Interventii PT Apartinatori

Critical Reviews in Oncology / Hematology 140 (2019) 52–66
Contents lists available at ScienceDirect
Critical Reviews in Oncology / Hematology

journal homepage: www.elsevier.com/locate/critrevonc
Review
Psychological interventions targeting partners of cancer patients: A T

systematic review
Anne-Kathrin Kleined,a,1, Nina Hallenslebena,1, Anja Mehnerta, Klaus Hönigb,c, Jochen Ernsta,
⁎
a
Department of Medical Psychology and Medical Sociology, University of Leipzig, Germany
b
Clinic of Psychosomatic Medicine and Psychotherapy, University of Ulm, Germany
c
Comprehensive Cancer Center Ulm (CCCU), Germany
d
Department of Organizational Psychology, University of Groningen, the Netherlands
ARTICLE INFO ABSTRACT
Keywords: Purpose: Cancer patients’ intimate partners often experience levels of psychological burden that are comparable
Cancer to or even exceed that of the patients, making it imperative that they too be provided with appropriate psy-
Partner chological support. This review aimed to present the content and the effects of interventions delivered to car-
Spouse egiving partners of cancer patients on both partners and patients. Furthermore, we provide information about
Review
the acceptability of the interventions and study quality.
Emotional distress
Methods: An initial search in Web of Science, PsycINFO, and PubMed databases was conducted. We included
Quality of life
Psycho-oncology RCTs as well as pre-post studies that focused on enhancing partners’ wellbeing or diminishing partners’ distress.
Intervention To be included, interventions had to have been offered to partners either only or predominantly. We included
studies published until December 2017. The methodological quality of the trials was assessed with the EPHPP
assessment tool.
Results: Nine studies met the inclusion criteria. Intervention topics included social support, short-term problem
solving, the marital relationship quality, role expectations, emotional resilience, and coping strategies. Positive
intervention effects were found with regard to social support, emotional distress, improved communication,
posttraumatic growth, self-efficacy, and coping. Despite considerably low response rates, the interventions were
generally well accepted. Most of the studies suffer limitations because of methodological flaws, the lack of
randomization, and small sample sizes.
Conclusion: Interventions delivered to partners of cancer patients may have positive effects on both partners and
patients. We derive several implications for future research: Intervention programs should be tailored to the
specific needs of caregiving partners with regard to the cancer trajectory and gender. Effort has to be made to
increase sample sizes as well as to include particularly burdened individuals. Selected measurement instruments
should be sensitive to specific intervention effects. Finally, information on both statistical as well as clinical
relevance of research findings should be provided.
1. Introduction Partners or spouses typically become the primary caregivers and the
most important source of support for cancer patients (McLean and
Almost half a million people in Germany are diagnosed with cancer Jones, 2007; Pitceathly and Maguire, 2003). This role entails intense
each year and have to deal with the overwhelming consequences of the challenges: in addition to dealing with feelings of uncertainty and an-
illness (Robert Koch-Institut [RKI], 2016). Suffering from cancer not only xiety arising from the cancer diagnosis, caretaking partners also have to
confronts patients with a variety of psychosocial challenges but also has a adjust to profound changes that affect almost all aspects of their lives.
