high level of problems with alcohol among perpetrators (35% cf. 5–23% of other subtypes). For sexual abuse,
the next most common perpetrator problems were mental health problems and physical health problems
(28–30%).
y For neglect, the most commonly associated problems were physical health problems (25%), followed by
financial problems and mental health problems (11% for each problem). The majority of main perpetrators
had at least one problem reported (76%) at a much higher frequency than perpetrators of other abuse
types. These findings suggest that for some neglect situations, these problems (particularly physical health
problems) create an inability to meet care needs rather than a wilful neglect of care needs.
y Associations between alcohol and drug use and perpetrator problems were evident for financial, physical and
psychological abuse.
Chapter 7: Who commits elder abuse? 75
In order to gain a better understanding of perpetrator dynamics, this section presents the analysis of elder abuse
committed by two perpetrator groups: adult children and friends. Adult children were the largest group among
familial perpetrators, and friends were the largest group among non-familial perpetrators. The following analysis
compares characteristics of older persons who experienced elder abuse committed by these two perpetrator groups,
as well as the perpetrator problems of the two perpetrators groups, for each abuse subtype. Due to small sample
sizes, this analysis excluded sexual abuse. The sample sizes for other perpetrator groups were in general too small.
For the sake of simplicity and clarity in the data, this analysis focuses on adult children and friend perpetrators
who were identified as main perpetrators if a participant reported having experienced an abuse subtype from
more than one perpetrator. In other words, older persons who experienced a subtype of abuse committed by
adult children and/or friends were mutually exclusive groups for the purpose of this analysis.
It is important to note at the outset that although adult children and friend perpetrators were relatively larger
groups, the sample size for each group was still very small and the results should be considered exploratory.
Because of small numbers in the two groups across abuse subtypes, no statistical test was applied. The
discussion below focuses on some general patterns.
Table 7.3 focuses on the selected characteristics of older persons who experienced specific abuse subtypes by
whether the abuse was committed by adult children or friends.
y There were no consistent patterns in terms of the intersection between gender and the two perpetrator
groups across the four abuse subtypes.
– Older people who experienced psychological abuse committed by adult children were more likely to be
older women, as opposed to older people who experienced such abuse committed by friends.
– Financial abuse committed by friends was more likely to be experienced by older men, while financial abuse
by adult children was somewhat more likely to be experienced by older women.
y Older people who experienced financial abuse committed by friends were older than those who experienced
such abuse by adult children and came from non-English speaking backgrounds.
y It appeared that older people who experience psychological, financial and physical abuse subtypes committed
by friends were more likely to be never married, compared to those who experienced such abuse by their
adult children.
y Older people who experienced financial abuse, physical abuse and neglect by friends reported poorer health
compared to those who experienced the same abuse subtypes committed by adult children.
The majority of friend perpetrators of psychological abuse, financial abuse and physical abuse were aged
65 years and older. This pattern suggests that friend perpetrators tended to be the peers of elder abuse victims.
76 National Elder Abuse Prevalence Study: Final Report
Table 7.3: Characteristics of older people who experienced elder abuse, by relationship with perpetrator and
abuse subtype
Notes: Weighted data for the statistics and unweighted sample sizes. a Other English speaking countries include: United
Kingdom, Republic of Ireland, United States of America, Canada, South Africa, New Zealand.
Table 7.4 shows whether the two perpetrator groups had problems, according to elder abuse victims.
y Adult children perpetrators were more likely than friend perpetrators to have at least one problem that the
elder abuse victim was aware of. This pattern was evident across all of the four abuse subtypes.
y The following problems were more common for adult children perpetrators than friend perpetrators:
– drug problems and mental health issues across the abuse subtypes with one exception (neglect), where
drug problems were similar between the two groups of perpetrators.
– financial problems and family issues in relation to psychological and financial abuse.
y In contrast, gambling problems were more common for friends than adult children perpetrators in relation to
financial and physical abuse, and neglect.
Chapter 7: Who commits elder abuse? 77
Table 7.4: Perpetrator problems, by relationship with perpetrator and abuse subtype
Note: Weighted data for the statistics and unweighted sample sizes.
In summary, comparing adult children and friend perpetrators across the four abuse subtypes highlights the
complex dynamics surrounding elder abuse. Drug problems, mental health, and financial problems were more
common among adult children perpetrators than friend perpetrators, while gambling problems were more
marked for friend perpetrators. In addition, elder abuse victims of friend perpetrators were older and more likely
to be from non-English speaking backgrounds and have poorer health. Never having been married also appeared
to be linked to elder abuse victims whose experience was inflicted by their friends.
The data appear to suggest that abuse by friends is linked with victim vulnerability to a greater extent than
abuse by adult children. In contrast, adult children are more likely to have problems than friend perpetrators, to
the extent the older person is aware of. This suggests that to some degree, victim-focused factors are important
in abuse by friends and perpetrator-focused factors are important in abuse by adult children. Moreover, in some
instances, problems, such as gambling, may have led friend perpetrators to target older persons who were
particularly vulnerable.
General patterns in the relationships between the perpetrators and the CALD victims were similar to those that
emerged for the whole sample. Familial perpetrators account for the majority of perpetrators for the CALD
subsample. Intergenerational dynamics are strongly evident with sons and daughters accounting for 13% of the
perpetrators. Sons- and daughters-in-law are almost as significant a category as sons and daughters (11%).
Among the group of perpetrators known to the older person as a friend, neighbour or acquaintance, friends
were the biggest group for the CALD subsample (18%), and they were also notably a bigger group than children.
By comparison, children were a larger perpetrator group than friends for the whole sample (see Figure 7.1).
Professional carers were a small group of perpetrators for the CALD subsample (2%).
78 National Elder Abuse Prevalence Study: Final Report
Figure 7.2: Survey of Older People: Relationship of perpetrators to older people who experienced at least one
type of abuse and are from a CALD background
Partner/spouse 8.1
Son/daughter a
12.5
Step-son/daughter 2.2
Grandson/daughter 2.7
Brother/sister 7.1
Brother/sister in-law 2.1
Son/daughter in-law 11.3
Other family member 4.6
Ex-partner/spouse 0.6
Friend 17.6
Neighbour 4.9
Professional carer 1.9
Service provider 6.2
Acquaintance 9.5
Other 2.2
0 5 10 15 20
Percentage
Notes: Weighted data and unweighted sample size (number of perpetrators (base) = 136). Multiple perpetrators could be
reported for each type of abuse. Values may not equal 100.0% due to rounding. a Includes biological/adopted children.
The main sources of this type of abuse were friends (33%), acquaintances (18%) and neighbours (12%).
Representation of professional carers (3%) and service providers (4%) was limited.
Table 7.5: Survey of Older People: Relationship of perpetrators to older people who are from a CALD
background and who experienced abuse relating to language and culture
Notes: Weighted data. Multiple perpetrators could be reported with only one reported as main perpetrator. Relationship
categories with no proportions are not shown. Values may not equal 100.0% due to rounding.
Chapter 7: Who commits elder abuse? 79
Figure 7.3 demonstrates that for concerns about emotional abuse, participants reported a son was the person
who mistreated the older person (26%). This was followed by another family member (24%) and a daughter
(21%). A son was also the most frequently reported person mistreating older people by taking advantage of
them financially (31%), again followed by another family member (28%) and daughter (23%). In contrast, for
concerns about failure to provide adequate care with routine activities and physical hurt, professional carers
were most commonly reported as a source of these concerns (27% and 23% respectively).
Figure 7.3: Survey of the General Community: Perpetrators identified for reported concerns by subtype
31.1
Son 15.4
25.8
21.8
23.1
10.7
Daughter 20.8
15.8
7.1
Partner/spouse 13.4
12.7
6.4
27.6
Other family 14.9
member 24.0
19.7
4.3
Professional 22.7
carer 8.9
26.9
3.1
Other 10.6
professional 4.5
7.3
5.1
Friend 2.7
2.6
1.0
1.3
Neighbour 2.0
0.7
0.7
4.3
3.9
Other person 4.7
3.4
0 10 20 30 40
Percentage
Notes: Weighted data and unweighted sample size (in order of bars n = 358, 128, 386, 258). Son and daughter (include step
and in-law). Other person includes colleagues and ex-partner/spouse. Percentages do not sum to 100.0% as multiple
options could be selected. Further analysis of concerns about sexual abuse and who mistreated them not reported
due to small sample sizes (n = 11).
80 National Elder Abuse Prevalence Study: Final Report
Where participants identified professional carers as causing concerns, they were asked an additional question
to ascertain where the mistreatment occurred. This information is reported in Table 7.6 for concerns about
emotional abuse and failure to provide adequate care only, due to small sample sizes in the other types (less than
30 observations).
The most commonly reported place of mistreatment was residential care (77% of participants who identified
emotional abuse and 69% of participants who identified failure to provide adequate care with routine activities).
The next most commonly reported place of mistreatment for each of these items was the older person’s home,
where the corresponding proportions were 27% and 25% respectively. Of note is that hospitals were identified by
6% of participants with concerns about failure to provide care.
Table 7.6: Survey of the General Community: Concern related to a professional carer – where the mistreatment
took place
Summary
The findings in this chapter establish the predominant relationship dynamics in elder abuse are familial and
intergenerational. For the overall sample, these relationships account for the largest perpetrator group, with
children representing just under one-fifth of perpetrators, and partners of children and grandchildren around
one-tenth taken together. The intergenerational relationship dynamic was particularly pronounced for financial
abuse, with one-third of perpetrators in this group being sons and daughters. With the exception of sexual
abuse and neglect, children are the biggest perpetrator groups for all the other abuse subtypes, though to lesser
extents than for financial abuse.
Intimate partners are also significant perpetrators, particularly for physical abuse, sexual abuse and psychological
abuse (around one-tenth for each). For neglect, intimate partners are almost on par with children (over one-fifth
for each group).
Abuse perpetrated by neighbours, acquaintances and friends was also significant, with these groups accounting
for approximately one-quarter of perpetrators combined. Together, this group of perpetrators was almost as
large as the intergenerational familial group. Friends are particularly likely to be implicated in financial abuse.
Neighbours and friends are also significant perpetrator groups for physical abuse. Friends (two in five) accounted
for the largest perpetrator group in relation to sexual abuse, followed by a much smaller group of acquaintances.
They were also well-represented as psychological abuse perpetrators (combined accounting for a little under
one-third of perpetrators).
Professionals were the smallest overall perpetrator group, with service providers and professional carers
combined accounting for around one-tenth. Service providers appeared to be especially likely to be associated
with financial abuse. Nevertheless, neglect was the abuse type most likely to be associated with service providers
and professional carers (over one-quarter taken together). However, the under-representation of people with a
need for a higher level of care in this sample is a limitation to consider.
Overall, men outweigh women as perpetrators of abuse by 10 percentage points (55% cf. 45%). They account
for more than three-quarters of abuse perpetrators for physical and sexual abuse and more than two-thirds for
financial abuse.
Consistent with some of the perpetrator-related risk factors identified in Storey’s (2020) systematic review
(chapter 2), perpetrators in this study are reported to have a range of problems, with around half having two
Chapter 7: Who commits elder abuse? 81
or more. Most commonly, they have mental health problems (more than a quarter) and financial problems
(nearly one in five). The most common problems associated with financial abuse are financial problems. Mental
health issues are the most commonly reported problems for physical and psychological abuse. For sexual abuse
perpetrators, problems with alcohol predominate.
Neglect is different from the other abuse subtypes, with physical health problems being the most common issue
associated with perpetrators.
Comparing adult children and friend perpetrators highlights the complex dynamics surrounding elder abuse. The
data appear to suggest that abuse by friends is linked with victim vulnerability to a greater extent than abuse by
adult children. In contrast, for abuse committed by adult children, perpetrator problems are a more significant
factor. Moreover, in some instances, problems, such as gambling, may have led friend perpetrators to target older
persons who were particularly vulnerable.
The findings from the SGC survey in relation to perpetrator identity are broadly in line with the patterns evident
in the SOP, with many concerns focusing on sons as perpetrators of financial abuse. Similarly, professional carers
are a significant source of concern about neglect.
CALD subsample
The pattern of elder abuse being familial and intergenerational generally also applied to the CALD subsample.
Intergenerational dynamics remain significant for this subsample but sons- and daughters-in-law are almost as
significant perpetrator groups as children.
Further, friends were the largest perpetrator group for this sample, while this was not the case for the overall sample.
In addition, friends were by far the largest source of abuse relating to language and culture (nearly one‑third).
82 National Elder Abuse Prevalence Study: Final Report
8 Seeking help
Key messages
f Where help or advice is sought from a third party, it is most often sought from family and friends
(approximately four in 10 respectively).
f Even without seeking third party support, a large majority of people who experience elder abuse
take action to try and stop the abuse (approximately eight in 10) mostly by either speaking to or
avoiding the perpetrator.
f The most common action taken is to speak to the perpetrator of the abuse (one-half).
f Action to avoid the perpetrator is also common, with four in 10 people who experienced abuse
breaking contact with the perpetrator and more than one-tenth withdrawing from social contact
more generally.
Introduction
This chapter examines the actions people take in response to elder abuse. It sets out the extent to which
participants in the SOP who experienced abuse sought assistance in addressing it. A wide range of potential
sources of help, from informal (family, friends) to formal (police, lawyers) is considered. The discussion sets out
the nature of the assistance sought and whether or not the person considered it to be effective. This topic is also
considered on the basis of the SGC findings, with the analysis examining whether the SGC participants who had
concerns about elder abuse in relation to someone known to them had sought assistance.
The first section in this chapter sets out SOP findings on the extent to which help and advice is sought in
relation to elder abuse. The second section examines SOP findings on actions taken in relation to elder abuse
and participants’ assessments of the effectiveness of these actions. The last part of the chapter examines what
actions SGC participants took to address concerns about elder abuse.
Chapter 8: Seeking help 83
Just over one-third (36%) reported that they had sought help or advice from a third party in relation to their
experience. An additional 8% experienced more than one subtype of abuse and sought help for one or more of
the subtypes rather than all the subtypes (data not shown).
Figure 8.1: Survey of Older People: Whether help or advice sought by persons who experienced each subtype
of elder abuse in the previous 12 months
Psychological:
27.7 71.8 0.5
Band 1 (low)
Psychological:
51.3 47.4 1.3
Band 2 & 3 (medium–high)
Neglect:
14.8 74.4 10.9
Band 1 (low)
Neglect:
28.5 64.4 7.1
Band 2 & 3 (medium–high)
0 25 50 75 100
Percentage
Yes No Unclear
Notes: Weighted statistics and unweighted sample sizes (in order of bars, n = 158, 127, 87, 843, 425, 418, 208, 130, 78, 1,423).
Excludes persons who answered the questions and who were not considered as having experienced a relevant type of
abuse by definition (see chapter 5). a Where a participant experienced two or more subtypes of abuse, the report for
each subtype was included; that is, a participant could have multiple records.
Help was most likely to be sought in relation to physical abuse (50%) and least likely to be sought in relation to
neglect (20%). For neglect in the low range, only 15% of those who experienced it sought help, compared with
29% in the medium to high ranges.
Two-fifths of the people who experienced psychological abuse sought help, with just over a quarter (28%) seeking
help for psychological abuse in the low range and just over half (51%) in the medium-to-high range seeking help.
Three in 10 people who experienced financial abuse sought help. Help for sexual abuse was sought less often,
with just under a quarter of those who reported experiencing sexual abuse also saying they sought help or
advice. This contrasts with evidence showing that 50% of women in the general population who experienced
sexual assault sought help and advice (AIHW, 2020).
Table 8.1: Survey of Older People: Sources of help or advice as reported by older persons who had sought help
for their experience of elder abuse, for each form of abuse and overall
Notes: Based on weighted data. Unweighted sample sizes. Multiple responses and sum may exceed 100.0%. Excludes persons
who answered the questions and who were not considered as having experienced a relevant type of abuse by definition
(see chapter 5). a Where a participant experienced two or more subtypes of abuse, a report for each subtype was
included as a separate case; that is, a participant may have multiple records.
Chapter 8: Seeking help 85
The seven most common sources of help overall were family members (41%) and friends (41%), followed by a GP
or nurse (29%), a professional carer (24%), the police (17%) and lawyers (15%).
The type of assistance sought varied according to the abuse type, although help seeking from family and friends was
consistently high compared to other sources (Table 8.1). The most common sources of help by abuse subtype was:
y With 30% of those experiencing financial abuse seeking help (Table 8.1), the most common forms of help
sought were: friends (54%), family members (53%), lawyers (33%), a GP or nurse (28%) and the police (25%).
y With 50% of those experiencing physical abuse seeking help (Table 8.1), the most common forms of help were:
the police (36%), family (33%), friends (28%), a professional carer or social worker (26%) and a GP or nurse
(22%). It is notable that this is the only time police or any other source of help exceeded the proportion of
family members or friends.
y With 24% of those who experienced sexual abuse (Table 8.1) seeking help, the main sources of help were: a
family member (37%), a GP or nurse (35%) a friend (34%) and a neighbour (18%). It is notable that police were
not a source of help at all for the sexual abuse group (see further chapter 14).
y With 40% of those who experienced psychological abuse seeking help, the most frequent sources of help
were: a friend (44%), a family member (40%), a GP or nurse (30%), a professional carer or social worker (22%),
a mental health professional (17%), a lawyer (17%) and the police (15%).
y With neglect the most under-reported form of abuse (20%, Table 8.1), the most common sources of help were:
a family member (50%), a GP or nurse (37%), a professional carer or social worker (33%) and a friend (22%).
Across the board, actions were taken to address the abuse in the majority of cases (82%), with variations in the
proportion who took action according to abuse subtypes (e.g. 63% in relation to neglect and 94% for physical
abuse). Notably, informal direct action (speaking to the person) was the most common form of action (53%). In
relation to financial abuse, this proportion was 59%. It was even lower in relation to neglect at 37%.
The proportions who took passive approaches (avoiding the person or withdrawing from social contact) raise
some concerns given the frequency with which family members were identified as perpetrators (see chapter 6).
Each of these actions could be seen to have adverse consequences that may compound the negative impact of
the abuse, particularly if avoiding the perpetrator means loss of contact with other family and social connections.
Avoiding the perpetrator was a frequent response, applying to 42%. Lower proportions indicated they withdrew
from social contact more generally (13%), a response that raises even greater concerns about the impact of social
isolation on wellbeing.
Table 8.2 details the actions taken according to the five abuse subtypes, with the final column depicting the
proportion of people who took the particular actions across all abuse subtypes.
y For psychological abuse, the direct action of speaking to the person was also most frequent, but lower at
52% compared with the other abuse subtypes. Breaking contact with the person was almost as common
at 49%. Seeking assistance from an intermediary to speak to the perpetrator was more common for
psychological abuse than for any other abuse type (24%).
Table 8.2: Survey of Older People: Older persons who experienced elder abuse and actions taken to stop
abusive behaviours from happening again, for each type of abuse and overall
Notes: Based on weighted data. Unweighted sample sizes. Multiple responses and sum may exceed 100.0%. Excludes persons
who answered the questions and who were not considered as having experienced a relevant type of abuse by definition
(see chapter 5). a Where a participant experienced two or more subtypes of abuse, a report for each subtype was
included as a separate case; that is, a participant may have multiple records.
The findings based on the yes/no assessments of effectiveness are depicted in Figure 8.2. The majority of elder
abuse victims who took actions felt their actions were effective in stopping their experience from happening
again (55–66%) regardless of the subtype of abuse. The proportions of actions that were reported as being
effective were similar across the five subtypes of abuse, though it appeared that effectiveness of actions taken
was slightly lower for financial abuse than for other types of abuse (55% vs 58–66%).
The fact that substantial minorities indicated the actions were not effective also merits attention. Particularly high
proportions provided negative responses for actions taken for financial abuse (37%), psychological abuse (35%) and
physical abuse (34%). More than a quarter provided negative evaluations for neglect (29%) and sexual abuse (27%).
The analysis of the qualitative data presented in the next section sheds light on the reasons behind these responses.
Chapter 8: Seeking help 87
Figure 8.2: Survey of Older People: Proportion of participants reported action taken was effective by type
of abuse
0 25 50 75 100
Percentage
Yes No Unclear
Notes: Based on weighted data. Unweighted sample sizes (in order of bars, n = 129, 115, 78, 717, 129). Excludes persons who
answered the questions and who were not considered as having experienced a relevant type of abuse by definition
(see chapter 5).
Qualitative insights
This section sets out further insights into participants’ views of the effectiveness of their actions in stopping
the abuse. This analysis is based on responses to an open-ended question that was asked of participants who
said they had taken action to stop the abuse. The question was ‘was/were the actions effective?’ The responses
provide insight into both effectiveness and lack of effectiveness. These are qualitative and reflect the responses
of participants who chose to answer the question. They do not reflect the experiences of all participants.
Effective responses
Substantial proportions of participants in the SOP who had provided extended responses about the action taken
following the experience of abuse, described the action as effective.
Around half of these participants described the action taken in response to financial abuse to be effective
(n = 63/129, 49%).
Most of the participants described the action taken with respect to each of the other forms of abuse as effective
(physical abuse: n = 71/115, 62%; sexual abuse: n = 44/78, 56%; psychological abuse: n = 34/71, 48%;13 neglect:
n = 81/129, 63%).
The majority of the participants who described taking effective action with respect to sexual abuse were women
(n = 36).
In relation to the raising of concerns about sexual abuse, in particular, some participants described undertaking
these discussions with the perpetrators (or their partners) in a public setting, with the perpetrators’ concern for
their reputation identified as supporting the effectiveness of these direct discussions.
Raising concerns about neglect in direct discussions was also described by some participants as enabling
any misunderstanding or underlying issues to be addressed. Some of these participants described how their
discussions provided an opportunity to reach a compromise arrangement in relation to their care needs, with
some also acknowledging constraints on the other party (n = 22).
13 Note that as there were 717 verbatim responses in relation to the reasons why action taken in response to psychological abuse was
effective, a random subsample (every 10th response) has been analysed.
88 National Elder Abuse Prevalence Study: Final Report
Some participants described how other people or professionals had spoken with perpetrators on their behalf with
positive effect (financial abuse: n = 5/63; physical abuse: n = 12/71; neglect: n = 9/54). In some instances, having
another person raise the concerns with the perpetrator helped them to see the effect of their abusive behaviour.
The extended responses also highlighted the effectiveness of engaging in discussions where they provided an
opportunity to identify and address underlying issues, including those requiring therapeutic or other support.
The data also indicate that older women were more likely to describe how they achieved an effective outcome when
concerns related to financial abuse or physical abuse were raised in direct discussions with the perpetrator (financial
abuse and physical abuse: n = 20) as compared to older men (financial abuse: n = 14; physical abuse: n = 13).
However, speaking directly with a perpetrator was not without risk. For example, one participant described their
discussions with the perpetrator as leading to their assault and hospitalisation; albeit, that this event operated as
the catalyst to stop the physical abuse that they had been experiencing.
In relation to counselling, participants described this as an effective step because it enabled them to receive
therapeutic assistance to deal with their experience of abuse. Mediation was also described as effective
where this provided an opportunity for facilitated discussions to address the older person’s experience, and
where perpetrators were described as genuinely engaging in the process and receptive to learning about and
addressing their behaviour.
Obtaining legal advice and support was described as effective where it supported the older person to clearly
communicate their views to the perpetrator and/or to take action to prevent further abuse. It is worth noting that
the receipt of professional advice was also identified as an effective response when it clarified behaviour that had
been interpreted as financial abuse. Taking legal action was also identified as an effective response where this
involved obtaining restraining orders, prosecution for assault, evictions, a separation/divorce or legal action to
prevent further access to financial accounts. Participants identified these actions as not only putting a stop to the
abuse but also providing a deterrent going forward.
Withdrawing contact
Some participants described the withdrawal of their contact temporarily or permanently as an effective response
to the abuse that they experienced (financial abuse: n = 15/63; physical abuse: n = 30/71; sexual abuse: n = 17/44;
psychological abuse: n = 17/34; neglect: n = 10/54). For some, the withdrawal of contact operated as a circuit
breaker, whereas for others, a complete withdrawal of contact was required to maintain the older person’s safety.
Some participants taking this action described how this made them feel more comfortable as they no longer had
to engage with the perpetrator. For other participants who described their withdrawal of contact as effective,
this nevertheless also meant withdrawing from their social life, which may have had other negative consequences
for the older person. Some participants who reported experiencing sexual abuse described, for example, how
they would avoid close contact with the perpetrator in social settings. In these circumstances, avoiding close
contact did not deprive them of their social engagement but was described as effective because the opportunity
to perpetrate the abuse was no longer available.
Ineffective responses
A substantial number of participants providing extended responses in the SOP described the action taken in
response to their abuse as ineffective (financial abuse: n = 49/129, 38%; physical abuse: n = 37/115, 32%; sexual
abuse: n = 17/78, 22%; psychological abuse: n = 34/71, 48% neglect: n = 39/129, 30%).
In most of these cases, the participating older people reported speaking or attempting to speak with the
perpetrator or that another person did so on their behalf (financial abuse: n = 39/49; physical abuse: n = 19/37;
sexual abuse: n = 8/17; psychological abuse: n = 16/34; neglect: n = 26/39).
Chapter 8: Seeking help 89
In these cases, the older participants described how these actions were not effective as the perpetrator was not
receptive to the concerns being raised. Reasons included indifference or lack of understanding on the part of the
perpetrator, perpetrators who were rigid in their views, who had unreasonable and undesirable personality traits
or were suffering from substance abuse, mental health issues or cognitive conditions. In these circumstances, the
perpetrators were not willing or able to acknowledge and address the concerns raised in the direct discussions,
or to acknowledge or address underlying issues.
Participants also described attempts at counselling or mediation that were ineffective because the perpetrator
would not listen, where this option did not address the pertinent issues or described unsatisfactory experiences
obtaining legal advice and support or taking legal action (financial abuse: n = 10/49; physical abuse: n = 5/37;
psychological abuse: n = 4/34).
In some cases, legal action was ignored or not understood by the perpetrator or restraining orders were flouted
and did not prevent the perpetrator from reoffending.
For other participants, steps taken to withdraw from their relationship with the perpetrator or from their social
engagement, more generally, were described as ineffective because they did not address the issues associated
with the abuse or the abusive behaviour remained unchanged (financial abuse: n = 3/49; physical abuse:
n = 12/37; sexual abuse: n = 9/17; psychological abuse: n = 14/34; neglect: n = 5/39). Attempts to withdraw were
also described as ineffective where this led to estrangement from family or friendship groups and/or to social
isolation, or where withdrawal of contact was unworkable in practice. Of note, where participants described
their withdrawal from their social life as an ineffective response to physical abuse, this was, in fact, described as
exacerbating their situation. In some instances, taking this action also meant that the older person did not receive
the care that they required.
Consistent with the findings in the SOP, the most common responses are the informal ones. The SGC participants
with concerns most frequently spoke to the person causing the concerns (67%), or another family member or
friend spoke to the person causing the concerns (61%).
Passive responses – breaking contact with or avoiding the source of the concerns – were also common, at 38%.
Formal intervention was reported by higher proportions of SGC participants compared with SOP participants.
In the SGC, 21% reported seeking legal advice, compared with 9% in the SOP. Mediation or counselling was also
reported more often in the SGC compared with the SOP (14% cf. 10%). SGC participants were also more likely to
report intervention by medical or social work professionals (24% in the SGC cf. 9% in the SOP).
These findings may indicate that abuse of a degree that comes to the attention of a third person may be of a
greater severity. It may also suggest that third parties could be less cautious (or fearful) about seeking help.
90 National Elder Abuse Prevalence Study: Final Report
Figure 8.3: Survey of the General Community: Where an action was taken in response to concerns, what action
was taken, by gender
66.6
I spoke to the person
66.2
causing the concerns
66.6
59.7
A family member or friend
spoke to the person causing 61.5
the concerns
60.9
39.0
I broke contact with or
avoided the person involved 36.8
causing the concerns 37.5
19.8
A professional
(social worker, doctor, nurse) 27.3
spoke to the person
24.3
5.3
A restraining order or safety
order was made against the 5.1
person causing the concerns 5.1
20.8
Other 25.2
23.5
0 10 20 30 40 50 60 70
Percentage
Gender
Men Women All
Notes: Percentages do not sum to 100.0% as multiple options could be selected. Four observations where gender
was reported as ‘Other’ not reported.
Summary
This chapter has examined actions taken and help-seeking responses in relation to elder abuse. The findings
indicate that help seeking from a third party was not a majority response on the part of older people who
experience elder abuse, with six in 10 people who experienced elder abuse not seeking help.
The most common forms of abuse for which help was sought from a third party were physical abuse, financial
abuse and psychological abuse. Help was least likely to be sought for neglect and sexual abuse. Notably, reliance
on criminal justice responses for sexual abuse was almost absent.
Where either help or advice was sought, or actions were taken to stop the abuse, the most commonly reported
avenues were informal. Help and advice was most commonly sought from family and friends. Older people most
frequently took it upon themselves to stop the abuse by speaking to the perpetrator.
A common means of stopping the abuse was breaking contact with or avoiding the perpetrator (42%). In
some cases, a more significant withdrawal from social life was reported (13%). These responses raise particular
Chapter 8: Seeking help 91
concerns, as they may have adverse consequences for the older person’s contact with other family, friends and
support networks. They also do not address perpetrator accountability.
Help and advice were sought more often from the helping professions (social workers, medical professionals) and
less often from lawyers and police. Actions to stop abuse involving quasi-legal services (mediation or counselling:
10%), legal advice (9%) or personal protection orders (4%) were not particularly common.
Where a third party had concerns about elder abuse, the SGC findings confirm that informal options for addressing
the abuse (speaking with the person causing the concerns) were the most common. Passive responses were also
reported by more than one-third of SGC participants (38%).
Notably, substantial minorities of older people who did report taking action to stop the abuse considered these
actions ineffective.
It is concerning that help and advice were sought only by a minority, that passive avoidant actions to stop abuse
were adopted by more than half of those who experienced abuse and that actions to stop abuse are evaluated as
ineffective by between a quarter and a third of older people who experience abuse.
These findings point to a need for an assessment of existing options for addressing elder abuse in general, as well
as particular subtypes, and whether these options are accessible, appropriate and adequate.
92 National Elder Abuse Prevalence Study: Final Report
9 Self-assessed seriousness
Key messages
f Being male, in the older age groups and living with the perpetrators were associated with lower
seriousness appraisals for financial and physical abuse.
f Financial abuse was accorded lower seriousness ratings if the perpetrators were family members.
Introduction
This chapter further examines elder abuse from the perspective of whether people who experience it consider
it is serious. First, the chapter sets out responses on how serious the abuse was from the perspective of those
who experience it. Second, the extent to which older persons’ assessments on seriousness varied according
to selected characteristics is examined. This chapter also examines the link between self-assessed seriousness
and help seeking.
Self-assessed seriousness
SOP participants who experienced abuse were asked to indicate how serious the abuse was for them. These
ratings demonstrate the participant’s subjective assessment of how serious the abuse was, based on their
understanding of the term ‘serious’. Response options were ‘very serious’, ‘somewhat serious’ and ‘not serious’.
As discussed in chapter 5, recent international research demonstrates that subjective seriousness ratings may be
influenced by the complex psychological dynamics associated with elder abuse (e.g. Burnes et al., 2019a).
Table 9.1 and Figure 9.1 present findings on ‘seriousness’ assessments. Overall, most participants assessed the
abuse as either very serious (28%) or somewhat serious (42%). Just over a quarter assessed the abuse as not
serious. In relation to the five subtypes of abuse, noteworthy findings are:
y Physical abuse was most likely to be rated as very serious (39%) or somewhat serious (40%), with only
20% rating it as not serious.
y Psychological abuse was the next most likely to be rated as very serious (32%) or somewhat serious
(46%), with only 21% rating it as not serious. Even at the low score (see chapter 5) range, nearly one in five
participants indicate the abuse was serious, with half nominating it as somewhat serious. In the medium to
high score range, 44% said the abuse was very serious and 41% somewhat serious.
y Financial abuse attracted the third strongest seriousness ratings, with 30% rating it as very serious and 35%
as somewhat serious. Close to a third rated it as not serious.
Chapter 9: Self-assessed seriousness 93
y Sexual abuse was most likely to be rated as somewhat serious (39%) or not serious (41%).
y The lowest seriousness assessments were accorded to neglect, with almost half (48%) of participants
indicating their experience was not serious. Not serious assessments were accorded to 57% of experiences in
the low score band and 33% in the medium/high score band. Three in 10 participants in the low score band
assessed their experience as somewhat/moderately serious and nearly four in 10 chose this assessment for
experiences in the medium/high band.
Figure 9.1: Survey of Older People: Reports of seriousness of main perpetrator’s behaviour, persons who
experience each subtype of elder abuse in the previous 12 months
30.0
Financial 34.5
31.9
3.7
38.5
40.4
Physical
19.8
1.3
20.0
38.8
Sexual
41.1
0.0
32.1
45.8
Psychological
21.0
1.1
19.2
Psychological: 50.6
Band 1 (low) 29.3
0.9
44.4
Psychological: 41.3
Band 2 & 3 (medium–high) 13.1
1.2
7.3
35.0
Neglect
47.6
10.0
1.8
Neglect: 30.3
Band 1 (low) 57.0
10.9
15.8
Neglect: 42.0
Band 2 & 3 (medium–high) 33.3
8.8
28.2
42.2
All types (combined) a
27.0
2.7
0 20 40 60
Percentage
Very serious Somewhat/moderately serious Not serious Other
Notes: Weighted statistics and unweighted sample bases. a Where a participant experienced two or more subtypes of abuse,
the report for each subtype was included; that is, a participant could have multiple records.
94 National Elder Abuse Prevalence Study: Final Report
Table 9.1: Survey of Older People: Reports of seriousness of main perpetrator’s behaviour, persons who
experience each subtype of elder abuse in the previous 12 months
Somewhat/ Number of
Very serious moderately Not serious Other Total participants
Abuse type (%) serious (%) (%) (%) (%) (base)
Financial 30.0 34.5 31.9 3.7 100.0 158
Physical 38.5 40.4 19.8 1.3 100.0 127
Sexual 20.0 38.8 41.1 0.0 100.0 87
Psychological 32.1 45.8 21.0 1.1 100.0 843
Band 1 (low) 19.2 50.6 29.3 0.9 100.0 425
Band 2 & 3 44.4 41.3 13.1 1.2 100.0 418
(medium – high)
Neglect 7.3 35.0 47.6 10.0 100.0 208
Band 1 (low) 1.8 30.3 57.0 10.9 100.0 130
Band 2 & 3 15.8 42.0 33.3 8.8 100.0 78
(medium – high)
All types (combined) a 28.2 42.2 27.0 2.7 100.0 1,423
Notes: Weighted statistics and unweighted sample bases. a Where a participant experienced two or more subtypes of abuse,
the report for each subtype was included; that is, a participant could have multiple records.
Table 9.2: Survey of Older People: Proportions of older persons who reported their elder abuse experience as
serious by selected characteristics, by type of elder abuse
Notes: Weighted statistics and unweighted sample bases. Son and daughter refers to biological/adopted children. Includes
a b
Regardless of the subtype, participants’ seriousness assessments were related to help seeking, with higher
seriousness assessments related to greater help seeking from a third party. However, even when abuse was
rated as very serious, sizable proportions of participants did not seek help. This was highest for sexual abuse
and neglect, with almost three in 10 participants who assessed sexual abuse as very serious and almost four in
10 participants who assessed neglect as very serious not seeking help. The sample sizes for sexual abuse and
neglect are not large and these insights should be considered exploratory.
96 National Elder Abuse Prevalence Study: Final Report
Table 9.3: Survey of Older People: Proportions of older persons who reported having sought help by
self‑assessed seriousness, by type of elder abuse
Somewhat/
moderately
Abuse type Very serious serious Not serious
Sought help from a third party %
Financial a 60.9 27.8 4.2
Physical 65.7 38.7 47.1
Sexual 28.8 16.1
Psychological a 59.0 39.7 11.9
Neglect 36.9 9.7
Number of participants (base) n
Financial 53 54 47
Physical 52 48 25
Sexual 64 33
Psychological 268 384 178
Neglect 84 104
Notes: Weighted statistics and unweighted sample bases. a The differences in help seeking across self-assessed seriousness
groups were statistically significant based on a design-based F test (p < .05)
Summary
This chapter has examined participants’ views on whether their experience of elder abuse was serious, as well as
setting out characteristics that were associated with higher or lower seriousness assessments. It also examined
connections between self-assessed seriousness and help seeking.
Overall, most participants rate their experience of abuse as very serious (28%) or somewhat serious (42%). The
relative distribution of seriousness ratings in this study is broadly consistent with those in the UK study, in which
33% rated the abuse as very serious, 43% as serious and 24% as not serious (O’Keeffe et al., 2007).
The two abuse subtypes most likely to be accorded a higher range of seriousness ratings are physical abuse
(39%) and psychological abuse (32%). Seriousness ratings for financial abuse are relatively evenly distributed
across the three seriousness bands.
Neglect attracts the lowest seriousness assessments with over half (58%) of participants rating their experience
as not serious.
The analyses suggests that older age and living with the perpetrator were associated with lower self-assessed
seriousness for financial and physical abuse. Older people were less likely to rate experiences of financial abuse
Chapter 9: Self-assessed seriousness 97
as serious if they lived with the perpetrator, or if the perpetrator was a family member. Men were less likely than
women to rate their financial, physical and sexual abuse as serious.
For all the subtypes of elder abuse, the more seriously older persons rated their elder abuse experience, the more
likely they reported having sought help. This pattern was less apparent for physical abuse. Physical abuse victims
with not serious assessment were more likely to seek help, compared to victims with not serious assessments
of other subtypes. It is worth noting that the findings are based on small samples for all subtypes except
psychological abuse, and these insights should be considered exploratory.
98 National Elder Abuse Prevalence Study: Final Report
10 Legal arrangements
Key messages
f A lower but still majority proportion have an enduring power of attorney, which can be used when
needed (51.6%).
f The findings suggest that having a will or an enduring power of attorney is associated with lower
reports of experiencing elder abuse.
f By contrast, the data show that only a small proportion of participants (3%) in the SOP had a family
agreement, with reports of having a family agreement associated with higher rates of elder abuse.
f The correlations between family agreements and elder abuse are as likely to be related to
socio‑economic status and reflective of relationship dynamics as they are to be related to the
uptake of family agreements.
Introduction
This chapter examines legal and financial arrangements involving participants in the SOP and the SGC. Specifically,
it sets out findings in relation to wills, powers of attorney and family agreements (these involve arrangements
where an interest in an asset, or another financial arrangement, may be exchanged for care, possibly including
a shared property arrangement). The discussion considers the extent to which older people who adopted these
legal mechanisms reported experiencing financial, physical, sexual or psychological abuse or neglect.
The discussion in this chapter relates to the following aspects of the research aims:
y Measure the proportion of people who provide assistance to older family members or friends and the types of
assistance they provide in relation to financial transactions and decision making.
y Assess the extent to which older Australians have adopted advance planning behaviours.
The following socio-demographic characteristics are included in the analyses: gender; age; religion; marital status;
whether the participant has children (biological, adopted or step); the participant’s household composition;
education; the applicable category according to the socio‑economic index for areas (SEIFA); whether the
participant lives in a major city, inner regional area or outer regional area; their health status and whether they
require assistance some or all of the time for one or more activities. In this chapter, ‘advance planning’ is used
as a general term to describe the three mechanisms considered rather than referring specifically to advance
care directives (see e.g. ALRC, 2017, p. 162). It is noted that the findings based on the SOP do not reflect the
experiences of older people who had insufficient cognitive capacity to participate in the survey, and who may be
at greater risk of experiencing the abuse or misuse of legal instruments.
Chapter 10: Legal arrangements 99
This chapter has four parts. The first part presents findings relating to wills; specifically, who has a will and the
prevalence of abuse for people with and without wills. The second section presents data relating to enduring
powers of attorney. It considers data on who has made an enduring power of attorney and who has been
appointed as an attorney, record-keeping practices by appointees and the prevalence of abuse for those with an
enduring power of attorney. The final section presents data relevant to family agreements. It looks at who enters
these agreements and the prevalence of abuse for those with a family agreement. The final section sets out some
relevant points from recent commentary and analysis in relation to proposals for greater protection for each of
the three types of arrangements considered in this chapter.
Participants in the SOP were asked whether they had executed a will. Consistent with the findings of Tilse,
Wilson, Settlerlund, and Rosenman (2005), Table 10.1 shows that the vast majority of older people (88%)
reported having a will at the time of interview. Slightly more women than men indicated having a will. Of note,
three‑quarters of these participants had discussed their will with someone in their family (data not shown in the
table). Most older people who reported they did not have a will, nevertheless reported having had discussions
with someone in their family about making a will (8% male; 7% female).
Table 10.1: Survey of Older People: Whether participant currently has a will, by gender
Notes: Weighted statistics and unweighted sample bases. a Total includes a small number of not-stated cases (0.1%).
Participants in the SOP were asked whether they had enduring powers of attorney in place. In this context, a
noteworthy finding reported in chapter 5 was that of the 2.1% of the sample who reported financial abuse, none
reported misuse of a power of attorney. Despite the absence of evidence of financial abuse being perpetrated
this way in this study of community dwelling older people, concerns about abuse in this context are evident in
some reports (see further below).
100 National Elder Abuse Prevalence Study: Final Report
As noted above, the findings based on the SOP in this discussion do not reflect the experiences of older people
who had insufficient cognitive capacity to participate in the survey, and who may be at greater risk of experiencing
the abuse or misuse of legal instruments. For example, older people with dementia have been identified as particularly
vulnerable to financial abuse perpetrated by people appointed under an enduring power of attorney (Miskovski, 2014).
Table 10.2 shows that just over half of older people (52%) indicated that they had granted an enduring
power of attorney, with 49% of older males and 54% of older females answering in the affirmative. More than
three‑quarters of these participants reported that their current power of attorney was valid (i.e. can be used
when needed) (86%), with the most common type of power of attorney dealing with both financial and medical
matters (79%; 78% of male and 81% female). Of those older people who reported that they did not have a power
of attorney, a minority reported having had discussions with their family about making a power of attorney (15%).
Table 10.2: Survey of Older People: Whether have ever given an enduring power of attorney to someone,
by gender
Notes: Weighted statistics and unweighted sample bases. a Includes a small number of participants who were unsure or
non‑responding (less than 1%). b A small proportion of participants (5%) did not know what type of power of attorney
they had and these data are not shown in the table, and hence the sum of the three types is not 100%.
What are family agreements and how common are these agreements?
A family agreement is defined as an agreement to provide care to an older person in return for financial support
or benefit (e.g. transfer of assets, property or finances) or bequest in a will. The ALRC reported a range of
concerns about family agreements; in particular, that they can have ‘serious consequences for the older person
if the promise of ongoing care is not fulfilled, or the relationship otherwise breaks down’ (ALRC, 2017, p. 203).
Concerns have been raised in relation to family agreements as potential enablers of elder abuse, particularly
where informal agreements and decisions are made typically without legal advice (Boersig & Illidge, 2018).
In this context, it should be noted that of the 2.1% of the sample who reported experiencing financial abuse,
41.5% indicated they had been pressured into loaning or giving money, possessions or property (chapter 5).
Further, chapter 11 reports an apparent association between receiving assistance with buying and selling a home
as a form of assistance with financial activities and experiencing financial abuse for both men and women.
Participants in the SOP were asked whether they had a family agreement in place and whether the family
agreement was written down. In contrast with the substantial proportions of older people who reported having
a will or having an enduring power of attorney, Table 10.3 presents a markedly different picture in relation to
family agreements, with only 3% of participating older people reporting that they had such an agreement. Just
under two-thirds of these older people reported that their family agreement was in writing. No differences of
note emerged for male or female older people; however, given the small numbers of participants with family
agreements, the findings on this issue should be interpreted with caution.
Chapter 10: Legal arrangements 101
Table 10.3: Survey of Older People: Whether have family agreements, by gender
Notes: Weighted statistics and unweighted sample bases. a Includes a small number of participants who were unsure or
non‑responding (approximate 1%).
Consistent with the response pattern identified in relation to wills, older people in the CALD subsample were less
likely to report having granted a power of attorney as compared to those in the non-CALD sample (see previous
Table 10.2: 54% vs 41%). This difference was also statistically significant and is consistent with prior research
focusing on intergenerational asset management that suggests a lack of knowledge or relevance, mutuality in
asset management and informal financial arrangements among extended families as explanations for the lower
uptake of formal advance planning mechanisms (King et al., 2011).
In contrast with the response pattern identified in relation to wills and powers of attorney, older people from the
CALD subsample were more likely to report having a family agreement as compared to those participating in
the SOP who spoke English only at home (Table 10.4: 4% vs 3%). However, this difference was not statistically
significant.
Table 10.4: Survey of Older People: Advance planning, by whether person speaks a language other than
English at home
Notes: Weighted data and unweighted sample sizes. The asterisks indicate that the difference between the two groups is
statistically significant based on a design based F-test (*p < 0.05)
Table 10.5: Survey of Older People: Proportion of people who have a will, by age
Notes: Weighted data and unweighted sample size. The asterisks indicate that the differences in reports of specific advance
planning type across the age groups were statistically significant based on a design based F-test (*p < 0.05; **p < .01;
***p < .001)
Having a will in place appeared to be associated with cultural backgrounds. Indigenous people were less likely than
the non-Indigenous groups to report having a will. A significantly lower proportion of older people who were born
in non-English speaking countries had a will compared to those who were born in Australia or in English speaking
countries (80% cf. 87–91%). Similar to this pattern, older people with no religion or a non‑Christian religion were
also associated with a lower proportion of having a will (data not shown). It is important to note that the religion
variable is likely to reflect different cultural backgrounds rather than religiosity per se.
An older person’s level of education also appeared to be associated with whether they reported having a will, with
older people without a post-school education less likely to report having a will (Year 12: 85%; Below Year 12: 87%) than
older people who had a certificate, diploma, trade (90%) or degree or higher (89%), to a statistically significant extent.
Older people who were either separated or divorced (77%) or had never married (76%) reported having a will
in lower proportions than those who were married (90%), whereas older people who were widowed reported
having a will to a greater extent (93%). These differences were statistically significant.
Older people living in inner or outer regional areas reported having a will in greater proportions than older
people in major cities.
An older person’s socio‑economic status appears to be associated with whether or not they reported having
a will. Those reporting the lowest socio‑economic status are least likely to report having a will (80%) and
those with the highest socio‑economic status are most likely to report having a will (91%). The data show
that reports of having a will increases with each socio‑economic category to a statistically significant extent,
although there were some variations in this pattern when the responses of male and female participants were
analysed separately. Where older people reported owning their own home outright, they were also more likely
to report having a will (93%) as compared with older people who owned their home with a mortgage (77%)
and older people who rented private or public housing (62% and 48% respectively) or who had other housing
arrangements (78%). The data also show an association between higher household income and having a will,
with the increases in reports of having a will generally increasing with participants’ household income, save for
the highest income quartile. These variations associated with home ownership and household income were also
statistically significant.
It appeared that advance planning involving a power of attorney was also associated with cultural backgrounds.
A greater proportion of older people with an active power of attorney were born in Australia than overseas.
A greater proportion of older people who were Christian reported having an active power of attorney (45%) as
compared to those who observed no religion (41%) or were of non-Christian religions (44%) (data not shown).
These differences with respect to country of birth and religion were statistically significant.
Chapter 10: Legal arrangements 103
Consistent with the findings in relation to the association between level of education and reports of having a will,
older people without a post-school education were also less likely to report having an active power of attorney
than older people with a certificate, diploma, trade or degree.
In terms of living arrangements, consistent with findings relating to the living arrangements of older people
with a will, older people who reported living in a household with any of their biological, step- or foster children
were less likely to report having an active power of attorney (34% cf. 46%). Both male and female older people
were more likely to report having an active power of attorney if they lived alone (49%) as compared to those
living with a partner (43%), who were in turn more likely to have an active power of attorney than older people
with other living arrangements (39%). Consistent with the response pattern for older people in relation to wills,
both male and female older people who were either separated or divorced (35%) or had never married (37%)
reported having an active power of attorney in lower proportions than those who were married (44%), whereas
older people who were widowed reported having an active power of attorney to a greater extent (56%). These
differences in relation to marital status were statistically significant.
In relation to geographical location, older people living in inner regional areas were more likely to report having
an active power of attorney than older people in major cities (43% cf. 48%), with lower reports from older people in
outer regional areas. The variations between inner regional areas and major cities were also statistically significant.
Consistent with the response patterns observed in relation to older people with wills, Table A10.5 shows that where
older people indicated that they did not have children, they reported having an active power of attorney in greater
proportions than those with biological or adopted children. A greater proportion of older people who reported
having step-children also reported having an active power of attorney, which may reflect more deliberate
advance planning among people with step-children, perhaps as a means of managing complex family dynamics.
Again consistent with the response patterns observed in relation to older people with wills, Table A10.5 shows
that the older person’s socio‑economic status also appeared to be associated with whether they had an active
power of attorney, with those reporting the lowest socio‑economic status less likely (to a statistically significant
extent) to report having an active power of attorney (41%) and those with the highest socio‑economic status
most likely to report having a power of attorney (47%). It is notable, however, that this pattern held for older
women in the highest socio‑economic status bracket (51%) rather than for older males in this bracket (41%).
Consistent with reports from older people with a will, where older people reported owning their own home outright,
they were more likely to report having a power of attorney (48%) as compared with older people who owned their
home with a mortgage (26%) and older people who rented private or public housing (28% and 22% respectively) or
who had other housing arrangements (50%). The data also show an association between higher household income
and having a power of attorney, with these reports generally increasing with participants’ household income, save
for the highest income quartile. These differences associated with home ownership and household income were also
statistically significant.
Also consistent with the data relating to older people with wills (see Appendix A, Table A10.4), there was a
negative association between poor/fair health (44%) or good health (42%) and reports of having an active
power of attorney as compared to those in excellent/very good health (46%). By way of contrast, older people
with a disability or long-term medical condition were more likely to report having an active power of attorney
(47%) than those without a disability or long-term medical condition (42%), as were older people who reported
needing help some or all of the time with one or more activities (50%) when compared with those who did not
require this assistance (40%) (data relating to requirement for assistance not shown). This finding may reflect the
differing nature of enduring powers of attorney, which are directed at facilitating decision making when a person
no longer has the capacity to make decisions affecting their wellbeing during their lifetime, whereas wills are
directed at implementing a person’s wishes with regard to their estate upon their death.
Consistent with the response patterns identified in relation to wills and active powers of attorney, the proportions
of older people with a family agreement increased with each age category. There were greater proportions of
people with a family agreement in the 70–74 years (3%), 75–79 years (3%) 80–84 years (4%) and 85+ (7%) age
categories as compared to people in the 65–69 years age category (1%).
104 National Elder Abuse Prevalence Study: Final Report
In contrast with the response patterns for older people with wills and with enduring powers of attorney, a greater
proportion of older people with family agreements were born in English speaking countries overseas (4%) as
compared to those born in Australia (3%) or in non-English speaking countries (2%).
A similarly contrasting response pattern emerged in relation to the level of education of people with a family
agreement. Older people without a post-school education were more likely to report having a family agreement
than older people with a certificate, diploma, trade or degree.
Consistent with the response options for those with wills and with enduring powers of attorney, a greater
proportion of older people who were Christian reported having a family agreement as compared to those who
observed no religion or were from non-Christian religions.
In terms of living arrangements, older people were more likely to report having a family agreement where they
lived alone or in another living arrangement as compared to those living with a partner. Older people who
were either separated or divorced (2%) or who were married (3%) reported having a family agreement in lower
proportions than those who were never married (4%) or widowed (4%).
Older people living in both outer regional areas (4%) and inner regional areas were more likely to report having a
family agreement than older people in major cities (3%). These patterns are different in some respects with those
emerging in relation to wills and enduring powers of attorney.
Also in contrast with the response patterns observed in relation to older people with wills and enduring powers
of attorney, where older people indicated that they did not have children living in or outside of their household,
they reported having a family agreement in higher proportions than those older people with children living
outside of their household. This finding prevailed whether those children were biological, adopted or step-
children. A smaller proportion of older people who reported having step-children (whether they lived in or
outside the household) also reported having a family agreement, with this pattern also reflected in the reports of
older people with biological or adoptive children.
The analysis suggested a different response pattern regarding socio‑economic status and older people with
family agreements than the pattern that emerged for those with wills and enduring powers of attorney. Rather
than high socio‑economic status being associated with the making of family agreements, those with mid-range
socio‑economic status were most likely to report having a family agreement (4%), and the group least likely
to report having a family agreement were older people in the second highest socio‑economic status (2%). It is
notable, however, that the pattern was somewhat different for older women when analysed separately, with
women in the lowest socio‑economic status (rather than mid-range socio‑economic status) having the highest
reports of family agreements.
In contrast with reports from older people with a will or enduring power of attorney, older people who rented
public housing (4%) or who had other housing arrangements (4%) were more likely to report having a family
agreement than those who owned their own home outright (3%) or who owned their home with a mortgage (1%).
Also in contrast with reports from older people with a will or enduring power of attorney, the data show an
association between lower household income and having a family agreement, with these reports generally
decreasing with participants’ household income.
Contrasting with the data relating to older people with wills and enduring powers of attorney, older people with
poor/fair health (5%) and good health (3%) were more likely to report a family agreement than those in excellent/
very good health (2%). Older people with a disability or long-term medical condition were more likely to report
having a family agreement (4%) than those without a disability or long-term medical condition (2%), as were
older people who reported needing help some or all of the time with one or more activities (5%) when compared
with those who did not require this assistance (2%).
Table 10.6: Survey of Older People: Older persons who had a valid/active power of attorney, who was granted
the power of attorney, by gender
Person granted power of attorney Males (%) Females (%) Total (%)
Partner/spouse 26.5 15.6 20.4
Son/daughter (biological/adopted) 64.0 74.8 70.1
Step-son/daughter 3.7 1.6 2.5
Grandson/daughter 0.7 0.5 0.6
Brother/sister 4.2 4.5 4.4
Brother/sister in-law 0.6 0.5 0.6
Son/daughter (in-law) 1.7 1.6 1.7
Other family member 5.0 3.7 4.3
Ex-partner/spouse 0.2 0.1 0.2
Friend 2.3 2.5 2.4
Neighbour 0.1 0.1 0.1
Professional carer 0.0 0.1 0.0
Other service provider 2.9 1.6 2.2
Other 0.4 0.3 0.3
Number of participants (base) 1,202 2,133 3,339
When considering the data from the SOP regarding appointments by reference to the age of the older person,
Table 10.7 shows that reports of appointments of biological and adopted children increase with the age of the
older person. By way of contrast, appointments of siblings and spouses under powers of attorney decrease with
the age of the older person. There is fluctuation in the age at which older people reported appointing other
family members. These response patterns hold for both older men and older women participating in the SOP
(data not shown, see Appendix A, Table A10.6).14
Table 10.7: Survey of Older People: Older persons who had an active power of attorney, who was granted the
power of attorney, by age
Relationships Age 65–69 (%) Age 70–74 (%) Age 75–79 (%) Age 80–84 (%) Age 85+ (%)
Partner/spouse a 30.7 25.0 17.0 15.6 4.6
Son/daughter 58.2 68.7 74.5 73.0 80.0
(biological/adopted) a
Note: aAge-related patterns were statistically significant (p < .05). Acquaintances category had no responses. Financial
planner is combined with other service provider.
14 There is a slight variation in this response pattern for women aged in the 75–79 and 80–84 age brackets (79% vs 78% respectively)
with respect to the appointment of biological children; however, this difference was not identified as statistically significant.
106 National Elder Abuse Prevalence Study: Final Report
Figure 10.1: Survey of Older People: Older persons with whom the family agreement arrangements were made,
by gender
27.1
Partner/spouse 7.7
16.6
33.2
Son/daughter a 46.9
41.3
2.7
Grandson/daughter 1.4
2.0
5.3
Brother/sister 4.3
4.7
1.4
Son/daughter in-law 1.5
1.4
5.9
3.0
Friend 6.4
4.8
0.6
Neighbour
0.3
10.2
Professional carer 12.7
11.3
9.4
Other service provider 16.4
12.9
0 10 20 30 40 50
Percentage
Gender
Men Women Total
Notes: Weighted data and unweighted sample sizes (men n = 86, women n = 125, total n = 213). Multiple responses allowed
so the sum may exceed 100%. a Includes biological/adopted children. Proportions for Other not shown (men = 0.4,
women = 0.3, total = 0.4).
Chapter 10: Legal arrangements 107
Table 10.8: Survey of Older People: Prevalence of elder abuse, by each type of advance planning
Total Number of
Advance planning Financial Physical Sexual Psychological Neglect (any type) participants
type by gender (%) (%) (%) (%) (%) (%) (base)
Whether currently *** ** *** ***
have a will
No 5.0 3.6 1.6 16.4 4.2 20.2 664
Yes 1.7 1.5 0.9 11.1 2.7 14.1 6,312
Whether have ** ** *
ever given an
enduring POA to
someone
Yes, active 1.3 1.3 0.8 10.3 2.9 13.4 3,339
Yes, not active 2.4 1.7 0.7 11.0 1.6 14.6 410
None 2.8 2.3 1.2 13.4 3.0 16.5 3,018
Where have family ** **
agreements
No 2.0 1.7 0.9 11.7 2.8 14.7 6,709
Yes 4.7 3.3 4.0 13.8 3.5 20.5 213
Notes: Weighted data and unweighted sample size. The asterisks indicate that the differences in the prevalence of abuse
(for each subtype and overall) and specific advance planning type were statistically significant based on a design-based
F test (*p < .05; **p < .01; ***p < .001).
The data show that not having a current will was associated with the experience of ‘any form of abuse’ and with
most abuse subtypes.
When considering reports of ‘any form of abuse’, a higher proportion (20.2%) of older people without a will reported
experiencing abuse than older people with a current will (14.1%), with this difference being statistically significant.
More specifically, a higher proportion of older people without a will reported experiencing financial abuse or
physical abuse (5.0% and 3.6% respectively) than older people with a will (1.7% in relation to financial and 1.5%
physical abuse), to a statistically significant extent. A similar pattern emerged in relation to psychological abuse:
16.4% of older people without a will reported experiencing this form of abuse, compared with 11.1% with a will.
Regarding sexual abuse and neglect, the patterns were also consistent, with a higher proportion of those without
a will reporting the experience of abuse, although the results were not significant.
Consistent with the observations made in relation to wills, Table 10.8 shows that not having an enduring power of
attorney was associated with the of experience of ‘any form of abuse’ and with most abuse subtypes.
Compared to older people with an enduring power of attorney (13%–15%) who reported experiencing ‘any form
of abuse’, a higher proportion (17%) of older people without a power of attorney reported experiencing ‘any form
of abuse’, to a statistically significant extent.
More specifically, 1–2% of older people with an enduring power of attorney reported experiencing financial
abuse or physical abuse. In contrast, a greater proportion of older people without a power of attorney reported
experiencing these forms of abuse. Older people with a power of attorney in place were also less likely to report
experiencing psychological abuse than older people without an enduring power of attorney (10–11% cf. 13%),
to a statistically significant extent. Older people with an enduring power of attorney were less likely to report
experiencing sexual abuse and neglect than older people without an enduring power of attorney, the differences
were not statistically significant.
108 National Elder Abuse Prevalence Study: Final Report
In contrast to the findings with respect to older people with wills and enduring powers of attorney, Table 10.8
shows that reports from older people that they had a family agreement were associated with greater chances of
experiences of abuse – each subtype and overall (any form), although the results were not significant for some
subtypes owing to small numbers of reports of family agreements and of reports of abuse.
Among participants in the SOP with family agreements, 21% reported experiencing ‘any form of abuse’ compared
with 15% who did not have family agreements. Findings about specific abuse types emerged in relation to
financial abuse and sexual abuse. For financial abuse, 5% of people with family agreements experienced this
form of abuse compared with 2% without. In relation to sexual abuse, 4% of people with family agreements
experienced this form of abuse compared with 1% without family agreements.
Older people with a family agreement were also more likely to report experiencing physical abuse, psychological
abuse and neglect than older people without a family agreement.
There are some complexities in interpreting these data, specifically in considering whether advance planning operates
as a protective mechanism against abuse. It is likely that the lower reports of abuse among people who have wills
or powers of attorney is tied to their socio‑economic status, rather than related to the advance planning behaviours
themselves. Chapter 5 establishes that people with lower socio‑economic status are more susceptible to abuse
than those in higher brackets. Since making a will or granting a power of attorney are more often the behaviours of
those with higher socio‑economic status, it may be this factor influencing the patterns reported in this chapter.
Wills
Broader community education to support older people to safely and effectively make a will is relevant in the
context of these data (Boersig & Illidge, 2018; Ries, 2019; Wendt et al., 2015). With appropriate financial and legal
advice and support, including for those from CALD backgrounds, more older people may take advantage of
the protection that a will can provide (ALRC, 2017; Boersig & Illidge, 2018; Senior Rights Victoria [SRV], 2017).
Research literature and case law evidences the impact of undue influence exerted upon people making a will
(see e.g. Darzins, Lowndes, & Weiner, 2009). Consideration of the experience of abuse other than abuse with
financial implications is also important because the exertion of pressure to make or change a will may not give
rise to a financial loss on the part of an older person but it may involve infringing a person’s right to choose how,
and to whom, they wish to distribute their assets (State Trustees, 2017).
Power of attorney
It is also noteworthy that research and commentary have identified the risks associated with the use of a power
of attorney to perpetrate elder abuse (Boersig & Illidge, 2018; Chesterman, 2016; Purser, Cockburn, Cross, &
Jacmon, 2018; Ries, 2019).
Analyses of elder abuse helpline data in Victoria and Queensland identified substantial reports of the abuse
or misuse or neglect of a power of attorney or enduring power of attorney (Vic. 11%; Qld 15%) (Elder Abuse
Prevention Unit, 2020; Joosten et al., 2020). Reports of older people being required to appoint a power of
attorney (Vic. 4%) or to modify an enduring power of attorney (Qld 5%) were also identified (Elder Abuse
Prevention Unit, 2020; Joosten et al., 2020).
Also described as ‘POA – facilitated financial exploitation’, this abuse may occur where the appointed person
uses their authority to access and use the assets for unauthorised, improper or illegal purposes (Jackson &
Hafemeister, 2012; Purser et al., 2018; Ries, 2019). Ries (2019, pp. 2–3) observed that powers of attorney ‘create
a situation of heightened vulnerability’ given that older people grant these powers in anticipation of future
incapacity and confer substantial authority with regard to asset management on the part of the appointed
person. Studies exploring the perpetration of abuse in this context have noted its frequency (see e.g. Clare,
Blundell, & Clare, 2011) and that it is exacerbated by socio-demographic trends identified in our SOP sample,
including wealth accumulation and the experience of disability or cognitive impairment (Ries, 2019).
Chapter 10: Legal arrangements 109
Ready access to financial instruments such as bank accounts and credit cards by appointees (see further
chapter 11) has been identified as a relevant factor in the perpetration of POA- facilitated abuse, particularly
where appointees are family members with a sense of entitlement to the older person’s assets or where there
is a lack of knowledge of their responsibilities on the part of appointees (Darzins et al., 2009; King et al., 2011;
Tilse, Wilson, White, Willmott, & McCawley, 2014; Tilse, Wilson, Settlerland, & Rosenman, 2005; Ries, 2019;
Wainer et al., 2010; Webb, 2018; Wilson, Tilse, Setterlund, & Rosenman, 2009).
Situations where the appointee has been named as a beneficiary in the older person’s will were also nominated
as relevant to the misuse of powers (Caxton Legal Centre, 2007; Purser et al., 2018). A continuum of behaviour
that depends on the existence of factors such as a lack of communication and planning, stress associated with
taking on financial responsibilities for older people, stereotypical views about family and gender roles through
to intentional predatory behaviour have been identified as relevant when considering abuse in the context of
powers of attorney (Ries, 2019). The existence of an enduring power of attorney has also been identified as
having the potential to create or exacerbate family tensions, which, in turn, may give rise to the perpetration of
this abuse (Purser et al., 2018; Ries, 2019). Abuse or misuse of an enduring power of attorney has been identified
in the analysis of elder abuse helpline data in Queensland as enabling the perpetration of other forms of abuse
such as neglect; for example, where the appointed person cancels needed home care services or prevents
visitations in residential aged care settings (Elder Abuse Prevention Unit, 2020).
To this end, recommendations have been made to educate and empower older people about powers of attorney and
to inform and support appointees to fulfil their responsibilities and to discourage misuse of authority (e.g. ALRC, 2017;
Boersig & Illidge, 2018; King et al., 2011; Purser et al., 2018; Ries, 2019; State Trustees, 2017; Tilse et al., 2005). Educating
and supporting older people to obtain independent advice and to appoint trusted people who have the skill and
capacity to appropriately undertake the role have been identified as important protective measures (Purser et al., 2018).
Measures that provide options for supported decision making rather than substituted decision making have also
been identified as potentially safeguarding against elder abuse in the Australian and international context (Office
of the Public Advocate Victoria, 2020; Roche, 2018). For example, the appointment of a supportive power of
attorney enables the appointment of a trusted person to help the older person to make their own decisions while
they still have decision-making capacity (Powers of Attorney Act 2014 (Vic.) Part 7).
The need for continuing professional education for financial and legal professionals to provide effective advice
about powers of attorney has also been highlighted (Purser et al., 2018; Ries, 2019). Given that the lack of scrutiny
of these legal arrangements has been identified as contributing to the vulnerability of older people using them
(Purser et al., 2018; Perios & Smyth, 2018), the establishment of a monitoring system has been recommended as
an important tool in mitigating abuse perpetrated in this context (ALRC, 2017; Darzins et al., 2009).
Specifically, the ALRC recommended the introduction of a national online register of enduring documents, together
with nationally consistent laws governing enduring powers of attorney (ALRC, 2017, Rec. 5-3). In addition to
improving access to legal remedies (ALRC, 2017, Rec. 5-2), the Commission recommended other safeguards against
the abuse and misuse of enduring powers including the accommodation of supported as well as substituted
decision making, enhancing witnessing requirements, clear articulation of obligations on appointed decision‑makers
and mandated record-keeping requirements (ALRC, 2017, Rec. 5-1). The Council of Attorney’s General are pursuing
a staged approach to enduring power of attorney reform to support the establishment of a mandatory national
register and access to justice arrangements (Department of Prime Minister and Cabinet, 2019).
Family agreements
As foreshadowed in the previous section, similar concerns have been raised in relation to family agreements as potential
enablers of elder abuse when exchanging assets for care, particularly where informal agreements and decisions
are made typically without legal advice (ALRC, 2017; Boersig & Illidge, 2018). Where agreements are not in writing
they may be unenforceable, with dire consequences for the older person. Where these agreements are in writing, they
are likely to be executed without legal advice and dispute resolution clauses (ALRC, 2017; Boersig & Illidge, 2018). Dire
consequences may include the older person not receiving the care that they were promised or expected or that
they are rendered homeless if the relationship with the person to whom they transferred their home breaks down
(ALRC, 2017; Board of Taxation, 2019; Boersig & Illidge, 2018; Elder Abuse Prevention Unit, 2020; Webb, 2018).
Factors that have been identified as deterring parties from executing formal family agreements include the
potential for social security and tax implications to arise when transferring property from the older person
to the party providing the care (Board of Taxation, 2019). The Board of Taxation in its Review of Granny
Flat Arrangements made recommendations to the Australian Government to address the capital gains tax
implications that arise in this context. These recommendations included exemptions for the capital gains tax
110 National Elder Abuse Prevalence Study: Final Report
events arising for the parties entering into family agreements, together with the potential for the transferee to
lose their main residence capital gains tax exemption should they later sell their home (Recommendations 1, 3
and 4) and measures to address tax implications where the family agreement involves ongoing rental payments
to the carer (Recommendation 2). In October 2020, consistent with the Board of Taxation’s recommendations,
the Australian Government announced that subject to the passage of legislation, capital gains tax will not apply
to the ‘creation, variation or termination of a formal written granny flat arrangement’ where these arrangements
are entered into due to familial or other personal relationships (Frydenberg & Sukker, 2020, p. 1).
Although formalising a family agreement can operate as a protective mechanism, it cannot remove all risks
associated with entering assets for care arrangements. Boersig and Illidge (2018) observed that if older people
are able to obtain early legal advice, this can support the provision of clearer and more timely explanations of
the risks and consequences of entering family agreements. They identified how a socio-legal model of service
delivery could reduce existing barriers by having social workers embedded in legal services and partnerships
with health services and providing outreach services where older people are located in the community (Boersig
& Illidge, 2018). This approach may support older people and their family members to give greater consideration
to the detail of the agreement, including the nature and level of care to be provided in return for assets (usually
registered legal title to real property) and mechanisms to be put in place to deal with potential changes in the
older person’s or family member’s circumstances (ALRC, 2017).
In addition to steps to support informed decision making by older people who are considering entering family
agreements and to encourage the execution of agreements in writing, access to effective dispute resolution
options has also been identified as particularly important. The ALRC has recommended that tribunals be given
jurisdiction to provide low cost, less formal and simpler, quicker and more flexible dispute resolution mechanisms
for families involved in assets for care arrangements, in addition to the existing legal and equitable remedies
(ALRC, 2017).15 A requirement for family agreements to be in writing in order to continue receipt of social
security benefits was also recommended by the ALRC.
Summary
This chapter has presented findings on the extent to which older people have adopted advance planning
mechanisms, specifically wills, powers of attorney and family agreements. It has also examined reports of
experiences of financial abuse, physical abuse, sexual abuse, psychological abuse or neglect in this context.
The data indicate that the vast majority of older people participating in the SOP had a will, and this was
particularly so for older people in the higher socio‑economic status categories and for people from non-CALD
backgrounds, with more women than men also reporting that they had a will. The data show that having a will
is associated with lower levels of abuse, with differences between reports of those with and without a will in
relation to financial abuse and physical abuse. These and other SOP data together suggest this association is
likely to be as related to socio‑economic status as it is to existence of the will itself.
In relation to enduring powers of attorney, the data show that just over half of the participants in the SOP
had made an appointment pursuant to an enduring power of attorney, with more women than men doing so.
Similar to having a will, having an enduring power of attorney was more likely for older people in the higher
socio‑economic status categories and for people from non-CALD backgrounds. Having an enduring power of
attorney is associated with lower levels of abuse, with differences between reports of those with and without a
power of attorney in relation to financial abuse, physical abuse and psychological abuse.
As with the findings in relation to having a will, these and other SOP findings also suggest that the association
between lower reports of abuse and having a power of attorney is likely to be as related to socio‑economic
status as it is related to having the power of attorney. In this regard it was noted that other empirical research
and commentary has illustrated the vulnerabilities of older people who have appointed enduring powers of
attorney and how they can be used to facilitate financial abuse.
The findings in relation to family agreements contrast starkly with those relating to wills and enduring powers of
attorney, with only a small proportion of participants in the SOP reporting that they had a family agreement. Of
note, having a family agreement was more likely for older people from lower socio‑economic status categories
and for those from a CALD background, and it was more likely to be associated with the experience of elder
abuse. When considered together with other SOP data, these findings are more likely to be associated with
socio‑economic status and to reflect other relationship dynamics than those related to the existence of a family
agreement per se.
15 This recommendation draws on the experience of the Victorian Civil and Administrative Tribunal to resolve disputes between
co‑owners of property and to make any order that the tribunal considers to be ‘just and fair’ (Property Law Act 1958 (Vic.) s 228).
Chapter 11: Giving and receiving support 111
Key messages
f Women were more likely than men to report receiving assistance with financial matters from children.
f About a quarter of SOP participants reported that a third party (mostly a spouse or child) had
access to bank accounts and PIN numbers.
f In some cases, these helping arrangements may facilitate financial abuse. One area evident in the
findings is in relation to participants who were assisted to buy, sell or manage a house.
f Among SGC participants who reported providing assistance with financial matters, only about half
reported keeping records about these transactions.
f Provision of care to older people, primarily assistance with daily activities, is common, with 80%
of SGC participants indicating they provided care to someone aged 65 or older. The predominant
relationship dynamic in this context is intergenerational.
Introduction
This chapter examines the provision of help to older people. It focuses on help in two areas. The first is providing
help in relation to financial matters, including the association between receiving such help and experiencing
financial abuse. The second is the provision of care, help or assistance to family members or others due to old
age or because of a disability, injury or illness. These findings provide further context for the findings in relation
to the perpetration of financial abuse (chapter 5) and the characteristics of people who experience abuse in
chapter 6.
The research aim that the discussion in this chapter addresses is:
y Measure the proportion of people who provide assistance to family members or friends, particularly in relation
to financial transactions and decision making, which could potentially enable abusive practices to occur.
This chapter first considers assistance with financial matters on the basis of SOP and SGC findings. This discussion
includes examining record-keeping practices. It then examines whether SOP participants who reported receiving
assistance with financial matters also reported experiencing financial abuse. The last part of the chapter sets out
findings on the extent to which SGC participants report providing care to others and the nature of this care.
112 National Elder Abuse Prevalence Study: Final Report
Figure 11.1: Survey of Older People: Proportion of participants who received each type of assistance with
financial matters a
Types of assistance received
Paying bills, budgeting 13.9
Taxes 27.2
Banking 10.3
0 20 40 60
Percentage
Notes: Based on weighted data. Unweighted sample size (n = 7,000). a For each item, a small number of cases who opted for
‘don’t know’ or refused to answer were included in the total when deriving the percentages (0.4%–1.1% across all the
items).
Chapter 11: Giving and receiving support 113
Figures 11.2 and 11.3 show that patterns in receiving various kinds of assistance were similar for older men and
women but varied according to the older person’s age.
These data show that in relation to the payment of bills and budgeting and banking, reports of receiving
assistance with these tasks increased as people aged, with higher reports in the 85 years and above age bracket
than in the 65–69 years age bracket. Paying bills and other transactions have increasingly shifted towards digital
formats and those in the higher age groups may have lower capability to manage this. There were some minor
fluctuations in the intervening age brackets.
By way of contrast, reports of assistance with managing pension or superannuation, taxes and managing
investments steadily declined with each age category. These response patterns generally held when the reports
of older women and older men were analysed separately.
Figure 11.2: Survey of Older People: Proportion of participants who received each type of assistance with
financial matters, by gender
Types of assistance received
14.1
Paying bills, budgeting
13.6
29.6
Taxes
25.2
10.7
Banking
9.9
2.6
Buying, selling or managing a business
2.1
19.6
Assistance received: one
20.3
0 20 40 60
Percentage
Gender
Men Women
Notes: Based on weighted data. Unweighted sample size (men n = 2,747, women n = 4,241).
114 National Elder Abuse Prevalence Study: Final Report
Figure 11.3: Survey of Older People: Proportion of participants who received each type of assistance with
financial matters, by age
13.0
14.5
Paying bills, budgeting 12.4
13.4
18.5
34.3
27.1
Taxes 24.6
20.3
21.8
9.0
9.7
Banking 9.5
11.4
15.5
23.3
Managing pension (or Centrelink-related 22.4
matters) or superannuation 18.4
15.6
11.2
6.0
Buying selling or managing your home 6.0
4.6
or other real estate 6.7
5.1
3.0
1.8
Buying, selling or managing a business 2.4
2.6
1.1
22.0
Managing investments such as shares, 20.0
trusts or mutual funds 21.2
16.3
15.6
0 10 20 30 40
Percentage
Age (years)
65–69 70–74 75–79 80–84 85+
Notes: Based on weighted data. Unweighted sample size (65–69 years n = 1,697; 70–74 years n = 1,927; 75–79 years n = 1,525;
80–84 years n = 1,105; 85+ years n = 746).
Chapter 11: Giving and receiving support 115
Figure 11.4: Survey of Older People: Older persons who received assistance in financial matters, who provided
assistance with financial matters, by gender
71.3
Financial planner 58.5
64.7
21.9
Partner/spouse 21.1
21.5
10.0
Son/daughter a 21.6
16.0
4.0
Other service provider 2.8
3.4
0.8
Son/daughter in-law 1.6
1.2
0.9
Other family member 1.3
1.1
0 25 50 75
Percentage
Gender
Men Women All
Notes: Weighted statistics and unweighted sample bases (men n = 1369, women n = 1,966, total n = 3,337).
a Includes biological/adopted children.
Figure 11.5 shows that reports of assistance from both financial planners and spouses/partners reduces with each
age category, with 71–72% of older people in the 65–69 years and 70–74 years age groups reporting that they
received assistance from their financial planner as compared to 45% in the 85 years and over age group. Similarly,
decreasing reports of assistance from spouses/partners with financial matters emerged, with 22% reporting that
they received this support from their spouse/partner in the 65–69 years age group as compared to 11% in those
85 years and over. Of note, the data in Figure 11.5 also show that reports of assistance provided by children and
other family members increase with age, with this response pattern present in both older men and older women
who reported receiving assistance from their children and other family members.
116 National Elder Abuse Prevalence Study: Final Report
Figure 11.5: Survey of Older People: Older persons who received assistance in financial matters, who provided
assistance with financial matters, by age
71.2
71.5
Financial planner 63.0
50.0
45.4
8.4
10.2
Son/daughter a 16.7
28.2
42.0
22.3
22.3
Partner/spouse 23.3
22.1
11.2
4.2
3.8
Other service provider 3.3
1.9
1.5
0.7
0.8
Son/daughter in-law 1.3
1.7
3.4
0.2
1.7
Other family member 0.1
1.9
3.8
0 25 50 75
Percentage
Age (years)
65–69 70–74 75–79 80–84 85+
Notes: Based on weighted data. Unweighted sample size. a Includes biological/adopted children.
Chapter 11: Giving and receiving support 117
Table 11.1: Survey of Older People: proportion of participants where someone else (not account holder) has
access to bank account or credit card by age and gender
Number of
Characteristic % participants (base)
All 27.5 7,000
Males 31.2*** 2,747
Females 24.2 4,241
All: age
65–69 years 27.4 1,697
70–74 years 28.9 1,927
75–79 years 25.4 1,525
80–84 years 25.5 1,105
85+ years 31.3 746
Males: age
65–69 years 31.9 725
70–74 years 35.1 763
75–79 years 30.5 589
80–84 years 26.9 404
85+ years 27.5 266
Females: age ***
65–69 years 23.6 969
70–74 years 23.7 1,161
75–79 years 20.9 934
80–84 years 24.1 699
85+ years 35.1 478
Notes: Based on weighted data. Unweighted sample size. The asterisks indicate that the differences in the proportions
reporting that other people had access to a PIN across the categories were statistically significant (***p < .001). A
small number of cases who opted for ‘don’t know’ or refused to answer were included in the total when deriving the
percentages (<1%).
Figure 11.6 shows that partners/spouses were the most likely nominated parties to have Personal Identification
Number (PIN) access to the bank accounts or credit card accounts of participating older people (69%). A higher
proportion of older male participants reported this arrangement (82%) than did older female participants (54%).
Older women were more likely to report their children having access to their PIN (42%) than were older men
(14%). The table of all relationship types is in Appendix A, Table A11.3.
Once again, as Figure 11.7 shows, while the reports of people regarding their spouse/partner’s access to their
PIN decreased with each age category (e.g. 65–69 years: 80%; 85+: 25%), the opposite response pattern is
observable in relation to the reports of children having PIN access (for e.g. 65–69 years: 16%; 85+: 66%). Again,
greater proportions of women in each age group reported their children having PIN access to their accounts
(female 85+: 80%; male 85+: 47%), whereas nearly half of the older men in the 85 years and older category
reported that their spouse had access to their PIN as compared to 9% of older women in this age bracket (see
the list of all relationship types in Appendix A, Table A11.4). It is noted that these variations were statistically
significant. The data also show that just under half of the children with PIN access to accounts were reported to
keep records, whereas just over half of spouses/partners with PIN access did so.
118 National Elder Abuse Prevalence Study: Final Report
Figure 11.6: Survey of Older People: Older persons whose PIN was accessible by someone else, who had access
to PIN, by gender of participant
82.2
Partner/spouse 54.1
69.0
14.3
Son/daughter a 41.7
27.2
1.1
Brother/sister 0.9
1.0
0.2
Son/daughter in-law 1.9
1.0
0.3
Grandson/daughter 1.6
0.9
0.9
Other family member 0.8
0.9
0 25 50 75 100
Percentage
Gender
Men Women All
Notes: Weighted statistics and unweighted sample bases (men n = 844, women n = 978, total n = 1,825). a Includes biological/
adopted children.
Chapter 11: Giving and receiving support 119
Figure 11.7: Survey of Older People: Older persons whose PIN was accessible by someone else, who had access
to PIN, by age of participant
80.4
79.1
Partner/spouse 73.5
52.7
25.4
15.9
18.1
Son/daughter a 24.6
42.6
65.9
1.9
0.2
Brother/sister 1.1
0.7
0.9
0.6
1.6
Son/daughter in-law 0.4
1.2
1.7
0.1
0.8
Grandson/daughter 1.0
1.7
2.3
0.8
0.0
Other family member 0.6
2.0
1.8
0 25 50 75 100
Percentage
Age (years)
65–69 70–74 75–79 80–84 85+
Notes: Based on weighted data. Unweighted sample size (in order of age, n = 438, 509, 367, 270, 241). a Includes biological/
adopted children.
Figure 11.9 shows that, of the 42% of participants who provided assistance with at least one financial matter,
more than half (56%) provided this assistance to parents (including step-parents). A wide range of relationships
formed the other recipients of assistance with financial matters, including parents-in-law (11%) and grandparents
(13%), followed by other relatives and friends to a less frequent extent.
120 National Elder Abuse Prevalence Study: Final Report
Figure 11.8: Survey of the General Community: Proportion of SGC participants who provided various types of
assistance with financial matters in the past 12 months
0 10 20 30 40 50
Percentage
Notes: Based on weighted data. Unweighted sample size, n = 3,400. Multiple responses could be selected.
Figure 11.9: Survey of the General Community: Participants who provided assistance with financial matters,
who received the assistance
Relationship
Parent including step-parent 55.8
Friend 6.7
Son/daughter including
5.9
step-son/step-daughter
Sibling including step-sibling 4.2
Partner/spouse 2.9
Other c 2.3
Two 11.8
0 20 40 60 80 100
Percentage
Notes: Based on weighted data. Unweighted sample size, n = 1,483 (relationship), 1,465 (number of relationships assisting),
total n = 3,400. Percentages may not total 100.0% due to rounding. Multiple responses could be selected. Excludes
don’t know/refused responses. a This option includes aunt/uncle, son/daughter-in-law, other family member, brother/
sister-in-law, grandparents-in-law. b This option includes neighbour, client/customer/patient, colleague. c This option
includes all other including a small proportion of ex-partner/spouse.
Chapter 11: Giving and receiving support 121
Where participants reported the oldest person they had assisted was over the age of 30, they were asked if
they had provided assistance with one or more specific financial matters within the last 12 months (as listed in
Figure 11.10). Half of these participants had provided assistance specifically with ‘electronic transactions such
as internet or telephone banking’ (52%) and ‘paying for something with your money first and subsequently
reimbursed’ (51%). Around one-third of these participants also provided assistance with ‘a financial form or
financial document’ (35%) and almost one-quarter reported needing to use someone else’s PIN (24%). A small
proportion (4%) ‘acted as an appointed administrator or financial manager’.
Figure 11.10 provides further detail about the types of financial assistance provided according to three age groups of
recipients the participants had helped: 30–64 years, 65–84 years and 85 years and older. Assistance recipients aged
over 85 years were more likely than assistance recipients of other age groups to be helped with having a financial
form/document signed and receiving authorisation to operate their bank account or credit card. The recipients aged
65–84 and 85 years and older were more likely than those aged 30–64 years to have been assisted with ‘needing
to make use of an existing power of attorney’, and ‘acting as an appointed administrator or financial manager’.
Figure 11.10: Survey of the General Community: Type of financial matter, by age of oldest person receiving assistance
20.3
Need to use their PIN 24.1
30.2
24.3
12.6
Help with Centrelink 13.3
payments as a nominee 17.8
14.0
79.6
At least one of the above 81.2
87.4
82.0
0 20 40 60 80 100
Percentage
Age (years)
30–64 65–84 85+ All
Notes: Based on weighted data. Unweighted sample size (30–64 years n = 293; 65–84 years n = 743; 85+ years n = 325;
all n = 1,361. Percentages may not total 100.0% due to rounding. Excludes don’t know/refused responses to either
question.
122 National Elder Abuse Prevalence Study: Final Report
By way of contrast, needing help ‘due to disability or poor health’ (24%) was the most frequently nominated
reason for assisting 30–64 year old people, and this was the highest frequency compared to other age groups
for this reason. Although only a small proportion of participants nominated ‘difficulties with reading or writing’
and ‘emotional and mental health concerns’ as reasons for assisting, this was more frequent for 30–64 year old
people than for older age groups.
Table 11.2: Survey of the General Community: Main reason the person needed help, by age of recipient assisted
Notes: Based on weighted data. Percentages may not total 100.0% due to rounding. a A small proportion of don’t know/refused
responses are excluded (1.5%, n = 21). b This question was only asked to those whose response to the question on the
oldest person they assisted was over 30. Therefore, responses covering people under 30 are excluded from this analysis.
Totals exclude ‘don’t know/refused’ responses.
Participants also reported keeping a range of records with more than two-thirds of participants keeping ‘receipts
or cash book’ and ‘bank statements, withdrawal slips or bank passport’ and more than half indicated that they
kept ‘key documents’. Multiple reporting arrangements also tended to be identified for involving or informing
the recipient or someone close to them about the assistance provided. Most participants said they answered
questions if asked, gave regular verbal updates and included the recipient or someone close to them in decision
making for key issues. Approximately three-quarters of these participants also indicated that they shared records
if requested and asked for advice on key issues.
Table 11.3: Survey of the General Community: Type of records kept and reporting arrangements where
assistance required with financial matters
Notes: Based on weighted data. Multiple responses could be selected for types of records. This question was only asked of
a
those who answered yes to assistance with any one specific financial matter in the past 12 months. b Multiple responses
could be selected for types of arrangements. This question was only asked of those who answered yes to assistance
with any one specific financial matter in the past 12 months. c Includes a small number of responses of ‘Don’t know’
or refusal.
Table 11.4 shows that the reports of participants keeping records or having reporting arrangements in place
for the person or someone close to them about the financial assistance provided, tended to increase with the
age group of the person being assisted. It is notable that keeping records and/or reporting arrangements were
most frequently in place when assisting people aged 85 years and older (cumulatively 49%). In contrast, neither
records nor reporting arrangements were in place for 67% of people aged 30–64 years old and 52% of 65–84
years old people when participants had assisted them with financial matters.
124 National Elder Abuse Prevalence Study: Final Report
Table 11.4: Survey of the General Community: Records and reporting arrangements, by age of person
receiving assistance
Notes: Based on weighted data. Percentages may not total 100.0% due to rounding. This question was only asked of those
who answered yes to assistance with any one specific financial matter in the past 12 months. Totals exclude ‘don’t know/
refused’ responses.
SGC participants were asked whether they were granted a power of attorney (POA) by someone. Table 11.5
focuses on whether record keeping and reporting arrangements differed among participants who provided
assistance in the last 12 months by whether they have a POA appointment (It is worth noting that for those
responses who were appointed pursuant to a POA, they may provide assistance with financial matters that are
not related to their POA and their record-keeping practice may not be related to their appointment as POA. Also
note that the large majority of participants who were appointed under a power of attorney indicated that they
had not used their POA responsibility). Overall, the findings suggest power of attorney arrangements seem to
encourage record-keeping arrangements; however, further encouragement of these practices is warranted.
y People appointed under a power of attorney are more likely to have kept records and/or made arrangements
on behalf of older people than people who are not appointed under a power of attorney. Of the people who
were appointed pursuant to a power of attorney and answered this question (n = 365), two-thirds reported
that they either kept records and made arrangements for the older person, that they made arrangements but
did not keep records or that they kept records but did not make arrangements (66%).
y As may be expected, the proportion of participants who reported that they did not keep records or make
arrangements was lower for participants who were appointed pursuant to a power of attorney (34% cf. 58%
of people not appointed under a power of attorney). It is noted that these data do not provide insight into the
nature and content of these records. As such, no inference can be made as to their accuracy or whether they
are sufficient in their coverage of all financial activities undertaken on behalf of the older person.
Table 11.5: Survey of the General Community: Records and arrangements, by whether appointed under a
power of attorney
The findings presented in this section extend these insights and suggest that older women are more vulnerable
to financial exploitation. Older women who received assistance with their financial matters were more likely to
report experiencing financial abuse compared to those without such assistance.
Chapter 11: Giving and receiving support 125
Figure 11.11 shows the analysis examining the levels of financial abuse reported by people who received assistance
with financial activities. Considering the data overall (men and women combined), there is no apparent
association between reports of the experience of financial abuse and receiving assistance on financial matters,
with similar proportions reporting experiencing financial abuse in both the group receiving financial assistance
and the one not receiving such assistance.
However, slightly different patterns emerged for men and women. The prevalence of financial abuse is marginally
lower for men receiving any assistance on financial matters, compared to those who did not receive any
assistance (1.8% cf. 2.4%). For women, receiving assistance on financial matters is associated with a slightly
higher likelihood of experiencing financial abuse (2.3% cf. 1.9%). It should be noted that these differences are
small. These results were not significant.
Figure 11.11: Survey of Older People: Proportion of participants who reported experience of financial abuse by
whether receiving assistance on financial matters
2.4
2.3
2.1
2.0
2 1.9
1.8
Percentage
0
Men Women All
Notes: Weighted data and unweighted sample size (men no assistance n = 1,378, assistance n = 1,369; women no assistance
n = 2,275, assistance n = 1,966; total no assistance n = 3,663, assistance n = 3,337).
Table 11.6 sets out more detailed analysis examining the extent to which receiving help with each of seven
specific activities is associated with experiences of financial abuse. The table depicts findings from three sets of
analysis: all participants, and men and women separately. It should be noted that some participants may receive
assistance with multiple financial activities.
The analysis of patterns based on the reports of all participants shows one activity was associated with a higher
likelihood of experiencing financial abuse: buying, selling or managing a home (4.7% cf. 1.9%).
Receiving assistance with the following financial activities was associated with a lower likelihood of experiencing
financial abuse: banking (1% cf. 2.2%) and managing pension or superannuation (1.3% cf. 2.3).
For men, these patterns were consistent but more apparent. Specifically, receiving assistance with all the
activities except buying, selling or managing their home or other real estate was associated with a lower
likelihood of experiencing financial abuse (0.5%–1.7% cf. 2.2%–2.6%). Men who were assisted with buying, selling
or managing their home or other real estate were three times more likely to experience financial abuse as men
who did not receive assistance with this financial activity (6.7% cf. 1.9%). By contrast, patterns for women were
less apparent.
126 National Elder Abuse Prevalence Study: Final Report
Table 11.6: Survey of Older People: Proportion of participants who reported experience of financial abuse by
whether receiving assistance on specific financial activities
Notes: Weighted data and unweighted sample sizes. The asterisks indicate that the differences in the prevalence of financial
abuse by whether or not receiving assistance for a specific activity was statistically significant based on a design-based
F test (*p < .05; **p < .01; ***p < .001).
Care was most frequently provided in an intergenerational context. Parents (including step-parents) (47%) and
grandparents (including step-grandparents) (23%) were the most frequent recipients of care. A wide range of
relationships formed the remaining people participants reported caring for. The provision of care to older people,
primarily assistance with daily activities, is common, with 80% of SGC participants who provided any care
indicating they provided care to someone aged 65 or older.
Chapter 11: Giving and receiving support 127
Table 11.7: Survey of the General Community: Provision of care and who receives their care
Recipient of care %
Whether care is provided
Yes 36.6
No 63.2
Total 100.0
If care provided, relationship between recipients and participants
Parent incl. step-parent 47.2
Grandparent incl. step-grandparent 22.5
Parent-in-law 11.0
Other relative 8.0
Unrelated person (e.g. neighbour) 4.8
Friend 4.6
Son/daughter incl. step-son/step-daughter 3.8
Sibling incl. step-sibling 3.6
Partner/spouse 3.5
Other unspecified (incl. small proportion of ex-partner/spouse) 1.3
Age of care recipients a
Care recipients aged 65+ 79.5
Number of participants (base), whether provided care 3,400
Number of participants (base), relationship with care recipients 1,245
Notes: Based on weighted data. Multiple responses could be selected. a This refers to the oldest care recipients if participant
provided care to multiple persons.
Most types of care were provided to a similar degree across the age groups of recipients. There was, however,
a significant difference in relation to ‘giving money to live on’: this was more likely for recipients aged 30–64
(38%). Financial support to older generations is relatively frequent, with more than a quarter of SGC participants
indicating they had provided money to live on to 65–84 year olds and 15% indicating they had provided this
support to those aged 85 and over.
Providing ‘help with day-to-day activities’ was the most frequent type of care provided overall (75%), with similar
patterns across age brackets. The next most common form of help was support with personal care, provided to
around a quarter of all care recipients across age brackets.
128 National Elder Abuse Prevalence Study: Final Report
Table 11.8: Survey of the General Community: Types of care provided, by age of person receiving care,
reported by participants who provided care
Notes: For each type of care, statistically significant differences across the three age groups are noted: ***p < .001.
Multiple responses could be selected.
Further analysis revealed that there were links between the provision of financial assistance and care and elder
abuse concerns for older family members or someone they knew (results are shown in Table 11.9).
y Elder abuse concerns were more prevalent among people who provided assistance with financial matters than
they were among people who did not provide such assistance (assistance to persons of any age: 25% cf. 12%;
assistance to older persons: 24% cf. 15%).
y Elder abuse concerns were more prevalent among people who reported provision of care compared to others
(care to persons of any age: 22% cf. 15%; care to older persons: 20% cf. 16%).
Table 11.9: Survey of the General Community: Proportion of participants who had concerns by providing
assistance with financial matters and providing care
Number of participants
Type of assistance % had any concerns (base)
Assistance with financial matters to persons of any age
No 11.9*** 1,917
Yes 24.9 1,483
Assistance with financial matters to persons 65+
No 14.8*** 2,331
Yes 23.5 1,069
Providing care
No 14.8*** 2,155
Yes 21.8 1,245
Providing care to persons 65+
No 16.4* 2,399
Yes 19.9 1,001
Notes: The asterisks indicate that the differences in having concerns by whether or not the participant provided specific
assistance for a specific activity was statistically significant based on a design-based F test (*p < .05; **p < .01;
***p < .001).
Chapter 11: Giving and receiving support 129
Summary
The findings in this chapter show that assistance with financial matters and care, more generally, is commonly
provided to people aged over 65. The findings indicate some gendered patterns in the management of financial
affairs that are likely to be associated with women being more susceptible to financial abuse from their children.
They also suggest that providing care affords opportunities to detect elder abuse.
In relation to financial matters, nearly half of SOP participants indicated receiving help to manage finances in the
past 12 months. This was most commonly in relation to tax, superannuation or a pension and investments.
The sources of assistance most commonly identified were a financial planner (65%) and partners and spouses
(22%). Older women were substantially more likely than older men to report receiving help with financial matters
from their children (22% cf. 10%). Analysis examining associations between receiving financial assistance and
experiencing financial abuse suggests that women appeared to be vulnerable to financial exploitation in this
context to a greater extent than men when a third party is helping them manage financial matters except in
matters relating to buying, selling and managing home or other real estates.
Furthermore, it appears that receiving assistance with buying or selling a home was associated with a higher
likelihood of experiencing financial abuse, with 5% of SOP participants who received this type of assistance
reporting financial abuse compared to 2% who didn’t receive assistance. These findings provide further context
for the findings reported in chapter 6 that show that compared with owning your own home outright, owning a
home with debt is associated with a greater likelihood of experiencing elder abuse overall (Appendix A, Table A6.1)
and financial abuse in particular (Appendix A, Table A6.2).
More than a quarter of all SOP participants indicated that a third party had access to bank accounts or credit
cards. The most frequent third party with access was a partner/spouse, and women were substantially more likely
than men to report that their children had access to their PIN. Around half of SOP participants indicated that
records of transactions were kept by the third party with PIN access.
The SGC data confirm that the provision of assistance with financial matters is common, with four in 10 SGC
participants indicating they had assisted someone aged 65 and over in this way. The most frequent types of
assistance were in relation to paying bills, budgeting and monitoring finances (73% of those who provided
assistance) and banking (63% of those who provided assistance).
Provisions of assistance with financial matters by SGC participants most commonly occurred in an
intergenerational context, with parents being the largest group of those assisted (56%).
Around one-half of SGC participants indicated they kept records or had arrangements to report to the person
they were assisting to keep track of their assistance with financial matters. Most commonly this involved
reporting to the person they were assisting (42%) rather than keeping written records (27%).
Around a third of SGC participants reported providing care to someone aged 65 years or over, most frequently
in an intergenerational context. Parents (47%) and grandparents (23%) were the most frequent receivers of care.
The most frequent type of care provided was help with day-to-day activities. Financial support (giving money to
live on) was not infrequent, with more than a quarter of SGC participants indicating they provided this assistance
to those aged 65–85 and 15% providing it to those aged 85 and over.
The findings demonstrate that SGC participants who reported providing financial assistance or care to those
aged 65 or over were also more likely to demonstrate concerns about elder abuse. These findings point to the
importance of the awareness of elder abuse among people who provide care and assistance to older people,
as means of supporting greater detection of elder abuse. They also reinforce a need to reduce social isolation,
which chapter 6 demonstrates is a serious problem for the older population, particularly those who experience
elder abuse.
130 National Elder Abuse Prevalence Study: Final Report
12 Attitudes, awareness
and understanding
Key messages
f Some socio-demographic characteristics are associated with lower levels of recognition, higher
levels of acceptance and more ageist attitudes. These insights hold to varying degrees for findings
from the SOP and SGC.
f Men and people born in non-English speaking countries are more likely to show higher levels of
acceptance of elder abuse and ageist attitudes.
f In general, psychological abuse and neglect have lower social recognition compared to physical,
financial and sexual abuse.
f There are links between levels of agreement with intergenerational support and attitudes indicative
of entitlement.
Introduction
This chapter sets out findings on the social context for elder abuse in Australia with a focus on views on elder
abuse, attitudes towards older people and attitudes to intergenerational support. These insights provide a basis
for understanding the wider social setting for elder abuse in Australia, consistent with the socio-ecological
and human rights approaches outlined in chapter 3. Further, as set out in chapter 2, identified risk factors for
experiencing and perpetrating elder abuse are attitudinal in nature.
The discussion in this chapter begins with a brief overview of the measures used to assess acceptance and
recognition of elder abuse, the prevalence of ageist attitudes and attitudes to intergeneration support. It then
sets out findings in relation to these issues for the SOP and SGC, comparing the overall patterns in findings
between these two studies. The third part of this chapter focuses on findings from the SGC to shed further light
on the association between ageist attitudes, intergenerational support and a sense of entitlement, and their
influence on the occurrence of elder abuse.
Chapter 12: Attitudes, awareness and understanding 131
The three items (items 3–5) form a scale, referred to as the Elder Abuse Acceptance scale. To generate the
scale, the item responses were first recoded (0 as strongly disagree, 25 as disagree, 50 as neither, 75 as
agree, and 100 as strongly agree); the mean of the recoded items formed the scale score. The scale has a
score range of 0–100, with higher scores indicating greater acceptance or condoning of elder abuse.
The findings from the SOP and the SGC show divergence in responses to the two statements, demonstrating
that SOP participants are more likely to hold elder abuse condoning attitudes than SGC participants. Twenty
per cent of SOP participants somewhat/strongly agreed with the statement ‘Abuse of older people can be
understandable if the person committing the abuse is under a lot of stress in their lives’, compared to 7% of SGC
participants. Twenty-five per cent of older participants somewhat/strongly agreed with the statement ‘Abuse can
be understandable if the older person is a difficult person to deal with’, compared to 7% of SGC participants.
Taken together, the findings in this section show that, overall, SGC participants are less accepting of elder abuse
than SOP participants and:
y Social norms are changing over time.
y Understanding the social norms that underlie attitudes and behaviours is potentially a driver for reducing the
occurrence of elder abuse.
y Given the low levels of acceptance of elder abuse in the SGC, the difference in views of SOP participants
warrants further consideration.
132 National Elder Abuse Prevalence Study: Final Report
75
Percentage
53.9 52.3
50
32.6
25 22.7
14.4
8.9 9.7 9.3 7.2 6.9
5.4 2.8 1.0 1.7 0.6 3.1 0.9
0
Strongly agree/agree Neither agree nor disagree Strongly disagree/disagree Refused/don’t know
Notes: Weighted data and unweighted sample size. Mean rating (based on 1 = Strongly disagree, 5 = Strongly agree) for SOP in
order of columns = 3.2, 3.1, 2.0, 2.1, 2.3). Non-responses (‘Don’t/know or refusal’) were excluded in computing mean ratings.
Figure 12.2 shows male participants to be more accepting of elder abuse compared to females (SOP means: 30.6
cf. 26.2; SGC means: 12.7 cf. 8.4) (higher scores indicating more accepting or excusable views towards elder abuse).
Taken together, the findings from SOP participants and the SGC indicate that the youngest people in the SGC
(aged 24 years or less) held more accepting views towards elder abuse compared to those aged 25–64 years
(Figure 12.3). This is evidenced by SGC participants under 25 being significantly more accepting of elder abuse
when compared to participants in the 25–64 age range (means: 14.8 cf. 8.9–10.3).
The SOP sample was more accepting towards elder abuse than the older SGC participants. People aged over
65 years had higher levels of accepting or condoning attitudes towards elder abuse as their age increased. SOP
participants in the 65–69 years range had the lowest mean score of all the age groups (22.7) while the oldest age
groups (85+ years) had the highest mean score (38.7).
Analysis also suggests that attitudes towards elder abuse were associated with socio-demographic
characteristics (see Appendix A, Table A12.1).
Findings relating to country of birth show a uniform pattern among participants. People born in non-English speaking
countries show higher acceptance of elder abuse (SOP mean = 32.2, SGC mean = 17.6) compared with those born
in English speaking countries (SOP mean = 25.4, SGC mean = 8.7) or in Australia (SOP mean = 27.7, SGC mean = 8.6).
A noteworthy point of difference between the SGC and SOP groups born overseas is that levels of acceptance
reduced significantly with longer periods of residence in Australia for general SGC participants. However, this
pattern was not evident for SOP participants with levels of acceptance remaining constant despite duration of
residence in Australia.
Chapter 12: Attitudes, awareness and understanding 133
Examination of the findings relating to education levels also indicates a consistent association with levels of
acceptance of elder abuse. SOP participants and SGC participants with an education below Year 12 showed
higher levels of acceptance (SOP mean = 31.7, SGC mean = 13.4) compared to people who had a degree or higher
(SOP mean = 22.1, SGC = 10.4). Similarly, unemployment is associated with higher levels of acceptance of elder
abuse (SOP mean = 28.1, SGC mean = 13.6) when compared to those employed (SOP mean = 22.9, SGC = 8.9).
Figure 12.2: SOP and SGC: Mean scores of scale of acceptance attitudes, by gender
35
Survey
30.6 SGC SOP
30
26.2
25
Mean score
20
15
12.7
10 8.4
0
Men Women*
Notes: Weighted data and unweighted sample size (SGC men n = 1,563, women n = 1,760; SOP men n = 2,530, women
n = 3,791). *The difference in mean scores between the category and the reference category (listed as the first
category) is a statistically significant difference at the 5% level – for both surveys. Statistical significance test is based
on bivariate regression.
Figure 12.3: SOP and SGC: Mean scores of scale of acceptance attitudes, by age
Survey
<25 14.8 SGC SOP
25–34* 10.0
35–44* 8.9
45–54* 9.2
55–64* 10.3
Years
65–69 22.7
70–74* 26.1
75–79* 30.6
80–84* 35.2
85+* 38.7
0 10 20 30 40
Mean score
Notes: Weighted data and unweighted sample size (SGC in order of age category n = 405, 535, 591, 768, 1,000; SOP in order
of age category n = 1,621, 1,781, 1,357, 955, 617). *The difference in mean scores between the category and the reference
category (listed as the first category) is a statistically significant difference at the 5% level for both surveys. Statistical
significance test is based on bivariate regression.
134 National Elder Abuse Prevalence Study: Final Report
Lower levels of household income are also associated with higher levels of acceptance for SOP participants and
SGC participants (SOP lowest household income quintile mean = 31.0 cf. middle quintile 27.9 and highest quintile:
22 and SGC lowest household income quintile mean = 13.3 cf. highest quintile mean = 6.6).
Findings from the SOP show attitudinal differences within and between groups from English and non-English
speaking countries of birth. There are lower levels of acceptance of elder abuse among participants born in New
Zealand compared to those born in Australia (mean 21 cf. 27.7). In comparison to the subsample born in Australia,
older participants born in South, East and South-East Europe (mean = 34.9) and South-East Asia (mean = 36.9)
were more accepting of elder abuse.
Similar to the findings from the SOP, findings from the SGC show associations between levels of acceptance
of elder abuse and country of birth. SGC participants born in Chinese Asia (mean = 24.3) and South-East Asia
(mean = 21.7) show higher levels of acceptance than those born in Australia (mean = 8.6). SGC participants born
in Africa also reported higher levels of acceptance of elder abuse (mean = 15.0) compared to Australian-born
participants. Consistent with the older people sampled, participants born in Australia are more accepting of elder
abuse (mean = 8.6) when compared to those born in the United Kingdom or Ireland (mean = 5.7).
Table 12.1: SOP and SGC: Mean scores of scale of acceptance attitudes towards elder abuse, by country
of birth
SOP SGC
Number of Number of
Country of birth Mean participants (base) Mean participants (base)
Australia 27.7 4,561 8.6 2,319
New Zealand 21.0* 128 13.1 94
UK/Ireland 26.2 850 5.7* 196
N&W Europe 27.6 211 .. 36
South, East, SE Europe 34.9* 260 8.5 66
Middle East, North Africa 29.0 28 13.1 57
South-East Asia 36.9* 77 21.7* 112
Chinese Asia 34.2 27 24.3* 120
South, Central Asia 30.2 46 17.6* 144
North America 24.0 45 .. 34
Africa 28.7 57 15.0* 76
Notes: *The difference in mean scores between the category and the reference category (born in Australia) is a statistically
significant difference at 5% level. Statistical significance test is based on bivariate regression.
Chapter 12: Attitudes, awareness and understanding 135
The responses were assigned to the score: 0 to ‘No’, 1 to ‘Yes, sometimes’, 2 ‘Yes, usually’, 3 ‘Yes, always’.
The sum of responses to all the items forms the scale score, and this score was recalibrated to a score
range of 0–100, with a higher score indicating a higher level of recognition. The scale was divided into
three groups based on quartiles: low recognition (lower quartile, scores of 0–79), moderate recognition
(two middle quartiles, scores of 80–97), and high recognition (upper quartile, scores of 99–100).
Participants in both the surveys were asked if they regarded a series of behaviours as abuse of an older person
using the following responses: ‘Yes, always’, ‘Yes, usually’, ‘Yes, sometimes’, ‘No’. Their responses are presented
in Figure 12.4. The following discussion focuses on the extent to which participants provided the response of
‘Yes, always’.
Items describing physically abusive behaviours and sexually abusive behaviours had the strongest recognition.
Pushing or shoving an older person was recognised as abuse of an older person most frequently (‘Yes, always’:
SOP 86%; SGC 91%). This is consistent with the National Community Attitudes towards Violence against Women
Survey (NCAS), which shows higher community recognition for physical violence compared with other types of
family violence (Webster et al., 2018, p. 41)
Although ‘talking to an older person in a sexual way when they do not want to’ was strongly recognised by the
general community (89%), a lower proportion of SOP participants (76%) recognised this as a form of elder abuse.
After physical and sexual abuse, the next strongest levels of recognition were accorded to financial abuse,
indicated with taking money from an older person without their consent (‘Yes always’: SOP 81%; SGC 86%). Some
variation in the recognition of other forms of financial abuse by SOP participants and the general community
were evident, with ‘Yes, always’ being nominated by the following proportions to:
y selling an older person’s home without their consent (SOP 76%; SGC 79%)
y not paying bills on the older person’s behalf when they said they would (SOP 75%; SGC 67%).
136 National Elder Abuse Prevalence Study: Final Report
Consistent levels of recognition for psychological abuse were evident for SOP and SGC participants. Highest
recognition was accorded to calling an older person hurtful names (SOP 78%, SGC 79%), deliberately embarrassing
an older person (SOP 73%; SGC 75%) and preventing an older person from having contact with the outside world
(SOP 68%; SGC 73%). Lowest recognition was evident for limiting contact with grandchildren (SOP 58%; SGC 49%)
and threatening to send the older person to a residential aged care facility (SOP 57%; SGC 60%).
In comparison with other forms of abuse, neglect was less well recognised by SOP and SGC participants but the
level of recognition from both was similar. Sixty-four per cent of SOP participants and 64% of SGC participants
answered ‘Yes, always’ to not providing help with personal activities such as dressing, washing, feeding when this
is normally expected or provided.
SGC SOP
Pushing or shoving an older person
91.2 86.1
4.2 4.4
Taking money from an older person 85.5 81.1
without their consent 6.8 6.8
79.4 78.0
Calling an older person hurtful names
7.9 6.0
Talking to an older person in a sexual way 89.0 76.2
when they do not want to
3.7 6.5
Selling an older person’s home without 79.3 76.0
their consent 8.4 6.5
Not paying bills on the older person’s 67.1 74.9
behalf when you said you would 12.0 6.9
0 25 50 75 100 0 25 50 75 100
Percentage Percentage
Notes: Weighted data and unweighted sample size (SGC n = 3,400; SOP n = 7,000). Data for ‘Yes, sometimes’, ‘No’ and
‘Refused/Don’t know’ not shown.
Chapter 12: Attitudes, awareness and understanding 137
Figure 12.5 shows that female participants are more likely to recognise elder abuse behaviours compared to male
participants. In relation to age (Figure 12.6), there was greater recognition of elder abuse in the 65–69 age range
(mean = 86.6) compared to older participants aged 75–79 years (mean = 82.2) and 85+ years (mean = 76.9). It is
noteworthy that a similar pattern was evident with levels of acceptance of elder abuse by SOP participants. Elder
abuse recognition tended to increase with the age of the community participant, ranging from a mean of 80.3 for
those under 25 years to 88.5 for 55–64 year olds.
The findings indicate some other socio-demographic characteristics were also associated with views on elder
abuse behaviours (full results in Appendix A, Table A12.2). Place of birth was associated with similar levels
of recognition of elder abuse for SOP participants and the general community. Compared with those born
in Australia (SOP mean = 84.1, SGC mean = 86.4), people born in English speaking countries show greater
recognition of elder abuse (SOP mean = 86.8, SGC mean =89.5), whereas those born in non-English speaking
countries show lower recognition (SOP mean = 81.7, SGC = 78.7). People living with a partner show higher levels
of recognition of elder abuse (SOP mean = 84.4, SGC mean = 86.5) compared to those not living with a partner
(SOP mean = 83.2, SGC mean = 82.8).
Some variation in levels of recognition of abuse are associated with education: lower recognition was evident
among SOP participants with an education at Year 12 (mean = 82.8) or below Year 12 attainment (83.1) compared
with those who had a degree or higher (86.5). In contrast, SGC participants who had a certificate/diploma/trade/
other education showed higher levels of recognition compared to SGC participants with a degree or higher
(mean = 87.4 cf. 84).
The findings in relation to income are similar. Older people and SGC participants in lower income households
show lower levels of recognition of elder abuse compared to those in higher income households.
Figure 12.5: SOP and SGC: Mean scores of scale of recognising abusive behaviours, by gender
Survey
SGC SOP
100
87.6 85.9
82.5 82.0
80
60
Mean score
40
20
0
Men Women*
Notes: Weighted data and unweighted sample size (SGC men n = 1,587, women n = 1,778; SOP men n = 2,662, women
n = 4,134). *The difference in mean scores between the category and the reference category (listed as the first
category) is a statistically significant difference at the 5% level for both surveys. Statistical significance test is based on
bivariate regression.
138 National Elder Abuse Prevalence Study: Final Report
Figure 12.6: SOP and SGC: Mean scores of scale of recognising abusive behaviours, by age
<25 80.3
25–34* 84.0
35–44* 85.8
45–54* 87.0
55–64* 88.5
Years
65–69 86.6
70–74* 86.5
75–79* 82.2
80–84* 81.0
85+* 76.9
0 25 50 75 100
Mean score
Survey
SGC SOP
Notes: Weighted data and unweighted sample size (SGC in order of age category n = 409, 542, 603, 773, 1,013; SOP in
order of age category n = 1,659, 1,891, 1,490, 1,066, 699. *The difference in mean scores between the category and the
reference category (listed as the first category in each survey) is a statistically significant difference at the 5% level.
Statistical significance test is based on bivariate regression.
Table 12.2: SOP and SGC: Mean scores for recognition of abusive behaviours, by country of birth
SOP SGC
Number of Number of
Country of birth Mean participants (base) Mean participants (base)
Australia 84.1 4,869 86.4 2,338
New Zealand 88.3* 132 88.9* 95
UK/Ireland 86.5* 923 90.5 197
N&W Europe 86.5 228 .. 36
South, East, SE Europe 82.8 308 86.6 70
Middle East, North Africa .. 36 85.8 61
South-East Asia 67.7* 78 76.4* 117
Chinese Asia 62.0* 29 69.6* 120
South, Central Asia 82.5 56 81.3* 147
North America 89.6* 44 .. 34
Africa 85.5 60 83.8 78
Notes: *The difference in mean scores between the category and the reference category (born in Australia) is statistically
significant difference at 5% level. Statistical significance test is based on bivariate regression.
Chapter 12: Attitudes, awareness and understanding 139
Older participants born in Australia showed higher levels of recognition of abuse (mean = 84.1) when compared
to older participants born in South-East Asia (mean = 67.7). However, older participants born in New Zealand
(mean = 88.3) and the United Kingdom/Ireland (mean = 86.5) showed higher levels of recognition of elder abuse
when compared to older participants born in Australia (mean = 84.1).
The findings for SGC participants show patterns generally consistent with the findings from the SOP. The highest
levels of recognition of elder abuse reported was by SGC participants born in the United Kingdom/Ireland
(mean = 90.5). Compared to SGC participants born in Australia (mean = 86.4), those born in Chinese Asia,
South‑East Asia and South, Central Asia showed less recognition of elder abuse (69.6, 76.4 and 81.3 respectively).
A report recently released by the World Health Organization ([WHO], 2021) found that ageism is widespread in
many parts of the world and it extends to various institutions and sectors of society. Attitudes towards older people
vary across countries and cultures. The WHO report found that regions such as the South-East Asian and Western
Pacific regions held the most negative attitudes towards older people, while those in anglophone cultures and
the European region hold the most positive views towards older people of the regions examined (pp. 32–33).
This section expands on the attitudinal findings previously presented by setting out findings from the general
community sample relating to attitudes of ageism, intergenerational support and sense of entitlement.
(See Box 12.3 for measures used in this analysis.)
The ratings to items 1, 2 and 5 were reversed. The mean of responses to the six items formed composite
scales as a continuous variable and this score was then recalibrated to a score range of 0–100. Higher
scores indicate higher levels of ageism towards older people.
16 It is worth noting that the original two subscales were positively correlated (r = .62). Factor analysis on these six items revealed only
one factor rather than two factors (or subscales). Given that this survey had a different target population in a different context and
the reduced number of items, it is not surprising that one factor emerged from the data. The reliability for the six items was robust
(Crobach’s alpha = .77).
140 National Elder Abuse Prevalence Study: Final Report
Ageism
Table 12.3 presents findings in relation to the items used to derive understanding of the extent of prejudiced
and discriminatory attitudes towards older people in the SGC sample. The majority of general SGC participants
reported low levels of ageist attitudes towards older people. An attitude of ageism towards older people was
indicated by a minority of general SGC participants who provided ratings of 7–10 (8–33%). Mean ratings across all
six items ranged from 2.2 to 4.9 (on a 0–10 scale) and the overall mean score of the ageism scale was 37.4 out of
100 (with higher ratings indicating higher levels of ageism towards older people).
The findings show that the highest level of agreement was in response to the statement ‘It is helpful to repeat
things to older people because they rarely understand the first time’ (mean = 4.9) and the lowest level of
agreement showed in response to the statement ‘Older people are a drain on the health care system and the
economy’ (mean = 2.2).
Slightly less than one-third of SGC participants expressed agreement with ‘It is helpful to repeat things to older
people because they rarely understand the first time’ and ‘Older people need to be protected from the harsh
realities of society’. Twenty-five per cent of SGC participants agreed ‘It is good to tell older people that they are
too old to do certain things; otherwise they might get their feelings hurt.’
Table 12.3: Survey of the General Community: Ageism scale mean ratings and distribution of item scores
Although the overall level of ageism evident in the SGC sample was low, some differences across these
characteristics were evident:
y Higher ageist levels present among male participants (mean = 39.2) compared with females (mean = 35.7).
y Younger SGC participants showed higher ageism levels than older participants with the highest being for
participants in the youngest age group (<25, mean = 39.9) and the lowest ageism level being for the oldest
age group (55–64 years, mean = 35.9). The 35–44 age group (mean = 36.9) and the 45–54 age group
(mean = 36.7) also showed lower levels of ageism when compared to SGC participants under 25 years of age.
y SGC participants born in non-English speaking countries showed higher levels of ageism (mean = 51) than
those born in Australia (mean = 33.3) and English speaking countries (mean = 34.9).
y Higher ageism levels were present for people not living with a partner (mean = 39) than those living with a
partner (mean = 36.3).
y SGC participants with no partner or children (mean = 39.8) showed higher ageism levels compared to those
partnered and with child/ren (mean = 37).
y SGC participants with an education below Year 12 (mean = 42.5) showed higher levels of ageism than
participants with a degree or higher (mean = 35.7).
y SGC participants not employed (mean = 41.8) showed higher levels of ageism than people employed full-time
(mean = 35.9).
y Ageism levels were significantly higher as participants’ levels of personal income lowered (lowest personal
income quintile 40.2 cf. highest personal income quintile 30.2).
y Ageism levels were significantly higher as participants’ levels of household income lowered (lowest household
income quintile mean = 40.8 cf. highest household income quintile mean = 29.6).
y Higher ageism levels also showed for SGC participants who rented from a public housing authority
(mean = 49.4), rented from a private landlord (mean = 40.3), or lived in other housing arrangements
(boarding, living at home, etc.) (mean = 39.8) compared to those who owned their home outright
(mean = 35.4).
y SGC participants in the lowest socio‑economic quintile (mean = 40.5) showed the highest levels of ageism
compared to those in the highest socio‑economic quintile (mean = 33.7).
y Higher levels of ageism showed for SGC participants who lived outside of a capital city (mean = 35.8 cf. 38.1
for capital city).
The patterns in the attitudes towards older people in relation to age and gender were consistent with the
research on ageism based on an online survey conducted by The Benevolent Society (2017). This research showed
that men were less likely to hold positive attitudes towards ageing and older people than women (p. 26). Further,
negative views towards ageing and older people declined with the increasing age of the participants (p. 111).
Figure 12.7: Survey of the General Community: Bivariate analysis of the Ageism scale by gender – Mean score
40 39.2
35.7
30
Mean score
20
10
0
Men Women*
Notes: Weighted data. Unweighted sample size (men n = 1,597, women n = 1,783). * The difference in mean scores between
the category and the reference category (listed as the first category) is a statistically significant difference at the 5%
level. Statistical significance test is based on univariate regression.
142 National Elder Abuse Prevalence Study: Final Report
Figure 12.8: Survey of the General Community: Ageism scale by age – Mean score
40 39.9
37.9 36.9 36.7 35.9
30
Mean score
20
10
0
<25 25–34 35–44* 45–54* 55–64*
Years
Notes: Weighted data. Unweighted sample size (In order of age category n = 408, 545, 606, 775, 1,021). * The difference in
mean scores between the category and the reference category (listed as the first category) is a statistically significant
difference at the 5% level. Statistical significance test is based on univariate regression.
The findings also show higher levels of ageism for those born in South and East Europe, Africa and the
Middle East and North Africa. Noting that the composition of Australia’s migrant intake has changed over time,
these results may be related to the length of time in Australia or age of the participant.
Table 12.4: Survey of the General Community: Ageism scale by country of birth
Notes: * The difference in mean scores between the category and the reference category (listed as the first category) is a
statistically significant difference at the 5% level. Statistical significance test is based on bivariate regression.
Chapter 12: Attitudes, awareness and understanding 143
Intergenerational support
Views regarding intergenerational support and sense of entitlement are shown in Table 12.5. There was general
agreement with intergenerational support among SGC participants (mean = 66.5). Regarding individual items,
the highest level of agreement was related to supporting parents financially (mean = 7.3) and with living with
ageing parents (mean = 7.0). In contrast, the lowest level of agreement related to adult children being supported
financially and with co-residence (mean = 5.5 and 6.8 respectively).
Sense of entitlement was measured using a single item and the findings are presented alongside the
findings on intergenerational support. Response options to the item were on a scale ranging from
0 (strongly disagree) to 10 (strongly agree) with higher levels indicating a stronger sense of entitlement.
Participants were asked to respond to the following statement:
y Other family members are entitled to some of the older person’s assets if they have assisted the older
person on a regular basis.
The findings on the general strong endorsement of intergenerational support were consistent with the work
of Weston and Qu (2016) that examined attitudes towards intergenerational support based on a sample aged
18 years and older.18 Their work showed most participants agreed with the provision of financial support to
ageing parents as well as to adult children. However, the authors indicated that there was greater support for
parents letting adult children live with them than the other way around. The SGC data showed similar but slightly
greater support for adult children letting their ageing parents live with them. While this difference may be
attributed to different methodologies between the SGC and the data on which this previous research was based
(e.g. sampling), further research could shed light on any change in attitudes towards intergenerational support in
the Australian context.
Table 12.5: Survey of the General Community: Distribution and mean ratings of intergenerational support and
sense of entitlement items
17 Four items capture views on intergenerational support. These items originated in the Australian Life Course Study conducted by AIFS
in 1997 and included in the Australian Survey of Social Attitudes (2012).
18 The research was based the survey Australian Survey of Social Attitudes (AuSSA) conducted in 2012 and the sample was recruited
from the Australian Electoral Roll.
144 National Elder Abuse Prevalence Study: Final Report
Analysis indicated that attitudes towards intergenerational support varied according to socio-demographic
characteristics (see Appendix A, Table A12.4). Some differences are evident in the extent to which SGC
participants agreed with intergenerational support:
y Country of birth was also associated with intergenerational support. SGC participants born in non-English
speaking countries showed higher levels of agreement with intergenerational support (mean = 73.2) than
those born in Australia (mean = 64.9).
y Participants who first arrived in Australia less than five years ago showed stronger intergenerational support
(mean = 75.2) compared to those who had been in Australia 15 years or more (mean = 64.7).
y SGC participants who identified with another religion (mean = 67.1) or no religion (mean = 73.4) showed
higher intergenerational support than participants who identified as Christian (mean = 64.2).
y SGC participants not living with a partner showed higher intergenerational support (mean = 68.1) compared to
those living with a partner (mean = 65.6).
y SGC participants living with a partner but no children showed the least intergenerational support
(mean = 62.7) compared to those partnered with child/ren (mean = 67.3).
y SGC participants who had low personal income levels showed higher levels of intergenerational support
(mean = 67.3) than those on the fourth highest household income quintile (mean = 63.9).
y SGC participants who had low household income levels showed higher levels of intergenerational support
(mean = 67.6) than those on the fourth highest household income quintile (mean = 63.7).
y SGC participants who owned their home (mean = 63.2) showed less intergenerational support when
compared to participants living in other types of housing situations and particularly those renting from a
public housing authority (mean = 74.6).
y Across geographic areas, participants living in a capital city showed higher intergenerational support
(mean = 67.6) than those participants living in the rest of the state (mean = 64).
Chapter 12: Attitudes, awareness and understanding 145
Figure 12.9: Survey of the General Community: Bivariate analysis of the Intergenerational Support scale by
gender – Mean score
75
67.4 65.6
50
Mean score
25
0
Men Women*
Notes: Weighted data. Unweighted sample size (men n = 1,597, women n = 1,777). * The difference in mean scores between
the category and the reference category (listed as the first category) is a statistically significant difference at the 5%
level. Statistical significance test is based on univariate regression.
Figure 12.10: Survey of the General Community: Bivariate analysis of the Intergenerational Support scale by
age – Mean score
75
70.3
68.6 66.9 65.5
61.2
50
Mean score
25
0
<25 25–34 35–44* 45–54* 55–64*
Years
Notes: Weighted data. Unweighted sample size (In order of age category n = 409, 546, 603, 773, 1,016). * The difference in
mean scores between the category and the reference category (listed as the first category) is a statistically significant
difference at the 5% level. Statistical significance test is based on univariate regression.
Figure 12.11: Survey of the General Community: Intergenerational Support Scale by country of birth – Mean score
Australia 64.9
Africa 67.9
0 25 50 75 100
Mean score
Notes: Weighted data and unweighted sample size (in order of bar, n = 2,339, 96, 197, 36, 72, 61, 121, 123, 148, 33, 78).
* The difference in mean scores between the category and the reference category (listed as the first category) is
a statistically significant difference at the 5% level. Statistical significance test is based on univariate regression.
Table 12.6: Survey of the General Community: Correlations between the Ageism scale, Intergenerational Support
scale and agreement/disagreement with entitlement of older persons’ assets for provision of regular assistance a
Agreement
rating with assets
Intergenerational entitlement for
Ageism scale Support scale regular assistance
(0–100) (0–100) (0–10)
Ageism scale 1.00
Intergenerational Support scale 0.29 1.00
Sense of entitlement 0.37 0.28 1.00
Note: a The statement was: ‘Other family members are entitled to some of the older person’s assets if they have assisted
the older person on a regular basis’.
Chapter 12: Attitudes, awareness and understanding 147
Further analysis was carried out to explore how attitudes toward older people are associated with attitudes
toward elder abuse. Table 12.7 shows:
y More ageist views are associated with more condoning attitudes towards elder abuse (r = .35).
y Higher levels of ageist attitudes towards older people are associated with lower levels of recognition of elder
abuse behaviours (r = -.28).
y Associations between attitudes towards elder abuse and intergenerational support were negligible.
y Associations between attitudes toward elder abuse and views on assets entitlement for regular assistance
were also negligible.
Table 12.7: Survey of the General Community: Correlations between attitudes toward elder abuse and the
Ageism scale, Intergenerational Support scale and agreement/disagreement with entitlement of older persons’
assets for provision of regular assistance a
Agreement
rating with assets
Intergenerational entitlement for
Ageism scale Support scale regular assistance
(0–100) (0–100) (0–10)
Elder abuse acceptance scale (0–100) 0.35 0.11 0.18
Elder abuse recognition scale (0–100) -0.28 -0.03 -0.13
Note: a The statement was: ‘Other family members are entitled to some of the older person’s assets if they have assisted the
older person on a regular basis’.
Summary
This chapter has examined the social values and norms that provide the context for elder abuse in Australia.
The chapter has specifically examined acceptance and recognition of elder abuse among older people and the
general community. It has also examined the prevalence of ageist views, as well as attitudes to intergenerational
support, among the general community. The analysis has highlighted variations in these areas connected to some
socio-demographic characteristics.
Overall, participants were not accepting of elder abuse but when subgroups in the sample were examined,
findings showed male participants and particularly those born in non-English speaking countries to be more
accepting of elder abuse than older females in the SOP. This gender difference held for the SGC.
Age of participants played a role in acceptance of elder abuse. Findings from the SOP sample showed increasing
acceptance of elder abuse as age increased whereas in the SGC sample, the youngest age group showed greater
acceptance of elder abuse.
SOP and SGC participants born in non-English speaking countries were more likely to be accepting of elder
abuse than those born in English speaking countries or Australia.
For both samples, particular demographic and socio‑economic characteristics were associated with a higher
degree of acceptance of elder abuse. These include identifying with lower education level, unemployment (or
retirement), low income and home ownership or rental.
When exploring the recognition of elder abuse subtypes, SOP and SGC participants were more likely to identify
financial and physical abuse when compared to psychological abuse and neglect. A strong proportion of SGC
participants also recognised sexual abuse. This was not the case with the SOP sample. This may be part of the
explanation for the under-reporting of this subtype of abuse (chapter 9).
SOP and SGC participants born in English speaking countries showed greater recognition of elder abuse
compared to those born in Australia and non-English speaking countries.
The findings in relation to ageism are generally consistent with the patterns emerging in the findings for
acceptance and recognition. Males showed a greater tendency toward ageist attitudes than females and younger
people were associated with greater ageist attitudes towards older people. Participants born in non-English
speaking countries showed greater ageist attitudes compared to those born in English speaking countries and in
Australia. These findings were similar to the factors that the WHO report (2021) identified as factors associated
with attitudes consistent with ageism.
148 National Elder Abuse Prevalence Study: Final Report
Particular demographic and socio‑economic characteristics were associated with a greater ageist attitudes and
these included lower education level, unemployment, low income and socio‑economic contexts.
SGC participants generally showed positive attitudes towards intergenerational support with regard to
supporting parents financially and with co-residence. Stronger acceptance of intergenerational support was
demonstrated by participants born in non-English speaking countries.
Intergenerational support was also associated with a stronger belief in family members being entitled to an older
person’s assets in return for their provision of regular assistance.
The findings presented in this chapter are consistent with the current directions in approaches to preventing
family violence premised on a socio-ecological approach (Cripps et al., 2019). As with greater recognition of
physical forms of family violence (Webster et al., 2018), this chapter establishes greater recognition of physical
forms of elder abuse and lower recognition of other forms such as neglect.
Findings about the lower recognition and awareness of elder abuse among some groups, including older men,
are consistent with findings in chapter 7 about men, and people in the 65–74 years age group being significant
among perpetrator groups. Similarly, these patterns provide further context for understanding findings about
men, in particular, being less inclined to seek help and advice about elder abuse in chapter 8.
A range of demographic factors extending beyond age and gender need to be considered in understanding
the possible drivers of elder abuse. Economic and social marginalisation along with cultural background
are characteristics of the people sampled that are associated with less consistent attitudes towards to the
recognition and disapproval of elder abuse. Similarly, findings from the National Community Attitudes Towards
Violence Against Women Survey (NCAS) showed that country of birth/length of time in Australia and English
language proficiency are the two strongest demographic predictors of attitudes supportive of violence against
women (Webster et al., 2018).
In summary, consistent with the direction of violence against women research (McCook & Powell, 2020),
understanding the normative processes underlying the social norms highlighted in this chapter and how
they influence family and intergenerational dynamics is important. These findings point to the need for the
development of an evidence-based prevention framework. They also suggest a need for further service
development to effectively respond to older people experiencing elder abuse.
Chapter 13: Elder abuse and CALD groups: synthesis 149
Key messages
f Abuse relating to language and culture was reported by 4% of the CALD subsample, with the most
commonly reported behaviour being ‘not respecting you when talking to you because of your culture’.
f CALD participants were more likely to identify son/daughter in-law, siblings and friends as perpetrators.
Introduction
This chapter draws together the findings of the CALD substudy from previous chapters to provide a synthesis
of the results in relation to the CALD subgroups in the SOP and SGC. These findings are based on the responses
of 608 SOP participants who spoke a language other than English at home, and 660 SGC participants who
were born in a non-English speaking country.19 For the SOP CALD subgroup, 85% were also born outside of
Australia in addition to speaking a language other than English at home. As such, these findings shed light on the
experience of elder abuse for people from non-English speaking backgrounds.
In the context of the methodology applied in this research, there are a number of issues that need to be taken
into account when considering the findings summarised in this chapter. First, each of the cultures represented
in the CALD subgroup has its own unique characteristics, including potentially diverse characteristics within
the subgroup. Comparisons between CALD subgroups in the SOP are not amenable to examination in this
methodology, largely due to small sample sizes. There are some limited comparisons available in the SGC findings
based on country of birth where sample sizes permitted.
Second, the social context for these findings is unique internationally, given the socio-demographic
characteristics of the CALD population in Australia and the varied make-up of the immigrant population. It
ranges from migrants from post-war Europe to recently arrived humanitarian migrants from countries such as
Iraq, Democratic Republic of Congo and Myanmar (Department of Home Affairs, 2019).
Third, this research has demonstrated that a range of issues, including socio‑economic issues, influence the
prevalence of elder abuse. It is therefore difficult to disentangle the influence of culture from other influences,
including the psycho-social and economic consequences of immigration and the experiences that preceded it. As
noted in chapter 4, it is likely that these samples under-represent people with limited proficiency in English, even though
146 SOP participants and 83 SGC participants were interviewed in languages other than English. For these reasons,
caution should be applied in interpreting the findings, particularly in attributing the findings to CALD status per se.
However, it is also evident that the findings in this chapter provide new insight into the way that being from a
CALD background influences the experience of elder abuse. Findings in relation to the CALD-specific abuse
subtype, abuse relating to language and culture, describe abuse dynamics arising particularly from CALD status.
This is evident in findings that this type of abuse is perpetrated less by family members (who are likely to share
the same cultural background as the victim) and more by friends, neighbours and acquaintances. This places a
focus on the intercultural dimensions of elder abuse, suggesting that being of CALD background is associated
with vulnerability to additional types of abuse. This indicates it is important to consider the influence of racist
attitudes, in addition to ageist attitudes, on the way CALD groups experience elder abuse.
With the exception of financial abuse, the prevalence of different types of abuse were similar for the CALD
subsample and the non-CALD sample (Table 5.7).
In relation to financial abuse, CALD participants were slightly less likely to report experiencing financial abuse
(1.6%) compared to the non-CALD sample (2.1%). This finding contrasts with some previous research that is
mainly qualitative, which has suggested that people from CALD groups may be at higher risk of financial abuse
because of language barriers (see summary in Dow & Brijnath, 2019). Findings in other areas, however, suggest
that some people from CALD backgrounds might be more susceptible to particular kinds of financial abuse.
Greater use of family agreements, especially among CALD older women, was reported in chapter 10, raising
concerns about abuse in this particular context (Boersig & Illidge, 2018).
As with the whole sample, psychological abuse was the most commonly reported of the five subtypes for the
CALD participants (11.9%, cf. 11.7% whole sample).
Reports of physical abuse among the CALD subsample (1.6%) were almost on par with prevalence rates in the
non-CALD sample (1.8%). The prevalence of sexual abuse for CALD and non-CALD participants was the same
(1.0%). Reports of neglect were also very similar between CALD (2.6%) and non-CALD (2.9%) participants.
The CALD-specific abuse subtype, abuse relating to language and culture, was reported by 4% of the CALD
subsample. For this subtype of abuse, the most common abuse item was ‘not respecting you when talking to you
because of your culture, race or ethnicity’ (67%). The next most common subtype item was being ‘made to feel
like you are just free labour’, reported by 45%.
The analysis of abuse subtypes and overall (experience of at least one form of abuse) by country of birth
(Australia, English speaking countries, non-English speaking countries) (Tables 6.2–6.8) indicates similar
prevalence patterns.
There is one area where a difference was evident: living with others was associated with a higher risk of
experiencing elder abuse for the non-CALD sample (22.8%) but this was not evident for CALD sample (14.1%).
It is worth noting that it is much less common for older people, in a community setting, to live with persons other
than their spouse/partner (less than 8%). The arrangements concerning living with others are diverse (e.g. with
adult children, other relatives, non-family members). As such, this finding should be treated with caution.
For abuse relating to language and culture, the socio-demographic characteristics associated with these
experiences are consistent with those outlined above. Prevalence of this kind of abuse was higher in the
younger age bracket (65–74 years: 5.4%) than the older age bracket (75–84 years: 2.1%). Living with children
is also associated with a slightly higher likelihood of experiencing this kind of abuse, with 4.4% of those living
with children reporting abuse relating to language and culture compared with 3.9% of those in other living
arrangements (Table 6.11).
Comparisons within the CALD subsample for CALD related abuse according to country of birth were not
possible due to small sample sizes.
However, for the CALD group, there are some differences in the types of family members identified as
perpetrators when compared to the non-CALD group. Even though children form a substantial proportion of all
main perpetrators overall, fewer children are perpetrators compared to the non-CALD sample (12.5% cf. 18.0%).
A slightly higher proportion of the CALD sample identified sons- and daughters-in-law as the main perpetrator
(11.3%, cf. 7.4% non-CALD). Furthermore, CALD participants identified their siblings as the main perpetrators at
a higher rate (CALD brother/sister 7.1% cf. 3.4% non-CALD), and they were also more likely to identify friends as
perpetrators (17.6% cf. 11.6% non-CALD).
The lower frequency of children as perpetrators compared to non-CALD participants and lower risk of elder
abuse when living with others are likely to reflect the different perpetrator dynamics evident for abuse relating to
language and culture compared with the other abuse subtypes. This type of abuse is not committed by children
and is less likely to be committed by other family members as well (8.8%). Instead, the main perpetrators of
this type of abuse are friends (30.6%), acquaintances (16.6%) and neighbours (11.2%) (Table A7.8). This finding
highlights the need to account for racism in the community, and the older person’s experience of racism, when
developing interventions to address elder abuse relating to language and culture.
In addition to the influence of elder abuse relating to language or culture in the findings above about
perpetrators, there are further considerations to explore about the frequency of children and other family
members as perpetrators of elder abuse with a different profile in this respect for the CALD subgroup compared
with the non-CALD subgroup.
First, the differences in the CALD subgroup identifying children as perpetrators do not appear to be a
consequence of living arrangements involving children. Living with children is not common for older people
overall, but somewhat greater proportions of CALD participants in the SOP live with children than the whole
sample (see Table 4.1 and Table 4.3 for weighted sample characteristics). This is particularly so for CALD couples
living with children (13.4% cf. 7.4% whole sample) but not substantially different to the Australian population
(15.5% Census CALD couples living with children).
Second, there may be complex intergenerational and extended family cultural dynamics potentially associated
with the different types of family members identified as the main perpetrator of elder abuse for the CALD
subgroup compared to the non-CALD group. Further research is needed to understand these family dynamics
and the extent to which demands and expectations upon the older person may be based in cultural norms.
Previous research has highlighted varying values about whether resources are considered communal or personal
within a family, and it is also evident that expectations are culturally determined (Miskovski, 2014; Wainer
et al., 2010). In terms of obligations toward older family members, international research draws attention to
the influence of ‘filial piety’ in East Asian cultures, which, it is argued, has the potential to create elder abuse
conducive dynamics by imposing unwanted care obligations on younger family members (Ho et al., 2017).
The somewhat greater presence of sons- and daughters-in-law and siblings as perpetrators for the CALD
subgroup raises the need for greater understanding of perpetrator dynamics, and how these might intersect
152 National Elder Abuse Prevalence Study: Final Report
with other perpetrator problems, such as mental health and financial problems (Table 7.2). For example, in
their small qualitative study Vrantsidis and colleagues (2016) reported that some participants from culturally
and linguistically diverse backgrounds experienced negative views about women associated with particular
cultures and strong views about parental and family obligations. A World Health Organization review (2002) has
highlighted particular issues that arise for mothers-in-law in some cultures, such as those in India, Austria and
Lebanon, that lead to higher rates of abuse against mothers-in-law.
A similar difference between the overall patterns was evident in relation to powers of attorney, with just over half
of the non-CALD sample having a power of attorney, compared with four in 10 of the CALD subsample.
Patterns in who was most frequently appointed as an attorney did not differ between the CALD subsample and
the whole sample, with about 70% in each sample appointing children. Family agreements were more common
among the CALD subsample, with 4.2% having them compared with 2.7% in the non-CALD sample (Appendix A,
Table A10.2).
SOP participants born in countries where English is not the primary language more frequently reported attitudes
indicative of accepting elder abuse than those born elsewhere, to a significant extent (mean 32.2 cf. 25.4
English‑speaking countries, 27.7 born in Australia). In relation to the acceptance of elder abuse and specific
regions of country of birth, SOP participants from South-East Europe (mean 34.9) and South-East Asia
(mean 36.9) were more likely than Australian-born participants to hold attitudes indicative of the greater
acceptance of elder abuse (mean 27.7, Table 12.1).
These findings were broadly consistent in the SGC sample in relation to region of country of birth. Participants
born in non-English speaking countries reported significantly greater acceptance of elder abuse (mean 17.6)
compared to those born elsewhere (8.7 English speaking countries, 8.6 Australia). Within specific regions, SGC
participants born in South-East Asia (mean 21.7) indicated greater elder abuse acceptance, along with those from
Chinese Asia (mean 24.3) and South and Central Asia (mean 17.6). In comparison, those SGC participants born
in Australia had a substantially lower acceptance score at 8.6. Notably, the findings show a consistent pattern
of lower acceptance of elder abuse among those in the SGC (up to 64 years of age) compared with the SOP
(age 65 and older), suggesting that increasing age and cultural factors may influence negative attitudes that are
more condoning of elder abuse.
Regression modelling using the SGC data showed that the greater acceptance of elder abuse for those born in
Asian countries held after controlling for gender, age and time since arrival in Australia. This means that cultural
experiences continue to be a significant factor in acceptance of elder abuse even after other socio-demographic
characteristics are accounted for.
Chapter 13: Elder abuse and CALD groups: synthesis 153
In the SGC, patterns were generally consistent in terms of differences among subsamples born in Australia
(mean 86.4), other English-speaking countries (mean 89.5) and non-English speaking countries (mean 78.7,
Appendix A, Table A12.2). Lower mean scores were evident for SGC participants from Chinese Asia (mean 69.6),
South‑East Asia (mean 76.4) and South and Central Asia (mean 81.3) compared with those born in Australia
(mean 86.4, Table 12.2). There was greater recognition of elder abuse among SGC participants compared with
SOP participants overall, although the differences were not as substantial between the surveys as they were for
acceptance of elder abuse.
Ageism (SGC)
Attitudes consistent with ageism were evident among some CALD subgroups in the SGC (this was not examined
in the SOP, Appendix A, Table A12.3). SGC participants born in Australia had a mean score on the ageism scale
of 33.3 (with lower scores indicating less ageist attitudes), whereas participants born in non-English speaking
countries indicated significantly higher scores (mean 51.0). For some specific regions, average ageism scores
were also higher compared to Australian-born participants, particularly for those born in Chinese Asia and
South and Central Asia (mean 55.7 and 55.6 respectively), South-East Asia (mean 52.7) and the Middle East and
North Africa (mean 51.6, Table 12.4).
A simple regression analysis showed that there were still significant differences by country of birth after
controlling for other socio-demographic characteristics, including gender, age and time in Australia. For specific
regions, this pattern of results was upheld for those born in Asian countries after controlling for gender, age and
time since arrival in Australia. Ageism is therefore highly likely to be associated with cultural experiences, which is
also consistent with the association between the acceptance of elder abuse and country of birth.
A regression model applied to the Intergenerational Support scale score showed that country of birth was still
a significant predictor after controlling for gender, age and time since arrival in Australia. This means cultural
experiences continue to be an important influence on positive agreement with intergenerational support even
when accounting for the influence of other socio-demographic characteristics.
154 National Elder Abuse Prevalence Study: Final Report
Summary
The findings in relation to the CALD subsample presented in this report provide some significant insights into
elder abuse as it is experienced by people born in non-English speaking countries who have immigrated to
Australia. Dow, and Brijnath (2019, p. 147) note that culture can mediate perceptions of elder abuse, types of
abuse and responses and that this varies widely between cultural groups. The findings from the SOP and SGC
discussed in this chapter provide insights into differences according to CALD and non-CALD background in
the prevalence of and dynamics of elder abuse. Cultural background was also shown to influence attitudes,
awareness and understanding of elder abuse according to country of birth.
It is notable that there is limited evidence of substantial differences in key patterns, including the overall
prevalence of elder abuse and the prevalence of different subtypes.
However, there are a number of areas where the findings point to a need to develop greater understanding of
the experience of elder abuse in the CALD context, in order to better understand dynamics so as to support
improved responses.
The first of these is in relation to the findings that point to an intergenerational familial dynamic that is weaker
for the CALD subgroup, with children being represented among perpetrators to a lesser extent than for the
non-CALD subsample. Further culturally sensitive research examining familial dynamics within different CALD
subgroups and how they influence the occurrence or non-occurrence of elder abuse would be helpful in further
understanding these dynamics.
The second related finding concerns the CALD-specific abuse subtype, abuse relating to language and culture,
measured in the SOP. This type of abuse is perpetrated much less by family members (apart from sons- and
daughters-in-law), who are likely to share the same cultural background, and more by friends, neighbours and
acquaintances, who are less likely to share the same cultural background. Further research examining links
between racism as a dimension of elder abuse is necessary.
These findings suggest that intergenerational dynamics may be less of an influence in the elder abuse
experienced by CALD groups but that racist dynamics in the community may be more of an influence.
Consequently, although prevalence levels are not substantially different between the CALD and non-CALD
groups in the SOP, the findings suggest the dynamics of elder abuse are different, at least to some extent.
A further area where the findings point to differences in dynamics between CALD and non-CALD groups is in the
area of awareness and attitudes (chapter 12). These findings highlight some consistent differences in the patterns
concerning awareness of elder abuse and recognition of elder abuse behaviours. Significantly, CALD participants
from some non-English speaking countries had lower levels of awareness and recognition than participants from
English speaking countries in both the SOP and the SGC, although this was stronger among SOP participants
than among SGC participants.
Similarly, ageist attitudes were stronger among SGC participants from the CALD subsample than they were
among the non-CALD sample. By contrast, positive attitudes to intergenerational support were stronger among
some of the CALD subsample than the non-CALD sample, particularly those born in South and Central Asia, the
Middle East and North Africa and South-East Asia.
Chapter 14: Summary and implications 155
This chapter provides an overview and summary of the main aspects of the findings in relation to elder abuse
set out in the preceding chapters of this report. The discussion also highlights similarities and differences
between the five abuse subtypes. This summary provides a basis for a discussion of the implications of the
findings for practice and policy in the second part of the chapter. The third section of the chapter provides
a high-level overview of the conceptual implications of the report.
Overview
Elder abuse overall
This report has estimated that 14.8% of the population aged 65 and over who live in the community (rather
than residential care settings) have experienced elder abuse in the past 12 months. Psychological abuse is the
most common subtype, followed by neglect, financial abuse, physical abuse and sexual abuse. The dynamics of
elder abuse are complex, with varied patterns in key areas – such as perpetrator profiles, factors that suggest
susceptibility and help-seeking behaviours – for the different subtypes.
of people aged
14.8 65+
have experienced
elder abuse
in the past 12 months
The analysis suggests that, overall, differences in prevalence of elder abuse between men and women are limited
(15.9% cf. 13.6%). Women were estimated to be slightly more likely to experience sexual abuse (1.2% cf. 0.7%) and
neglect (3.5% cf. 2.2%). Although gendered dynamics are not particularly evident in the prevalence estimates,
some subtle gender-related patterns are relevant in other areas (see further below).
Socio‑economic status, relationship type and housing type are particularly associated with higher or lower risks
of experiencing abuse.
People who are living with a partner are less likely to experience abuse. Those who are divorced or separated are
more likely. Having step-children is also associated with a greater likelihood of experiencing abuse.
The findings in relation to socio‑economic status and elder abuse point to some complexities in understanding
how socio‑economic status and elder abuse intersect. Financial, sexual and psychological abuse subtypes
are more likely to be experienced by people from relatively disadvantaged socio‑economic areas (based on
Socio‑Economic Indexes for Areas (SEIFA) – the Index of Relative Socio‑Economic Disadvantage). Consistent
with this, those who own their own home with debt or who rent housing have a higher rate of experiencing
psychological abuse compared to those who own their home without debt, even after controlling for other socio-
demographic characteristics. On the other hand, participants with a degree or higher level of education are more
likely to report experiencing psychological abuse, which may reflect a higher level of awareness of elder abuse
and a great capacity to identity such behaviours.
156 National Elder Abuse Prevalence Study: Final Report
Family members, especially sons and daughters, are the largest perpetrator group (nearly one in five). About one
in 10 elder abuse perpetrators are intimate partners.
1 in 10
elder abuse perpetrators are
intimate partners
Together, friends, neighbours and acquaintances (e.g. co‑workers) also reflect about a quarter of all perpetrators.
Professionals and carers are not significant elder abuse perpetrators, although they are implicated in neglect to a
higher extent than other types of abuse.
Perpetrator profiles are an area where gendered patterns are evident, with men outnumbering women as
perpetrators by 10 percentage points.
percentage
men outnumber women points
as perpetrators by
Elder abuse is assessed as very serious or somewhat serious by three-quarters of those who experience it. The
findings indicate that older people with poorer health or disability were more likely to experience elder abuse.
Older people with poor mental health also had a greater likelihood of experiencing elder abuse. Furthermore, less
frequent contact with family members and friends and a lower sense of social support were associated with the
elevated likelihood of experiencing elder abuse.
Most people who experience elder abuse try to manage it themselves and don’t seek help or advice. However,
eight in 10 do take some action to stop the abuse, most commonly speaking directly with the perpetrator
(approximately one-half). Another common action is breaking contact with or avoiding the perpetrator
(approximately two in five). This is concerning, since it may compound the effects of the abuse and reflect
a lack of effective ways of addressing abuse. It also means that perpetrators are not held to account.
8 in 10
take some action to
of those who
experience elder abuse
Where help and assistance are sought, it most often involves family members and friends. Otherwise, health
professionals, including GPs and nurses, are the most common sources of assistance. Legal responses are
more commonly accessed for some abuse subtypes (physical and financial abuse) than others but are not
a predominant source of support.
Chapter 14: Summary and implications 157
The estimates suggest psychological abuse is the most common form of abuse, affecting 11.7% of the sample,
with around half of these experiences in the low score range, more than one-third in the high score range and
14% in the mid-score range. There are three particularly common forms of psychological abuse occurring in
nearly commensurate proportions (between 46% and 49%): being insulted, called names or sworn at in an
aggressive or offensive way; being excluded or repeatedly ignored; and being undermined or belittled.
Women are more likely to experience psychological abuse than men (12.6% cf. 10.7%). Those in the lower age
range (65–69) are also more vulnerable in this community dwelling sample.
As with financial abuse, psychological abuse is most likely to be committed by family members, with children
accounting for 18% of perpetrators and other relationships of this type taking the total to around 30%. Friends
(10%), neighbours (8%) and acquaintances (12%) are also well-represented in the perpetrator groups. Partners
accounted for 8% of all psychological abuse perpetrators.
After physical abuse, psychological abuse was the abuse type for which help and advice was most likely to be
sought (40%). Apart from friends and family, the most common sources of formal help were a GP or nurse (30%)
and a professional carer or social worker (22%). Help from lawyers or police was not uncommon, at 15% and 17%
respectively for those who sought help.
Taking direct action to stop the psychological abuse by speaking to the perpetrator was only a little more
common than breaking contact with them (52% cf. 49%).
Of those experiencing
psychological abuse
took direct action to
stop the abuse
by speaking to the perpetrator
After physical abuse, psychological abuse attracted the highest seriousness ratings of the abuse subtypes from
participants who experienced it, with only 21% saying it was not serious, 32% saying it was very serious and 46%
saying it was somewhat serious. Those who reported psychological abuse were four times more likely to have
psychological distress scores indicating a probable serious mental illness than those who did not experience
psychological abuse (9.6% cf. 2.4%). Low social support scores were particularly evident for psychological abuse,
possibly reflecting the consequence of breaking off contact with perpetrators.
There were similar views on psychological abuse between SGC and SOP participants. Similar proportions
strongly agreed that calling an older person hurtful names was always abuse (79% SGC and 78% SOP), for
example. The main difference in views was on limiting contact with grandchildren, with SOP participants more
likely to strongly agree that this was elder abuse (58% cf. 49%).
158 National Elder Abuse Prevalence Study: Final Report
Neglect 2.9%
of
sample
The findings set out in this report indicate that neglect, as an omission rather than an action, is different in character
from other subtypes of abuse. After psychological abuse, the analysis suggests neglect is the most common
subtype of abuse, with 2.9% of the sample reporting neglect across three score bands: low at 60%, medium
at 24% and high at 16%. The most common forms of neglect involved failing to provide assistance with routine
housework (80%), travel or transport (69%), shopping for groceries or clothes (57%) and preparing meals (52%).
Women are more likely than men to experience neglect (3.5% cf. 2.2%). Participants in the lower age bracket
(65–69) and those over 80 are also more likely to experience neglect.
Intimate partners are the largest perpetrator group (25%), followed closely by children (24%). Unlike the other
abuse subtypes, service providers and professionals are significant among perpetrators (13–14%).
Most of the participants who reported experiencing neglect did not seek help or support (70%). Where help or
support was sought, the most common sources, apart from family and friends, were a GP or nurse (37%) and a
professional carer or social worker (33%). After speaking to the person causing the neglect directly (48%), the
most common action taken to stop the neglect was withdrawing from social life (14%), which may in itself be a
reflection of an absence of support for social activity.
of those who
experienced neglect
did not seek
help or support
Neglect was the most likely subgroup to be self-rated as ‘not serious’ (48%). Those who experienced it also had
much lower psychological and physical wellbeing, lower social support and higher levels of disability than those
who didn’t. For example, psychological distress scores indicative of probable serious mental illness were five
times more likely for the neglect group compared with participants who did not experience neglect (15% cf. 3%).
These findings are likely to reflect the issues that gave rise for the need for care – they also reinforce the negative
implications of a failure to provide care for an inherently vulnerable group.
It is also notable that this research has revealed the existence of a group of older people who experience neglect
but are not included in the prevalence estimate because they have no one to help them and are therefore not
captured by the conceptual and operational definitions applied. This group represents 6% of the total sample.
Views on neglect were similar between the SOP and SGC samples, with the same proportions (64%) strongly
endorsing not providing help with personal care when this is normally expected or provided as a form of abuse.
Chapter 14: Summary and implications 159
The estimates suggest financial abuse is the third most common abuse subtype, reported by 2.1% of the
sample. There are three particularly common forms of this abuse: being pressured into giving or loaning money,
possessions or property (42%), having money, possessions or property taken without consent (34%), and the
perpetrator not contributing to household expenses or aged care/home service fees where this had previously
been agreed (31%).
This form of abuse is mainly committed by children (33%) or other family members. However, friends (9%),
neighbours (5%) and acquaintances (3%) are also perpertators, to a smaller extent.
Financial abuse
is mainly committed by
children
Just under one-third of the sample that experienced financial abuse sought help and advice. Where this involved
someone other than a family member or friend, this was most commonly a professional carer or social worker
(34%), GP or nurse (28%), or a lawyer (33%). Police were involved in a quarter of the cases where help or advice
were sought. Apart from speaking to the person directly (59%), the most common action taken to stop the abuse
was breaking contact with the perpetrator (30%).
Just under
one-third ofexperienced
those who
financial abuse
sought
help or advice
Close to a third of the participants who experienced financial abuse indicated it was not serious, approximately
one-third rated it as somewhat serious and another third said it was very serious. Levels of psychological
wellbeing scores indicating probable mental illness were close to one-tenth for financial abuse participants,
compared with 2.4% for participants who did not experience financial abuse. The participants who experienced
financial abuse reported a low sense of social support compared to those without such an experience.
There were some differences between generations in the strength of recognition of financial abuse, with
recognition being stronger among SGC participants.
160 National Elder Abuse Prevalence Study: Final Report
The estimates indicate physical abuse is the second least common subtype of abuse (1.8% of the sample). Of
those who experienced physical abuse, threats of harm is the most common form (61%), followed by being
grabbed, pushed or shoved (47%), and hit, punched, kicked and slapped (22%).
Men are slightly more likely than women to experience physical abuse (2% cf. 1.6%), which is also particularly
associated with those in the 65–69 years age bracket.
Family members, especially children (17%), are prominent perpetrators but friends (10%), neighbours (12%) and
acquaintances (9%) are also well-represented among those who commit physical abuse. Notably, 12% of physical
abuse perpetrators are spouses.
physical abuse
of
is committed by children
The physical abuse subgroup was the most likely of all the abuse subtypes to seek help or advice, with around
half doing so. This group was also the most likely to engage with police (36%), the only abuse subtype for which
police as a source of help or advice outweighed family members (33%). Six in 10 participants who took action
to stop the physical abuse spoke directly to the person, and breaking contact (54%) was most common for this
abuse subtype.
The
physical abuse
subgroup was the most likely
of all the abuse subtypes to seek
help or advice
Of all the abuse subtypes, physical abuse was most likely to be rated as very serious (39%) and least likely
to be rated as not serious (20%). Participants who experienced physical abuse were almost five times more
likely to have a psychological distress score indicating probable serious mental illness than those who did not
(14.5% cf. 3%). They were also more likely to rate their physical health as ‘poor’ compared with those who did not
experience physical abuse (14% cf. 7%). Disability was reported by more than two-thirds of the physical abuse
subgroup, compared with less than half who did not experience physical abuse.
Attitudes in relation to physical abuse differed slightly between the SOP and the SGC samples. SGC participants
were more likely to strongly agree that pushing or shoving an older person was elder abuse (91% cf. 86%).
Chapter 14: Summary and implications 161
Sexual abuse 1%
of
sample
The analysis indicates sexual abuse is the least common subtype of abuse, applying to just 1% of the sample.
The most common type of sexual abuse was being spoken ‘to in a sexual way’ when this was unwanted (76%).
Unwanted sexual touching was reported by almost two-thirds of those who experienced sexual abuse.
Women were more likely to experience sexual abuse than men (1.2% cf. 0.7%), with those in the 65–69 years old
age bracket more likely to experience it than those in the older age brackets.
Unlike the other abuse subtypes, sexual abuse is unlikely to be committed by family members (other than intimate
partners). The biggest perpetrator groups are friends (42%), acquaintances (13%), neighbours (9%) and spouses (9%).
The biggest
sexual abuse
perpetrator group was friends
Help was least likely to be sought for sexual abuse out of all the subtypes, with three-quarters not seeking help.
Of those who did seek help, the most common sources of help were family members, a GP or nurse, a friend and
a neighbour. The most common actions taken to stop sexual abuse were speaking to the person and breaking
contact with them. Police and legal responses were almost never invoked.
After neglect, sexual abuse was also the form of abuse most likely to be rated as ‘not serious’ by those who
experienced it. This contrasts with participants who experienced sexual abuse having psychological distress
scores similar to those who experienced physical or financial abuse. Of the sexual abuse subgroup, the same
proportion (14%) had scores indicative of probable serious mental illness as was present for physical abuse.
Attitudes to sexual abuse varied between the SOP and SGC samples, with the SGC sample more likely to agree
that talking in a sexual way to an older person against their will was sexual abuse (89% cf. 76%).
Based on the estimates in chapter 5, sexual abuse is just under half as common as financial abuse, with an
estimated 39,472 people aged 65 and over affected by sexual abuse, compared with 83,787 affected by
financial abuse. However, financial abuse receives much greater attention and is perceived to be the most
common form of elder abuse (e.g. ALRC, 2017, p. 160).
The prevalence of sexual abuse revealed in this study is higher than indications from other Australian
studies. Hill and Katz’ 2019 analysis of the PSS (chapter 2) suggested between 0.1% and 0.4% of women
aged 55 and over had experienced sexual violence (using narrower measures of forced sexual activity or
threats of forced sexual activity (chapter 5)) in the preceding 12 months.
162 National Elder Abuse Prevalence Study: Final Report
The measures used in the SOP are wider and consistent with some international studies on elder abuse,
namely Ireland (Naughton et al., 2010), the UK (O’Keeffe et al., 2007) and Canada (National Initiative for
the Care of Elderly, 2016). These studies revealed prevalence rates of 0.5%, 0.2% and 1.5% respectively.
Yon’s meta-analysis (2019) indicates a pooled prevalence of sexual abuse across 50 studies of 2.2%.
The profile of common sexual abuse perpetrators being friends in the SOP is consistent with the Canadian
findings (National Initiative for the Care of Elderly, 2016). It is also consistent with other international
literature (Band-Winterstein et al., 2019).
The association between the perpetration of sexual abuse and alcohol misuse by perpetrators reported in
chapter 7 is consistent with evidence about alcohol or another substance being involved in about half of
cases involving sexual offending from the PSS (AIHW, 2020).
In the context of all forms of elder abuse being under-reported, sexual abuse is particularly under‑reported
when considered in the context of sexual assault reporting generally. Around half of women in the general
population who experience sexual assault seek help or advice (AIHW, 2020). The most common source
of help and advice for women who experience sexual assault in the broader community are friends and
family members (seven in 10), health professionals (four in 10), counsellors, support workers or telephone
helplines (one‑quarter) and police (one in six) (AIHW, 2020).
The findings suggest that complex family dynamics underpin elder abuse to a significant extent, with a
complicated relationship history involving divorce and re-partnering with a partner who had children, is
associated with an elevated likelihood of abuse. Conversely, being in a relationship with a partner lowers the risk.
Similarly, and likely related to some extent, socio‑economic disadvantage is associated with elder abuse, as
evidenced through the links with housing type and living in a disadvantaged area with an elevated risk of some
subtypes of elder abuse. Conversely, owning your own home outright is associated with a lower risk of elder abuse.
The findings set out in chapter 6 indicate relative rankings of these issues with the experience of elder abuse
overall and in relation to specific subtypes, in the context of analysis assessing the relative importance of
14 different variables. For all but sexual abuse, psychological distress and social support are in the top four
explanatory variables. For sexual abuse, psychological distress and social support are in the top four.
Chapter 14: Summary and implications 163
Legal arrangements
The analysis indicates that advance planning, particularly in relation to wills, is common among those aged over
65. The majority have wills (88%) and around half have executed power of attorney instruments allowing others
to make financial and lifestyle decisions on their behalf. People from higher SEIFA brackets are more likely to
have made these two kinds of arrangements than those from lower SEIFA brackets.
SGC participants were more likely than those in the SOP to believe that elder abuse is common in the community
(54% cf. 43%). They were also more likely to believe that elder abuse is ignored (52% cf. 41%). Similarly, they were
less likely to believe that elder abuse is a private matter that should be handled in the family (9% cf. 15%).
The SGC participants were substantially less likely than SOP participants to agree that abuse is understandable if
the perpetrator is experiencing stress (7% cf. 20%). They were even more unlikely to endorse the proposition that
elder abuse is understandable if the older person is difficult to deal with (7% cf. 25%).
Overall, the data indicated modest levels of ageist attitudes towards older people in the general community.
The participants in the SGC largely held positive attitudes towards older people. A benevolent ageism measure
– that ‘it is helpful to repeat things to older people because they rarely understand the first time’ – drew the
strongest levels of endorsement, with one-third of participants providing a rating of 7–10, on a rating scale from 0
strongly disagree to 10 strongly agree. The vast majority rejected the notion that ‘older people are a drain on the
health care system and the economy’.
The findings on intergenerational support suggest that this is a majority social value among people in the
SGC with a mean score of 66.5 (on a score range of 0–100, a higher score indicating a stronger sense of
intergenerational support) and stronger endorsement for children supporting parents than parents supporting
children. A sense of entitlement to older people’s assets where there has been regular assistance to that person is
164 National Elder Abuse Prevalence Study: Final Report
not insignificant, with over one-quarter agreeing (with ratings of 7–10, on a rating scale from 0 strongly disagree
to 10 strongly agree) that other family members held this view. Agreement with intergenerational support was
linked with attitudes indicative of entitlement.
Analysis assessing links between a lower recognition of elder abuse and condoning elder abuse found ageist
views were linked with more condoning attitudes towards elder abuse and lower recognition of elder abuse
behaviours. Links between condoning attitudes to elder abuse and lack of recognition of elder abuse behaviours
and support for intergenerational support and entitlement were not evident.
Together these findings suggest that values and attitudes, particularly ageism, are relevant to the wider social
context in which elder abuse occurs to some extent. Although ageism is not common (mean score of 37.4),
ageist values are evident to a not insignificant extent.
a single or repeated act or failure to act, including threats, that results in harm or distress to an older
person. These occur where there is an expectation of trust and/or where there is a power imbalance
between the party responsible and the older person.
This conceptual definition was operationalised into the technical analytic definition applied to deriving prevalence
estimates as set out in chapter 5.
The findings in this report enable an assessment of the appropriateness of the conceptual definition. In broad
terms, these findings indicate that this definition does represent an appropriate characterisation of elder abuse
for research purposes.
However, findings in relation to neglect that fell within the definition (chapter 5) and neglect that was excluded
from the definition because the person experiencing neglect had no-one to help require further consideration.
The following discussion deals with each of these issues in turn. It should be noted that this discussion refers
to a research definition. Different considerations may apply in relation to legal, policy and practice definitions,
depending on the purpose that they serve (Kaspiew et al., 2016; Kaspiew et al., 2019).
Conceptual definition
Findings in two areas set out in the report provide broad support for the approach to the definition. First, the
relevance of the concept of a power imbalance (as an addition to or replacement for) a relationship of trust is
confirmed by the findings in chapter 6. These findings (summarised above) confirm that people who experience
elder abuse are considerably more likely than those who don’t to report: psychological ill-health, physical ill-
health, disability and social isolation. Further, the analysis suggests indications of vulnerability are associated
with people who experience abuse from friends (chapter 6) to a greater extent than those who experience
abuse from adult children. These findings indicate that power imbalance is a useful and relevant concept in the
understanding of elder abuse, alongside a relationship of trust.
The second area where the research findings support the conceptual definition are in the elder abuse recognition
findings set out in chapter 12. These findings (also summarised above), demonstrate clear majority support
among SOP and SGC participants for the following actions to be considered elder abuse:
y Not providing help with personal activities such as dressing, washing and feeding when this is normally
expected or provided
y Talking to an older person in a sexual way when they do not want to
y Taking money from an older person without their consent
y Calling an older person hurtful names.
This level of community agreement with key measures that relate to the conceptual definition indicate that the
approach taken is in step with community values.
However, the definition should be tested against the experiences of groups not covered in this study: Aboriginal
and Torres Strait Islander people, LGBTIQ groups, people with cognitive decline and people living in residential
care settings (see ‘Further Research’, below).
Chapter 14: Summary and implications 165
Neglect
The construct of neglect warrants further consideration for two reasons. First, because this research has
highlighted a group whose experiences fall outside the conceptual and technical definitions but who nonetheless
have an unmet need for help (chapter 6). Second, the findings indicate that as a construct, neglect has some
features that set it apart from other abuse subtypes.
Chapter 6 highlighted a group whose experiences would have amounted to neglect had there been a person
responsible for the activities neglected (‘the no-one to help’ group). As a result of the application of the conceptual
and technical definition, this group was excluded from the neglect prevalence estimate in chapter 5 because there
was no one to help. This finding underscores the complexity of neglect as a construct, in that it reflects an omission.
In practical terms, the experience of the no-one to help group amounts to a double omission in that a need for
help was not fulfilled because there was no-one to complete the required actions. The need for an actor (and a
relationship of trust or power imbalance) in the conceptual definition is inconsistent with this circumstance.
Another area where the findings distinguish neglect from other types of abuse is in relation to perpetrator
characteristics as reported by participants in the SOP and set out in chapter 7. For the neglect group,
perpetrators’ problems with health, including physical health, were evident to a greater extent than for other
abuse subtypes, suggesting that an inability to provide care contributed to the occurrence of neglect.
For these reasons, it is desirable to distinguish neglect from other subtypes of abuse, in recognition of its
distinctive characteristics. Accordingly, it is suggested that rather than the term ‘elder abuse’ being used, the
term ‘abuse and neglect of older people’ should be adopted, to encourage identification of abuse and neglect
as distinct concepts and to raise awareness of neglect. This terminology is also consistent with sensitivities
in relation to the use of the term ‘elder’ in some contexts, particularly for Aboriginal and Torres Strait islander
groups (Clare et al., 2014; Kaspiew et al., 2016).
In developing definitions in legal, policy and pactice frameworks, careful consideration should be applied to the
definition of neglect in those frameworks. In some instances, it may well be inappropriate to apply a definition of
neglect that excludes people in the position of the ‘no-one to help’ group in this research. An example of this may
be in frameworks that determine access to services based on need. In other contexts – such as the determination
of civil or criminal liability – a different definition may be appropriate.
Implications
This section sets out the implications of the findings in this report for further policy and program development.
The discussion identifies a need for development in three related areas: preventing, identifying and responding to
elder abuse.
Two broad points are important for the discussion. First, the findings summarised above reinforce the point that
elder abuse is a complex phenomenon, particularly since it involves family members and social connections to
a very significant extent. The findings suggest that each form of abuse involves different dynamics, meaning
that policy and practice responses need to be tailored for different subtypes. This insight is consistent with
local and international literature, which recognises that blanket approaches are likely to be ineffective and that
interventions need to be carefully designed for the specific subtype (Conrad, Liu, & Iris, 2019; DeLiema et al.,
2018; Dow & Brijnath, 2019; Santos et al., 2019).
Second, the findings indicate a need for increased focus on neglect and sexual abuse, in particular, given levels of
prevalence and lower levels of awareness and help seeking in these areas. Further, as psychological abuse is the
most common form of abuse and attracts the second highest self-assessments of seriousness, this form of abuse
also warrants more attention. This is not to say that focus should be diverted from financial abuse and physical
abuse. However, with financial abuse commonly considered to be the most common form (e.g. ALRC, 2017;
Dow & Brijnath, 2019), it is evident that other types of abuse may not have received sufficient attention to date.
In the context of these two points, the discussion about policy and program development that follows is general
in nature and should be considered in light of specific and detailed findings set out in this report.
166 National Elder Abuse Prevalence Study: Final Report
Background
In Australia, the development of responses to elder abuse has been organic and iterative, with agencies in
different states and territories that provide services to older people developing responses as different needs have
been identified (e.g. Dow et al., 2020).
The stocktake of elder abuse prevention and awareness responses across Australia highlights a varied range of
strategies including awareness raising campaigns, moves toward safeguarding mechanisms at state and territory
level, and the development of multi-agency, multi-disciplinary responses such as Health Justice Partnerships
(Commonwealth of Australia, 2019).
Locally and internationally, the evidence base on what works to address elder abuse is undeveloped and there is
limited empirical evidence on the effectiveness of particular approaches (Baker, Francis, Hairi, Othman, & Choo,
2016; Dow & Brijnath, 2019; Teresi et al., 2016). However, insights from developments in allied areas, such as family
violence, are considered useful in informing thinking in approaches in relation to elder abuse (e.g. Ng, Lim, &
Kadir, 2020; Teresi et al., 2016; Warren & Blundell, 2018b). Notably, there is local and international recognition of
the potential value of applying a public health approach to interventions across three levels: primary (preventing
elder abuse from happening), secondary (identifying situations where elder abuse may be happening and
intervening to stop it) and tertiary (addressing elder abuse when it does happen) (e.g. Clare et al., 2014; Forum
on Global Violence Prevention, Board on Global Health, Institute of Medicine, National Research Council, 2014;
Lord, McMahon, & Nivelle, n.d.; Mann et al., 2014).
With the National Plan (CAG, 2019) based on a clear recognition of the need for responses to elder abuse to be
evidence-based, this research highlights some significant areas where there is a need and opportunity for policy
and program development.
A prevention framework
This research establishes that elder abuse is a significant problem in Australia. In the context of population
dynamics described in chapter 1, it will become an even larger problem as the proportion of the population that
is over 65 increases.
The findings set out in this report highlight significant correlations between lower psychological, physical
and social wellbeing and elder abuse, reinforcing the importance of addressing this problem in the interests
of supporting the health and wellbeing of the population aged 65 and over. They also identify a range of
socio‑demographic and economic characteristics that are associated with a greater likelihood of experiencing
elder abuse.
With prevention science increasingly applied in the development of strategies to reduce other social problems,
such as family violence, in the context of a public health approach (e.g. Our Watch et al., 2015), the development
of a rigorous prevention framework could assist in reducing the prevalence of elder abuse in Australia into the
future. If such an approach was to be developed, it would be important to focus on using tested and effective
prevention frameworks to implement evidence-based programs and policies that reduce risk factors associated
with elder abuse over the life course (see e.g. Teresi et al., 2016).
Notably, prevention efforts in elder abuse are in their early stages internationally (Teresi et al., 2016) but as a
world leader in primary prevention for family violence (Kinnersley, 2020), Australia is well-placed to develop allied
approaches in the context of elder abuse. Notwithstanding the lack of evidence-based approaches, international
elder abuse experts confirm that ‘the seriousness and scope of the problem’ requires preventative actions
(Pillemer, Burnes, Riffin, & Lachs, 2016, p. s200).
A range of insights set out in this report would inform the development of such a framework. They include
evidence of greater susceptibility to abuse among those with complex relationship histories, people who are
renting or paying off homes, people with psychological and/or physical health problems and disability and those
who are socially isolated (chapter 6). In addition, chapter 8 indicates that attention to financial management
practices, particularly improving approaches to financial management among older women, and accountability
practices where support in financial management is provided would be important.
Chapter 14: Summary and implications 167
The findings of this research suggest greater recognition of elder abuse among the SGC participants in
comparison with the SOP participants. Chapter 12 also highlights greater recognition of elder abuse among those
who provide support to people aged 65 and over. These findings indicate that social attitudes to elder abuse
are amenable to change and that connections between older people and those around them are important to
detecting and addressing elder abuse.
There is a need for awareness-raising measures with several strands. First, there needs to be awareness raised
of what elder abuse is, both among older people and among others in the community, such as the family and
friends who support them. It is clear that family and friends are important sources of help and support for older
people experiencing abuse (even accounting for some being perpetrators of abuse). They need to be alert to
indications that elder abuse may be occurring and equipped to respond appropriately if disclosures are made.
Some international research highlights the potentially important role that family and friends can play as ‘a bridge’
to formal avenues of support (Dominguez et al., 2019, p. 10).
A further strand of awareness raising should focus on raising awareness of services and strategies available
to support people experiencing elder abuse. A lack of awareness that assistance is available for elder abuse
(both on the part of older people and the people they turn to for help on an informal basis) may contribute to
the help‑seeking patterns described in chapter 8.
This underscores both the necessity of and the opportunities for systematic identification mechanisms. Although
those who experience elder abuse may be reluctant to reveal it without prompting, engagement with medical
practitioners and other helping professionals offers an important opportunity for elder abuse experiences to
be identified and addressed. Dong (2015) argues that early detection and interventions, such as incorporating
effective treatment of underlying problems, providing community-based services, and appropriately involving
family, may help delay or prevent elder abuse.
However, recent analyses in Australia suggest that in order to support better screening and identification, other
improvements in responses are also required (Dow & Brijnath, 2019). According to their analysis, among the
issues that need to be addressed to support effective screening are:
y improvements in levels of knowledge about elder abuse among health professionals
y better training on signs that may indicate elder abuse is occurring
y access to effective screening and assessment tools
y organisational support to manage elder abuse cases where they are identified (p. 152).
The complex dynamics and potential risks involved in disclosing abuse need to be carefully managed, with
Dominguez and colleagues (2019, p. 11) identifying a need for ‘victim centred interventions’ based on ‘an
assessment of the clients’ needs and wishes, including the exploration and limiting of interventions that a victim
perceives as negative to encourage engagement with services’.
In this context, it is significant that chapter 9 demonstrated that legal and justice responses were relied on for
support to a lesser extent than responses involving health and helping professionals. This finding is in line with
research showing that even for legal problems more broadly, legal mechanisms are relied on for advice to a lesser
extent than non-legal sources of advice (Coumarelos et al., 2012, p. 135). This indicates that health professionals
are the key frontline point for identifying elder abuse.
168 National Elder Abuse Prevalence Study: Final Report
There is significant focus on financial abuse in existing responses to elder abuse (e.g. Commonwealth of Australia,
2019). This study demonstrates a need for a systematic assessment of whether existing options for addressing
elder abuse overall, and each subtype, are adequate, appropriate and accessible. Further, indications of the role
of perpetrator problems such as mental and physical ill-health and financial strain suggest a need to focus on
perpetrators as well as victims in service responses.
Psychological abuse is the most prevalent form of elder abuse and assessments of those who experience it
indicate it is the second most serious form. There needs to be consideration of whether existing mechanisms to
identify and address this form of abuse are adequate and sufficient.
Similarly, help seeking for sexual abuse and neglect are very limited. Sexual abuse is substantially under-reported
with almost no reliance on justice responses among older people. Neglect is also substantially under-reported.
The findings suggest that identification and response strategies for these two abuse subtypes require further
development, including for the group that was not captured by the neglect definition because they had no-one
to help (chapter 6).
Overall, the evidence base on the coverage and effectiveness of existing interventions is very limited. In general
terms, there are two approaches that are recognised as promising, albeit with limited evidence in support of
them at this stage.
One is family mediation, which is low-conflict, non-adversarial and may provide an opportunity for family
relationships to be repaired (Dow & Brijnath, 2019). Research by Adams, Bagshaw, Wendt, and Zannettino (2014)
has highlighted support for mediation in the context of financial abuse but concern about the limited number of
appropriately trained mediators in Australia. Examples of mediation approaches include Relationships Australia’s
Elder Relationships Service, which includes therapeutic family counselling and mediation (Commonwealth of
Australia, 2019).
The other is multidisciplinary approaches that may include legal, therapeutic and health-based elements
(Joosten et al., 2017). These approaches are applied internationally (e.g. in Singapore – see Ng et al., 2020) and
are gaining increasing momentum in Australia through initiatives such as the Elder Abuse Service Trials currently
being trialled (Commonwealth of Australia, 2019).
Although, the findings of this research point to a need for a comprehensive assessment of the adequacy of
responses to elder abuse, some particular gaps have already been identified in recent analyses.
In their systematic literature review research, Warren and Blundell (2019, 2018a, 2018b) examined the state of
Australian and international social policy and Indigenous and non-Indigenous prevention and service responses
to elder abuse in rural and remote communities. They concluded that elder abuse in rural and remote areas
presents specific challenges that require specialised policy, prevention and service responses compared to urban
areas, including because of geographic isolation, high service demand, poor resourcing, lack of transportation,
and issues with confidentiality and privacy, as well as historical and cultural experiences of Indigenous people
living in those areas (Warren & Blundell, 2019).
Warren and Blundell’s review found consideration of elder abuse in the context of rural and remote areas to be
inadequate across social policy literature, with only superficial regard to the vulnerabilities associated with these
communities, particularly for Indigenous peoples (Warren & Blundell ,2018a). This is consistent with identified
problems in relation to health services (AIHW, 2019b) and legal services (Coumelaros et al., 2012).
Prevention and service responses were better represented in the literature; however, evaluation of these
responses was lacking (Warren & Blundell, 2019). The lack of evaluation raised questions about the extent to
which delivery of elder abuse prevention and service responses was actually occurring in rural and remote areas
(Warren & Blundell ,2019). Warren and Blundell (2018b) also assessed overlapping literature between elder
abuse and family violence and identified commonalities across the two sectors that could be utilised for better
collaboration in service delivery and approaches to addressing abuse. They observed that formal and informal
collaboration could benefit resource limited services in rural and remote areas and improve responses to family
violence and abuse across the life course.
Chapter 14: Summary and implications 169
It should also be noted that the Department of Communities WA has commissioned the University of Western
Australia to examine elder abuse in Western Australia, including assessment of current responses for older
people, their families and perpetrators. This research is due for completion in June 2021.
Further research
This study represents a highly significant advance in knowledge about elder abuse in Australia. However, it also
highlights the need for further research in a number of areas, including in relation to groups in the community
who were not included in sufficient numbers in the SOP sample to support analysis (see further below).
In particular, the findings point to a need to better understand the context for the complicated family dynamics,
in particular, intergenerational dynamics, these findings have highlighted as being significant in the occurrence
of elder abuse. Similarly, with friends, neighbours and associates significant among perpetrators, further
examination of social dynamics is warranted.
The findings of significant associations between a relationship history involving relationship breakdown and
step-children, together with likely related dynamics in relation to financial and housing stress, suggest that it is
important to understand how these and other complicated relationship dynamics influence the occurrence of
elder abuse. A life-course approach, which could not be substantially implemented as a theoretical framework for
this study (chapter 3), would be an important framework for further research.
Further, the importance of longitudinal research that can support further examination of causal factors in the
occurrence of elder abuse, to further explore correlations identified in cross sectional research, is increasingly
recognised internationally (e.g. Acierno, Hernandez-Tejada, Anetzberger, Loew, & Muzzy, 2017; Dong, 2015;
Williams et al., 2017).
A longitudinal study that examines the wellbeing of older Australians would have significant value in
understanding a range of issues identified as relevant to elder abuse in this research. These include:
y the health and aged care dimensions of elder abuse including: health problems as causes and consequences
of elder abuse, the demands placed on the health care system as a result of elder abuse, whether elder abuse
is a factor that precipitates entry into aged care
y events that are associated with a worsening or improvement in intergenerational relationship dynamics
(e.g. relationship breakdown, partnering and re-partnering among parents and adult children)
y the influence of financial/economic events such as unemployment and retirement on relationship dynamics
y intergenerational support and financial management practices in families
y historical and contemporary patterns in the causes and management of family conflict
y community-level factors associated with the experience of elder abuse committed by friends, neighbours
and acquaintances
y histories of exposure to child abuse and family violence
y the impact of physical and mental health and substance misuse problems on intra- and intergenerational
relationship dynamics
y how sustained the experience of elder abuse is over time
y whether certain types of elder abuse (e.g. psychological abuse and neglect) are precursors to other types
y the causal relationship between elder abuse and psychological distress, social isolation and disability.
Further research, including using qualitative methods, on elder abuse in some specific contexts would aid in
developing improved responses.
As noted in chapter 4, an important group not covered in this methodology is people with cognitive
decline. Given the evidence of links between cognitive decline and a higher risk of experiencing elder abuse
(Storey, 2020), and indications that elder abuse experiences are different for those experiencing cognitive
decline (Gillbard, 2019), this is an important area for further research. However, it is also methodologically and
ethically complex (Williams et al., 2017) and would require a specialised approach.
Although data collection from those with cognitive impairment requires particular care, it should not be
considered unfeasible. Cognitive impairment does not inevitably equate to incapacity to consent to participation
in research. It is recognised that people with dementia may want to participate in research and they have a
right to be heard in matters about them (see review by Mayo & Wallhagen, 2009; Chesterman & Bedson, 2017;
Slaughter, Cole, Jennings, & Reimer, 2007). Further, research assessing elder abuse in the context of cognitive
impairment can also be based on the observations of third parties, such as family members, carers and
professionals (see e.g. studies cited in Ho et al., 2017 and Touza & Prado, 2019).
170 National Elder Abuse Prevalence Study: Final Report
For Aboriginal and Torres Strait Islander groups, the understanding of elder abuse is situated within the history
of colonisation and its consequences, including dispossession from traditional lands, removal of children and the
disruption of cultural norms in relation to respect and care for elders (Gooda, 2012). Research on elder abuse
among Aboriginal Torres Strait Islander communities is scarce but existing sources have drawn attention to
cultural norms concerning resource sharing being distorted as a lever for financial abuse (Gooda, 2012; Western
Australia Office of the Public Advocate, 2005; Western Australia Parliament, 2018).The ‘I never thought it would
happen to me’ report concluded that Aboriginal and Torres Strait Islander older people are at ‘greater risk’ of
elder abuse and that it may occur at a younger age for these groups.
Following recommendations for research from the inquiry’s report (Western Australia Parliament, 2018, Rec 4),
the WA Department of Communities commissioned IPS Management Consultants to examine elder abuse among
older Aboriginal and Torres Strait Islander people in metropolitan, regional and remote areas in Western Australia
in early 2021. Further research on elder abuse among Aboriginal and Torres Strait Islander groups outside of
WA is also required, including research that takes into account the diverse circumstances of communities in
rural, regional and remote areas (Warren & Blundell, 2019). In keeping with recognition of the need for policy
and services to be developed in partnership with Aboriginal and Torres Strait Islanders in a culturally safe way
(e.g. Western Australia Parliament, 2018, Rec. 5), such research should be led by Aboriginal and Torres Strait
Islander communities (National Health and Medical Research Council [NHMRC], 2018).
Locally and internationally, there is very limited empirical evidence on elder abuse in LGBTIQ contexts (Dow et al.,
2020, Westwood, 2019). The limited existing evidence suggests that LGBTIQ groups are at potentially higher
risk of experiencing elder abuse, including in forms ‘relating to their minority sexualities and/or gender identities,
while also being at greater risk of non-disclosure due to their social marginalisation’ (Westwood, 2019, p. 110).
Consistent with Westwood’s observation, Seniors Rights Victoria points to a range of factors that may feed
into experiences of elder abuse among LGBTIQ people, including homophobia and transphobia and a potential
history of estrangement from family and friends (Seniors Rights Victoria, 2018), possibly leading to heightened
levels of social isolation.
Research specifically focusing on elder abuse experiences among LGBTIQ people is needed. Such research
should examine the form elder abuse takes in this context, its impact and what effective service responses look
like from an LGBTIQ perspective.
It is also important to further understand neglect, including the experiences of the people in the ‘no-one to
help’ group (chapter 6). Given that neglect has characteristics that set it apart from other forms of abuse, more
in‑depth understanding of how it arises and how it may be addressed, including through overcoming barriers to
help seeking, is warranted.
It is important to note in this context that the Royal Commission into Aged Care Quality and Safety found that
the aged care system was difficult to enter, navigate and access (summary report, pp. 65–66). It also noted that
wait times to receive care at home could be lengthy, ranging from seven to 34 months depending on the level of
care required. Even when care packages were assigned, some people received less care than required and did
not have access to the specific services they needed.
Summary
In summary, this report has set out the findings of the largest study to date on elder abuse in Australia. The
research suggests that one in six Australians aged 65 and over and living in the community has experienced
elder abuse in the preceding 12 months.
In applying a socio-ecological lens, the research has established factors correlated with elder abuse at the
individual, family and community level. It has also highlighted challenges for the service system and areas where
further policy development is required. Additionally, it has identified a need for further research to continue to
develop the evidence base.
Figure 14.1 depicts the main aspects of the findings using a socio-ecological framework with insights particularly
well-developed at the individual, family and community level. The following paragraphs describe the main points
in the figure.
Although the research shows that anyone can experience elder abuse, the study indicates that factors associated
with greater vulnerability for victims at the individual level are social isolation, mental ill-health, physical ill-health,
disability and renting or paying off a home. Factors associated with perpetrators (according to victims), are
mental ill-health, physical ill-health, substance misuse and unemployment. Issues common to both victims and
perpetrators are dependence (in either direction or at a mutual level) and shared residence.
Chapter 14: Summary and implications 171
These findings, together with insights from other research (ABS, 2020b; Acierno et al., 2010; National Initiative
for the Care of the Elderly, 2015), indicate a need to understand complex victim–perpetrator dynamics and how
histories of child abuse, family violence and other trauma may influence susceptibility to experiencing elder
abuse at this later life stage or indeed susceptibility to perpetrating elder abuse.
Given that a significant proportion of elder abuse involves family members, the family-level factors indicated in
this research are also significant. The association between a history of relationship breakdown and vulnerability
to experiencing elder abuse, together with findings on financial stress, point to the significance of both
relationship issues and financial stress in the constellation of issues that underpin elder abuse. It is also significant
that the relational context in which elder abuse occurs inhibits disclosure and action and potentially compounds
its impact. Family-level influences relevant to perpetration include entitlement, the opportunity afforded by being
in a position of trust and the fact that very limited help-seeking behaviour means that often there may be no or
limited adverse consequences to deter perpetration.
Findings on financial management practices and the susceptibility of women who receive help in managing
financial matters to financial abuse point to a need to increase financial literacy among older women to reduce
opportunities for abuse. Better understanding of norms and practices around intergenerational support, as well
as factors that support family cohesion across the life course, would support insight into these dynamics within
families and how positive practices could be enhanced.
Community dynamics are significant in a number of ways. First, members of an older person’s community, such
as neighbours, friends and acquaintances, are a significant group of perpetrators. In this context, vulnerability
stemming from ill-health and isolation on the part of the victim are important in considering dynamics in this
context. Additionally, attitude-related issues, including ageism and lack of recognition of abusive behaviours,
together with the secrecy that surrounds abuse, are significant issues in abuse conducive circumstances.
Second, next to family members, friends are the group to whom an older person is most likely to disclose
abuse. These findings reinforce the need to raise awareness at the community level of what elder abuse is, how
to support a person who discloses abuse, and what help is available for people experiencing abuse. They also
point to the importance of a community role in looking out for older people and ‘calling out’ abusive behaviour
when seen.
Although this research was not intended to directly examine the service system, the findings about levels of
non‑disclosure and reliance on avoidant strategies, demonstrate a need to assess whether the mechanisms
available to identify and address elder abuse are appropriate, adequate and accessible. Importantly, in
developing further responses, the question that should be asked is ‘what do people who experience elder abuse
want from services?’
Further development of responses needs to appropriately manage the complex relationship contexts within
which elder abuse occurs so that the potential negative consequences of disclosure do not place the older
person in a worse position than they would have been in had disclosure not occurred. This complex relationship
context, the possibility that in many cases the older person may want to maintain an ongoing relationship with
the perpetrator and the role that perpetrator-related problems such as mental and physical ill-health play in
driving abuse, place focus on the need to reduce opportunity and address perpetrator problems, as well as
improve understanding on what drives perpetrator behaviour.
172 National Elder Abuse Prevalence Study: Final Report
Research to understand elder abuse for Aboriginal and Torres Strait Islander people,
LGBTIQ, CALD (more detail) and those with cognitive impairment
References
Acierno, R., Hernandez-Tejada, M., Anetzberger, G., Loew, D., & Muzzy, W. (2017). The National Elder Mistreatment Study:
An 8-year longitudinal study of outcomes. Journal of Elder Abuse and Neglect, 29(4), 254–269.
Acierno, R., Hernandez-Tejada, M., & Kilpatrick, D. (2010). Prevalence and correlates of emotional, physical, sexual and financial
abuse and potential neglect in the United States: The National Elder Mistreatment Study. American Journal of Public Health,
100(2), 292–297.
Acierno, R., Hernandez-Tejada, M., Muzzy, W., & Steve, K. (2009) National Elder Mistreatment Study. Washington: National
Institute of Justice. Retrieved from www.ojp.gov/pdffiles1/nij/grants/226456.pdf.
Acierno, R., Steedly, M., Hernandez-Tejada, M., Frook, G., Watkins, J., & Muzzy, W. (2020). Relevance of perpetrator identity to
reporting of elder financial and emotional mistreatment. Journal of Applied Gerontology, 39(2), 221–225.
Adams, V. M., Bagshaw, D., Wendt, S., & Zannettino, L. (2014). Financial abuse of older people by a family member: A difficult
terrain for service providers in Australia. Journal of Elder Abuse & Neglect, 26(3), 270–290.
Ahnlund, P., Andersson, T., Snellman, F., Sundström, M., & Heimer, G. (2020). Prevalence and correlates of sexual, physical, and
psychological violence against women and men of 60 to 74 years in Sweden. Journal of Interpersonal Violence, 35(5–6),
1539–1561.
Australian Bureau of Statistics (ABS). (2012). Information Paper: Use of the Kessler Psychological Distress Scale in ABS Health
Surveys, Australia, 200708. Canberra: ABS.
Australian Bureau of Statistics. (2013). Statistical Language – Statistical Language Glossary. Canberra: ABS.
Australian Bureau of Statistics. (2014). Australian historical population statistics, 2014. ABS cat. no. 3105.0.65.001. Canberra: ABS
Australian Bureau of Statistics. (2017a). Australian Demographic Statistics, Jun 2016. ABS cat. no. 3101.0. Canberra: ABS.
Australian Bureau of Statistics. (2017b). Census of Population and Housing: Reflecting Australia – Stories from the Census, 2016.
Cultural diversity in Australia, 2016. Canberra: ABS.
Australian Bureau of Statistics. (2017c). Personal safety, Australia: Methodology. Canberra: ABS. Retrieved from
www.abs.gov.au/methodologies/personal-safety-australia-methodology/2016
Australian Bureau of Statistics. (2017d). Personal safety, Australia: Statistics for family, domestic, sexual violence, physical assault,
partner emotional abuse, child abuse, sexual harassment, stalking and safety. Canberra: ABS. Retrieved from
www.abs.gov.au/statistics/people/crime-and-justice/personal-safety-australia/latest-release accessed 24/10/2020
Australian Bureau of Statistics. (2018a). Socio‑economic Indexes for Areas (SEIFA) – Technical paper (Catalogue no.
2033.0.55.001). Canberra: ABS. Retrieved from www.ausstats.abs.gov.au/ausstats/subscriber.nsf/0/756EE3DBEFA869EFCA2
58259000BA746/$File/SEIFA%202016%20Technical%20Paper.pdf
Australian Bureau of Statistics. (2018b). Population Projections, Australia 2017–2066 (Catalogue no. 3222.0). Canberra: ABS.
Australian Bureau of Statistics. (2019a). Disability, Ageing and Carers, Australia: Summary of findings. Canberra: ABS.
Australian Bureau of Statistics. (2019b). Characteristics and outcomes of childhood abuse. Canberra: ABS.
Australian Bureau of Statistics. (2020a). Basic survey design. Canberra: ABS. Retrieved from www.abs.gov.au/websitedbs/
D3310114.nsf/home/Basic+Survey+Design+-+Errors+in+Statistical+Data
Australian Bureau of Statistics. (2020b). National, state and territory population, June 2020. Canberra: ABS. Retrieved from
www.abs.gov.au/ausstats/abs@.nsf/Lookup/by%20Subject/4125.0~Sep%202017~Main%20Features~Selected%20Highlights~2
Australian Institute of Family Studies and Social Research Centre. (2019). Elder Abuse National Research Stage 1 – Strengthening
the evidence base. Technical report – Survey design (unpublished).
Australian Institute of Health and Welfare (AIHW). (2018). Older Australian at a glance. Canberra: AIHW.
Australian Institute of Health and Welfare. (2019a). Insights into vulnerabilities of Aboriginal and Torres Strait Islander people
aged 50 and over. Canberra: AIHW.
Australian Institute of Health and Welfare. (2019b). Rural & remote health. Cat. no. PHE 255. Canberra: AIHW. Retrieved from
www.aihw.gov.au/reports/rural-remote-australians/rural-remote-health
Australian Institute of Health and Welfare. (2020). Sexual assault in Australia. Canberra: AIHW.
Australian Law Reform Commission (ALRC). (2016). Elder abuse: Discussion paper 83 (DP 83). Sydney: ALRC.
174 National Elder Abuse Prevalence Study: Final Report
Australian Law Reform Commission. (2017). Elder abuse: A national legal response. Sydney: ALRC. Retrieved from
www.alrc.gov.au/publications/elder-abuse-report
Bagshaw, D., Wendt, S., & Zannettino, L. (2009). Preventing the abuse of older people by their family members (Stakeholder
Paper 7). Sydney: Domestic Violence Clearing House.
Bagshaw, D., Wendt, S., Zannettino, L., & Adams, V. (2013) Financial abuse of older people by family members: views and
experiences of older Australians and their family members. Australian Social Work, 66(1) 86–103.
Baker, P., Francis. D., Hairi, N., Othman, S., & Choo, W. (2016). Interventions for preventing abuse in the elderly. Cochrane
Database of Systematic Reviews, 8.
Band-Winterstein, T., Goldblatt, H., & Lev, S. (2019). Breaking the taboo: Sexual assault in late life as a multifaceted
phenomenon‑toward an integrative theoretical framework. Trauma, Violence & Abuse, 1–13.
Batchelor, F., Savvas, S., Dang, C., Goh, A.M.Y., Levinger, P., Peck, A. et al. (2020). Inside the system: Aged care residents’
perspectives. Parkville, Vic.: National Ageing Research Institute.
The Benevolent Society. (2017). The drivers of ageism. Sydney: The Benevolent Society. Retrieved from
www.benevolent.org.au/ArticleDocuments/340/Ageism_Full_Report_Final.pdf.aspx
Biggs, S., & Haapala, I. (2013). Elder mistreatment, ageism, and human rights. International Psychogeriatrics, 25(8), 1299–1306.
Board of Taxation. (2019). Review of Granny Flat Arrangements. Canberra: Commonwealth of Australia. Retrieved from
taxboard.gov.au/sites/taxboard.gov.au/files/2020-10/Final%20-%20Review%20of%20granny%20flat%20arrangements_0.pdf
Boersig, J., & Illidge, D. (2018). Addressing elder abuse: Perspectives from the community legal sector in the ACT. Macquarie Law
Journal, 18, 93–113.
Bows, H. (2018). Sexual Violence Against Older People: A Review of the Empirical Literature. Trauma, Violence & Abuse, 19(5),
567–583.
Bronfenbrenner, U. (1986). Ecology of the family as a context for human development: Research perspectives. Developmental
Psychology, 22(6), 723–742.
Blundell, B., Clare, J., Moir, E., Clare, M., & Webb, E. (2017). Review into the prevalence and characteristics of elder abuse in
Queensland. Perth, WA: Curtain University and Murdoch University.
Burnes, D., Lachs, M. S., & Pillemer, K. (2018). Addressing the measurement challenge in elder abuse interventions: Need for a
severity framework. Journal of Elder Abuse & Neglect, 30(5), 402–407.
Burnes, D., Acierno, R., & Hernandez-Tajada, M., (2019b). Help-seeking among victims of elder abuse: Findings from the National
Elder Mistreatment Study. Journals of Gerontology: Social Sciences, 5(74), 891–896.
Burnes, D., Lachs, M., Burnette, D., & Pillemer, K., (2019a). Varying appraisals of elder mistreatment among victims: Findings from
a population study. Journals of Gerontology: Social Sciences, 5(74), 881–890. doi:10.1093/geronb/gbx005
Burnight, K., & Mosqueda, L. (2011). Theoretical model development in elder mistreatment, Final report submitted to the National
Institute of Justice, grant number 2005-IJ-CX-0048, May 2011, NCJ 234488. Retrieved from www.ncjrs.gov/pdffiles1/nij/
grants/234488.pdf
Cary, L. A., Chasteen, A. L., & Remedios, J. (2017). The ambivalent ageism scale: Developing and validating a scale to measure
benevolent and hostile ageism. The Gerontologist, 57(2), e27–36.
Caxton Legal Centre Inc. (2007, February). Submission 112, Inquiry into older people and the law. Retrieved from www.aph.gov.
au/Parliamentary_Business/Committees/House_of_Representatives_Committees?url=/laca/olderpeople/subs/sub112.pdf
Chesterman, J. (2016). Taking control: Putting older people at the centre of elder abuse response strategies. Australian Social
Work, 69(1), 115–124.
Chesterman, J. & Bedson, L. (2017). Are national elder abuse prevalence studies inclusive of the experiences of people with
cognitive impairment? Findings and recommendations for future research. Melbourne: Office of the Public Advocate.
Clare, M., Blundell, B., & Clare, J. (2011). Examination of the extent of elder abuse in Western Australia: A qualitative and
quantitative investigation of existing agency policy, service responses and recorded data. Perth: Advocare Inc.
Clare, M., Clare, B., Blundell, B., & Clare, J. (2014). Conceptualising elder abuse: Does this label fit? Communities, Children and
Family Australia, 8(1), 37–48.
Commonwealth of Australia. (2019). Everybody’s business: Stocktake of elder abuse awareness, prevention and response
activities in Australia. Canberra: Attorney-General’s Department.
Conrad, K., Liu, P.-J., Iris, M. (2019). Examining the role of substance abuse in elder mistreatment: Results from mistreatment
investigations. (2019). Journal of Interpersonal Violence, 34(2) 366–391.
Cook., J., Dinnen, S., & O’Donnell, C. (2011). Older women survivors of physical and sexual violence: A systematic review of the
quantitative literature. Journal of Women’s Health, 7(20) 1075–1081.
Coumarelos, C., Macourt, D., People, J., McDonald, H., Wei, Z., Iriana, R. & Ramsey, S. (2012). Legal Australia-Wide Survey: Legal
Need in Australia. Sydney, NSW: Law and Justice Foundation of NSW.
Council of Attorneys General (CAG). (2019). National Plan to Respond to the Abuse of Older Australian (Elder Abuse) 2019–2023.
Canberra: Attorney General’s Department.
Cripps, K., Diemer, K., Honey, N., Mickle, J., Morgan, J., Parkes, A. et al. (2019). Attitudes towards violence against women and
gender equality among Aboriginal people and Torres Strait Islanders: Findings from the 2017 National Community Attitudes
towards Violence against Women Survey (NCAS) (ANROWS Insights, Issue 03/2019). Sydney: ANROWS.
Darzins, P., Lowndes , G., & Wainer, J. (2009). Financial Abuse of Elders: A Review of the Evidence Protecting Elders Assets Study
Melbourne. Melbourne: Monash University.
References 175
DeLiema, M., Yonashiro-Cho, J., Gassoumis, Z. D., Yon, Y., & Conrad, K. J. (2018). Using latent class analysis to identify profiles of
elder abuse perpetrators. Journals of Gerontology Series B: Psychological Sciences & Social Sciences, 73(5), e49–e58.
Department of Home Affairs. (2019). Australia’s Migration Trends. Canberra: Commonwealth of Australia.
Department of Prime Minister and Cabinet. (2019). Regulation Impact Statement Update. Retrieved from
ris.pmc.gov.au/2019/12/12/national-register-enduring-power-attorney-instruments
Dominguez, S., Storey, J., & Glorney, E. (2019). Help-seeking behaviour in victims of elder abuse: A systematic review.
Trauma, Violence and Abuse, 1–15.
Dong, X. (2015). Elder abuse: Systematic review and implications for practice. Journal of the American Geriatrics Society, 63(6),
1214–1238.
Dow, B., & Brijnath, B. (2019). Elder abuse: Context, concepts and challenges. Australia’s welfare 2019: Data insights. Canberra,
ACT: Australian Institute of Health and Welfare.
Dow, B., Vrantsidis, F., Joosten, M., & Gahan, L. (2020). Elder abuse in Australia. In M. Shankardass (Ed.), International Handbook
of Elder Abuse and Mistreatment (pp. 559–574). Singapore: Springer Nature.
Elder Abuse Prevention Unit (EAPU). (2019). Year in review. Brisbane: UnitingCare.
Elder Abuse Prevention Unit. (2020). Year in review. Brisbane: UnitingCare.
EKOS Research Associates Inc. (2011). Follow up and final awareness and perceptions of elder abuse survey 2010. Ottawa:
Justice Canada.
Family Law Council. (2012a). Improving the family law system for Aboriginal and Torres Strait Islander clients. Canberra:
Commonwealth of Australia.
Family Law Council. (2012b). Improving the family law system for culturally and linguistically diverse clients. Canberra:
Commonwealth of Australia.
Forum on Global Violence Prevention; Board on Global Health; Institute of Medicine; National Research Council. (2014).
Elder Abuse and Its Prevention: Workshop Summary. Washington (DC): National Academies Press (US).
Frydenberg, J. & Sukker, M. (2020, 5 October). Removing capital gains tax for granny flats (Media Release). Retrieved from
ministers.treasury.gov.au/ministers/josh-frydenberg-2018/media-releases/removing-capital-gains-tax-granny-flats
Gil, A., Kislaya, I., Santos, A., Nunes, B., Nicolau, R., & Fernandes, A. (2015). Elder abuse in Portugal. Journal of Elder Abuse and
Neglect, 27, 174–194.
Gillbard, A. (2019). Differences in risks and experiences of familial elder abuse for victims with cognitive impairments,
Family & Relationship Services Australia National Conference, Lovedale NSW. Retrieved from frsa.org.au/wp-content/
uploads/2019/11/2019-FRSA-ejournal_web.pdf
Gooda, M. (2012). AAG Elder Abuse and Neglect Conference. Canberra: Australian Human Rights Commission. Retrieved from
humanrights.gov.au/about/news/speeches/aag-elder-abuse-and-neglect-conference#fn13
Herrenkohl, T., Fedina, L., Roberto, K., Raquet, K., Hu, R., Rousson, A., & Mason, A. (2020). Child maltreatment, youth violence,
intimate partner violence, and elder mistreatment: A review and theoretical analysis of research on violence across the life
course. trauma, violence and abuse. (Online publication). doi.org/10.1177/1524838020939119
Hightower J., Smith, M., & Hightower, H. (2006). Hearing the voices of abused older women. Journal of Gerontological Social
Work, 46(3–4), 205–227.
Hill, T., & Katz, I. (2019). Insights into abuse of older people: Analysis of Australian Bureau of Statistics datasets (SPRC report).
Sydney, NSW: Social Policy Research Centre UNSW Sydney.
Ho, C., Wong, S., & Ho, R. (2017). Global analysis of elder abuse: A meta-analysis and meta-regression. East Asian Archives of
Psychiatry, 27, 43–55.
Jackson, S. L., & Hafemeister, T. L. (2012). Pure financial exploitation vs. hybrid financial exploitation co-occurring with physical
abuse and/or neglect of elderly persons. Psychology of Violence, 2(3), 285–296.
Jackson, S. L., & Hafemeister, T. L. (2013). Research in brief: Understanding elder abuse. New directions for developing theories
of elder abuse occurring in domestic settings. Washington, D.C.: National Institute of Justice, Office of Justice Programs, U.S.
Department of Justice. Retrieved from www.ncjrs.gov
Jeon, G.-S., Cho, S.-I., Choi, K., & Jang, K.-S. (2019). Gender differences in the prevalence and correlates of elder abuse in a
community dwelling older population in Korea. International Journal of Environmental Research and Public Health, 16, 100.
doi.org/10.3390/ijerph16010100
Joosten, M., Dow, B., & Gally, E. (2019). Violence against older women: Intergenerational family violence. In H. Bows (Ed.),
Violence Against Older Women, Volume 1, (pp. 97–111). Palgrave Studies in Victims and Victimology.
Joosten, M., Gartoulla, P., Feldman, P., Brijnath, B., & Dow, B. (2020). Seven years of elder abuse data in Victoria (2012–2019).
Melbourne: National Ageing Research Institute and Seniors Rights Victoria.
Joosten, M., Vrantsidis, F., & Dow, B. (2017). Understanding elder abuse: A scoping study. Melbourne: University of Melbourne and
the National Ageing Research Institute.
Kaspiew, R., Carson, R., & Rhoades, H. (2016). Elder abuse: Understanding issues, frameworks and responses (Research Report
No. 35). Melbourne: Australian Institute of Family Studies.
Kaspiew, R., Carson, R., Dow, B., Qu, L., Hand, K., Roopani, D. et al. (2019). Elder Abuse National Research – Strengthening the
Evidence Base: Research definition background paper. Melbourne: Australian Institute of Family Studies.
Kessler, R. C., Barker, P. R., Colpe, L. J., Epstein, J. F., Gfroerer, J. C., Hiripi, E. et al. (2003). Screening for serious mental illness in
the general population. Archives of General Psychiatry 60(2), 184–189.
176 National Elder Abuse Prevalence Study: Final Report
Kessler, R. C., Green, J. G., Gruber, M. J., Sampson, N. A., Bromet, E., Cuitan, M. et al. (2010). Screening for serious mental illness in
the general population with the K6 screening scale: results from the WHO World Mental Health survey initiative. International
Journal of Methods in Psychiatric Research, 19(S1), 4–22.
King, C., Wainer, J., Lowndes, G., Darzins, P., & Owada, K. (2011). For love or money: Intergenerational management of older
Victorians’ assets. Protecting Elders Assets Study. Melbourne: Monash University.
Kinnersley, P. (2020). Hansard House of Representative Standing Committee on Social Policy and Legal Affairs, Family domestic
and sexual violence inquiry, Hearing 7, September 2020, 9 18.
Koga, C., Hanazato, M., Tsuji, T., Suzuki, N., & Kondo, K. (2019). Elder abuse and social capital in older adults: The Japan
Gerontological Evaluation Study. Gerontology, 1–11.
Kulakçı Altıntas, H., & Korkmaz Aslan, G. (2020). Prevalence of elder abuse among community-dwelling older adults in Turkey
and its associated factors. Psychogeriatrics, 20(1), 3–10.
Lacey, W. (2014). Neglectful to the point of cruelty? Elder abuse and the rights of older persons in Australia. Sydney Law Review,
36, 99–130.
Lacey, W., Middleton, H., Bryant, L., & Garnham, B. (2017). Prevalence of elder abuse in South Australia: Final report. Current data
collection practices of key agencies. Retrieved from www.sahealth.sa.gov.au/wps/wcm/connect/96f22500421782fe9d2bff40
535c9bd4/Prevalence+of+Elder+Abuse+in+South+Australia+-+Final+Report_February+2017.pdf?MOD=AJPERES&CACHEID
=96f22500421782fe9d2bff40535c9bd4
Lawton, M. P., & Brody, E. M. (1969). Assessment of older people: Self-maintaining and instrumental activities of daily living.
The Gerontologist, 9(3, Pt 1), 179–186.
Lifespan of Greater Rochester Inc. (2011). Under the Radar: New York State Elder Abuse Prevalence Study. Self-reported
prevalence and documented case surveys (final report). New York: Weill Cornell Medical Center of Cornell University &
New York City Department of the Aging.
Lord, M., McMahon, K., & Nivelle, S. (n.d.) Preventing Elder Abuse: A Literature Review for the SMPCP Elder Abuse Prevention
Network. Melbourne: SMCP EAPN Steering Committee.
Mann, R., Horsley, P., Barrett, C., & Tinney, J. (2014). Norma’s Project: A research study into the sexual assault of older women in
Australia (ARCSHS Monograph Series No. 98). Melbourne: Australian Research Centre in Sex, Health and Society, La Trobe
University.
Mayo, A. M., & Wallhagen, M. I. (2009). Considerations of informed consent and decision-making competence in older adults
with cognitive impairment. Research in Gerontological Nursing, 2(2), 103–111.
McCook, S., & Powell, A. (2020). Feasibility study into the possible inclusion of social norms measures within the 2021 National
Community Attitudes towards Violence against Women Survey (NCAS) (Research report, 02/2020). Sydney, NSW: ANROWS.
Mysyuk, Y., Gerardus, R., Westendorp, R. G., & Lindenberg, J. (2016). How older persons explain why they became victims of
abuse. Age and Ageing, 45(5), 696–702.
McDonald, L., & Thomas, C. (2013). Elder abuse through a life course lens. International Psychogeriatrics, 25(8), 1235–1243.
Miskovski, K. (2014). Preventing financial abuse of people with dementia. Sydney: Alzheimers Australia NSW. Retrieved from
nsw.fightdementia.org.au/sites/default/files/20140618-NSW-Pub-DiscussionPaperFinancialAbuse.pdf
National Clearinghouse on Abuse in Later Life (NCALL). (2011). Abuse in Later Life Power & Control Wheel. Madison, WI: NCALL.
Retrieved from www.ncall.us//FileStream.aspx?FileID=27
National Health and Medical Research Council. (2018). Keeping research on track II. Canberra: Commonwealth of Australia.
National Initiative for the Care of the Elderly. (2015). Into the light: National Survey on the Mistreatment of Older Canadians 2015.
Toronto: National Initiative for the Care of the Elderly.
Naughton, C., Drennan, J., Treacy, M. P., Lafferty, A., Lyons, I., Phelan, A. et al. (2010). Abuse and neglect of older people in Ireland:
Report on the National Study of Elder Abuse and Neglect. Dublin, Ireland: National Centre for the Protection of Older People,
University College Dublin. Retrieved from www.ncpop.ie/userfiles/file/Prevalence%20study%20summary%20report.pdf
New South Wales Parliament, Legislative Council. General Purpose Standing Committee No 2. (2016). Elder abuse in New South
Wales. Sydney: NSW Parliament.
Ng, W., Lim, Z., & Kadir, M. (2020) A Multidisciplinary Care Management Approach to Preventing and Managing Elder Abuse:
The Singapore Experience. In M. Shankardass (Ed.), International Handbook of Elder Abuse and Mistreatment, (pp. 559–574).
Singapore: Springer Nature.
Office of the Public Advocate (Vic.). (2020). Future planning for decision-making and the law in Victoria. A resource for lawyers
who practise in future planning. Melbourne: Office for the Public Advocate, Victoria.
Office of the Public Advocate (Qld) & Queensland Law Society. (2010). Elder Abuse: How well does the law in Queensland cope?
Qld: Office of the Public Advocate (Qld) & Queensland Law Society.
O’Keeffe, M., Hills, A., Doyle, M., McCreadie, C., Scholes, S., Constantine, R. et al. (2007). UK Study of Abuse and Neglect of Older
People: Prevalence survey report. London: National Centre for Social Research.
Oltmann, S.M. (2016). Qualitative interviews: A methodological discussion of the interviewer and respondent context. Qualitative
Social Research, 17(2), art.15.
Our Watch, Australia’s National Research Organisation for Women’s Safety (ANROWS) and VicHealth. (2015). Change the
story: A shared framework for the primary prevention of violence against women and their children in Australia. Melbourne:
Our Watch.
Pence, E., & Paymar, M. (1993). Education groups for men who batter: The Duluth model. Springer Publishing Company.
doi.org/10.1891/9780826179913
References 177
Peiros, K., & Smyth, C. (2018). Professionals’ role in preventing elder financial abuse. Taxation in Australia, 52(9), 510.
Piggott., J. (2016). Population ageing in Australia: National policy challenges and future directions. In H. Kendig., P. McDonald &
J. Piggot (Eds.), Population ageing and Australia’s future. Canberra: ANU Press.
Pillemer, K., Burnes, D., Riffin, C., & Lachs, M. (2016). Elder Abuse: Global Situation, Risk Factors, and Prevention Strategies.
Gerontologist, 56(2), 194–205.
Purser, K., Cockburn, T., Cross, C., & Jacmon, H. (2018). Alleged financial abuse of those under an enduring power of attorney:
An exploratory study. British Journal of Social Work, 48(4), 887–905.
Qu, L., Carson, R., Kaspiew, R., Weston, R., Commerford, J., Dean, A. et al. (2017). Elder abuse prevalence scoping study. A report
to the Australian Government Attorney-General’s Department (unpublished). Melbourne: Australian Institute of Family Studies.
Ries, N. (2019). Enduring Powers of Attorney and Financial Exploitation of Older People: A Conceptual Analysis and Strategies
for Prevention. Journal of Aging and Social Policy. doi: 10.1080/08959420.2019.1704143
Roche, B. (2018). To investigate if supported decision making will reduce elder financial abuse in Australia. Acton, ACT:
The Winston Churchill Memorial Trust of Australia.
Royal Commission into Aged Care Quality and Safety. (2019). Interim Report: Neglect. Adelaide: Royal Commission into Aged
Care Quality and Safety.
Royal Commission into Aged Care Quality and Safety. (2020). Counsel assisting’s final submissions. Adelaide: Royal Commission
into Aged Care Quality and Safety.
Royal Commission into Aged Care Quality and Safety. (2020). Experimental Estimates of the Prevalence of Elder Abuse in
Australian Aged Care Facilities. Adelaide: Royal Commission into Aged Care Quality and Safety.
Royal Commission into Aged Care Quality and Safety. (2021). Final Report: Care, Dignity and Respect. Adelaide: Royal
Commission into Aged Care Quality and Safety.
Royal Commission into Aged Care Quality and Safety. (2021). Final Report: Summary. Adelaide: Royal Commission into Aged
Care Quality and Safety.
Roberto, K. A., & Teaster, P. B. (2017). Theorizing elder abuse. In X. Dong (Ed.), Elder Abuse: Research, Policy and Practice.
Springer International Publishing.
Sandmoe, A., Wentzel-Larsen, W., & Hjemdal, K. (2017). Violence and abuse against elderly people in Norway: A national
prevalence study (A/S Report No. 9/2017) Oslo: Norwegian Centre for Violence and Traumatic Stress Studies.
Schober, M. F. (2018). The future of face-to-face interview. Quality Assurance in Education, 26(2), 290–302.
Seniors Rights Victoria. (2017). Submission 117, Australian Law Reform Commission Inquiry into Protecting the Rights of Older
Australians from Abuse.
Seniors Rights Victoria. (2018). Elder abuse, gender and sexuality. Melbourne, Vic.: Seniors Rights Victoria. Retrieved from
seniorsrights.org.au/wp-content/uploads/2018/05/Elder-Abuse-Gender-and-Sexuality.pdf
Santos, A. J., Nunes, B., Kislaya, I., Gil, A. P., & Ribeiro, O. (2019). Elder abuse victimization patterns: Latent class analysis using
perpetrators and abusive behaviours, BMC Geriatrics, 19, doi: 10.1186/s12877-019-1111-5
Slaughter, S., Cole, D., Jennings, E., & Reimer, M. A. (2007). Consent and assent to participate in research from people with
dementia. Nursing Ethics, 14(1), 27–40.
Spike, C. (2015). The EAPU helpline: Results of an investigation of five years of call data. Report for the International Association
of Gerontology and Geriatrics Asia & Oceania Regional Congress 2015. Chermside Central, Qld: Elder Abuse Prevention Unit,
UnitingCare Community.
State Trustees. (2017). Financial elder abuse final report. Melbourne: State Trustees.
State Trustees. (2016). Financial abuse family legacy and elder abuse. Melbourne: State Trustees.
Stodolska, A., Parnicka, A., Tobiasz-Adamczyk, B., & Grodzicki, T. (2020). Exploring elder neglect: New theoretical perspectives
and diagnostic challenges. The Gerontologist, 60(6).
Storey, J. E. (2020). Risk factors for elder abuse and neglect: A review of the literature. Aggression & Violent Behavior, 50, 101–339.
Swift, H., Abrams, D., Marques, S., Vauclair, C., Bratt, C., Lima, M. (2018). Ageism in the European Region: Finding from the
European Social Survey in Perspectives on Elder Abuse and Mistreatment from Selected Countries and Regions. Springer
International Publishing.
Teresi, J., Burnes, D., Skowron, E., Dutton, M., Mosqueda, L., Lachs, M., & Pillemer, K. (2016). State-of-the-science on prevention
or elder abuse and lessons learned from child abuse and domestic violence prevention: Towards a conceptual framework for
research. Journal of Elder Abuse and Neglect, 28(4–5), 263–300.
Tilse, C., Setterlund, D., Wilson, J., & Rosenman, L. (2005). Minding the money: A growing responsibility for informal carers.
Ageing and Society, 25(2), 215–227.
Tilse, C., Wilson, J., Settlerlund, D. S., & Rosenman, L. S. (2005). Older people’s assets: A contested site. Australasian Journal on
Ageing, 24, S51–S56.
Tilse, C., Wilson, J., White, B., Willmott, L., & McCawley, A. L. (2014). Enduring powers of attorney: Promoting attorneys’
accountability as substitute decision makers. Australasian Journal on Ageing, 33(3), 193–197.
Touza, C., & Prado, C. (2019). Risk factors for domestic elder abuse and cognitive decline: A comparative study with a Spanish
sample. Journal of Family Studies, 25(4), 460.
Vittinghoff, E., & McCulloch, C. E. (2006). Relaxing the rule of ten events per variable in logistic and cox regression. American
Journal of Epidemiology, 165(6), 710–718.
178 National Elder Abuse Prevalence Study: Final Report
Vrantsidis F., Dow B. Joosten M., Walmsley M., & Blakey J. (2016). The Older Person’s Experience: Outcomes of Interventions into
Elder Abuse. Melbourne: National Ageing Research Institute and Senior’s Rights Victoria. Retrieved from
seniorsrights.org.au/wp-content/uploads/2016/06/SRV-NARI-Outcomes-Report-2016-FINAL-Web-6-June-2016.pdf
Wainer, J., Darzins, P., & Owada, K. (2010). Prevalence of financial elder abuse in Victoria. Melbourne: Monash University. Retrieved
from www.eapu.com.au/uploads/ research_resources/VIC Prevalence_of_FInancial_Elder_Abuse_MAY_2010-Monash.pdf
Wainer, J., Owada, K., Lowndes, G., & Darzins, P. (2011). Diversity and financial elder abuse in Victoria: Protecting Elders’ Assets
Study. Melbourne: Faculty of Medicine, Nursing and Health Sciences, Monash University. Retrieved from eccv.org.au/library/
file/projects/Monash_University_Diversity_and_financial_elder_abuse_in_Victoria.pdf
Warren, A., & Blundell, B. (2018a). Elder abuse in rural & remote communities: Social policy, prevention and responses. Perth, WA:
Curtin University and the Older Person’s Advocacy Network.
Warren, A., & Blundell, B. (2018b). Bridging the gap between responses to elder abuse and responses to family and domestic
violence in rural and remote communities. In Be the Change: Leaving No One Behind, Family & Relationship Services Australia
National Conference e-Journal, edition 3-November 2018.
Warren, A., & Blundell, B. (2019). Addressing elder abuse in rural and remote communities: Social policy, prevention and
responses, Journal of Elder Abuse & Neglect, 31(4–5), 424–436.
Webb, E. (2018). Housing an ageing Australia: the ideal of security of tenure and the undermining effect of elder abuse.
Macquarie Law Journal, 18, 57–78.
Webster, K., Diemer, K., Honey, N., Mannix, S., Mickle, J., Morgan, J. et al. (2018). Australians’ attitudes to violence against women
and gender equality: Findings from the 2017 National Community Attitudes towards Violence against Women Survey.
Sydney: ANROWS.
Wendt, S., Bagshaw, D., Zannettino, L., & Adams, V. (2015). Financial abuse of older people: A case study. International Social
Work, 58(2), 287–296. doi:10.1177/0020872813477882.
Western Australia, Office of the Public Advocate. (2005). Mistreatment of older people in Aboriginal communities project:
An investigation into elder abuse in Aboriginal communities. Perth: Office of the Public Advocate.
Western Australia Parliament. (2018). ‘I Never Thought It Would Happen to Me’: When Trust is Broken: Final Report of the Select
Committee into Elder Abuse. Perth: WA Parliament.
Weston, R., & Qu, L. (2016). Attitudes towards intergenerational support (Australian Family Trends No. 11). Melbourne: Australian
Institute of Family Studies.
Westwood, S. (2019). Abuse and older lesbian, gay bisexual, and trans (LGBT) people: A commentary and research agenda.
Journal of Elder Abuse & Neglect, 31(2), 97–114.
Williams, J. L., Davis, M., & Acierno, R. (2017). Global prevalence of elder abuse in the community. In X. Dong (Ed.), Elder abuse:
Research, practice and policy (pp. 45–65). Cham: Springer International Publishing.
Wilson, J., Tilse, C., Setterlund, D., & Rosenman, L. (2009). Older people and their assets: A range of roles and issues for social
workers. Australian Social Work, 62(2), 155–167.
Wong, J. S., Breslau, H., McSorley, V. E., Wroblewski, K. E., Howe, M. J. K., Waite, L. J. (2020). The Social Relationship Context of
Elder Mistreatment. The Gerontologist, 60(6), 1029–1039. doi.org/10.1093/geront/gnz154
World Health Organization. (2021). Global report on ageism. Geneva: World Health Organization. Retrieved from
cdn.who.int/media/docs/default-source/2021-dha-docs/9789240016866-eng.pdf?sfvrsn=7375d0b8_7&download=true
World Health Organization and International Network for the Prevention of Elder Abuse. (2002). Missing voices: views of older
persons on elder abuse. Geneva: World Health Organization and International Network for the Prevention of Elder Abuse.
Yon, Y., Mikton, C., Gassoumis, Z., & Wilber, K. (2017) Elder abuse prevalence in community settings: A systematic review and
meta-analysis. Lancet Global Health, 1, 356–365.
Yon, Y., Mikton, C., Gassoumis, Z., & Wilber, K. (2019). The prevalence of self-reported elder abuse among older women in
community settings: A systematic review and meta-analysis. Trauma, Violence & Abuse, 20(2), 245–259.
Zheng, S., Li, M., Kong, D., & Dong, X. (2019). Sources and variations in social support and risk for elder mistreatment in a
US Chinese population. Journal of the American Geriatrics Society, 67(S3), S499–S505.
Appendix A: Additional tables 179
Chapter 4
Table A4.1: Survey of Older People: Characteristics of the SOP CALD sample
Education
Degree or higher 27.8 16.3 11.8
Certificate/Diploma 14.3 17.9 23.7 c
Trade/Other 7.4 9.4
Year 12 12.0 14.1 13.0
Below Year 12 35.7 39.7 40.8
Not stated 2.5 10.7
Home ownership
Own outright 75.7 76.1 65.3
Own, paying off mortgage 7.7 7.9 14.9
Rent from private landlord 5.6 5.4 8.3
Rent from public housing authority 4.3 4.2 5.6
Other (boarding, living at home, etc.) 5.3 4.4 2.4
SEIFA Quintile (IRSD 2016)
1 – Lowest socio‑economic status 14.8 15.6 20.6
2 15.6 17.3 13.8
3 17.3 16.8 18.2
4 20.1 19.7 22.0
5 – Highest socio‑economic status 32.2 30.6 25.6
State
NSW 34.9 38.6 39.6
Vic. 37.0 37.2 34.6
QLD 8.9 7.2 9.0
SA 7.9 6.1 6.9
WA 8.9 8.3 7.4
Tas. 0.7 0.7 0.6
NT 0.7 0.4 0.6
ACT 1.2 1.5 1.5
Region
Major cities 86.2 88.3 91.2
Inner regional 8.9 7.7 5.2
Outer regional/Remote/Very remote 4.9 3.9 3.6
Notes: 2016 Census data includes population over 65 residing in a private dwelling and reported speaking a language other
than English at home (excluding not stated). Refused/Don’t know responses included in calculation of proportions for
unweighted/weighted and 2016 Census data but not shown in the table above. This explains why the total proportions
for some characteristics do not sum to 100.0%. Due to high proportion of Refused/Don’t know in 2016 Census for
education, all information for this characteristic is reported (not stated/inadequately described). a Proportion in 2016
Census for categories: With grandchildren/Other relatives with/without unrelated persons / Unrelated persons / Other
combined. b Proportion in 2016 Census for categories: not employed and retired combined. c Proportion in 2016 Census
for categories: Certificate/Diploma and Trade/other combined. Other includes no educational attainment (Census 2016).
Appendix A: Additional tables 181
Notes: Weighted data for statistics and unweighted sample sizes. A small number of participants whose country of birth were
not stated are excluded from the three groups.
Chapter 6
182
Table A6.1: Survey of Older People: Proportion of participants who reported experience of elder abuse by socio‑demographic characteristics, men, women
and overall
70–74 years 12.1 9.6–14.6 763 17.3 14.7–20.0 1,161 14.9 13.1–16.8 1,927
75–79 years 10.6 7.7–13.4 589 12.4 10.1–14.7 934 11.5 9.6–13.3 1,525
80–84 years 11.1 7.2–15.0 404 11.1 8.4–13.7 699 11.1 8.7–13.4 1,105
85+ years 11.8 6.5–17.1 266 7.3 5.0–9.7 478 9.4 6.6–12.3 746
Indigenous status
Non-Indigenous 13.5 11.9–15.1 2,705 15.8 14.5–17.2 4,179 14.7 13.7–15.8 6,892
Indigenous .. .. .. 23.1 8.5–37.8 53 23.3 12.4–34.3 85
Country of birth
Australia 13.4 11.6–15.2 1,909 16 14.5–17.5 3,076 14.8 13.6–16.0 4,991
English speaking countries 15.9 11.8–20.0 480 14.3 11.2–17.3 689 15.1 12.5–17.6 1,171
Non-English speaking countries 12.4 8.3–16.5 358 16.8 12.8–20.9 470 14.7 11.9–17.6 832
Education ** *
Degree or higher 15.4 12.2–18.5 816 21.5 17.9–25.0 907 18.2 15.8–20.6 1,725
Certificate/Diploma/Trade 14.3 11.5–17.2 856 17.9 15.0–20.9 998 16 13.9–18.0 1,857
Year 12 13.6 9.1–18.1 289 13.7 10.0–17.5 474 13.7 10.8–16.6 764
Below Year 12 12.6 9.8–15.3 716 13.9 12.0–15.9 1,721 13.4 11.8–15.0 2,441
Living arrangements
Living with a partner 12.3 10.5–14.1 1,918 15.1 13.2–17.0 1,858 13.6 12.3–14.9 3,781
Living alone 16.0 12.6–19.3 691 16.7 14.6–18.8 1,965 16.4 14.7–18.2 2,660
Other 24.2 15.1–33.4 138 18.4 14.1–22.7 418 20.2 16.0–24.3 559
Men Women All
% experience 95% Number of % experience 95% Number of % experience 95% Number of
of abuse confidence participants of abuse confidence participants of abuse confidence participants
(any form) interval (base) (any form) interval (base) (any form) interval (base)
Marital status *** *** ***
Married (this includes registered 11.9 10.1–13.6 1,870 14.6 12.7–16.5 1,836 13.1 11.8–14.4 3,712
marriage only)
Separated/divorced 26 19.9–32.1 331 26.7 22.6–30.7 692 26.4 22.9–29.9 1,023
Widowed 12.7 8.7–16.7 330 13.1 10.9–15.3 1,456 13 11.1–15.0 1,790
Never married 13.5 8.4–18.5 199 16 10.2–21.9 230 14.8 10.9–18.6 430
Whether having any children (of any age) in household (bio/step/foster)
No 13.4 11.8–15.1 2,447 15.5 14.1–17.0 3,786 14.5 13.4–15.6 6,243
Yes 14.7 9.7–19.8 300 19.3 14.6–23.9 434 17.2 13.7–20.6 736
Any biological/adopted children inside/outside the household?
No 14.0 10.0–18.0 435 14.6 10.9–18.2 566 14.2 11.5–16.9 1004
Yes 13.5 11.8–15.3 2,306 16.1 14.7–17.6 3,665 14.9 13.8–16.1 5,980
Any children (including step-children and adopted children)
Biological or adopted children 13.0 11.2–14.7 2,128 15.5 14.0–17.0 3,453 14.3 13.2–15.5 5,590
Step-children 14.1 3.6–24.5 61 19.5 7.4–31.6 62 16.2 8.2–24.2 123
Both biological/adopted AND 20.5 12.6–28.4 177 25.2 17.9–32.5 211 22.9 17.5–28.2 388
step-children
None 14.0 9.7–18.3 374 13.9 10.1–17.7 504 13.9 11.1–16.8 881
Household annual income quintile (before tax)
Lowest household income 18.7 13.5–24.0 363 20.2 16.7–23.6 767 19.6 16.6–22.5 1,134
quintile (incl. neg./zero income
9.3 4.9–13.7 221 16.8 11.6–22.0 337 12.9 9.5–16.4 558
14.4 10.5–18.3 494 17.4 13.3–21.6 540 15.8 12.9–18.6 1,035
11.9 8.0–15.8 389 14.4 9.8–19.1 346 13 10.0–16.0 735
Highest household income 13.2 9.5–17.0 475 19.2 13.2–25.2 248 15.1 11.9–18.3 724
quintile
Housing
Own outright 12.2 10.5–13.9 2,160 14.3 12.8–15.7 3,269 13.3 12.2–14.4 5,436
Appendix A: Additional tables
Own, paying off mortgage 16.4 10.3–22.6 195 32.9 25.4–40.3 238 24 19.1–28.9 434
Rent from private landlord 23.6 15.1–32.1 160 17.9 11.4–24.4 248 20.6 15.3–25.8 409
183
Rent from public housing 12.1 4.7–19.4 73 27.4 18.9–35.9 150 20.5 14.5–26.4 225
authority
Other (boarding, living at home etc.) 21.6 12.9–30.2 134 14.5 9.3–19.8 278 17.4 12.7–22.1 412
Men Women All
184
Inner regional 12.7 9.6–15.7 640 17.7 14.8–20.5 1,035 15.3 13.2–17.4 1,678
Outer regional/Remote/Very 18.3 12.7–24.0 297 14.0 9.9–18.2 432 16.1 12.6–19.6 731
remote
Notes: Weighted data for the statistics and unweighted sample sizes. The asterisks indicate that the differences in the prevalence across categories in the variable are statistically significant
based on a design-based F test (*p < .05; **p < .01; ***p < .001)
Table A6.2: Survey of Older People: Proportion of participants who report experience of financial abuse by socio‑demographic characteristics, men, women and overall
income
3.3 0.7–5.9 221 1.8 0.4–3.3 337 2.6 1.1–4.1 558
1.3 0.2–2.5 494 2.8 1.1–4.6 540 2 1.0–3.0 1,035
2.5 0.2–4.7 389 1.5 0.1–2.9 346 2.1 0.6–3.5 735
Highest household income 1.1 0.3–1.9 475 3.7 1.1–6.4 248 1.9 0.9–2.9 724
quintile
Housing *** * ***
Own outright 1.8 1.1–2.4 2,160 1.7 1.2–2.2 3,269 1.7 1.3–2.1 5,436
Own, paying off mortgage 1.0 -0.2–2.2 195 4.6 1.5–7.7 238 2.7 1.1–4.3 434
Rent from private landlord 8.5 2.4–14.5 160 3.3 -0.4–6.9 248 5.7 2.3–9.2 409
Rent from public housing 2.0 -0.6–4.6 73 4.7 0.6–8.8 150 3.4 0.9–5.9 225
authority
Other (boarding, living at 2.3 -0.6–5.2 134 1.3 -0.4–3.0 278 1.7 0.1–3.3 412
home, etc.)
SEIFA Quintile (IRSD 2016)
1 – Lowest socio‑economic 4.1 1.7–6.5 397 2.0 0.9–3.0 662 3 1.7–4.2 1,063
status
2 1.6 0.4–2.8 545 2.6 1.5–3.8 823 2.1 1.3–2.9 1,369
3 1.8 0.5–3.1 531 1.4 0.5–2.2 810 1.6 0.8–2.3 1,345
4 2.6 0.8–4.3 515 2.1 0.8–3.4 819 2.3 1.3–3.4 1,336
5 – Highest 1.2 0.5–1.9 759 2.1 1.0–3.2 1,124 1.7 1.0–2.4 1,884
socio‑economic status
Men Women All
% experience 95% Number of % experience 95% Number of % experience 95% Number of
of abuse confidence participants of abuse confidence participants of abuse confidence participants
(financial) interval (base) (financial) interval (base) (financial) interval (base)
Region
Major city 2.0 1.2–2.7 1,810 1.9 1.3–2.5 2,774 1.9 1.5–2.4 4,591
Inner regional 2.5 0.8–4.3 640 2.5 1.2–3.8 1,035 2.5 1.4–3.6 1,678
Outer regional/Remote/ 2.1 0.5–3.7 297 2.0 0.7–3.3 432 2.1 1.0–3.1 731
Very remote
Notes: Based on weighted data. Participants who could not be classified regarding experience of abuse due to item non-response were included in the total when deriving the prevalence
figures. The asterisks indicate that the differences in the prevalence of this subtype across that specific characteristic categories (separately for men, women, and all) were statistically
significant based on a design-based F test (*p < .05; **p < .01; ***p < .001).
Table A6.3: Survey of Older People: Proportion of participants who report experience of physical abuse by socio‑demographic characteristics, males, females and
overall
Australia 2.3 1.4–3.2 1,909 1.7 1.1–2.3 3,076 2.0 1.5–2.5 4,991
English speaking countries 1.0 0.3–1.8 480 1.8 0.8–2.9 689 1.4 0.8–2.1 1,171
187
Non-English speaking 1.6 0.2–3.1 358 1.2 0.0–2.3 470 1.4 0.5–2.3 832
countries
Men Women All
188
Marital status * * **
Married (this includes 1.6 0.9–2.3 1,870 1.2 0.6–1.8 1,836 1.4 0.9–1.9 3,712
registered marriage only)
Separated/divorced 4.3 1.3–7.3 331 3.3 1.6–5.0 692 3.7 2.1–5.3 1,023
Widowed 1.3 0.0–2.5 330 1.5 0.6–2.3 1,456 1.4 0.7–2.1 1,790
Never married 3.4 0.9–6.0 199 2.8 -0.4–6.1 230 3.3 1.2–5.3 430
Whether having any children (of any age) in household (bio/step/foster)
No 1.9 1.3–2.6 2,447 1.4 0.9–1.8 3,786 1.6 1.3–2.0 6,243
Yes 2.1 -0.1–4.3 300 3.6 1.1–6.2 434 2.9 1.2–4.6 736
Any children (including step-children and adopted children) *
Biological or adopted 1.5 0.9–2.2 2,128 1.6 1.1–2.1 3,453 1.6 1.2–2.0 5,590
children
Step-children 3.9 -3.6–11.4 61 3.7 -1.9–9.3 62 3.8 -1.3–8.9 123
Both biological / adopted 4.6 0.3–9.0 177 3.4 0.2–6.6 211 4.0 1.3–6.7 388
AND step-children
None 2.8 1.1–4.4 374 0.8 0.1–1.5 504 1.8 0.9–2.7 881
Household annual income quintile (before tax)
Lowest household income 3.4 0.7–6.0 363 2.6 1.4–3.9 767 3.0 1.7–4.3 1,134
quintile (incl. neg./zero
income
1.7 -0.6–3.9 221 1.8 0.0–3.5 337 1.7 0.3–3.1 558
Men Women All
% experience 95% Number of % experience 95% Number of % experience 95% Number of
of abuse confidence participants of abuse confidence participants of abuse confidence participants
(physical) interval (base) (physical) interval (base) (physical) interval (base)
1.6 0.3–2.9 494 1.0 -0.1–2.0 540 1.3 0.5–2.2 1,035
1.9 0.0–3.8 389 2.5 0.0–5.1 346 2.2 0.6–3.7 735
Highest household income 0.9 0.2–1.7 475 2.0 -0.1–4.2 248 1.3 0.4–2.1 724
quintile
Housing ** **
Own outright 1.8 1.0–2.5 2,160 1.3 0.8–1.7 3,269 1.5 1.1–1.9 5,436
Own, paying off mortgage 3.7 0.7–6.6 195 4.7 1.2–8.3 238 4.2 1.9–6.4 434
Rent from private landlord 2.4 -1.5–6.2 160 3.7 0.3–7.1 248 3.1 0.5–5.6 409
Rent from public housing 3.0 0.3–5.7 73 3.2 0.7–5.8 150 3.3 1.4–5.2 225
authority
Other (boarding, living at 1.5 -0.1–3.2 134 0.3 -0.3–0.9 278 0.8 0.1–1.5 412
home, etc.)
SEIFA Quintile (IRSD 2016) ** * **
1 – Lowest socio‑economic 4.8 1.8–7.8 397 1.7 0.4–3.1 662 3.2 1.6–4.8 1,063
status
2 1.3 0.4–2.1 545 2.9 1.5–4.3 823 2.1 1.3–2.9 1,369
3 2.2 0.9–3.6 531 1.2 0.5–2.0 810 1.7 1.0–2.5 1,345
4 1.3 0.4–2.3 515 0.5 0.0–1.1 819 0.9 0.4–1.4 1,336
5 – Highest 0.9 0.2–1.7 759 1.7 0.6–2.8 1,124 1.3 0.7–2.0 1,884
socio‑economic status
Region
Major city 1.4 0.8–2.0 1,810 1.6 1.0–2.2 2,774 1.5 1.1–1.9 4,591
Inner regional 3.1 1.3–4.8 640 1.7 0.9–2.6 1,035 2.4 1.4–3.3 1,678
Outer regional/Remote/ 3.0 0.3–5.7 297 1.7 0.1–3.3 432 2.4 0.8–3.9 731
Very remote
Notes: Based on weighted data. Participants who could not be classified regarding experience of abuse due to item non-response were included in the total when deriving the prevalence
figures. The asterisks indicate that the differences in the prevalence of this subtype across the specific characteristic categories (separately for men, women, and all) were statistically
significant based on a design-based F test (*p < .05; **p < .01; ***p < .001).
Appendix A: Additional tables
189
Table A6.4: Survey of Older People: Proportion of participants who report experience of sexual abuse by socio‑demographic characteristics, males, females and overall
190
85+ years 0.8 -0.1–1.7 266 0.0 0.0–0.0 478 0.4 -0.1–0.8 746
Indigenous status
Non-Indigenous 0.7 0.3–1.1 2,705 1.2 0.8–1.6 4,179 1.0 0.7–1.2 6,892
Indigenous .. .. .. 0.7 -0.7–2.1 53 1.6 -0.9–4.1 85
Country of birth
Australia 0.8 0.3–1.2 1,909 1.2 0.8–1.7 3,076 1.0 0.7–1.3 4,991
English speaking countries 0.7 -0.2–1.6 480 1.4 0.5–2.3 689 1.1 0.4–1.7 1,171
Non-English speaking countries 0.5 -0.5–1.5 358 1.0 0.0–2.0 470 0.8 0.1–1.5 832
Education ***
Degree or higher 0.5 0.0–0.9 816 2.9 1.6–4.3 907 1.6 0.9–2.3 1,725
Certificate/Diploma/Trade 0.6 -0.1–1.3 856 1.7 0.8–2.7 998 1.1 0.6–1.7 1,857
Year 12 1.2 -0.1–2.4 289 0.7 0.1–1.3 474 1.0 0.3–1.6 764
Below Year 12 0.7 0.0–1.5 716 0.6 0.2–1.0 1,721 0.7 0.3–1.0 2,441
Living arrangements * * **
Living with a partner 0.4 0.0–0.8 1,918 0.8 0.3–1.3 1,858 0.6 0.3–0.9 3,781
Living alone 1.8 0.4–3.3 691 1.8 1.2–2.5 1,965 1.9 1.2–2.5 2,660
Other 0.8 -0.4–2.0 138 1.3 0.3–2.3 418 1.2 0.4–1.9 559
Men Women All
% experience 95% Number of % experience 95% Number of % experience 95% Number of
of abuse confidence participants of abuse confidence participants of abuse confidence participants
(sexual) interval (base) (sexual) interval (base) (sexual) interval (base)
Marital status * *** ***
Married (this includes registered 0.4 0.0–0.7 1,870 0.7 0.3–1.2 1,836 0.5 0.2–0.8 3,712
marriage only)
Separated/divorced 1.8 -0.3–3.9 331 3.0 1.6–4.4 692 2.5 1.3–3.7 1,023
Widowed 1.5 0.2–2.9 330 0.9 0.4–1.4 1,456 1.0 0.6–1.5 1,790
Never married 1.8 -0.6–4.2 199 2.8 0.7–4.8 230 2.4 0.7–4.0 430
Whether having any children (of any age) in household (bio/step/foster)
No 0.7 0.3–1.2 2,447 1.2 0.8–1.5 3,786 1.0 0.7–1.2 6,243
Yes 0.4 -0.2–1.0 300 1.5 -0.2–3.3 434 1.0 0.0–2.0 736
Any children (including step-children and adopted children)
Biological or adopted children 0.7 0.2–1.2 2,128 1.2 0.7–1.6 3,453 0.9 0.6–1.3 5,590
Step-children 0.0 0.0–0.0 61 4.5 -1.8–10.8 62 1.7 -0.8–4.2 123
Both biological / adopted AND 0.0 0.0–0.0 177 0.7 -0.1–1.6 211 0.4 -0.1–0.8 388
step-children
None 1.3 0.1–2.5 374 1.5 0.6–2.4 504 1.5 0.7–2.2 881
Household annual income quintile (before tax)
Lowest household income 1.6 -0.2–3.5 363 1.0 0.2–1.7 767 1.3 0.4–2.2 1,134
quintile (incl. neg./zero income
0.0 0.0–0.0 221 1.5 0.2–2.9 337 0.7 0.1–1.4 558
0.8 -0.1–1.7 494 1.8 0.6–3.0 540 1.2 0.5–2.0 1,035
0.5 -0.3–1.2 389 3.3 0.7–6.0 346 1.7 0.4–2.9 735
Highest household income quintile 0.4 -0.1–1.0 475 0.7 -0.2–1.6 248 0.5 0.1–1.0 724
Housing
Own outright 0.7 0.3–1.2 2,160 1.1 0.7–1.5 3,269 0.9 0.6–1.3 5,436
Own, paying off mortgage 0.1 -0.1–0.4 195 2.3 0.4–4.2 238 1.1 0.3–2.0 434
Rent from private landlord 0.6 -0.3–1.6 160 0.3 -0.1–0.8 248 0.5 0.0–1.0 409
Rent from public housing 0.8 -0.8–2.4 73 2.4 0.5–4.4 150 1.9 0.5–3.2 225
authority
Appendix A: Additional tables
Other (boarding, living at home, 1.2 -0.5–2.9 134 1.7 -0.4–3.8 278 1.5 0.1–2.9 412
etc.)
191
Men Women All
192
Inner regional 0.9 -0.2–1.9 640 0.7 0.2–1.1 1,035 0.8 0.3–1.3 1,678
Outer regional/Remote/ 0.9 -0.4–2.1 297 0.7 -0.1–1.6 432 0.8 0.0–1.6 731
Very remote
Notes: Based on weighted data. Participants who could not be classified regarding experience of abuse due to item non-response were included in the total when deriving the prevalence
figures. The asterisks indicate that the differences in the prevalence of this subtype across that specific characteristic categories (separately for men, women, and all) were statistically
significant based on a design-based F test (*p < .05; **p < .01; ***p < .001).
Table A6.5: Survey of Older People: Proportion of participants who report experience of psychological abuse by socio‑demographic characteristics, males, females
and overall
Yes 12.7 7.8–17.6 300 15.3 11.1–19.4 434 14.1 10.9–17.3 736
Any children (including step-children and adopted children) * **
193
Men Women All
194
Notes: Based on weighted data. Participants who could not be classified regarding experience of abuse due to item non-response were included in the total when deriving the prevalence
figures. The asterisks indicate that the differences in the prevalence of this subtype across that specific characteristic categories (separately for men, women, and all) were statistically
significant based on design-based F test (*p < .05; **p < .01; ***p .001).
Table A6.6: Survey of Older People: Proportion of participants who report experience of neglect by socio‑demographic characteristics, males, females and overall
English speaking countries 2.0 0.9–3.2 480 3.7 1.9–5.5 689 2.9 1.8–3.9 1,171
Non-English speaking countries 2.4 0.4–4.4 358 3.5 1.5–5.6 470 3.0 1.6–4.4 832
195
Men Women All
196
Marital status * **
Married (this includes registered 2.1 1.3–2.9 1,870 3.2 2.2–4.2 1,836 2.6 2.0–3.2 3,712
marriage only)
Separated/divorced 4.2 1.0–7.4 331 5.5 3.6–7.4 692 5.0 3.2–6.7 1,023
Widowed 1.1 0.1–2.2 330 2.8 1.7–3.9 1,456 2.4 1.5–3.3 1,790
Never married 0.4 -0.2–0.9 199 4.3 1.0–7.6 230 2.1 0.6–3.7 430
Whether having any children (of any age) in household (bio/step/foster)
No 2.1 1.4–2.9 2,447 3.2 2.5–3.9 3,786 2.7 2.2–3.2 6,243
Yes 2.6 0.8–4.3 300 5.5 2.6–8.4 434 4.2 2.4–5.9 736
Any children (including step-children and adopted children)
Biological or adopted children 2.0 1.2–2.8 2,128 3.6 2.8–4.4 3,453 2.9 2.3–3.4 5,590
Step-children 2.8 -1.3–6.9 61 0.9 -0.9–2.7 62 2.1 -0.5–4.7 123
Both biological / adopted AND 2.9 0.3–5.6 177 5.0 1.3–8.7 211 4.0 1.7–6.3 388
step-children
None 2.9 0.5–5.4 374 1.9 0.7–3.1 504 2.4 1.0–3.8 881
Household annual income quintile (before tax) *
Lowest household income 3.8 0.9–6.6 363 4.6 2.7–6.4 767 4.3 2.7–5.8 1,134
quintile (incl. neg./zero income
1.5 -0.2–3.2 221 2.6 1.0–4.1 337 2.0 0.9–3.2 558
1.4 0.0–2.9 494 3.9 1.5–6.3 540 2.6 1.2–3.9 1,035
2.0 0.1–3.9 389 2.4 0.7–4.2 346 2.2 0.8–3.5 735
Men Women All
% experience 95% Number of % experience 95% Number of % experience 95% Number of
of abuse confidence participants of abuse confidence participants of abuse confidence participants
(neglect) interval (base) (neglect) interval (base) (neglect) interval (base)
Highest household income 1.5 0.1–2.9 475 1.3 -0.2–2.7 248 1.4 0.3–2.5 724
quintile
Housing ** *** ***
Own outright 2.1 1.3–2.8 2,160 2.7 2.0–3.4 3,269 2.4 1.9–2.9 5,436
Own, paying off mortgage 1.4 -0.3–3.0 195 9.9 5.0–14.7 238 5.3 2.8–7.8 434
Rent from private landlord 1.9 0.0–3.7 160 3.3 0.7–5.9 248 2.6 1.0–4.3 409
Rent from public housing 1.1 -0.6–2.8 73 10.4 4.3–16.4 150 6.2 2.8–9.6 225
authority
Other (boarding, living at home, 7.8 1.2–14.4 134 3.5 0.1–6.8 278 5.2 1.9–8.6 412
etc.)
SEIFA Quintile (IRSD 2016)
1 - Lowest socio‑economic status 1.8 0.1–3.6 397 4.2 2.3–6.1 662 3.1 1.8–4.4 1,063
2 1.6 0.6–2.5 545 2.5 1.5–3.6 823 2.1 1.3–2.8 1,369
3 1.5 0.2–2.9 531 4.7 2.4–6.9 810 3.2 1.8–4.5 1,345
4 3.9 1.4–6.3 515 3.0 1.7–4.3 819 3.4 2.1–4.7 1,336
5 – Highest socio‑economic 2.2 1.0–3.4 759 3.1 1.9–4.3 1,124 2.7 1.9–3.6 1,884
status
Region * *
Major city 2.2 1.3–3.1 1,810 3.6 2.7–4.5 2,774 3.0 2.3–3.6 4,591
Inner regional 2.8 1.1–4.5 640 4.0 2.6–5.3 1,035 3.4 2.3–4.5 1,678
Outer regional/Remote/Very 0.8 -0.3–1.9 297 1.2 0.2–2.2 432 1.0 0.3–1.7 731
remote
Notes: Based on weighted data. Participants who could not be classified regarding experience of abuse due to item non-response were included in the total when deriving the prevalence
figures. The asterisks indicate that the differences in the prevalence of this subtype across thespecific characteristic categories (separately for men, women, and all) were statistically
significant based on a design-based F test (*p < .05; **p < .01; ***p < .001).
Appendix A: Additional tables
197
198 National Elder Abuse Prevalence Study: Final Report
Regression analysis
Regression analysis is applied to the experience of any form of abuse, and each of the five subtypes to examine
whether any association between a specific elder abuse variable and characteristic continues to hold. Given that
these elder abuse variables are binary (i.e. experiencing abuse, or not expiring abuse), logistic regression is used.
It is worth noting that sexual abuse involved a small number of occurrences. Although numbers of participants
who experience subtypes are small, the logistic regression can be applied according to the research by
Vittinghoff and McCulloch (2007).20 Table A6.7 presents the odds of logistic regression analysis for overall elder
abuse and the five subtypes.
The results of logistic regression analysis are presented in the form of odds ratios in Table A6.8. For categorical
variables, an odds ratio for a specific category of a variable is relative to the variable’s reference group (shown in
brackets). Levels of statistical significance are marked by asterisks. When an odds ratio of one does not affect the
outcome variable (here the experience of elder abuse), an odds ratio above one means an association with higher
odds of experiencing elder abuse, and an odds ratio below one indicates lower odds of experiencing elder abuse.
For example, for abuse of any form, the odds ratio of 1.179 for females indicates that the odds of experiencing
elder abuse is 17.9% higher for women than for men, controlling other characteristics, and the asterisk next to it
shows that the difference is statistically significant (p < .05).
Table A6.7: Odds ratios of logistic regression of elder abuse, overall (any form) and subtypes
20 Vittinghoff and McCulloch (2007) indicate that logistic regression can be applied where the number of events per predictor variable is
fewer than 10, in particular with adequate control of confounding factors.
Appendix A: Additional tables 199
Notes: For each categorical variable, the reference category is in the brackets. The asterisks indicate that the coefficient is
statistically significantly different from the reference group (*p < .05, **p < .01, ***p < .001).
Table A6.8 shows the regression analysis that further considers variables social contact, sense of social support
(or isolation) and mental health (as measure by Kessler 6 psychological distress). In this analysis, sense of social
support and psychological distress variables were considered as continuous variables, the interpretation of
odds ratios is the same as above. For example, for abuse of any form, the odds ratio for sense of social support
is 0.98, indicating with each unit increase in the score, the odds of experiencing elder abuse were lower by 2%.
As outlined in the report, the analysis is correlational and doesn’t suggest any causal direction. For example,
psychological distress could make older people more vulnerable to elder abuse, and elder abuse could lead to
psychological distress or a low sense of social support (or higher level of social isolation).
Table A6.8: Odds of logistic regression of elder abuse, overall (any form) and subtypes, with sense of social
support and psychological distress scales as part of correlates
Note: For each categorical variable, the reference category is in the brackets. The asterisks indicate that the coefficient is
statistically significantly different from the reference group (*p < .05, **p < .01, ***p < .001).
Appendix A: Additional tables 201
Chapter 7
Table A7.1: Survey of Older People: Relationship of perpetrators to the persons who experienced at least one
type of abuse
Notes: Weighted data. Multiple perpetrators could be reported for each type of abuse. Values may not equal 100.0% due to
rounding. a Includes where gender not reported (includes: other, don’t know and refused). b Includes don’t know/refused
responses and missing responses to any form of abuse.
Table A7.2: Survey of Older People: Number of perpetrators reported, by all types and subtype of abuse
Number of
perpetrators Financial (%) Physical (%) Sexual (%) Psychological (%) Neglect (%)
Multiple perpetrators 3.4 6.2 2.9 8.2 21.1
reported (2+)
2 2.7 6.2 2.9 7.6 16.9
3 0.6 – – 0.6 2.9
4 – – – – 0.6
5 – – – – 0.7
Base 142 113 82 764 208
Notes: Multiple perpetrators could be reported for each type of abuse. The proportions for multiple perpetrators are derived
from individual perpetrators before categories were combined. The proportions exclude don’t know/refused responses
and missing responses.
202 National Elder Abuse Prevalence Study: Final Report
Table A7.3: Survey of Older People: Relationship of perpetrators and main perpetrators to the participant
(people who experienced financial, physical or sexual abuse)
Notes: Weighted data. Multiple perpetrators could be reported for each type of abuse. Values may not equal 100.0% due to
rounding. a Includes where gender not reported (includes: other, don’t know and refused). b Includes don’t know/refused
responses and missing responses.
Table A7.4: Survey of Older People: Relationship of all perpetrators and main perpetrator to participant
(person who experienced neglect)
Notes: Weighted data. Multiple perpetrators could be reported for each type of abuse. Values may not equal 100.0% due to
rounding. a Includes where gender not reported (includes: other, don’t know and refused). b Includes don’t know/refused
responses and missing responses.
Notes: Carers include all persons who care for the participant (including the persons who are responsible for neglect). Multiple
carers could be selected. a It is noted that gender information is not available for carers and the data are not separated
for sons and daughters. b Includes don’t know/refused responses and missing responses.
Table A7.6: Survey of Older People: Relationship of all perpetrators and main perpetrators to participant
(person who experienced psychological abuse)
Medium-high
score bands:
All (as % of Low score band: Main perpetrator
Perpetrator participants) Main perpetrator Main perpetrator (Medium-high)
Partner/spouse 8.4 7.9 7.4 8.4
Son/daughter (biological/ 19.2 17.3 13.2 21.3
adopted) a
Son – 8.8 6.4 11.1
Daughter – 6.8 5.5 8.1
Step-son/daughter 2.6 1.8 1.1 2.4
Grandson/daughter 4.3 4.0 4.0 4.0
Brother/sister 3.6 3.3 3.2 3.4
204 National Elder Abuse Prevalence Study: Final Report
Medium-high
score bands:
All (as % of Low score band: Main perpetrator
Perpetrator participants) Main perpetrator Main perpetrator (Medium-high)
Brother/sister in-law 1.4 1.0 1.1 1.0
Son/daughter (in-law) 10.9 9.7 7.9 11.5
Other family members 7.2 6.4 5.8 7.0
Ex-partner/spouse 1.8 1.4 0.6 2.2
Friend 10.4 12.0 15.0 9.1
Neighbour 8.1 7.7 8.4 7.0
Professional carer 0.7 0.7 0.6 0.7
Service providers 3.9 2.7 4.0 1.5
Acquaintances 13.2 11.0 11.6 10.4
(incl. colleagues, clients)
Other 2.9 2.4 3.6 1.2
Unspecified b 9.5 10.7 12.6 9.0
Total – 100.0 100.0 100.0
Base 843 843 425 418
Notes: Weighted data. Multiple perpetrators could be reported for each type of abuse. Values may not equal 100.0% due to
rounding. a Includes where gender not reported (includes: other, don’t know and refused). b Includes don’t know/refused
responses and missing responses.
Table A7.7: Survey of Older People: Relationship of perpetrators to older people who experienced at least one
type of abuse and are from a CALD background
Notes: Weighted data. Multiple perpetrators could be reported for each type of abuse. Values may not equal 100.0% due to
rounding. a Includes where gender not reported (includes: other, don’t know and refused). b Includes don’t know/refused
responses to any form of abuse.
Appendix A: Additional tables 205
Table A7.8: Survey of Older People: Relationship of perpetrators to older people who experienced abuse
relating to language and culture
Notes: Weighted data. Multiple perpetrators could be reported with only one reported for main perpetrator. Relationship
categories with no proportions are not shown. Values may not equal 100.0% due to rounding. a Includes don’t know/
refused responses and missing responses.
Table A7.9: Survey of the General Community: Where each concern reported, who mistreated the older person
Failed to provide
adequate care with
Taken advantage Physically hurt or Abused them routine activities
of them financially mistreated them emotionally or personal care
(%) (%) (%) when needed (%)
Perpetrator Men Women Men Women Men Women Men Women
Son 30.5 30.9 19.6 12.1 26.1 24.6 26.7 17.5
Daughter 19.1 26.4 12.3 9.6 21.0 20.9 17.4 14.7
Partner/Spouse 7.3 7.0 9.8 17.0 7.2 16.6 5.4 7.6
Other family member 31.4 25.1 13.6 16.6 26.7 22.4 18.6 21.2
Professional carer 3.2 5.2 22.0 21.0 8.9 9.0 24.8 27.3
Other professional 4.6 2.1 10.6 10.9 2.3 6.1 4.8 9.9
Friends 6.9 3.8 5.2 2.6 2.6 1.0 1.1
Neighbours 2.2 3.9 1.2 1.5
Other person 3.7 4.8 6.0 2.3 4.5 4.9 4.0 3.0
Number of persons who 146 210 55 72 152 233 116 141
mistreated older person
Notes: Son and daughter (includes step and in-law). Other person includes colleagues and ex-partner/spouse. Percentages
do not sum to 100.0% as multiple options could be selected. Further analysis of concerns about sexual abuse and who
mistreated them not reported due to small sample sizes (n = 11).
206 National Elder Abuse Prevalence Study: Final Report
Chapter 10
Survey of Older People
Table A10.1: Survey of Older People: Whether participant has had discussion with someone in the family about
current will, by gender
Notes: Weighted statistics and unweighted sample bases. a Total includes a small number of not-stated cases (0.5%).
Table A10.2: Survey of Older People: Advance planning, by whether person speaks a language other than
English at home
Speak a language
Speak only English other than English at
Type of advance planning, by gender at home (%) home (%)
Males
Having a will 89.4 73.1
Having ever granted POA to someone 51.0 40.1
Having a family agreement 2.8 ^3.9
Number of participants (base) 2,475 272
Females
Having a will 90.8 79.6
Having ever granted POA to someone 56.3 42.5
Having a family agreement 2.6 ^4.4
Number of participants (base) 3,909 332
Table A10.3: Survey of Older People: Older persons whose PIN was accessible by someone else, who had
access to PIN, by gender and age
Men Women
65–69 70–74 75–79 80–84 85+ 65–69 70–74 75–79 80–84 85+
years years years years years years years years years years
(%) (%) (%) (%) (%) (%) (%) (%) (%) (%)
Partner/spouse 87.0 87.3 87.6 74.4 48.1 72.9 69.3 54.5 29.8 8.6
Son/daughter 6.5 10.3 11.3 24.1 46.6 26.7 27.4 42.6 62.1 80.1
(biological/adopted)
Step-son/daughter 0.4 0.5 0.6 2.8 1.8 0.3 0.0 0.5 1.2 3.1
Grandson/daughter 0.1 0.3 0.6 0.0 1.1 0.0 1.5 1.6 3.5 3.3
Brother/sister 2.4 0.2 1.3 0.0 0.9 1.0 0.3 0.9 1.4 0.8
Brother/sister in-law 0.0 0.0 0.0 0.0 0.0 0.0 0.4 0.3 0.0 0.6
Son/daughter (in-law) 0.1 0.6 0.1 0.0 0.0 1.0 2.8 0.7 2.4 3.0
Other family members 1.5 0.1 1.0 1.0 0.7 0.0 0.0 0.1 3.0 2.7
Ex-partner/spouse 1.8 0.2 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0
Friend 0.7 1.1 0.0 0.0 0.0 1.1 1.9 0.6 1.1 0.6
Appendix A: Additional tables 207
Men Women
65–69 70–74 75–79 80–84 85+ 65–69 70–74 75–79 80–84 85+
years years years years years years years years years years
(%) (%) (%) (%) (%) (%) (%) (%) (%) (%)
Neighbour 0.0 0.0 0.0 0.0 0.0 0.0 0.7 0.0 0.0 0.0
Professional carer 0.0 0.0 0.0 0.0 0.4 0.0 0.0 0.3 0.0 0.0
Financial planner 1.5 0.6 0.9 0.0 0.3 0.0 0.6 0.0 0.0 0.0
Other service providers 0.0 0.4 0.3 0.0 1.9 0.0 0.0 0.1 0.4 0.0
Acquaintances 0.9 1.2 0.0 0.0 0.0 0.5 0.0 0.0 0.0 1.1
Other 0.5 0.5 1.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Unclear 0.0 1.0 0.0 1.2 0.0 0.8 0.7 0.3 0.9 1.2
Number of participants 233 254 181 102 74 204 254 186 167 167
Table A10.4: Survey of Older People: Proportion of people who have a will, by socio‑demographic characteristics
Notes: Weighted data and unweighted sample size. The asterisks indicate that the differences in the proportion of having a
will across the categories of specific characteristic variables are statistically significant based on a design-based F test
(*p < .05; **p <.01; ***p < .001).
Appendix A: Additional tables 209
Table A10.5: Survey of Older People: Proportion of people who have a valid/active power of attorney, by
socio‑demographic characteristics
Notes: Weighted data and unweighted sample size. The asterisks indicate that the differences in the proportion of having POA
across the categories of specific characteristic variables are statistically significant based on a design-based F test
(*p < .05; **p < .01; ***p < .001)
Table A10.6: Survey of Older People: Older persons who had an active power of attorney, who was granted the
power of attorney, by gender and age
Men Women
65–69 70–74 75–79 80–84 85+ 65–69 70–74 75–79 80–84 85+
years years years years years years years years years years
(%) (%) (%) (%) (%) (%) (%) (%) (%) (%)
Partner/spouse 33.8 31.3 23.5 25.4 9.1 28.4 20.3 11.4 7.6 0.9
Son/daughter 52.1 63.0 69.6 67.4 75.9 62.8 73.0 78.9 77.7 83.3
(biological/
adopted)
Step-son/ 4.0 2.5 2.0 5.0 5.4 2.1 1.3 3.0 2.0 0.9
daughter
Grandson/ 0.0 0.1 0.4 0.9 3.9 0.0 0.1 0.7 0.5 1.6
daughter
Brother/sister 6.9 5.3 4.1 0.6 1.4 8.9 4.7 3.2 1.6 1.1
Brother/sister 1.6 1.2 1.0 0.0 0.0 1.0 0.7 0.5 0.5 0.5
in-law
Appendix A: Additional tables 211
Men Women
65–69 70–74 75–79 80–84 85+ 65–69 70–74 75–79 80–84 85+
years years years years years years years years years years
(%) (%) (%) (%) (%) (%) (%) (%) (%) (%)
Son/daughter 1.8 1.1 2.2 2.3 1.3 1.5 1.8 1.0 2.1 2.2
(in-law)
Other family 7.6 2.7 3.5 4.4 5.8 4.3 3.0 3.1 3.8 6.9
members
Ex-partner/ 0.8 0.1 0.2 0.0 0.0 0.5 0.2 0.0 0.0 0.0
spouse
Friend 3.8 2.1 2.4 2.2 1.6 1.3 2.3 2.4 2.5 4.6
Neighbour 0.0 0.0 0.0 0.4 0.7 0.0 0.2 0.0 0.0 0.1
Professional 0.0 0.0 0.0 0.0 0.2 0.1 0.0 0.2 0.1 0.0
carer
Other service 2.7 2.8 2.6 5.0 1.4 2.6 1.4 1.1 1.6 1.8
providers
Other 0.8 0.4 0.3 0.0 0.0 0.6 0.1 0.0 0.1 0.5
Number of 279 372 339 247 184 411 639 552 494 362
participants
(base)
Notes: Based on weighted data. Unweighted sample size. No responses for acquaintance category. Financial planner is
combined with other.
Table A10.7: Survey of Older People: Older persons who had an active power of attorney, who was granted the
power of attorney, by whether speak a language other than English at home
Speak a language
Speak only English other than English
Person granted power of attorney at home (%) at homes (%)
Partner/spouse 21.2 15.2
Son/daughter (biological/adopted) 70.4 68.5
Step-son/daughter 2.5 2.8
Grandson/daughter 0.7 0.0
Brother/sister 4.3 5.0
Brother/sister in-law 0.7 0.0
Son/daughter (in-law) 1.6 2.3
Other family members 4.4 3.8
Ex-partner/spouse 0.1 0.5
Friend 2.5 2.1
Neighbour 0.1 0.2
Professional carer 0.04 0.0
Other service providers 2.1 2.7
Other 0.3 0.6
Number of participants (base) 3,121 218
Table A10.8: Survey of Older People: Proportion of people who have a family agreement, by
socio‑demographic characteristics
Notes: Weighted data and unweighted sample size. The asterisks indicate that the differences in the proportion of having family
agreements across the categories of specific characteristic variables are statistically significant based on a design-based
F test (*p < .05; **p < .01; ***p < .001)
214 National Elder Abuse Prevalence Study: Final Report
Table A10.9: Survey of Older People: Prevalence of elder abuse, by each type of advance planning
Notes: Weighted data and unweighted sample size. The asterisks indicate that the prevalence of abuse type between the
categories and the reference (listed as first category) is statistically significant based on bivariate logistic regression
(*p < .05; **p < .01; ***p < .001).
Number of
Power of attorney % participants (base)
Whether appointed under an existing power of attorney
Yes 19.8 806
No 78.5 2,537
Not stated 1.7 57
Total 100.0 3,400
If power of attorney in place (n = 806): type of power of attorney
Financial 14.3 115
Medical 7.2 60
Both 75.9 611
Not stated 2.6 20
Appendix A: Additional tables 215
Number of
Power of attorney % participants (base)
If power of attorney in place (n = 806): whether power of attorney used
Yes 27.9 236
No 70.7 562
Not stated 1.4 8
If no power of attorney (n = 2,594): whether had discussions with someone in family to be
appointed under a power of attorney
Yes 24.7 732
No 74.3 1,838
Not stated 1.0 24
Notes: Based on weighted data. Row percentages may not total 100.0% due to rounding.
Table A10.11: Survey of the General Community: Participants who were appointed under a power of attorney,
proportion of how persons who granted the power of attorney were related to the participants
Relationships %
Parents (incl. step-parents and in-law) 78.7
Partner/spouse 7.8
Sibling incl. step-sibling 5.2
Friend 3.5
Other relative 3.0
Grandparent incl. step-grandparent 2.6
Son/daughter incl. step-son/step-daughter 0.9
Unrelated person (e.g. neighbour) 0.4
Other unspecified (incl. small proportion of 1.3
ex-partner/spouse)
Don’t know/Refused 1.6
Number of participants (base) 806
Table A10.12: Survey of the General Community: Socio-demographic characteristics of people appointed
pursuant to a power of attorney
Number of participants
Family agreements % (base)
Whether family agreement in place
Yes 3.0 93
No 96.2 3,283
Don’t know/refused ^0.8 24
Where family agreement present (n = 93): Whether family agreement is written down
Yes 54.6 58
No 40.2 31
Don’t know/refused 5.2 4
Notes: Based on weighted data. Row percentages may not total 100.0% due to rounding.
Table A10.14: Survey of the General Community: Participants with a family agreement, how parties to the
family agreement were related
%
Relationships (n = 93)
Parents (incl. step, in-law) 70.8
Grandparent incl. step-grandparent 15.2
Other relative 7.1
Don’t know/refused 5.8
Sibling incl. step-sibling 3.2
Unrelated person (e.g. neighbour) 2.8
Partner/spouse 2.7
Other unspecified (incl. small proportion of ex-partner/spouse) 1.4
Friend 0.5
Son/daughter (incl. step-son/step-daughter) 0.4
Table A10.15: Survey of the General Community: Characteristics of participants who had entered a family
agreement compared with power of attorney
Notes: Based on weighted data. Excludes don’t know/refused responses for both categories.
Appendix A: Additional tables 219
Chapter 11
Table A11.1: Survey of Older People: Number of types of assistance received with financial matters, by age
Age
65–69 years 70–74 years 75–79 years 80–84 years 85+ years
Types of financial assistance (%) (%) (%) (%) (%)
Number of types of assistance received
None 48.9 52.8 54.2 59.8 56.4
One 21.2 20.1 20.9 16.6 19.3
Two 13.4 12.0 13.0 12.2 12.6
Three 9.0 7.5 6.0 5.3 6.7
Four or more 7.5 7.6 5.9 6.0 5.1
Total 100.0 100.0 100.0 100.0 100.0
Number of participants (base) 1,697 1,927 1,525 1,105 746
Table A11.2: Survey of Older People: Older persons who received assistance in financial matters, who provided
assistance with financial matters, by gender
Table A11.3: Survey of Older People: Older persons whose PIN was accessible by someone else, who had
access to PIN, by gender of participant
Table A11.4: Survey of Older People: Older persons whose PIN was accessible by someone else, who had
access to PIN, by age of participant
65–69 years 70–74 years 75–79 years 80–84 years 85+ years
Access to PIN (%) (%) (%) (%) (%)
Partner/spouse 80.4 79.1 73.5 52.7 25.4
Son/daughter (biological/adopted) 15.9 18.1 24.6 42.6 65.9
Step-son/daughter 0.4 0.3 0.6 2.0 2.6
Grandson/daughter 0.1 0.8 1.0 1.7 2.3
Brother/sister 1.9 0.2 1.1 0.7 0.9
Brother/sister in-law 0.0 0.2 0.1 0.0 0.3
Son/daughter (in-law) 0.6 1.6 0.4 1.2 1.7
Other family member 0.8 0.0 0.6 2.0 1.8
Ex-partner/spouse 1.0 0.1 0.0 0.0 0.0
Friend 0.9 1.4 0.3 0.5 0.3
Neighbour 0.0 0.3 0.0 0.0 0.0
Professional carer 0.0 0.0 0.1 0.0 0.2
Financial planner 0.8 0.6 0.5 0.0 0.1
Other service provider 0.0 0.2 0.2 0.2 0.8
Acquaintances 0.7 0.7 0.0 0.0 0.6
Other 0.3 0.3 0.8 0.0 0.0
Number of participants (base) 438 509 367 270 241
Table A11.5: Survey of the General Community: Older person to whom SGC participants provided assistance
with financial matters
Notes: Based on weighted data. Percentages may not total 100.0% due to rounding. a This option includes: aunt/uncle, son/
daughter-in-law, other family member, brother/sister-in-law, grandparents-in-law. b This option includes: neighbour,
client/customer/patient, colleague. c This option includes all other including a small proportion of ex-partner/spouse.
Chapter 12
Table A12.1: SOP and SGC: Mean scores of scale of acceptance attitudes, by socio‑demographic characteristics
SOP SGC
Number of Number of
participants participants
Characteristic Mean (base) Mean (base)
Indigenous status ..
Non-Indigenous 28.2 6,238 10.4 3,239
Indigenous 28.6 76 16 77
Country of birth
Australia 27.7 4,561 8.6 2,319
English speaking countries 25.4* 1,065 8.7 367
Non-English speaking countries 32.2* 699 17.6* 629
Years since first arrived in Australia
<5 years 6 18.2 192
5–14 years 32.4 39 15.8 362
15+ years 29.1 1,701 11.7* 422
Religion, 3 categories
Christianity 29.8 4,181 10.2 1,415
Other religion 30.7 532 15.8* 373
No religion 22.9* 1,530 9.6 1,490
Education
Degree or higher 22.1 1,628 10.4 1,395
Certificate/Diploma/Trade/Other 26.9* 1,706 8.8 991
Year 12 27.7* 696 12 498
Below Year 12 31.7* 2,128 13.4* 408
Employment, detailed
Full-time 22.9 317 8.9 1,631
Part-time 24.5 507 11.0* 846
222 National Elder Abuse Prevalence Study: Final Report
SOP SGC
Number of Number of
participants participants
Characteristic Mean (base) Mean (base)
Not employed a 28.9* 5,496 13.6* 839
Partnering
Living with partner 27.8 3,509 8.7 1,981
Not living with partner 29.3* 2,802 13.6 1,273
Whether having any children (of any age) in household (bio/step/foster)
No 28.3 5,646 12 1,894
Yes 27.6 666 8.8* 1,411
Family type
Partnered with child/ren 27.1 361 8.8 1,157
Partnered no children 27.9 3,148 8.3 854
No partner with child/ren 28.6 305 8.4 254
No partner no child/ren 29.3 2,492 14.5* 1,024
Before tax or deductions, personal annual income ..
quintile
Lowest personal income quintile (incl. neg./zero income) .. 13.4 754
2 .. 9.2* 401
3 .. 10.2* 468
4 .. 7.2* 708
Highest personal income quintile .. 8.6* 178
Before tax or deductions, household annual income quintile
Lowest household income quintile (incl. neg./zero income) 31 997 13.3 455
2 29.2 523 11.1 453
3 27.9* 982 8.2* 426
4 25.2* 702 7.5* 328
Highest household income quintile 22.0* 702 6.6* 374
Housing
Own outright 28.6 4,925 10.7 805
Own, paying off mortgage 23.6* 414 8.6* 1,204
Rent from private landlord 28.6 367 12.4 845
Rent from public housing authority 30.8 197 18.5* 126
Other (boarding, living at home, etc.) 28.5 359 12.1 288
SEIFA Quintile (IRSD 2016)
1 Lowest socio‑economic status 29.2 953 14.2 475
2 28.1 1,227 11.0* 551
3 28.4 1,213 8.6* 643
4 28.6 1,202 11.2* 698
5 Highest socio‑economic status 27.5 1,733 9.2* 956
Capital city/Rest of state
Capital city 28.5 3,767 11.2 2,296
Rest of state 27.8 2,564 9.2* 1,032
Kessler 6 mental health
No probable serious mental illness 27.9 6,157 ..
Probable serious mental illness 40.0* 167 ..
Social support scale a
Appendix A: Additional tables 223
SOP SGC
Number of Number of
participants participants
Characteristic Mean (base) Mean (base)
Lower quartile 36.1 1,615 ..
Middle quartiles 27.8* 3,422 ..
Upper quartiles 18.9* 1,241 ..
Notes: Weighted data and unweighted sample size. *The difference in mean scores between the category and the reference
category (listed as the first category) is a statistically significant difference at the 5% level. Statistical significance test is
based on bivariate regression. a This category includes unemployed, retired and not in the labour force.
Table A12.2: SOP and SGC: Mean scores of scale of recognising abusive behaviours, by socio‑demographic
characteristics
SOP SGC
Number of Number of
participants participants
Characteristic Mean (base) Mean (base)
Indigenous status
Non-Indigenous 84.1 6,704 85.2 3,281
Indigenous 83.2 82 81.8 77
Country of birth
Australia 84.1 4,869 86.4 2,338
English speaking countries 86.8* 1,143 89.5* 370
Non-English speaking countries 81.7* 788 78.7* 648
Years since first arrived in Australia
<5 years 6 75.5 196
5–14 years 85.3 42 80.8* 365
15+ years 84 1,865 86.3* 436
Religion, 3 categories
Christianity 83.6 4,535 86.2 1,430
Other religion 80.2* 570 82.2* 385
No religion 86.6* 1,603 84.8 1,504
Education
Degree or higher 86.5 1,691 84 1,412
Certificate/Diploma/Trade/Other 85.2 1,811 87.4* 1008
Year 12 82.8* 738 83.7 499
Below Year 12 83.1* 2,373 84.4 418
Employment, detailed
Full-time 85.6 336 86.1 1,650
Part-time 87 522 84.8 851
Not employed a 83.7 5,934 83.0* 857
Partnering
Living with partner 84.4 3,692 86.5 2,002
Not living with partner 83.2* 3,090 82.8* 1,292
Whether having any children (of any age) in household (bio/step/foster)
No 84 6,080 83.9 1,918
Yes 84.1 706 86.4* 1,431
Family type
Partnered with child/ren 84.3 376 86.1 1,176
224 National Elder Abuse Prevalence Study: Final Report
SOP SGC
Number of Number of
participants participants
Characteristic Mean (base) Mean (base)
Partnered no children 84.4 3,316 87 857
No partner with child/ren 83.7 330 88.3 255
No partner no child/ren 83.3 2,756 81.7* 1,042
Before tax or deductions, personal annual income quintile
Lowest personal income quintile (incl. neg./zero income) .. 82.8 764
2 .. 86.8* 402
3 .. 86.2* 473
4 .. 87.0* 711
Highest personal income quintile .. 84.9 177
Before tax or deductions, household annual income quintile
Lowest household income quintile (incl. neg./zero income) 82.9 1,103 84.3 465
2 84.6 547 85.9 454
3 85.9* 1,022 86.3 431
4 85.1 725 86.2 328
Highest household income quintile 86.4* 717 87.2 376
Housing
Own outright 84.2 5,287 87.1 812
Own, paying off mortgage 86.5* 429 86.2 1,218
Rent from private landlord 82.4 395 82.7* 860
Rent from public housing authority 81.9 219 79.4* 127
Other (boarding, living at home, etc.) 82.7 401 81.7* 290
SEIFA Quintile (IRSD 2016)
1 Lowest socio‑economic status 83.1 1,025 84.8 485
2 83.9 1,327 85.3 554
3 84.4 1,313 85.9 654
4 84.5 1,303 84.4 703
5 Highest socio‑economic status 84.1 1,834 84.8 970
Capital city/Rest of state
Capital city 83.7 4,049 84.0 2,334
Rest of state 84.5 2,756 86.7* 1,036
Kessler 6 mental health
No probable serious mental illness 84 6,612
Probable serious mental illness 82.6 185
Social support scale
Lower quartile 79.1 1,766
Middle quartiles 84.7* 3,685
Upper quartiles 89.5* 1,282
Notes: Weighted data and unweighted sample size. * The difference in mean scores between the category and the reference
category (listed as the first category) is a statistically significant difference at the 5% level. Statistical significance test is
based on bivariate regression. a This category includes unemployed, retired and not in the labour force.
Appendix A: Additional tables 225
Table A12.3: Survey of the General Community: Bivariate analysis of the Ageism scale by socio‑demographic
characteristics
Number of
Lower Middle Upper participants
Characteristic Mean quartile (%) quartiles (%) quartile (%) Total (%) (base)
Gender
Male 39.2 22.0 52.0 26.0 100.0 1,597
Female 35.7* 33.0 43.6 23.5 100.0 1,783
Age (years)
<25 years 39.9 17.6 58.0 24.4 100.0 408
25–34 years 37.9 25.9 48.3 25.7 100.0 545
35–44 years 36.9* 29.1 47.2 23.8 100.0 606
45–54 years 36.7* 29.3 45.8 24.8 100.0 775
55–64 years 35.9* 34.0 41.7 24.4 100.0 1,021
Indigenous status
Non-Indigenous 37.2 27.4 48.3 24.4 100.0 3,293
Indigenous 42.7 32.4 30.9 36.8 100.0 80
Country of birth
Australia 33.3 32.4 51.3 16.3 100.0 2,349
English speaking 34.9 34.1 44.7 21.2 100.0 371
countries
Non-English speaking 51.0* 9.6 39.3 51.1 100.0 652
countries
Years since first arrived in Australia
<5 years 52.5 6.1 42.9 51.0 100.0 197
5–14 years 48.2* 15.5 36.5 48.0 100.0 369
15+ years 40.0* 24.7 45.7 29.6 100.0 438
Religion, 3 categories
Christianity 38.5 26.1 47.0 26.9 100.0 1,433
Other religion 49.3* 13.5 38.9 47.6 100.0 388
No religion 33.2* 32.7 50.7 16.6 100.0 1,510
Education
Degree or higher 35.7 31.8 43.5 24.7 100.0 1,411
Certificate/Diploma/ 36.5 27.3 50.8 21.9 100.0 1,012
Trade/Other
Year 12 37.9 24.9 52.7 22.4 100.0 503
Below Year 12 42.5* 22.3 42.5 35.2 100.0 425
Employment, detailed
Full-time 35.9 28.9 49.9 21.2 100.0 1,656
Part-time 36.3 30.6 44.9 24.5 100.0 853
Not employed 41.8* 21.0 47.2 31.8 100.0 863
Partnering
Living with partner 36.3 29.8 46.4 23.9 100.0 2,012
Not living with partner 39.0* 23.8 50.4 25.7 100.0 1,297
Whether having any children (of any age) in household (bio/step/foster)
No 38.0 26.1 49.4 24.5 100.0 1,923
Yes 36.7 29.1 46.1 24.8 100.0 1,441
226 National Elder Abuse Prevalence Study: Final Report
Number of
Lower Middle Upper participants
Characteristic Mean quartile (%) quartiles (%) quartile (%) Total (%) (base)
Family type
Partnered with child/ 37.0 28.0 46.6 25.4 100.0 1,184
ren
Partnered no children 35.1 32.8 46.3 20.8 100.0 859
No partner with child/ 34.9 35.5 43.6 20.9 100.0 257
ren
No partner no child/ 39.8* 21.6 51.9 26.6 100.0 1,045
ren
Before tax or deductions, personal annual income quintile
Lowest personal 40.2 22.4 49.0 28.6 100.0 765
income quintile (incl.
neg./zero income)
2 35.9* 31.1 49.0 19.9 100.0 403
3 36.8* 26.6 50.3 23.0 100.0 474
4 34.7* 28.5 51.4 20.1 100.0 713
Highest personal 29.6* 43.7 43.7 12.6 100.0 178
income quintile
Before tax or deductions, household annual income quintile
Lowest household 40.8 23.8 47.0 29.2 100.0 466
income quintile (incl.
neg./zero income)
2 37.0* 26.7 52.6 20.7 100.0 457
3 35.5* 27.8 49.7 22.5 100.0 432
4 32.1* 34.0 50.5 15.4 100.0 329
Highest household 30.2* 39.6 46.9 13.5 100.0 376
income quintile
Housing
Own outright 35.4 31.6 47.5 21.0 100.0 816
Own, paying off 35.7 30.3 46.9 22.8 100.0 1,219
mortgage
Rent from private 40.3* 23.1 47.1 29.8 100.0 868
landlord
Rent from public 49.4* 10.8 47.8 41.4 100.0 132
housing authority
Other (boarding, living 39.8* 18.6 55.6 25.7 100.0 289
at home, etc.)
SEIFA Quintile (IRSD 2016)
1 Lowest 40.5 25.2 42.8 32.0 100.0 488
socio‑economic status
2 38.8 22.1 54.2 23.7 100.0 557
3 37.7 27.8 46.6 25.7 100.0 655
4 38.1 27.7 46.4 25.9 100.0 712
5 Highest 33.7* 32.2 48.9 18.8 100.0 968
socio‑economic status
Capital city/Rest of state
Capital city 38.1 26.8 46.3 26.9 100.0 2,341
Rest of state 35.8* 28.8 51.5 19.7 100.0 1,044
Notes: * The difference in mean scores between the category and the reference category (listed as the first category) is a
statistically significant difference at the 5% level. Statistical significance test is based on univariate regression.
Appendix A: Additional tables 227
Table A12.4: Survey of the General Community: Bivariate analysis of the Intergenerational Support scale by
socio‑demographic characteristics
Middle Number of
Lower quartiles Upper participants
Characteristic Mean quartile (%) (%) quartile (%) Total (%) (base)
Indigenous status
Non-Indigenous 66.6 28.1 49.9 22.1 100.0 3,290
Indigenous 67.2 31.9 40.5 27.6 100.0 77
Country of birth
Australia 64.9 31.9 49.6 18.5 100.0 2,339
English speaking countries 62.6 31.1 54.6 14.3 100.0 370
Non-English speaking 73.2* 15.9 48.0 36.0 100.0 656
countries
Years since first arrived in Australia
<5 years 75.2 11.9 45.8 42.3 100.0 198
5–14 years 72.5 18.3 48.7 32.9 100.0 369
15+ years 64.7* 27.6 53.0 19.4 100.0 439
Religion, 3 categories
Christianity 64.2 32.4 49.3 18.3 100.0 1,501
Other religion 67.1* 27.3 49.9 22.8 100.0 1,435
No religion 73.4* 16.5 48.5 35.0 100.0 390
Education
Degree or higher 66.5 26.4 53.9 19.7 100.0 1,407
Certificate/Diploma/Trade/ 66.7 29.3 47.2 23.6 100.0 1,008
Other
Year 12 68.0 24.7 51.3 24.0 100.0 503
Below Year 12 64.6 33.3 46.0 20.7 100.0 426
Employment, detailed
Full-time 66.0 29.6 49.8 20.6 100.0 1,651
Part-time 66.5 27.5 50.6 21.8 100.0 853
Not employed 67.6 26.3 48.5 25.2 100.0 862
Partnering
Living with partner 65.6 30.3 49.1 20.5 100.0 2,010
Not living with partner 68.1* 24.2 51.3 24.5 100.0 1,292
Whether having any children (of any age) in household (bio/step/foster)
No 66.0 28.6 50.1 21.3 100.0 1,918
Yes 67.2 27.5 49.5 22.9 100.0 1,439
Family type
Partnered with child/ren 67.3 27.4 49.9 22.7 100.0 1,182
Partnered no children 62.7* 35.5 47.9 16.6 100.0 859
No partner with child/ren 66.9 28.4 47.3 24.3 100.0 257
No partner no child/ren 68.2 23.8 51.7 24.4 100.0 1,040
Before tax or deductions, personal annual income quintile
Lowest personal income 67.3 26.1 50.7 23.2 100.0 761
quintile (incl. neg./zero
income)
2 66.9 27.6 47.6 24.8 100.0 402
3 67.3 25.4 51.0 23.6 100.0 469
4 63.9* 33.8 50.2 16.0 100.0 713
228 National Elder Abuse Prevalence Study: Final Report
Middle Number of
Lower quartiles Upper participants
Characteristic Mean quartile (%) (%) quartile (%) Total (%) (base)
Highest personal income 64.4 32.1 55.2 12.8 100.0 178
quintile
Before tax or deductions, household annual income quintile
Lowest household income 67.6 25.8 49.4 24.8 100.0 462
quintile (incl. neg./zero
income)
2 64.9 34.7 42.1 23.2 100.0 455
3 64.8 29.0 53.0 18.0 100.0 429
4 63.7* 34.1 51.3 14.6 100.0 329
Highest household income 66.2 26.9 55.5 17.5 100.0 375
quintile
Housing
Own outright 63.2 32.7 50.2 17.1 100.0 816
Own, paying off mortgage 65.5* 30.6 49.7 19.7 100.0 1,216
Rent from private landlord 69.8* 23.6 48.0 28.4 100.0 862
Rent from public housing 74.6* 17.0 42.2 40.8 100.0 132
authority
Other (boarding, living at 69.1* 20.8 55.0 24.2 100.0 290
home, etc.)
SEIFA Quintile (IRSD 2016)
1 Lowest socio‑economic 67.8 29.3 43.0 27.7 100.0 489
status
2 66.9 28.1 49.1 22.8 100.0 555
3 66.5 28.7 48.0 23.3 100.0 652
4 66.5 27.7 51.8 20.5 100.0 712
5 Highest socio‑economic 65.4 28.0 53.7 18.3 100.0 966
status
Capital city/Rest of state
Capital city 67.6 25.6 50.7 23.7 100.0 2,341
Rest of state 64.0* 34.0 47.6 18.4 100.0 1,038
Notes: * The difference in mean scores between the category and the reference category (listed as the first category) is a
statistically significant difference at the 5% level. Statistical significance test is based on univariate regression.
Appendix B: Methodology 229
Appendix B: Methodology
Relevant questions in
Topic Survey of Older People Sources Rationale
Demographics (e.g. members Section A Australian surveys (e.g. Capture data relevant to
of household & housing, HILDA; Census; The Survey the characteristics of the
education, marital status, of Foster and Relative/ older person and potential
country of birth, ATSI Kinship Carers) risk/mitigating factors
status, sexual orientation, of abuse; data relevant
employment, income) to potential vulnerability
including social isolation
Social networks and social Section B, including the Household, Income and
or ill-health; understanding
support social support scale Labour Dynamics in
contextual factors that are
Australia (HILDA) survey;
correlates or moderators
UK (2007)
of abuse
Relevant questions in
Topic Survey of Older People Sources Rationale
Perpetrator characteristics D2–D6 Canada (2015); UK (2007); Collect data on the
(i.e. gender, living E9–E14; Ireland (2010) perpetrators of abuse and
arrangements, age, relevant characteristics/
employment, specific F2–F7; risk factors associated with
problems such as G2–G7; abuse
substance abuse) H2–H7 Where multiple perpetrators,
participants are asked
about the person who has
affected them the most to
reduce participant burden
Seriousness of event for D7; E15; F8; G8; H8 Ireland (2010) To collect data on the
participant impact of the event on the
participant
Help-seeking or reporting D8–D11; E16–E19; F9–F12; UK (2007); Ireland (2010) To collect data on
behaviour G9–G12; H9–H12 help‑seeking or reporting
behaviours where available
Overall experience of abuse D12–D13; E20–E21; F13–F14; Canada (2015) To understand participant’s
or neglect G13–G14; H13–H14 subjective experience of
abuse or neglect since
turning 65
It is worth noting that research on elder abuse communities is often qualitative and, at the time of survey development,
AIFS was not aware of any surveys with a focus on CALD communities. Consequently, items were specifically
created for the Survey of Older Persons. They apply to those participants who were born in non‑English speaking
countries. Consistent with other areas of the SOP survey, additional CALD questions assess the frequency of the abuse
experience and the relationship between the participant and the source of the behaviour (SOP Questionnaire Section I).
Table B2: Sources of relevant questions in the Survey of the General Community
Instrument drafting
Preparatory work was undertaken for instrument development and this involved a review of international
prevalence studies and existing literature to articulate the core concepts and constructs that could be measured
by the survey instruments. Identification of relevant scales and items was also undertaken. Following this, survey
instruments were drafted.
21 The original scale contained 13 items. Due to limitations associated with the survey length, the SGC includes six items based on factor
loadings. The six items cover two subscales, three items on benevolent ageism and three on hostile ageism.
Appendix B: Methodology 233
The scales and measures on elder abuse used in the draft surveys were all derived from existing international
prevalence studies; more specifically, the following studies: Canada (2015), New York, USA (2011), Ireland
(2010) and UK (2007) (refer to Table 1 in section 1 for the methodology for these and other international elder
abuse prevalence studies). Other questions on socio‑demographic backgrounds or questions important for
understanding the context of elder abuse were derived from other existing studies (e.g. Household, Income and
Labour Dynamics in Australia Survey (HILDA), Census).
Instrument testing
Testing for the Older Person and General Community surveys involved three main elements, described below.
Cognitive testing
The Social Research Centre (SRC) conducted 18 cognitive interviews for both surveys in August 2018, with
people aged 62+ for the Older People instrument and people ranging from 27–57 for the General Community
instrument. Cognitive testing also included participants from CALD backgrounds. Participant feedback led to the
revision of both questionnaires and included:
y reduction in the length of the questions
y revision of the wording for some questions
y revised response options.
More specifically, feed back from cognitive testing for the Older People instrument included that:
y Sensitivities in subject matter and wording need to be foreshadowed prior to asking relevant questions.
y All jargon should be avoided and the language used in the survey must be readily understood and accessible
for use in a telephone format.
y Alternative wording for some questions was recommended.
y Participants reported great value in the survey, with some noting that although the survey covered sensitive
issues, it was important to ask these difficult questions to gauge the extent of elder abuse in Australia.
y Some participants reported experiencing fatigue by the end of the survey and suggested strategies whereby
interviewers could allow for breaks or call back at alternate times as potential solutions.
Scale validation
SRC conducted online surveys to test scales from the two questionnaires in September 2018. Data were collected
from 800 participants from an online panel. Scale validation testing used Factor and Rasch analyses. This process
aimed to reduce the number of items measuring a particular scale; for example, attitudes to elder abuse, health,
strength of social networks, ageism.
The data collected were used to validate the pre-identified scales. Initially, exploratory analyses were conducted
to identify the main groups of items measuring domains of interest (e.g. ageism), followed by confirmatory
analyses to test the reliability of the relationship between the items within the groups identified. The results of
scale validation informed the further refinement of the two questionnaires, which resulted in a reduction in the
length of the survey due to the removal of certain items. The draft questionnaires were then pilot tested.
The pilot testing involved the use of SRC’s established ‘call alert’ protocols for logging any participant concerns
or issues. The protocols require interviewers to fill out a ‘call alert’ form when they have any concerns about the
real or perceived physical or emotional welfare of participants.
The ‘call alert’ form provides a direct link to field operations of any situation that might require follow-up by
the project lead or relevant authorities, if disclosure of imminent self-harm or harm to others is recorded.
This escalation process also helps interviewers in processing difficult or distressing calls, as well as ensuring
234 National Elder Abuse Prevalence Study: Final Report
interviewer welfare is maintained. In preparation for the ‘call alert’ process, interviewers were trained in how to
deal with sensitive subject matter and potentially distressed opponents.
This secondary work involved testing the structure and content of the questionnaires (e.g. survey introduction/
consent script, survey questions, response options, flows, skips, etc.). It also replicated all aspects of the main
survey processes, including recruitment using the random digit dialling approach, letters where required,
fieldwork procedures and supporting documentation. One hundred telephone interviews were conducted for the
SOP and 20 telephone interviews were conducted for the SGC.
SRC conducted debriefing sessions with interviewers following the pilot testing of each survey. Following
feedback from SRC, changes were made to the questionnaires, including:
y refining introductory and consent scripts
y adding interviewer notes and instructions
y refining question wording to improve comprehension
y providing additional response options to minimise the volume of responses going through ‘other’.
After these initial interviews, a raw data file was produced and feedback from interviewer monitoring and
debriefing collated to support finalising the questionnaires and fieldwork methodology. This process enabled
the research team to make any final minor revisions to the questionnaire to adjust for length and phrasing of
questions, to assist with the recruitment of participants through the telephone introduction and consent script,
and to improve the overall flow of the questionnaires.
Fieldwork
This section outlines the sampling recruitment approach and data collection activities for the SOP and SGC. The
SOP and SGC Technical Reports provided by SRC provide more detailed information on the sample recruitment
and fieldwork.
Appendix B: Methodology 235
A primary approach letter was also sent to each record for which an address could be established by SamplePages
using the Sensis’ MacroMatch service (for approximately 30% of the total landline sample of the SGC). For both
surveys, mobile phone numbers selected to take part were sent a pre-approach text message informing them
that they might be contacted for the research and offering them a way to opt out if they wanted to.
Potential participants who had not received explanatory materials via SMS or mail were provided with the
opportunity to receive the explanatory materials (either a copy of the information sheet or direction to a weblink
with the information sheet), and to then be recontacted by SRC at a later stage. This approach ensured that all
participants were in the same position and had been given time to consider the decision to participate rather
than feeling rushed or pressured into making a decision.
Participants were offered a call back to undertake the interview at a convenient time, if they were unable to do so
at the time of the initial call.
Interview
Interviewers were selected on the basis of their experience and demonstrated skill in projects with similar
sensitivities and requirements. Training was also provided by SRC prior to fieldwork so interviewers were briefed
on the project background, privacy and confidentiality issues, mandatory reporting and on the administration of
the survey. In-field quality monitoring techniques were also applied to the survey and these techniques included
monitoring each interviewer within the first three shifts and regular field-team debriefing.
In all cases, an oral consent script contained in the SOP and SGC questionnaires was read out to participants
by interviewers prior to the commencement of the substantive telephone interview. Time was provided to
participants to consider this information and to ask any questions if they wished to do so. The introduction script
was constructed in a way to provide ‘explain if necessary’ text and interviewer briefing notes in the event the
participant had further questions. Screening protocols for interviewers were also developed for any potential
participants contacted who were living in care facilities.
Participants who wished to be involved in an interview were generally required to have sufficient fluency in
English to understand the nature of the research project, to give informed consent and to understand and
respond to the survey questions. However, bilingual interviewers were available for a number of languages
(as determined prior to fieldwork) for participants who wished to participate in the study but who were not
fluent speakers of English.
Where a participant was insufficiently fluent in English to be interviewed, the language was recorded and the
call followed-up by the relevant bilingual interviewer where available. Informed consent was obtained in the
same way by interviewers: by reading out the translated consent script, providing the opportunity to receive
a translated information sheet and obtaining oral consent from the participant, which was recorded by the
interviewer in the program.
The summary statistics shown in Table 3 indicate budgetary and sampling design targets were met for the SOP.
The in-scope population for the survey was the non-institutionalised population of Australia, aged 65 years and
over, and it excluded residents of institutional premises (prisons, nursing homes, etc.) and military bases. The
response rate (RR3) for the survey was 15.5% and the average interview length was 27.9 minutes. The sample
blend used for the target interview completes of 7,000 was 80% landline numbers and 20% mobile phone
numbers. Taking into account cost and sampling considerations (e.g. older people being more likely to have
landline numbers) the aim of this blend was to ensure higher response rates.
For the landline sample, the participant was selected from all residents of the household aged 65 years or over.
In the case of the mobile sample, the qualifying participant was the telephone answerer, if aged 65 years or over.
Participants were offered a call back to undertake the interview at a convenient time, if they were unable to at
the time of the initial call.
The landline sample was stratified in proportion to the population based on state and capital city/rest of state
divisions. Given the lack of geographic information available for the mobile sample, the mobile phone stratum was
not geographically stratified and allowed to ‘fall naturally’. Final allocations to 15 geographic strata were based on
the postcode/location information provided by the landline and mobile sample. Acknowledging there are a range
of factors that may affect the accuracy of survey results, the sample size achieved of 7,000 for the SGC means
smaller sampling errors and greater opportunity for subgroup analysis (dependent on the variable of focus).
Sample frame
Key SOP fieldwork statistics Total Landline Mobile
Interviews completed 7,000 5,600 1,400
Response rate (RR3) 15.5% 15.3% 16.1%
Average interview length (minutes) 27.9 28.1 26.8
LOTE interviews completed (from total sample) 840
As indicated in section 2, some potential participants were excluded from taking part in the SOP (referred
to as coverage error). In this instance, persons without telephones (2.0% of adults) and those living in care
accommodation (3.2% of those aged 65+ years) had no chance of being selected to participate. While these
subgroups of the population may have different characteristics and outcomes from survey participants, the
overall extent of non-coverage is very low and is not expected to produce a responding dataset with notable
bias. In addition, since weighting was performed to match ABS benchmarks, which have extremely low levels
of coverage error, it is expected that weighted estimates made from the survey dataset will provide a good
representation of the older population. For further information about the approach to weighting see section 7.
A total of 200,041 sample records were generated and attempted for the survey, and 194,388 phone numbers
were called to complete the 7,000 interviews. In total, 433,102 calls were placed to the sample records. This
equates to an interview every 61.9 calls (65.0 calls per interview for landline numbers and 49.3 calls per interview
for mobile numbers). The average number of calls made to each sample member was 2.2 (2.2 calls per sample
member for the landline frame and 2.2 calls per sample member for the mobile frame).
Appendix B: Methodology 237
The landline sample was stratified in proportion to the population based on state and capital city/rest of state
divisions. Given the lack of geographic information available for the mobile sample, the mobile phone stratum
was not geographically stratified and allowed to ‘fall naturally’. Final allocations to 15 geographic strata were
based on the postcode/location information provided by the landline and mobile sample.
Sample frame
Key SGC fieldwork statistics Total Landline Mobile
Interviews completed 3,400 387 3,013
Response rate (RR3) 11.6% 13.2% 11.4%
Average interview length (minutes) 17.6 18.0 17.6
LOTE interviews completed (from total sample) 83
A total of 77,494 sample records were generated and attempted for the survey, and 67,165 phone numbers were
called to complete the 3,400 interviews. In total, 184,169 calls were placed to the sample records. This equates to
an interview every 54.2 calls (134.4 calls per interview for landline numbers and 43.9 calls per interview for mobile
numbers). The average number of calls made to each sample record was 2.7 (3.2 calls per sample record for the
landline frame and 2.6 calls per record for the mobile frame).
Ethical considerations
The AIFS Human Research Ethics Committee provided ethical review and clearance for this research. As
the surveys involved engagement with older people and people from the community, it was acknowledged
that during the course of their participation in a telephone interview and, in particular, when considering the
potentially sensitive questions included in the instruments, some participants may reflect on experiences of
violence/abuse and/or on complex family and other relationship dynamics, and that this may present risk issues.
These circumstances raised significant ethical complexities for the research team to address in order for the
Ethics Committee to provide ethical clearance. These complexities included:
y ensuring that data from a potentially vulnerable population who may have experienced significant levels of
trauma were collected sensitively, and without causing further trauma
y ensuring that appropriate interview protocols were in place to assist interviewers to sensitively collect the data
and, at the same time, support those participants who may become distressed during and after the interview.
First, all interviewers were selected on the basis of their experience and demonstrated skill in projects with similar
sensitivities and requirements. All selected interviewers attended a comprehensive briefing session covering the
project background, objectives and procedures; all aspects of administering the survey questionnaire, including
specific data quality issues; an overview of participant liaison issues (including sensitive issues that may arise);
and practice interviewing.
Second, protocols relating to capacity to consent, privacy and confidentiality, and sensitive issues were cleared
as part of the ethical clearance process. Mandatory reporting obligations were also in place.
Lastly, significant care was also taken to ensure that the data were stored in accordance with the approach
outlined in the application for ethical clearance (including electronic material stored on password-protected
servers, and files and hard copy documents in locked cabinets).
Data weighting
Surveys are a commonly used tool for collecting data that allow for making inferences about a population. To
be able to do so, however, requires a probability sample – one in which every element of the population has a
known, non-zero chance of selection. Some units in the population may not have a chance of selection (such
as persons who do not have a telephone so cannot be part of a telephone survey), and there may be different
rates of response across unit characteristics. Therefore, many sample surveys yield subsets that imperfectly cover
their target populations despite the best possible sample design and data collection practices. In such situations,
weighting can reduce the extent of any biases introduced through non-coverage.
Appendix B: Methodology 239
The approach for deriving weights generally consists of the following steps:
1. computing a design weight for each participant as the inverse of their chance of selection
2. adjusting the design weights so they match population distributions across a range of participant characteristics.
The first step is essential to providing the statistical framework necessary for making population inferences from
a sample survey. The second step aims to reduce non-response bias and to ensure that survey estimates are
consistent with other sources.
Design weights
The design weights account for differences in the probabilities of participants taking part in the survey, noting
that participants may have a chance of being selected in both the landline and the mobile sample.
Each participant’s weight is the inverse of their probability of selection where the chance of selection is
calculated via the following formula:
The design weight for person K is then the inverse of their selection probability:
1
dk
pk
Adjusted weights
To ensure that estimates made from the dataset are representative of the target population, the design weights
will be adjusted so that they match external benchmarks of key demographic parameters likely to be correlated
with the survey outcomes and the likelihood of response. Such auxiliary parameters vary from survey to survey
but typically include age, gender, education, geographic location and telephone status. Specific variables used
for weighting are confirmed after exploring the final dataset. Two things are looked at:
1. Which (potential) weighting variables are distributed very differently from the population? We will adjust for
any variables with notable non-response bias.
2. Which (potential) weighting variables are most correlated with the survey’s outcome variables? We will adjust
for any variables that are highly correlated with our outcomes.
Included in this exploration are any and all survey items for which there are reliable benchmarks. This would
include at least the items in the demographic sections (Parts 1 and 2), plus any other items that we can match to
the Census, the National Health Survey, etc.
The method for adjusting the design weights is generalised regression (GREG) weighting, which uses non-linear
optimisation to minimise the distance between the design and adjusted weights subject to the weights meeting
the benchmarks.
240 National Elder Abuse Prevalence Study: Final Report
Consistent with all the international prevalence studies, people who live in non-private dwellings, such as
residential aged care, and who are in private dwellings but do not have the cognitive capacity to engage in
meaningful interviews were excluded from the sample. Research considering the prevalence of abuse among
these vulnerable groups would be more appropriately examined through targeted research.22
Moreover, the larger the sample, the greater the opportunity to derive reliable estimates of the prevalence
of abuse in different subgroups where possible. Analysis of differences and similarities in response patterns
among subgroups may also reveal insights into attitudes and behaviours that shed light on, for example, factors
that increase the risk of elder abuse. Key subgroups for the Prevalence Study are based on demographic
characteristics (e.g. age groups, gender) and socio‑economic status (e.g. education and employment). Cultural
backgrounds (e.g. Indigenous, CALD) are other important subgroups.
The CALD substudy depends on the subsamples from the two surveys. A limitation of this approach is that
it necessitates an analytic approach based on the CALD sample as a whole, with limited ability to work with
specific subsamples. This will mean that differences among different subsamples will not be amenable to
in‑depth exploration. It is possible that the approach will attract criticism for not being sufficiently sensitive
to differences in experience between different CALD groups.
Data collection through telephone interviews can be achieved with a simpler sampling design at a much lower cost
compared with face‑to‑face data collection. Telephone interviewing also permits a large number of interviews to
be completed within a reasonably short time frame, and also allows a greater number of questions to be asked.
The relative anonymity of a telephone interview may make it easier for participants to disclose any experience of
abuse (Oltmann, 2016; Schober, 2018). Like face‑to‑face interviews, telephone interviews involve direct interaction
with participants, thereby enabling any participant queries/concerns to be immediately addressed.
22 Data from the 2016 Census showed that only 8% of older people aged 65 years and older were living in non-private dwellings and just
one-quarter of people aged 85 years and older were living in non-private dwellings. In other words, focusing on people aged 65 years
and older who live in private dwellings represents the experiences of the vast majority.
Appendix B: Methodology 241
Although the prospect of sampling from an administrative dataset such as the Medicare Enrolment Database
was initially raised, a probability sample was ultimately obtained by screening the wider population using a
dual‑frame (landline and mobile) random digit dialling methodology. The RDD approach is a robust method that
has been used in a number of international studies to provide research evidence on elder abuse. Dual-frame RDD
refers to the method of obtaining a sample of the target population by random screening of the wider population.
There are several disadvantages associated with the dual-frame random digit dialling approach and the fieldwork
company responsible for the data collection provided advice as to how to address these. For example, people
increasingly screen out unrecognised telephone numbers, and the response rate is likely to be lower and have
greater non-response bias compared to the approach of using a database as the sampling frame. Advice from
the fieldwork company also indicated older people were more likely to have landlines. Older people living in
remote areas that have a high proportion of the Indigenous population are also less likely to have access to
landline phones or have mobile phones compared to those living in non-remote areas. Both options would likely
result in under-representation of this segment of the older population.
The following methods were employed to improve sample precision and address the limitations of this approach:
y Data were weighted to reduce sample bias arising from differential non-response rates and/or coverage
(such as age, gender, region) and phone status (mobiles vs landlines).
y Stratified random sampling was used to ensure an adequate representation of subgroups of the population.23
The landline sample was stratified by state/territory and regional areas (metropolitan vs non-metropolitan
area) with no stratification for mobile phones (given mobile phones do not have prefixed area codes).
y Pre-contact SMS messages were also sent to potential mobile participants to alert them about the study prior
to receiving the call from SRC, with the view to increasing response rates.
23 Stratification can be achieved when the sampling frame includes information required for stratification, so that the sample can be
divided into discrete (i.e. non-overlapping) strata.
242 National Elder Abuse Prevalence Study: Final Report
ANSM1. Good morning/afternoon/evening. My name is <SAY NAME> calling on behalf of the Australian Institute
of Family Studies from the Social Research Centre. We are conducting an important national study about the
wellbeing of older people in Australia and how they are treated by others. If you would like to take part in this
study, please call our hotline number: 1800 083 037 and we will call you back at a time that is convenient to you.
Thank you and we look forward to hearing from you.
ANSM2. Good morning/afternoon/evening. My name is <SAY NAME> calling on behalf of the Australian Institute
of Family Studies from the Social Research Centre. We left a message recently on your answering machine/voice
mail regarding an important national study about the wellbeing of older people in Australia and how they are
treated by others. If you would like to take part in this study, please call our hotline number: 1800 083 037 and
we will call you back at a time that is convenient to you. Thank you and we look forward to hearing from you.
WELCOME SCREEN
Good (morning/afternoon/evening). My name is (...). I’m calling from the Social Research Centre, on behalf of the
Australian Institute of Family Studies.
We are conducting an important national study with people aged 65 years and older, about their wellbeing and
how they are treated by others.
This study is funded by the Australian Government Attorney-General’s Department. Its results will be used to
support policy development to better meet the needs of older people in Australia.
INTRODUCTION
*(ALL)
*(SAMPLETYPE=RDD, MOBILE)
INTRO2 For this survey, we are interested in talking to people aged 65 or over. Can I check, are you aged
65 years or over?
REINTRODUCE IF NECESSARY: Good (morning/afternoon/evening). My name is (...). I’m calling
from the Social Research Centre on behalf of the Australian Institute of Family Studies.
We are conducting an important national study with people aged 65 years and older about their
wellbeing and how they are treated by others.
IF NEEDED: This research is being funded by the Australian Government Attorney-General’s
Department. Results will be used to support government’s policy development to better meet
the needs of older people.
1. Yes – Continue
2. No – Under 65 years of age (GO TO TERM1)
3. Language difficulty (GO TO PLOTE)
3. Queried about how telephone number was obtained (GO TO ATELQ)
4. Respondent refusal (GO TO RR1)
*(SAMPLETYPE=1, LANDLINE)
A3a To see if your household qualifies for the study, could I first ask – including yourself, how many
people in your household are aged 65 or over?
1. None (GO TO TERM 1)
2 Number given (Specify) (_____, WHOLE NUMBERS, RANGE 1 TO 20)
3. Household refusal (GO TO RR1)
4. Language difficulty (GO TO PLOTE)
*(SAMPLETYPE=RDD, LANDLINE)
SELINTRO Good (morning/afternoon/evening). My name is (...). I’m calling from the Social Research Centre
on behalf of the Australian Institute of Family Studies.
We are conducting a study about the wellbeing of older people and how they are treated by
others. We are interested in speaking to people aged 65 years and older.
IF NEEDED: This study is being funded by the Australian Government Attorney-General’s
Department. Results will be used to support government’s policy development to better meet
the needs of older people.
1. Continue
2. Respondent refusal (GO TO RR1)
3. Language difficulty (GO TO PLOTE)
4. Queried about how telephone number was obtained (GO TO ATELQ)
*(SAMPLETYPE=RDD, MOBILE)
SAFE May I just check if it is safe for you to take this call at the moment. If not, I am happy to call you
back when it is more convenient for you.
1. Safe to take call
2. Not safe to take call
3. Respondent refusal (GO TO RR1)
MOB_APPT Do you want me to call you back on this number or would you prefer I call back on another
phone?
1. This number (STOP, MAKE APPOINTMENT)
2. Other number (STOP, MAKE APPOINTMENT, RECORD OTHER PHONE NUMBER)
3. Respondent refusal (GO TO RR1)
*(SAMPLETYPE=RDD, MOBILE)
MOB_APPT_A Just so I know your time zone, which state or territory do you live in?
1. NSW
2. VIC
3. QLD
4. SA
5. WA
6. TAS
7. NT
8. ACT
9. (Refused) (GO TO RR1)
*TIMESTAMP
*CONSENT
*(ALL)
CONSENT This survey will take about 25 minutes depending on your answers.
Participation is voluntary. You can finish the interview at another time or stop if you choose.
Some of the questions may seem quite sensitive. Please let me know if there are any you would
prefer not to answer and I will skip over them.
Any information you provide is protected by Commonwealth privacy laws and our Privacy
Policy, which is available on our website. Your information will be kept confidential, as far as
allowed by law. However, if you tell us about unreported abuse or neglect or a risk of harm to
yourself or other people, we may be required by law to report this to the relevant authorities.
The data we collect will be given to the Australian Institute of Family Studies who will use it to
write a research report for the Australian Government Attorney-General’s Department. We will
remove any information from the data that can identify you or your individual responses. If you
wish to withdraw from the study, you can do so up until we remove any identifiable information
from the dataset.
After this project is completed, these data will be kept at the Australian Data Archive at the
Australian National University and may be used for future research.
If you have any questions about the information I have provided, just let me know. If not are you
happy to begin the survey now?
IF RESPONDENT IS SUSPICIOUS OR DOUBTFUL: If you want to verify that the survey is
legitimate, or would like more information or to review our privacy policy, please call the Social
Research Centre’s 1800 number (1800 083 037), or visit our website at www.srcentre.com.au
IF RESPONDENT NEEDS MORE ASSURANCE: Or you could check the project information on
our website www.srcentre.com.au/sop
INTERVIEWER NOTE: If you have concerns respondent is not able to answer freely, arrange
call back
1. Yes, continue (GO TO MON)
2. Not a convenient time (MAKE APPOINTMENT)
246 National Elder Abuse Prevalence Study: Final Report
ATELQ [DISPLAY IF SAMPLETYPE=RDD, LANDLINE] Your telephone number has been chosen at
random from all possible telephone numbers in your area. We find this is the best way to obtain
a representative sample of people across Australia.
[DISPLAY IF SAMPLETYPE=RDD, MOBILE] Your number was randomly generated by a
computer. We’re calling mobile phones as well as landlines because lots of people have mobile
phones but don’t have landlines. So, we call mobile phones as well as landlines so we can get a
representative sample of people across Australia.
MON This call may be monitored or recorded for quality assurance purposes. Is that OK?
1. Yes
2. No
[PROGRAMMER NOTE: DISPLAY ALL CODE FRAME CHRONOLOGICALLY TO INTERVIEWERS. CODE FRAME
NUMBERS IN THIS DOC ARE FOR DATA]
*(TIMESTAMP)
DEMOGRAPHICS – Part 1
*(ALL)
*(ALL)
*(ALL)
DEM16am What is the name of the suburb or town where you live?
IF NEEDED: In order to make sure the study covers people living in all areas, we need to ask
where people live.
SELECT LOCALITY FROM LOCALITIES SPECIFIC TO SELECTED POSTCODE
INCLUDE SPECIFIED OTHER OPTION
88888 Don’t know
99999 Refused
*(ALL)
A3b Including yourself, how many people in your household are aged 65 or over?
1. SPECIFY (_____, WHOLE NUMBERS, RANGE 1 TO 20)
98. (Don’t know)
99. (Refused)
6. Grandson(s)/daughter(s)
7. Aunt(s) / uncle(s)
8. Brother(s) / sister(s)
9. Brother(s) / sister(s) (in-law)
10. Son(s) / daughter(s) (biological or adopted)
11. Son(s) / daughter(s) (step)
12. Son(s) / daughter(s) (in-law)
13. Other family member(s)
14. Friend(s)
15. Other person(s) not related to you (specify)
16. (Respondent lives in aged / respite care or nursing home) (TERM4)
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
*(ALL)
A6 Do you have any children (including step-children and adopted children) not living with you?
(IF YES, PROBE TO FRAME)
1. Yes, biological or adopted children
2. Yes, step-children
3. Yes, both biological / adopted AND step-children
4. No
98. (Don’t know)
99. (Refused)
*(ALL)
*(ALL)
16. Malta
8. New Zealand
9. Philippines
10. South Africa
11. United Kingdom (England, Scotland, Wales, Nth Ireland)
12. United States of America
13. Vietnam
14. Other (specify)
98. (Don’t know)
99. (Refused)
A9a In what year did you first arrive in Australia to live here for one or more years?
INTERVIEWER NOTE: If lived in Australia on more than one occasion, record the year they first
arrived to live for one year or more.
1. Record year (RANGE 1918 – 2020)
98. (Don’t know)
99. (Refused)
*(ALL)
*(TIMESTAMP)
*(ALL)
PREB1 The next few questions are about family, friends and support.
*(ALL)
B1 Approximately how often do you spend time with family members or friends living outside your
household? (Would it be …)
(READ OUT IF REQUIRED)
1. Every day
2. Several times a week
3. About once a week
4. 2 or 3 times a month
5. About once a month
6. Once or twice every 3 months
7. Less often than once every 3 months
98. (Don’t know)
99. (Refused)
250 National Elder Abuse Prevalence Study: Final Report
*(ALL)
B3 How often do you use the telephone, computer or similar devices to stay in touch with family
members or friends living outside your household?
(READ OUT IF REQUIRED)
1. Every day
2. Several times a week
3. About once a week
4. 2 or 3 times a month
5. About once a month
6. Once or twice every 3 months
7. Less often than once every 3 months
98. (Don’t know)
99. (Refused)
*(ALL)
B5 How often do you attend community activities such as social events or religious services?
(READ OUT IF REQUIRED)
1. Every day
2. Several times a week
3. About once a week
4. 2 or 3 times a month
5. About once a month
6. Once or twice every 3 months
7. Less often than once every 3 months
98. (Don’t know)
99. (Refused)
*(ALL)
B6 I’m going to read out some statements that can be used to describe how much support people
get from others.
For each one please tell me whether you strongly agree, agree, neither agree nor disagree,
disagree or strongly disagree.
These are your opinions, there are no right or wrong answers.
INTERVIEWER NOTE: Repeat scale as needed.
*(STATEMENTS)
*(ROTATE)
a) I often need help from other people but can’t get it.
1. Strongly agree
2. Agree
4. Disagree
5. Strongly disagree
99. (Refused)
*(TIMESTAMP)
PERSONAL WELLBEING
*(ALL)
*(ALL)
*(ALL)
C2 Do you have any medical conditions or disabilities that have lasted or are likely to last for six
months or more and cause you any difficulties in everyday life?
1. Yes
2. No
98. (Don’t know)
99. (Refused)
*(ALL)
1. Yes
2. No
99. (Refused)
*(ALL)
C5 The following questions are about your feelings in the past four weeks. In the last four weeks,
how often did you feel …
*(STATEMENTS)
*(ROTATE)
a) Nervous
b) Hopeless
c) Restless or fidgety
252 National Elder Abuse Prevalence Study: Final Report
f) Worthless
*(RESPONSE FRAME)
(READ OUT)
99. (Refused)
*(ALL)
C7 The next questions are about how people treat each other. If you feel uncomfortable with any
question, just tell me and I will move onto the next one.
I am going to read out some statements about the abuse of older people. For each one please
tell me whether you strongly agree, agree, neither agree nor disagree, disagree or strongly
disagree.
(IF NECESSARY: There are no right or wrong answers, only opinions)
*(STATEMENTS)
*(ROTATE)
d) Abuse of older people can be understandable if the person committing the abuse is under
a lot of stress in their lives.
e) Abuse can be understandable if the older person is a difficult person to deal with.
*(RESPONSE FRAME)
(PROBE IF NECESSARY)
1. Strongly agree
2. Agree
3. Neither agree nor disagree
4. Disagree
5. Strongly disagree
98. (Don’t know)
99. (Refused)
*(ALL)
C8 Now, some questions about what you understand about the abuse of older people. These are
your opinions, there are no right or wrong answers.
People think of different things when they hear about the abuse of older people. I’d like you to
tell me whether you regard the following sorts of behaviour as abuse of an older person.
Just tell me yes or no as I read them out.
*(STATEMENTS)
Appendix C: Survey questionnaires 253
*(ROTATE)
c) Not paying bills on the older person’s behalf when you said you would
g) Not providing help with personal activities such as dressing, washing, feeding when this is
normally expected/provided
i) Preventing an older person from having contact with the outside world
*(TIMESTAMP)
PREVALENCE – Neglect
*(ALL)
D1 I am going to ask you some questions about how often you required help with daily activities
from a person or people over the past 12 months.
INTERVIEWER NOTE: We’re interested in finding out about people that you might get help
from, not a company, institution or agency.
During the last 12 months, how often did you require help with …
*(STATEMENTS)
a) Preparing meals
d) Travel or transport
e) Personal care such as washing or bathing (including getting in and out of bath or shower),
dressing or undressing
D1x I’d like to ask you some questions about the time someone you know was helping you with
[INSERT D1 STATEMENT].
D1ba Who is mainly responsible for helping you with [INSERT D1 STATEMENT]?
IF NEEDED: What is their relationship to you?
INTERVIEWER NOTE: Respondent can provide a maximum of two people. Record first person
here and second person on next page.
(PROBE AS NECESSARY)
(RESPONSE FRAME)(SINGLE REPONSE)
1. Partner / spouse
2. Ex-partner / spouse
3. Parent
4. Step-parent
5. Parent in-law
6. Grandson(s)/daughter(s)
7. Aunt / uncle
8. Brother / sister
9. Brother / sister (in-law)
10. Son / daughter (biological or adopted)
11. Step-son / daughter
12. Son / daughter (in-law)
13. Other family member
14. Friend
15. Neighbour
16. Professional carer
17. Medical professional (e.g. doctor, nurse)
18. Financial professional (e.g. account, financial advisor, bank employee)
19. Other professional (e.g. lawyer)
20. Other (specify)
21. No one *(SINGLE RESPONSE)
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
D1bb And is there another person who is mainly responsible for helping you with [INSERT D1
STATEMENT]?
IF NEEDED: What is their relationship to you?
(PROBE AS NECESSARY)
(SINGLE RESPONSE)(RESPONSE FRAME)
1. Partner / spouse
2. Ex-partner / spouse
3. Parent
4. Step-parent
5. Parent-in-law
6. Grandson(s)/daughter(s)
7. Aunt / uncle
8. Brother / sister
9. Brother / sister (in-law)
10. Son / daughter (biological or adopted)
11. Step-son / daughter
12. Son / daughter (in-law)
13. Other family member
14. Friend
15. Neighbour
16. Professional carer
17. Medical professional (e.g. doctor, nurse)
18. Financial professional (e.g. account, financial advisor, bank employee)
19. Other professional (e.g. lawyer)
20. Other (specify)
97. No one else
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
*(PROGRAMMER NOTE: WITHIN EACH STATEMENT AND IF D1Ba=1 TO 20, SHOW D1C FOR EACH PERSON
SELECTED AT D1Ba AND D1Bb)
D1c In the last 12 months, did you always get the help when you needed it from your [DISPLAY IF
D1Ba=1 TO 19 / D1bb=1 TO 19: D1B RESPONSE; ELSE: INSERT TEXT FROM D1Ba/D1bb=20]?
1. Yes
2. No
98. (Don’t know)
99. (Refused)
*(PROGRAMMER NOTE: WITHIN EACH STATEMENT AND IF D1Ba=1 TO 20 / D1bb=1 TO 20 AND D1C=2, SHOW
D1D FOR EACH PERSON SELECTED AT D1Ba and D1bb)
D1d How many times in the last 12 months did you not receive help from your [DISPLAY IF D1Ba=1
TO 19 / D1bb=1 TO 19: D1B RESPONSE; ELSE: INSERT TEXT FROM D1Ba/D1bb=20] when you
needed it? Was that …
256 National Elder Abuse Prevalence Study: Final Report
(READ OUT)
1. Once
2. A few times
3. Many times
4. Every day or almost every day
98. (Don’t know)
99. (Refused)
NEGLECT_COUNT
PERPETRATOR_NEGLECT
1. NEGLECT_COUNT=2 OR 3
2. ALL OTHERS
PERPETRATOR_COUNT_NEGLECT
1. IF PERPETRATOR_NEGLECT=1
0. IF PERPETRATOR_NEGLECT=2
D2 I’d like to know a bit more about the person who did not always provide help when you needed
it (in the last 12 months), and as a result, affected your daily life the most, can I just check who
is the person among those you mentioned?
IF NEEDED: Who is the person who affected your daily life the most, when they didn’t help you
with activities you needed help with.
*(PROGRAMMER NOTE: WHERE NEGLECT_COUNT=2 – ONE PERSON SELECTED FOR
NEGLECT, AUTOFILL RELATIONSHIP FROM D1ba
*(PROGRAMMER NOTE: DISPLAY SELECTIONS FROM D1ba AND D1bb ONLY)
1. Partner / spouse
2. Ex-partner / spouse
3. Parent
4. Step-parent
5. Parent in-law
6. Grandson(s) / daughter(s)
7. Aunt / uncle
8. Brother / sister
9. Brother / sister (in-law)
10. Son / daughter (biological or adopted)
11. Step-son / daughter
12. Son / daughter (in-law)
Appendix C: Survey questionnaires 257
D3 I’d like to know a bit more about the person who did not always provide help when you needed
it this is the person who, by not providing you help when you needed it, affected your daily life
the most (in the last 12 months).
Firstly, how old was this person (at the last time you needed help but did not receive it)?
1. SPECIFY (____, WHOLE NUMBERS, RANGE: 15 TO 100)
998. (Don’t know)
999. (Refused)
D3a Could you tell me which of the following age groups he/she was in?
(READ OUT)
1. Under 18
2. 18 – 24
3. 25 – 34
4. 35 – 44
5. 45 – 54
6. 55 – 64
7. 65 – 74
8. 75 and over
98. (Don’t know)
99. (Refused)
D5 Thinking about the last time you needed help but did not receive it, at the time did this person
live with you?
1. Yes
258 National Elder Abuse Prevalence Study: Final Report
2. No
98. (Don’t know)
99. (Refused)
D6 To the best of your knowledge, when this last happened did this person have problems with …?
(READ OUT ONE BY ONE)
*(ACCEPT MULTIPLES)
1. Alcohol
2. Drugs
3. Gambling
4. Mental health problems
5. Physical health problems
6. Financial problems
7. Other (specify)
8. (None of these) *(SINGLE RESPONSE)
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
D7 Because the person did not always provide help with some activities when you needed, how
serious was this for you? Was it …
(READ OUT)
1. Very serious
2. Somewhat/moderately serious
3. Not serious
98. (Don’t know)
99. (Refused)
1. A family member
2. A friend
3. Professional carer or social worker
4. GP or nurse
5. Hospital emergency department
6. The police
7. Helpline
8. Local council or authority
9. Lawyer
10. Neighbour
11. Your religious group
12. Other (please specify)
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
D10 Did you take any of the following actions in response to the behaviour?
(READ OUT ONE BY ONE)
*(ACCEPT MULTIPLES)
1. I spoke to the person.
2. A family member or friend spoke to the person on my behalf.
3. A professional (social worker, doctor, nurse) spoke to the person on my behalf.
4. I sought mediation or counselling.
5. I sought out a legal advice service.
6. I broke contact with or avoided the person involved.
7. I stopped going out or withdrew from social life in general.
8. A restraining order or safety order was made against the person.
9. Other (please specify)
10. (Nothing) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
D12 Now, thinking more generally … Sometimes people do not get help when they need it. One type
of neglect is when a person responsible for helping an older person does not help that person
with daily activities. Since turning 65 do you feel you were neglected in this way?
1. Yes
2. No
98. (Don’t know)
99. (Refused)
D13 How many times do you feel you were neglected since turning 65? Was it …
(READ OUT)
1. Once
2. A few times
3. Many times
98. (Don’t know)
99. (Refused)
*(TIMESTAMP)
*(STATE=NT)
*(ALL)
E1 I’d like to ask you some questions about how you manage your finances. We’re not interested
in any personal information about your finances, just about help you might have received to
manage them.
In the last 12 months, did anyone help you in any of the following ways with managing your money?
*(STATEMENTS)
*(ROTATE)
b) Paying bills, budgeting/accounting/monitoring finances
c) Taxes (e.g. tax returns)
d) Banking (e.g. withdrawing money from ATMs, using internet banking, etc.)
e) Managing pension (or Centrelink-related matters) or superannuation
f) Buying, selling or managing your home or other real estate
g) Buying, selling or managing a business
h) Managing investments such as shares, trusts or mutual funds
*(RESPONSE FRAME)
Appendix C: Survey questionnaires 261
1. Yes
2. No
98. (Don’t know)
99. (Refused)
*(ALL)
E3 Is someone other than yourself authorised to use your bank account or credit card (e.g. access
to PIN or internet banking details)?
IF NECESSARY: Please include any accounts that are in your name, including joint accounts.
Exclude persons who hold any joint accounts with you.
1. Yes
2. No
8. (Don’t know)
9. (Refused)
E4a To your knowledge, does your [DISPLAY E4 RESPONSE] keep any records of when they have
used your account?
INTERVIEWER NOTE: If more than one person in the category selected in E4 (e.g. two sons),
ask if either person has kept records.
1. Yes
2. No
98. (Don’t know)
99. (Refused)
*(ALL)
E5b Have you ever had discussions with someone in your family about making a will?
1. Yes
2. No
98. (Don’t know)
99. (Refused)
E5c Have you ever had discussions with someone in your family about your current will (e.g. change
your will)?
1. Yes
2. No
98. (Don’t know)
99. (Refused)
*(ALL)
E6e And is this power of attorney currently active, which means that it can be used anytime if needed?
(IF NECESSARY: a legal agreement between you and another person which enables another
person to make financial and/or medical decisions on your behalf)
1. Yes
2. No
98. (Don’t know)
99. (Refused)
E6b Have you ever had discussions with someone in your family about making an enduring power of
attorney?
1. Yes
2. No
98. (Don’t know)
99. (Refused)
99. (Refused)
*(ALL)
E7c Do you have an arrangement or agreement to provide financial support or benefit to someone
(e.g. transfer of assets, property or finance) in return for care and assistance?
1. Yes
2. No
98. (Don’t know)
99. (Refused)
*(ALL)
E8a Now some questions about your experiences in financial dealings with others.
If you are uncomfortable answering these questions, or you believe that answering may place
you or another person at risk, you can choose to skip them at any time.
(INTERVIEWER NOTE: This could include a family member, friend or professional caregiver, or
someone you depend on for other services (e.g. lawyer or financial professional))
In the past 12 months has someone you know …?
*(STATEMENTS)
*(ROTATE ALL EXCEPT STATEMENT H)
b) Deliberately prevented you from accessing or using your money, possessions or property
266 National Elder Abuse Prevalence Study: Final Report
g) Not contributed to household expenses such as rent or food, or aged care/home service
fees where this was previously agreed
E8X I’d like to ask you some questions about the time someone you know [INSERT E8A STATEMENT].
E8b How many times did this happen over the last 12 months? Was it …
(READ OUT)
1. Once
2. Few times
3. Many times
4. Every day or almost every day
98. (Don’t know)
99. (Refused)
E10
i. I’d like to ask a few more questions about this person, please let me know if this is the same
person you told me about who didn’t provide help with some daily activities when you needed.
I can skip the question. Just to check was this person …?
1. Male
2. Female
3. (Other)
99. (Refused)
*(E10 NOT 97, IDENTIFIED PERSON WHO AFFECTED THEM FINANCIALLY) (E10=97, GO TO E15)
E11 Did he/she live with you when this last happened (that is, the behaviours that affected you
financially)?
PROGRAMMER: IF E10=97. AUTOFILL RESPONSE TO D5.
1. Yes
2. No
98. (Don’t know)
99. (Refused)
E12 What age was he/she when this last happened (that is, the behaviours that affected you
financially)?
PROGRAMMER: IF E10=97. AUTOFILL RESPONSE TO D3.
1. SPECIFY (____, WHOLE NUMBERS, RANGE: 15 TO 100)
998. (Don’t know)
999. (Refused)
E12a Could you tell me which of the following age groups he/she was in when this last happened
(that is, the behaviours that affected you financially)?
PROGRAMMER: IF E10=97. AUTOFILL RESPONSE TO D3a.
1. Under 18
2. 18 – 24
3. 25 – 34
4. 35 – 44
5. 45 – 54
6. 55 – 64
7. 65 – 74
8. 75 and over
Appendix C: Survey questionnaires 269
E13 Was he/she employed when this last happened (that is, the behaviours that affected you
financially)?
PROGRAMMER: IF E10=97. AUTOFILL RESPONSE TO D5a.
3. Yes
4. No
98. (Don’t know)
99. (Refused)
E14 To the best of your knowledge, when this last happened did he/she have problems with …?
PROGRAMMER: IF E10=97. AUTOFILL RESPONSE TO D6.
(READ OUT ONE BY ONE)
*(ACCEPT MULTIPLES)
1. Alcohol
2. Drugs
3. Gambling
4. Mental health problems
5. Physical health problems
6. Financial problems
7. Other (please specify)
8. (None of these) *(SINGLE RESPONSE)
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
E15 How serious was this (that is, the behaviours that affected you financially) for you? Was it …
(READ OUT)
1. Very serious
2. Somewhat/moderately serious
3. Not serious
98. (Don’t know)
99. (Refused)
E18 Did you take any of the following actions to stop this from happening again?
(READ OUT ONE BY ONE)
*(ACCEPT MULTIPLES)
1. I spoke to the person causing the abuse.
2. A family member or friend spoke to the person on my behalf.
3. A professional (social worker, doctor, nurse) spoke to the person on my behalf.
4. I sought mediation or counselling.
5. I sought out a legal advice service.
6. I broke contact with or avoided the person involved.
7. I withdrew or stopped going out from social life in general.
8. A restraining order or safety order was made against the person.
9. Other (please specify)
10. Nothing *(SINGLE RESPONSE)
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
E20 Financial abuse happens when someone has taken control over or prevented you from
accessing your money, possessions or property against your will. Since turning 65 do you feel
this has happened to you?
1. Yes
2. No
98. (Don’t know)
99. (Refused)
E21 How many times do you feel that happened since turning 65? Was it …
(READ OUT)
1. Once
2. A few times
3. Many times
98. (Don’t know)
99. (Refused)
*(PROGRAMMER NOTE: COMPUTE DUMMY VARIABLE)
PERPETRATOR_FIN
1. E9A=1 TO 20, STATED PERPETRATOR OF FINANCIAL ABUSE AND E10 NOT 97
2. ALL OTHERS
7. E10=97
*(PROGRAMMER NOTE: COMPUTE DUMMY VARIABLE)
PERPETRATOR_COUNT_FIN
2. PERPETRATOR_NEGLECT=1 AND PERPETRATOR_FIN=1
1. PERPETRATOR_NEGLECT=1 AND PERPETRATOR_FIN=2 (OR PERPETRATOR_FIN=7)
0. PERPETRATOR_NEGLECT=2 AND PERPETRATOR_FIN=2
*(TIMESTAMP)
*(STATE=NT OR NSW)
*(TIMESTAMP)
272 National Elder Abuse Prevalence Study: Final Report
*(ALL)
F1a Now I’d like to ask you some questions about the ways you have been treated by people.
If you are not comfortable answering these questions, or you believe that answering may place
you or another person at risk, you can choose to skip them at any time.
(INTERVIEWER NOTE: This could include a family member, friend or professional caregiver, or
someone you depend on for other services (e.g. lawyer or financial professional).)
In the past 12 months, has someone you know …?
*(STATEMENTS)
*(ROTATE ALL EXCEPT STATEMENT G)
a) Insulted you, called you names or swore at you in a way you found offensive or aggressive
d) Threatened to harm others that you care about (e.g. pets, children, relatives, friends)
f) Prevented you from seeing or contacting family members (e.g. grandchildren) or your
doctor/nurse
g) Done anything else to cause you emotional distress
*(RESPONSE FRAME)
1. Yes
2. No
98. (Don’t know)
99. (Refused)
*(PROGRAMMER NOTE: FOR EACH STATEMENT WHERE F1A=1, SHOW F1x to F1b)
F1x I’d like to ask you some questions about the time someone you know [INSERT F1A STATEMENT].
F1b How many times did this happen over the last 12 months?
*(RESPONSE FRAME)
(READ OUT)
1. Once
2. Few times
3. Many times
4. Every day or almost every day
98. (Don’t know)
99. (Refused)
*(ACCEPT MULTIPLES)
1. Partner / spouse
2. Ex-partner / spouse
3. Parent
4. Step-parent
5. Parent in-law
6. Grandson(s) / daughter(s)
7. Aunt / uncle
8. Brother / sister
9. Brother / sister (in-law)
10. Son / daughter (biological or adopted)
11. Step-son / daughter
12. Son / daughter (in-law)
13. Other family member
14. Friend
15. Neighbour
16. Professional carer
17. Medical professional (e.g. doctor, nurse)
18. Financial professional (e.g. account, financial advisor, bank employee)
19. Other professional (e.g. lawyer)
20. Other (specify)
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
F2b I have already asked you questions about someone you know who may have affected you in the
past. Could I confirm, is this the same person who has affected you in the past?
*(PROGRAMMER NOTE: IF PERPETRATOR_COUNT_FIN=2, DISPLAY NEGLECT AND
FINANCIAL ABUSE AND RELATIONSHIP TO PERSON. ELSE IF PERPETRATOR_NEGLECT=1
DISPLAY NEGLECT AND D2 RESPONSE. ELSE IF PERPETRATOR_FIN=1, SHOW FINANCIAL
ABUSE AND E9A RESPONSE)
99. (Refused)
F4 Did he/she live with you when this last happened, (that is, did those behaviours towards you)?
1. Yes
2. No
98. (Don’t know)
99. (Refused)
F5 What age was he/she when this last happened (that is, did those behaviours towards you)?
1. SPECIFY (____, WHOLE NUMBERS, RANGE: 15 TO 100)
998. (Don’t know)
999. (Refused)
F5a Could you tell me which of the following age groups he/she was in when this last happened
(that is, did those behaviours towards you)?
1. Under 18
2. 18 – 24
3. 25 – 34
4. 35 – 44
5. 45 – 54
6. 55 – 64
7. 65 – 74
8. 75 and over
98. (Don’t know)
99. (Refused)
F6 Was he/she employed when this last happened (that is, did those behaviours towards you)?
1. Yes
2. No
98. (Don’t know)
99. (Refused)
F7 To the best of your knowledge, when this last happened (that is, did those behaviours towards
you), did he/she have problems with …?
(READ OUT ONE BY ONE)
*(ACCEPT MULTIPLES)
1. Alcohol
2. Drugs
3. Gambling
4. Mental health problems
5. Physical health problems
6. Financial problems
7. Other (specify)
8. (None of these) *(SINGLE RESPONSE)
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
F8 How serious was this (that is, those behaviours) for you?
1. Very serious
2. Somewhat/moderately serious
3. Not serious
98. (Don’t know)
99. (Refused)
F11 Did you take any of the following actions to stop this from happening again?
(READ OUT ONE BY ONE)
*(ACCEPT MULTIPLES)
1. I spoke to the person causing the abuse.
2. A family member or friend spoke to the person on my behalf.
3. A professional (social worker, doctor, nurse) spoke to the person on my behalf.
4. I sought mediation or counselling.
5. I sought out a legal advice service.
6. I broke contact with or avoided the person involved.
7. I withdrew or stopped going out from social life in general.
8. A restraining order or safety order was made against the person.
9. Other (please specify)
10. Nothing *(SINGLE RESPONSE)
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
2. No
98. (Don’t know)
99. (Refused)
*(F12=1 OR 2)(ACTION EFFECTIVE/INEFFECTIVE)
F13 Sometimes actions such as these can result in emotional harm or psychological abuse. Since
turning 65 do you feel you have experienced psychological abuse?
1. Yes
2. No
98. (Don’t know)
99. (Refused)
F14 How many times do you feel that happened since turning 65? Was it …
(READ OUT)
1. Once
2. A few times
3. Many times
98. (Don’t know)
99. (Refused)
*(PROGRAMMER NOTE: COMPUTE DUMMY VARIABLE)
PERPETRATOR_PSY
1. F2B=3 OR (F2A=1-20 AND PERPETRATOR_COUNT_FIN=0) Diff person to previously mentioned
2. ALL OTHERS
7. F2B=1 TO 2 – same as neglect OR financial
*(PROGRAMMER NOTE: COMPUTE DUMMY VARIABLE)
PERPETRATOR_COUNT_PSY
3. PERPETRATOR_COUNT_FIN=2 AND PERPETRATOR_PSY=1
*(TIMESTAMP)
*(STATE=NT OR NSW)
278 National Elder Abuse Prevalence Study: Final Report
*(ALL)
G1a Now I’d like to ask you about some behaviours that may cause physical harm.
If you are not comfortable answering these questions, or believe that answering may place you
or another person at risk, you can choose to skip them at any time.
In the past 12 months, has someone you know …?
*(STATEMENTS)
*(ROTATE ALL EXCEPT STATEMENT H)
a) Tied or held you down, or restrained you in any other way, or locked you up
b) Grabbed, pushed or shoved you
c) Hit, punched, kicked or slapped you
d) Threatened you with a weapon (e.g. knife, gun or any other objects)
e) Injured you with a weapon (e.g. knife, gun or any other objects)
f) Threatened to harm you in any way
g) Given you drugs or too much medicine in order to control you / make you docile
h) Hurt you physically in any other way
*(RESPONSE FRAME)
1. Yes
2. No
98. (Don’t know)
99. (Refused)
*(PROGRAMMER NOTE: FOR EACH STATEMENT WHERE G1A=1, SHOW G1x TO G1b)
G1x I’d like to ask you some questions about the time someone you know has [INSERT G1a
STATEMENT].
G1b How many times did this happen over the last 12 months? Was it …?
(READ OUT)
1. Once
2. Few times
3. Many times
4. Every day or almost every day
98. (Don’t know)
Appendix C: Survey questionnaires 279
99. (Refused)
G2a Who has affected you the most (in terms of physical harm)?
*(PROGRAMMER NOTE: DISPLAY SELECTIONS FROM G2 ONLY)
*(PROGRAMMER NOTE: IF ONLY 1 PERSON SELECTED AT G2, AUTOCODE TO G2A)
(READ OUT)
1. Partner / spouse
2. Ex-partner / spouse
3. Parent
4. Step-parent
5. Parent in-law
6. Grandson(s) / daughter(s)
7. Aunt / uncle
8. Brother / sister
9. Brother / sister (in-law)
280 National Elder Abuse Prevalence Study: Final Report
G2b I have already asked you questions about someone you know who may have affected you in the
past. Could I confirm, is this the same person who has affected you in the past?
*(PROGRAMMER NOTE: IF PERPETRATOR_COUNT_PSY=3, DISPLAY NEGLECT, FINANCIAL
AND PSYCHOLOGICAL ABUSE AND RELATIONSHIP TO PERSON. ELSE DISPLAY ALL OTHER
TYPES OF ABUSE AND RELATIONSHIP TO PERSON AS APPROPRIATE)
G4 Did he/she live with you when this last happened, (that is, he/she had the last violent episode
to you)?
1. Yes
2. No
98. (Don’t know)
99. (Refused)
G5 What age was he/she when this last happened (that is, he/she had the last violent episode
to you)?
1. SPECIFY (____, WHOLE NUMBERS, RANGE: 15 TO 100)
998. (Don’t know)
999. (Refused)
G5a Could you tell me which of the following age groups he/she was in when this last happened
(that is, he/she had the last violent episode to you)?
1. Under 18
2. 18 – 24
3. 25 – 34
4. 35 – 44
5. 45 – 54
6. 55 – 64
7. 65 – 74
8. 75 and over
98. (Don’t know)
99. (Refused)
G6 Was he/she employed when this last happened (that is, he/she had the last violent episode
to you)?
1. Yes
2. No
98. (Don’t know)
99. (Refused)
G7 To the best of your knowledge, when this last happened (that is, he/she had the last violent
episode to you), did he/she have problems with …?
(READ OUT ONE BY ONE)
*(ACCEPT MULTIPLES)
1. Alcohol
2. Drugs
3. Gambling
4. Mental health problems
5. Physical health problems
6. Financial problems
7. Other (specify)
8. (None of these) *(SINGLE RESPONSE)
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
G8 How serious was this (that is, the last violent episode he/she had to you) for you?
1. Very serious
2. Somewhat/moderately serious
3. Not serious
98. (Don’t know)
99. (Refused)
G11 Did you take any of the following actions to stop this from happening again?
(READ OUT ONE BY ONE)
*(ACCEPT MULTIPLES)
1. I spoke to the person causing the abuse.
2. A family member or friend spoke to the person on my behalf.
3. A professional (social worker, doctor, nurse) spoke to the person on my behalf.
4. I sought mediation or counselling.
5. I sought out a legal advice service.
6. I broke contact with or avoided the person involved.
7. I withdrew or stopped going out from social life in general.
8. A restraining order or safety order was made against the person.
9. Other (please specify)
10. Nothing *(SINGLE RESPONSE)
Appendix C: Survey questionnaires 283
G13 Physical abuse happens when a person intentionally causes bodily injury to another person.
Since turning 65 do you feel you experienced physical abuse?
1. Yes
2. No
98. (Don’t know)
99. (Refused)
G14 How many times do you feel that happened since turning 65? Was it …
(READ OUT)
1. Once
2. A few times
3. Many times
98. (Don’t know)
99. (Refused)
PERPETRATOR_PHY
PERPETRATOR_COUNT_PHY
*(TIMESTAMP)
*(STATE=NT OR NSW)
*(TIMESTAMP)
*(ALL)
H1a Now I’m going to ask some questions about any unwanted sexual experiences you may have
had in the last 12 months.
If you are not comfortable answering these questions or believe that answering may place you
or another person at risk, you can skip them at any time.
If you would prefer to speak to a <male / female> interviewer, I can arrange a call back at a time
that’s convenient to you.
In the past 12 months has someone you know tried to or has … ?
STATEMENTS
a) Touched you in a sexual way against your will
b) Made you watch pornography against your will
c) Forced you to engage in sexual acts against your will
d) Talked to you in a sexual way when you did not want to
e) Tried to engage you in any other unwanted sexual experience/s
RESPONSE FRAME
1. Yes
2. No
98. (Don’t know)
99. (Refused)
*(PROGRAMMER NOTE: FOR EACH STATEMENT WHERE H1A=1, SHOW H1x TO H1b)
H1x I’d like to ask you some questions about the time someone you know [INSERT H1a STATEMENT].
H1b How many times did this happen over the last 12 months? Was it …
(READ OUT)
1. Once
2. Few times
3. Many times
4. Every day or almost every day
98. (Don’t know)
99. (Refused)
(READ OUT)
1. Partner / spouse
2. Ex-partner / spouse
3. Parent
4. Step-parent
5. Parent in-law
6. Grandson(s) / daughter(s)
7. Aunt / uncle
8. Brother / sister
9. Brother / sister (in-law)
10. Son / daughter (biological or adopted)
11. Step-son / daughter
12. Son / daughter (in-law)
13. Other family member
14. Friend
15. Neighbour
16. Professional carer
17. Medical professional (e.g. doctor, nurse)
18. Financial professional (e.g. account, financial advisor, bank employee)
19. Other professional (e.g. lawyer)
20. Other (specify)
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
H2b I have already asked you questions about someone you know who may have affected you in the
past. Could I confirm, is this the same person who has affected you in the past?
*(PROGRAMMER NOTE: IF PERPETRATOR_COUNT_PHY=4, DISPLAY NEGLECT, FINANCIAL,
PSYCHOLOGICAL AND PHYSICAL ABUSE AND RELATIONSHIP TO PERSON. ELSE DISPLAY
ALL OTHER TYPES OF ABUSE AND RELATIONSHIP TO PERSON AS APPROPRIATE)
3. (Other)
98. (Don’t know)
99. (Refused)
H4 Did he/she live with you when this last happened (that is, he/she made unwanted sexual
advances)?
1. Yes
2. No
98. (Don’t know)
99. (Refused)
H5 What age was he/she when this last happened (that is, he/she made unwanted sexual advances)?
1. SPECIFY (____, WHOLE NUMBERS, RANGE: 15 TO 100)
998. (Don’t know)
999. (Refused)
H5a Could you tell me which of the following age groups he/she was in when this last happened
(that is, he/she made unwanted sexual advances)?
1. Under 18
2. 18 – 24
3. 25 – 34
4. 35 – 44
5. 45 – 54
6. 55 – 64
7. 65 – 74
8. 75 and over
98. (Don’t know)
99. (Refused)
H6 Was he/she employed when this last happened (that is, he/she made unwanted sexual
advances)?
1. Yes
2. No
98. (Don’t know)
99. (Refused)
H7 To the best of your knowledge, when this last happened (that is, he/she made unwanted sexual
advances), did he/she have problems with …?
(READ OUT ONE BY ONE)
*(ACCEPT MULTIPLES)
288 National Elder Abuse Prevalence Study: Final Report
1. Alcohol
2. Drugs
3. Gambling
4. Mental health problems
5. Physical health problems
6. Financial problems
7. Other (specify)
8. (None of these) *(SINGLE RESPONSE)
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
H8 How serious was this (that is, when he/she made unwanted sexual advances) for you?
1. Very serious
2. Somewhat/moderately serious
3. Not serious
98. (Don’t know)
99. (Refused)
H11 Did you take any of the following actions to stop this from happening again?
(READ OUT ONE BY ONE)
*(ACCEPT MULTIPLES)
1. I spoke to the person causing the abuse.
2. A family member or friend spoke to the person on my behalf.
3. A professional (social worker, doctor, nurse) spoke to the person on my behalf.
4. I sought mediation or counselling.
5. I sought out a legal advice service.
6. I broke contact with or avoided the person involved.
7. I withdrew or stopped going out from social life in general.
8. A restraining order or safety order was made against the person.
9. Other (please specify)
10. Nothing *(SINGLE RESPONSE)
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
H13 Sexual abuse happens when a person forces undesired sexual behaviour upon you against your
will. Since turning 65 do you feel you experienced this type of behaviour?
1. Yes
2. No
98. (Don’t know)
99. (Refused)
H14 How many times do you feel that happened since turning 65? Was it …
(READ OUT)
1. Once
2. A few times
290 National Elder Abuse Prevalence Study: Final Report
3. Many times
98. (Don’t know)
99. (Refused)
PERPETRATOR_SEX
PERPETRATOR_COUNT_SEX
*(TIMESTAMP)
*(STATE=NT OR NSW)
*(TIMESTAMP)
*(ALL)(HIDDEN)
I1 I am going to ask you some questions about the ways you have been treated by people you know.
If you are not comfortable answering these questions, or believe answering may place you or
another person at risk, you can choose to skip them at any time.
(INTERVIEWER NOTE: This could include family members, friends or professional caregiver, or
someone you depend on for other services (e.g. lawyer, financial or other professionals).
In the past 12 months, has someone you know …?
(STATEMENTS)
a) Not respected you when talking to you because of your culture, race or ethnicity
b) Mis-translated between English and your preferred language on purpose, when talking about
financial or legal documents
d) Made you feel that you are just free labour (e.g. doing most of household work, or providing
child care, etc.)
e) Limited or restricted your contact with your friends or others from the same cultural
background, including friends and relatives overseas
g) Limited or restricted your access to culturally familiar activities (e.g. attending certain events,
watching or listening to shows in your preferred language)
(RESPONSE FRAME)
1. Yes
2. No
3. (Don’t Know)
4. (Refused)
*(PROGRAMMER NOTE: FOR EACH STATEMENT WHERE I1=1, SHOW I1x to I2)
I1x I’d like to ask you some questions about the time someone you know [INSERT I1 STATEMENT].
4. Step-parent
5. Parent in-law
6. Grandson(s) / daughter(s)
7. Aunt / uncle
8. Brother / sister
9. Brother / sister (in-law)
10. Son / daughter (biological or adopted)
11. Step-son / daughter
12. Son / daughter (in-law)
13. Other family member
14. Friend
15. Neighbour
16. Professional carer
17. Medical professional (e.g. doctor, nurse)
18. Financial professional (e.g. account, financial advisor, bank employee)
19. Other professional (e.g. lawyer)
20. Other (specify)
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
I4 I have already asked you questions about someone you know who may have affected you in the
past. Could I confirm, is this the same person who has affected you in the past?
*(PROGRAMMER NOTE: IF PERPETRATOR_COUNT_SEX=5, DISPLAY NEGLECT, FINANCIAL,
PSYCHOLOGICAL, PHYSICAL AND SEXUAL ABUSE AND RELATIONSHIP TO PERSON. ELSE
DISPLAY ALL OTHER TYPES OF ABUSE AND RELATIONSHIP TO PERSON AS APPROPRIATE)
1. [DISPLAY IF PERPETRATOR_NEGLECT=1: Yes, same person for neglect]
2. [DISPLAY IF PERPETRATOR_FIN=1: Yes, same person for financial abuse]
3. [DISPLAY IF PERPETRATOR_PSY=1: Yes, same person for psychological abuse]
4. [DISPLAY IF PERPETRATOR_PHY=1: Yes, same person for physical abuse]
5. [DISPLAY IF PERPETRATOR_SEX=1: Yes, same person for sexual abuse]
6. No, different person
98. (Don’t know)
99. (Refused)
I5 How serious was this (that is, those behaviours ) for you?
1. Very serious
2. Somewhat/moderately serious
3. Not serious
98. (Don’t know)
99. (Refused)
DEMOGRAPHICS – Part 2
*(ALL)
*(ALL)
*(ALL)
*(ALL)
*(ALL)
INTERVIEWER NOTE: If completed Year 12 or less, confirm if completed any post-school TAFE,
trade or certificates.
(READ OUT IF REQUIRED)
1. Year 9 or below
2. Year 10, form 4, intermediate
3. Year 11, form 5, leaving
4. Year 12, form 6, matriculation, HSC
5. Trade/apprenticeship
6. Certificate (business college, TAFE)
7. Diploma (business college, TAFE)
8. Degree (bachelor)
9. Post-graduate (PhD, masters, post-grad dip.)
10. Other (Specify_____)
98. (Don’t know)
99. (Refused)
*(ALL)
13. Judaism
14. Sikhism
11. Other
98. (Don’t know)
99. (Refused)
*(ALL)
A16 Thinking about your own income. Do you currently receive income from any of the following
sources?
(READ OUT, WAIT FOR YES/NO BEFORE PROCEEDING TO NEXT CODE)
*(ACCEPT MULTIPLES)
1. Wage or salary (paid job)
2. Your own business or a share in partnership
3. Any government pension, benefits or allowance
4. Superannuation
5. Income from investment property (e.g. rental income)
6. Income from other investments (e.g. shares, funds)
7. Any other regular source
8. (No income) *(SINGLE RESPONSE)
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
A17 Before tax or other deductions, what is your annual household income?
Please include wages and salaries, government pensions, benefits and allowances and income
from interest, dividends, or other sources.
INTERVIEWER NOTE: Seeking estimate only – especially if unsure of income of other household
members.
1. $1 – $7,799 per year ($1 – $149 per week)
2. $7,800 – $15,599 per year ($150 – $299 per week)
3. $15,600 – $20,799 per year ($300 – $399 per week)
4. $20,800 – $25,999 per year ($400 – $499 per week)
5. $26,000 – $33,799 per year ($500 – $649 per week)
6. $33,800 – $41,599 per year ($650 – $799 per week)
7. $41,600 – $51,999 per year ($800 – $999 per week)
8. $52,000 – $64,999 per year ($1,000 – $1,249 per week)
9. $65,000 – $77,999 per year ($1,250 – $1,499 per week)
10. $78,000 – $90,999 per year ($1,500 – $1,749 per week)
11. $91,000 – $103,999 per year ($1,750 – $1,999 per week)
12. $104,000 – $155,999 per year ($2,000 – $2,999 per week)
13. $156,000 – $181,999 per year ($3,000 – $3,499 per week)
14. $182,000 – $207,999 per year ($3,500 – $3,999 per week)
15. $208,000 or more per year ($4,000 or more per week)
16. Nil income
17. Negative income
296 National Elder Abuse Prevalence Study: Final Report
*(ALL)
A19a What is the name of the suburb or town where you live?
INT NOTE: In order to analyse results, we need to ask where people live.
SELECT LOCALITY FROM LOCALITIES SPECIFIC TO SELECTED POSTCODE
INCLUDE SPECIFIED OTHER OPTION
IF DK OR REF – USE POSTCODE FROM SAMPLE
*(SAMTYP=2, MOBILE)
DEM20 Now just a question or two about your use of telephone services.
Is there at least one working fixed line (landline) telephone inside your home that is used for
making and receiving calls?
1. Yes
2. No
98. (Don’t know)
99. (Refused)
Appendix C: Survey questionnaires 297
*(ALL)
DEM20b How many mobile phones, in total, do you have that you receive calls on?
1. Number of mobile phone numbers given*(DISPLAY UNLIKELY RESPONSE IF >5) (IF
SAMPLETYPE=LANDLINE: RANGE 0 TO 9; IF SAMPLETYPE=MOBILE: RANGE 1 TO 9)
88888 Don’t know
99999 Refused
*(TIMESTAMP)
CLOSE
*(ALL)
IF YES
y If you have any concerns about potential or actual elder abuse, please contact 1800
ELDERHelp (1800 353 374) to be redirected to the existing phone line service in your
jurisdiction (free call).
y If you or someone you know have experienced violence or sexual assault and require
immediate or ongoing assistance, contact 1800 RESPECT (1800 737 732) to talk to a
counsellor from the National Sexual Assault and Domestic Violence hotline.
y For confidential support and information, contact Safe Steps’ 24/7 family violence response
line on 1800 015 188 or the Men’s Referral Service on 1300 766 491.
y If you are concerned for your safety or that of someone else, please contact the police in your
local area, or call 000 for emergency assistance.
y For a confidential discussion with an experienced counsellor you can call Lifeline on 13 11 14.
IF NO This completes the survey. From the Social Research Centre and the Australian Institute of
Family Studies, thank you very much for your time and assistance. Your co-operation is greatly
appreciated.
298 National Elder Abuse Prevalence Study: Final Report
MANDATORY REPORTING
*(STATE=NT OR NSW)
MAND [DISPLAY IF STATE=NT] Did the respondent mention anything not elsewhere captured about any
domestic violence (neglect, emotional/psychological, physical, sexual or financial abuse) that
they may have experienced from someone they live with or who cares for them in their home?
1. Yes (Specify)
2. No (Continue
[DISPLAY IF STATE=NSW] Did the respondent mention anything not elsewhere captured that
may constitute a criminal offence?
1. Yes (Specify)
2. No (Continue)
INT_MAND
MAND_REP
TERMINATION SCRIPTS
TERM1 Thank you for your time but we need to speak with people aged 65 and over.
TERM4 Thank you for your time, unfortunately we need to speak to people that are not currently in
aged or respite care.
ANSM1.Good morning/afternoon/evening. My name is <SAY NAME> calling on behalf of the Australian Institute of
Family Studies from the Social Research Centre. We are conducting an important national study about attitudes
towards older people in Australia and how they are treated by others. If you would like to take part in this study,
please call our hotline number: 1800 083 037 and we will call you back at a time that is convenient to you. Thank
you and we look forward to hearing from you.
ANSM2.Good morning/afternoon/evening. My name is <SAY NAME> calling on behalf of the Australian Institute
of Family Studies from the Social Research Centre. We left a message recently on your answering machine/voice
mail regarding an important national study about attitudes towards older people in Australia and how they are
treated by others. If you would like to take part in this study, please call our hotline number: 1800 083 037 and
we will call you back at a time that is convenient to you. Thank you and we look forward to hearing from you.
WELCOME SCREEN
Good (morning/afternoon/evening). My name is (...). I’m calling on behalf of the Australian Institute of Family
Studies from the Social Research Centre.
We are conducting an important national study with people aged between 18 to 64 years about attitudes
towards older people in Australia and how they are treated by others.
This research is being funded by the Australian Government Attorney-General’s Department. Results from the
study will be used to support government’s policy development to better address the needs of older people.
INTRODUCTION
*(ALL)
*(SAMPLETYPE=MOBILE)
INTRO2 For this survey, we are interested in talking to people aged 18 to 64, can I check, are you within
that age range?
REINTRODUCE IF NECESSARY: Good (morning/afternoon/evening). My name is (...). I’m calling
on behalf of the Australian Institute of Family Studies from the Social Research Centre. We are
300 National Elder Abuse Prevalence Study: Final Report
conducting a study about attitudes towards older people in Australia and how they are treated
by others. We are interested in speaking to people aged between 18 to 64 years.
IF NEEDED: This research is being funded by the Australian Government Attorney-General’s
Department. Results from the study will be used to support government’s policy development
to better address the needs of older people.
1. Yes – Continue
2. No – Under 18 years of age OR over 64 years of age (GO TO TERM1)
3. Language difficulty (GO TO PLOTE)
4. Queried about how telephone number was obtained (GO TO ATELQ)
5. Respondent refusal (GO TO RR1)
*(SAMPLETYPE=LANDLINE)
INTRO3 To help with this important study we’d like to arrange a short interview with the person aged
between 18 to 64 years in your household who is going to have the next birthday.
Would that be yourself or someone else?
REINTRODUCE IF NECESSARY: Good (morning/afternoon/evening). My name is (...). I’m calling
on behalf of the Australian Institute of Family Studies from the Social Research Centre. We are
conducting a study about attitudes towards older people in Australia and how they are treated
by others. We are interested in speaking to people aged between 18 to 64 years.
IF NEEDED: This research is being funded by the Australian Government Attorney-General’s
Department. Results from the study will be used to support government’s policy development
to better address the needs of older people.
1. Yes, me – Continue
2. No – Another household member is eligible (GO TO SELINTRO)
3. Phone answerer refused to pass over the selected respondent (ATTEMPT CONVERSION)
(GO TO RR1)
4. No – Nobody over 18 years of age OR over 64 years of age (GO TO TERM1)
5. Selected respondent does not speak English (GO TO PLOTE)
6. Queried about how telephone number was obtained (GO TO ATELQ)
7. Respondent refusal (GO TO RR1)
SELINTRO Good (morning/afternoon/evening). My name is (...). I’m calling on behalf of the Australian
Institute of Family Studies from the Social Research Centre. We are conducting a study about
older people in Australia and how they are treated by others. We are interested in speaking to
people aged between 18 and 64 years.
IF NEEDED: This research is being funded by the Australian Government Attorney-General’s
Department. Results from the study will be used to support government’s policy development
to better address the needs of older people.
1. Continue
2. Respondent refusal (GO TO RR1)
3. Selected respondent does not speak English (GO TO PLOTE)
4. Queried about how telephone number was obtained (GO TO ATELQ)
*(SAMPLETYPE=MOBILE)
SAFE May I just check whether or not it is safe for you to take this call at the moment. If not, I am
happy to call you back when it is more convenient for you.
1. Safe to take call
2. Not safe to take call
Appendix C: Survey questionnaires 301
MOB_APPT Do you want me to call you back on this number or would you prefer I call back on another
phone?
1. This number (STOP, MAKE APPOINTMENT)
2. Other number (STOP, MAKE APPOINTMENT, RECORD OTHER PHONE NUMBER)
3. Respondent refusal (GO TO RR1)
*(SAMPLETYPE=MOBILE)
MOB_APPT_A Just so I know your time zone, can you tell me which state you’re in?
1. NSW
2. VIC
3. QLD
4. SA
5. WA
6. TAS
7. NT
8. ACT
9. (Refused) (GO TO TERM2)
*CONSENT
*(ALL)
CONSENT This survey will take about 15 minutes depending on your answers.
Participation is voluntary. You can finish the interview at another time or stop if you choose.
Some of the questions may seem quite sensitive. Please let me know if there are any you would
prefer not to answer and I will skip over them.
Any information you provide is protected by Commonwealth privacy laws and our Privacy
Policy, which is available on our website. Your information will be kept confidential, as far as
allowed by law. However, if you tell us about unreported abuse or neglect or a risk of harm to
yourself or other people, we may be required by law to report this to the relevant authorities.
The data we collect will be given to the Australian Institute of Family Studies who will use it to
write a research report for the Australian Government Attorney-General’s Department. We will
remove any information from the data that can identify you or your individual responses. If you
wish to withdraw from the study, you can do so up until we remove any identifiable information
from the dataset.
After this project is completed, these data will be kept at the Australian Data Archive at the
Australian National University and may be used for future research.
If you have any questions about the information I have provided, just let me know, if not, are you
happy to begin the survey now?
IF RESPONDENT IS SUSPICIOUS OR DOUBTFUL: If you want to verify that the survey is
legitimate, or if you would like more information or to review our privacy policy, you can call the
Social Research Centre’s 1800 number (1800 083 037) during business hours, or you can check
our website at www.srcentre.com.au
IF RESPONDENT NEEDS MORE ASSURANCE1: Or you could check the project information on
our website www.srcentre.com.au/generalcommunity
INTERVIEWER NOTE: If you have concerns respondent is not able to answer freely arrange
call back.
IF RESPONDENT WANTS MORE INFORMATION ABOUT USE OF PERSONAL INFORMATION/
ARCHIVING (INTERVIEWER TO PROVIDE INFORMATION AS REQUIRED): After you complete
a telephone interview, the data will be anonymised, which means that no one will be able to
identify you from your answers. Any personal information held by the Social Research Centre,
such as your telephone number, will be destroyed after the project is completed.
After this project is completed, the anonymised data will be securely kept at the Australian Data
Archive at the Australian National University. This means that other research organisations or
individuals may be able to access the data for future research after this project. However, no one
will be able to identify you or your individual responses.
Further information about how personal information is handled, including how to make a
privacy complaint or request access to your information, can be found in our Privacy Policy,
which is available on our website at [www.srcentre.com.au].
1. Yes, continue (GO TO MON)
2. Not a convenient time (MAKE APPOINTMENT)
3. Refused (GO TO RR1)
ATELQ [DISPLAY IF SAMPLETYPE=LANDLINE] Your telephone number has been chosen at random
from all possible telephone numbers in your area. We find that this is the best way to obtain a
representative sample of people across Australia.
[DISPLAY IF SAMPLETYPE=MOBILE] Your number was randomly generated by a computer.
We’re calling mobile phones as well as landlines because lots of people have mobile phones
but don’t have landlines. So, we call mobile phones as well as landlines so we can get a
representative sample of people across Australia.
Appendix C: Survey questionnaires 303
*(ALL)
MON Thank you. This call may be monitored or recorded for quality assurance purposes. Is that OK?
1. Yes
2. No
[PROGRAMMER NOTE: DISPLAY ALL CODE FRAME CHRONOLOGICALLY TO INTERVIEWERS.
CODE FRAME NUMBERS IN THIS DOC ARE FOR DATA]
*(TIMESTAMP)
DEMOGRAPHICS – Part 1
*(ALL)
PREDEM1 Firstly, I’d like to start by asking some questions about you.
*(ALL)
A1a Could you please tell me which of the following age groups you are in?
(READ OUT)
1. Under 18 (GO TO TERM1)
2. 18 – 29 years
3. 30 – 39 years
4. 40 – 49 years
5. 50 – 59 years
6. 60 – 64 years
7. Over 65 (GO TO TERM1)
99. (Refused) (GO TO TERM1)
*(ALL)
DEM16am What is the name of the suburb or town where you live?
IF NEEDED: In order to make sure the study covers people living in all areas, we need to ask
where people live.
SELECT LOCALITY FROM LOCALITIES SPECIFIC TO SELECTED POSTCODE
INCLUDE SPECIFIED OTHER OPTION
88888 Don’t know
99999 Refused
*PROGRAMMER NOTE: IF DEM16am IS DK OR REF GO TO TERM 13
*(ALL)
*(ALL)
*(ALL)
2. Separated
3. Divorced
4. Widowed
5. Never married
98. (Don’t know)
99. (Refused)
*(ALL)
A9a In what year did you first arrive in Australia to live here for one or more years?
INTERVIEWER NOTE: If lived in Australia on more than one occasion, record the year they first
arrived to live for one year or more.
1. Record year (RANGE 1918 – 2019)
98. (Don’t know)
99. (Refused)
*(ALL)
*(TIMESTAMP)
306 National Elder Abuse Prevalence Study: Final Report
*(ALL)
B1 I’m going to read out some statements and I would like you to tell me how much you agree or
disagree with each one, on a scale from 0 to 10, where 0 is strongly disagree and 10 is strongly
agree.
There are no right or wrong answers, only opinions.
*PROGRAMMER DISPLAY ON SCREEN WITH G) INTERVIEWER NOTE: Ageist means showing
prejudice against older people.
*(STATEMENTS)
*(ROTATE)
a) It is good to tell older people that they are too old to do certain things; otherwise they might
get their feelings hurt when they eventually fail.
b) Even if they want to, older people shouldn’t be allowed to work because they have already
paid their debt to society.
e) It is helpful to repeat things to older people because they rarely understand the first time.
h) Older people are a drain on the health care system and the economy.
i) Parents should help their adult children financially if they need it.
j) Adult children should help their ageing parents financially if they need it.
k) Adult children should let their ageing parents live with them if they need to.
l) Parents should let their adult children live with them if they need to.
m) Other family members are entitled to some of the older person’s assets (IF NEEDED:
e.g. money, property, shares) if they have assisted the older person on a regular basis.
*(RESPONSE FRAME)
1. SPECIFY (_____, WHOLE NUMBERS, RANGE 0 TO 10)
98. (Don’t know)
99. (Refused)
*(TIMESTAMP)
*(ALL)
C1 I’d like to ask you some questions about helping family members or others who require
assistance with their finances due to old age or because of a disability, injury or illness.
Within the last 12 months have you assisted anyone with any of the following?
INTERVIEWER NOTE: If needed: Have you assisted anyone with this due to their old age,
disability, injury or illness in the last twelve months.
*(STATEMENTS)
*(ROTATE)
b) Paying bills, budgeting/accounting/monitoring finances
c) Taxes (e.g. tax returns)
d) Banking (e.g. withdrawing money from ATMs, using internet banking. etc.)
Appendix C: Survey questionnaires 307
C3a We are interested in the oldest person that you assisted in the last 12 months. Is that your …?
*(PROGRAMMER NOTE: DISPLAY SELECTIONS FROM C2 ONLY)
*(PROGRAMMER NOTE: IF ONLY 1 PERSON SELECTED AT C2, AUTOCODE TO C3A)
(READ OUT)
1. Partner / spouse
2. Ex-partner / spouse
3. Parent(s)
4. Parent(s) in-law
5. Grandparent(s)
6. Aunt / uncle
7. Brother / sister
8. Son / daughter
308 National Elder Abuse Prevalence Study: Final Report
C4 Thinking of the [IF MULTIPLE PERSONS SELECTED AT C2: oldest] person you assisted (in the
last 12 months), can you tell me how old this person is?
IF NEEDED: We are referring to your [DISPLAY C3a RESPONSE]
(PROBE TO FRAME)
1. Under 30
2. 30 – 44
3. 45 – 54
4. 55 – 64
5. 65 – 74
6. 75 – 80
7. 80 – 85
8. 85+ years
98. (Don’t know)
99. (Refused)
C5 Which of the following is the main reason why this person needed help?
INTERVIEWER NOTE: We’re referring to the types of help you mentioned you provided them.
IF NEEDED: We are referring to your [DISPLAY C3a RESPONSE]
(READ OUT)
1. Not interested in doing it themselves
2. Lacks confidence in doing it themselves
3. Due to dementia or confusion
4. Due to disability or poor health
5. Due to being old and frail
6. Because English is not their first language
7. Because of difficulty with reading or writing
8. Other (specify)
98. (Don’t know)
99. (Refused)
C6 We’re interested in the types of assistance you provided this person. In the last 12 months, did
you …?
IF NEEDED: We are referring to your [DISPLAY C3a RESPONSE]
*(STATEMENTS)
Appendix C: Survey questionnaires 309
a) Assist them with electronic transactions such as internet banking, telephone banking
b) Help fill in a financial form or prepare a financial document and then get him/her to sign it
c) Receive authorisation to operate their bank account or credit card through an arrangement
with the bank
f) Pay for something with your money first and were subsequently reimbursed
g) Need to make use of an existing power of attorney granted to you by that person
(IF NECESSARY: a legal agreement between you and another person which enables you
to make financial and/or property decisions on behalf of the person)
C7 Do you regularly keep any records for what you are doing for this person?
IF NEEDED: We are referring to your [DISPLAY C3a RESPONSE]
1. Yes
2. No
98. (Don’t know)
99. (Refused)
C8 Do you have any arrangements for involving or informing this person or someone close to this
person of the assistance you have been providing them?
IF NEEDED: We are referring to your [DISPLAY C3a RESPONSE]
1. Yes
2. No
310 National Elder Abuse Prevalence Study: Final Report
C8a What are the arrangements for involving or informing this person or someone close to this
person of what you are doing?
(READ OUT ONE BY ONE)
*(ACCEPT MULTIPLES)
1. Answer their questions if requested
2. Share my records if requested
3. Give regular verbal updates
4. Give regular written updates without records
5. Ask for their advice on key issues
6. Include them in decision-making for key issues
7. Other
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
*(TIMESTAMP)
ADVANCED PLANNING
*(ALL)
*(ALL)
C9 Have you been appointed by someone to make decisions under an existing power of attorney?
1. Yes
2. No
98. (Don’t know)
99. (Refused)
C9x Have you ever had discussions with someone in your family to be appointed under a power of
attorney?
1. Yes
2. No
98. (Don’t know)
99. (Refused)
C9a What type of power of attorney have you been appointed under?
(READ OUT)
1. Financial
2. Medical
3. Both
Appendix C: Survey questionnaires 311
C11 Who is the person who has granted you a power of attorney?
INTERVIEWER NOTE: PROBE TO FRAME
*(ACCEPT MULTIPLES)
1. Partner / spouse
2. Ex-partner / spouse
3. Parent(s)
4. Parent(s) in-law
5. Grandparent(s)
6. Aunt / uncle
7. Brother / sister
8. Son / daughter
9. Son / daughter (in-law)
10. Other family member
11. Friend
12. Neighbour
13. Other (specify)
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
*(ALL)
C13 Do you have an arrangement or agreement to provide care to an older person in return for
financial support or benefit (e.g. transfer of assets, property or finances) or bequest in a will?
1. Yes
312 National Elder Abuse Prevalence Study: Final Report
2. No
98. (Don’t know)
99. (Refused)
*(TIMESTAMP)
*(TIMESTAMP)
*(ALL)
D2 I am going to read out some statements about the abuse of older people. For each one please
tell me whether you strongly agree, somewhat agree, neither agree nor disagree, somewhat
disagree or strongly disagree.
(IF NECESSARY: There are no right or wrong answers, only opinions)
*(STATEMENTS)
*(ROTATE)
a) Abuse of older people is common in our community.
b) Most people turn a blind eye to or ignore abuse of older people.
c) Abuse of older people is a private matter to be handled in the family.
Appendix C: Survey questionnaires 313
d) Abuse of older people can be understandable if the person committing the abuse is under a
lot of stress in their lives.
e) Abuse can be understandable if the older person is a difficult person to deal with.
*(RESPONSE FRAME)
(PROBE IF NECESSARY)
1. Strongly agree
2. Somewhat agree
3. Neither agree nor disagree
4. Somewhat disagree
5. Strongly disagree
98. (Don’t know)
99. (Refused)
*(ALL)
D3 People think of different things when they hear about the abuse of older people. I’d like you to
tell me whether you regard the following sorts of behaviour as abuse of an older person.
*(STATEMENTS)
*(ROTATE)
a) Selling an older person’s home without their consent
b) Taking money from an older person without their consent
c) Not paying bills on the older person’s behalf when you said you would
d) Deliberately embarrassing an older person
e) Calling an older person hurtful names
f) Pushing or shoving an older person
g) Not providing help with personal activities such as dressing, washing, feeding when this is
normally expected/provided
h) Limiting contact with grandchildren
i) Preventing an older person from having contact with the outside world
j) Talking to an older person in a sexual way when they do not want to
k) Threatening to send them to a residential aged care facility
*(RESPONSE FRAME)
(IF YES, PROBE: Would you say that is always abuse, usually abuse, or just sometimes abuse of
an older person?)
1. Yes, always
2. Yes, usually
3. Yes, sometimes
4. No
98. (Don’t know)
99. (Refused)
*(TIMESTAMP)
314 National Elder Abuse Prevalence Study: Final Report
*(ALL)
E1 Thinking about any older person/s you know personally over the age of 65 such as relatives or
friends, in the last 12 months, have you had any concerns that someone else in their family, their
carer, or a person they trusted has …
INTERVIEWER NOTE: This older person/s could be your parents, grandparents, other close
older relatives or friends, colleagues or people you know from community or religious groups.
*(STATEMENTS)
*(ROTATE)
a) Taken advantage of them financially (e.g. coercing them for money or assets, stealing money
or assets, preventing them from accessing money or assets)
c) Abused them emotionally (e.g. intimidating, threatening them, preventing access to family
and others)
e) Failed to provide adequate care with routine activities or personal care when needed
(e.g. shopping, transport, washing/dressing)
*(RESPONSE FRAME)
1. Yes
2. No
98. (Don’t know)
99. (Refused)
*(PROGRAMMER NOTE: FOR EACH STATEMENT WHERE E1=1, SHOW E2x TO E3a)
E2X I’d like to ask you some questions about your concerns that someone has [INSERT E1
STATEMENT].
13. Neighbour
14. Partner / spouse
15. Ex-partner / spouse
16. Other (specify)
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
5. Other
98. (Don’t know)
99. (Refused)
E4b Did you take any of the following actions to stop this from happening again?
(READ OUT ONE BY ONE)
*(ACCEPT MULTIPLES)
1. I spoke to the person causing the concerns.
2. A family member or friend spoke to the person causing the concerns.
3. A professional (social worker, doctor, nurse) spoke to the person.
4. I sought mediation or counselling for the person I’m concerned about.
5. I sought out a legal advice service for the person I’m concerned about.
6. I broke contact with or avoided the person involved causing the concerns.
7. A restraining order or safety order was made against the person causing the concerns.
8. Other
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
*(TIMESTAMP)
PROVISION OF CARE
*(ALL)
PREF1 I’d like to ask you some further questions about providing assistance to family members or
others who require assistance due to old age or because of a disability, injury or illness.
*(ALL)
F1 Do you provide care, help or assistance to family members or others due to old age or because
of a disability, injury or illness?
This includes assisting with personal care and hygiene (such as washing, bathing, dressing or
eating), and other activities (such as shopping, housework and cooking or giving money to live on).
(IF NECESSARY: this can be informal or formal care, help or assistance.)
INTERVIEWER NOTE: If necessary, clarify that this section is different to assistance with
financial matters as asked about earlier.
1. Yes
2. No
98. (Don’t know)
99. (Refused)
F2a We are interested in the oldest person that you assisted in the last 12 months. Is that your …?
*(PROGRAMMER NOTE: DISPLAY SELECTIONS FROM F2 ONLY)
*(PROGRAMMER NOTE: IF ONLY 1 PERSON SELECTED AT F2, AUTOCODE TO F2A)
(READ OUT)
1. Partner / spouse
2. Ex-partner / spouse
3. Parent(s)
4. Parent(s) in-law
5. Grandparent(s)
6. Aunt / uncle
7. Brother / sister
8. Son / daughter
9. Son / daughter (in-law)
10. Other family member
11. Friend
12. Neighbour
13. Other (PROGRAMMER DISPLAY RESPONSE FROM F2_13)
98. (Don’t know) *(SINGLE RESPONSE)
99. (Refused) *(SINGLE RESPONSE)
F3 Thinking about the [IF MULTIPLE PEOPLE SELECTED AT F2: oldest] person you assist, how old
is this person?
IF NEEDED: We are referring to your [DISPLAY F2a RESPONSE]
(PROBE TO FRAME)
1. Under 30
2. 30 – 44
3. 45 – 54
4. 55 – 64
5. 65 – 74
6. 75 – 80
7. 80 – 85
8. 85+
98. (Don’t know)
99. (Refused)
*(TIMESTAMP)
DEMOGRAPHICS – Part 2
*(ALL)
PREDEM2 Finally, I’d just like to conclude by asking a few more questions about you.
*(ALL)
*(ALL)
*(ALL)
7. Islam
8. Greek Orthodox
9. Baptist
10. Hinduism
12. Christian (No further information)
13. Judaism
14. Sikhism
11. Other
98. (Don’t know)
99. (Refused)
*(ALL)
A13 Before tax or other deductions, what is your personal annual income (excluding anyone else in
your household)?
Please include wages and salaries, government pensions, benefits and allowances and income
from interest, dividends, or other sources.
1. $1 – $7,800 per year ($1 – $149 per week)
2. $7,800 – $15,599 per year ($150 – $299 per week)
3. $15,600 – $20,799 per year ($300 – $399 per week)
4. $20,800 – $25,999 per year ($400 – $499 per week)
5. $26,000 – $33,799 per year ($500 – $649 per week)
6. $33,800 – $41,599 per year ($650 – $799 per week)
7. $41,600 – $51,999 per year ($800 – $999 per week)
8. $52,000 – $64,999 per year ($1,000 – $1,249 per week)
9. $65,000 – $77,999 per year ($1,250 – $1,499 per week)
10. $78,000 – $90,999 per year ($1,500 – $1,749 per week)
11. $91,000 – $103,999 per year ($1,750 – $1,999 per week)
12. $104,000 – $155,999 per year ($2,000 – $2,999 per week)
13. $156,000 – $181,999 per year ($3,000 – $3,499 per week)
14. $182,000 – $207,999) per year ($3,500 – $3,999 per week)
15. $208,000 or more per year ($4,000 or more per week)
16. Nil income
17. Negative income
98. (Don’t know)
99. (Refused)
*(ALL)
A14 Now including everyone in your household, before tax or other deductions, what is your annual
household income?
Please include wages and salaries, government pensions, benefits and allowances and income
from interest, dividends, or other sources.
INTERVIEWER NOTE: Includes you and anyone else in your household. Seeking estimate only –
especially if unsure of income of other household members.
1. $1– $7,800 per year ($1 – $149 per week)
2. $7,800 – $15,599 per year ($150 – $299 per week)
3. $15,600 – $20,799 per year ($300 – $399 per week)
Appendix C: Survey questionnaires 321
*(ALL)
*(MOBILE=2)(LANDLINE SAMPLE)
DEM16a What is the name of the suburb or town where you live?
IF NEEDED: In order to make sure the study covers people living in all areas, we need to ask
where people live.
SELECT LOCALITY FROM LOCALITIES SPECIFIC TO SELECTED POSTCODEINCLUDE
SPECIFIED OTHER OPTION
PROGRAMMER: USE POSTCODE FROM SAMPLE IF SUBURB REFUSED OR DK
*(ALL)
DEM20 Now just a question or two about your use of telephone services.
How many residential telephone numbers do you have?
Do not include mobile phone numbers or dedicated FAX numbers or modems.
*(ALL)
*TIMESTAMP
CLOSE
*(ALL)
y If you have any concerns about potential or actual elder abuse, please contact
1800 ELDERHelp (1800 353 374) to be redirected to the existing phone line service in
your jurisdiction (free call).
Appendix C: Survey questionnaires 323
y If you or someone you know have experienced violence or sexual assault and require
immediate or ongoing assistance, contact 1800 RESPECT (1800 737 732) to talk to a
counsellor from the National Sexual Assault and Domestic Violence hotline.
y For confidential support and information, contact Safe Steps’ 24/7 family violence response
line on 1800 015 188 or the Men’s Referral Service on 1300 766 491.
y If you are concerned for your safety or that of someone else, please contact the police in your
local area, or call 000 for emergency assistance.
y For a confidential discussion with an experienced counsellor you can call Lifeline on 13 11 14.
IF NO This completes the survey. From the Social Research Centre and the Australian Institute of Family
Studies, thank you very much for your time and assistance. Your co-operation is greatly appreciated.
*(TIMESTAMP)
TERMINATION SCRIPTS
TERM1 Thank you for your time but we need to speak with people aged between 18 and 64 years.
TERM2 Thanks for your time today but we are unable to go ahead without confirming which state you
live in.
AAPOR
All- Client SUR Client SUR SRC SUR detailed
term Definition Description category description netting outcome AAPOR category
1 INTRO1=2 Household all Ineligible HH under 18 Screen 4.7 – No Cat 4 – Not
under 18 or or over 64 outs eligible Eligible
over 64 years respondent
of age
2 INTRO1=4 Household Refusal Refused Refusals 3.21 – No Cat 3 – Unknown
refusal screener eligibility, non-
completed, interview
residential and
live contact
made
3 INTRO2=2 Mobile Ineligible Mobile under Screen 4.7 – No Cat 4 – Not
respondent 18 or over 64 outs eligible Eligible
under 18 or respondent
over 64 years
of age
4 INTRO2=5 Mobile Refusal Refused Refusals 3.21 – No Cat 3 – Unknown
respondent screener eligibility, non-
refusal completed, interview
residential and
live contact
made
5 INTRO3=3 Household all Ineligible HH under 18 Screen 4.7 – No Cat 4 – Not
under 18 or or over 64 outs eligible Eligible
over 64 years respondent
of age
6 INTRO3=3 Household Refusal Refused Refusals 2.111 – Cat 2 – Eligible,
refusal Household- non-interview
level refusal
7 INTRO3=4 Household all Ineligible HH under 18 Screen 4.7 – No Cat 4 – Not
under 18 or or over 64 outs eligible Eligible
over 64 years respondent
of age
8 INTRO3=7 Respondent Refusal Refused Refusals 2.112 – Known Cat 2 – Eligible,
refusal respondent non-interview
refusal
324 National Elder Abuse Prevalence Study: Final Report
AAPOR
All- Client SUR Client SUR SRC SUR detailed
term Definition Description category description netting outcome AAPOR category
9 SELINTRO=2 Respondent Refusal Refused Refusals 2.112 – Known Cat 2 – Eligible,
refusal respondent non-interview
refusal
10 SAFE=3 Respondent Refusal Refused Refusals 2.112 – Known Cat 2 – Eligible,
refusal respondent non-interview
refusal
11 MOB_APPT=3 Respondent Refusal Refused Refusals 2.112 – Known Cat 2 – Eligible,
refusal respondent non-interview
refusal
12 MOB_APPT_ Mobile – Refusal Refused Refusals 2.112 – Known Cat 2 – Eligible,
A=9 refused state respondent non-interview
refusal
13 CONSENT=2 Refused Refusal Refused Refusals 2.112 – Known Cat 2 – Eligible,
consent respondent non-interview
refusal
14 A1a=1 Respondent Ineligible Respondent Screen 4.7 – No Cat 4 – Not
under 18 under 18 outs eligible Eligible
respondent
15 A1a=7 Respondent Ineligible Respondent Screen 4.7 – No Cat 4 – Not
over 64 over 64 outs eligible Eligible
respondent
16 A1a=99 Refused to Refusal Refused Refusals 2.112 – Known Cat 2 – Eligible,
provide age respondent non-interview
group refusal
17 DEM16am= Unable to Refusal Terminated Refusals 2.11 – Refusal Cat 2 – Eligible,
88888 provide during survey non-interview
suburb or
town
18 DEM16am= Refused Refusal Terminated Refusals 2.11 – Refusal Cat 2 – Eligible,
99999 to provide during survey non-interview
suburb or
town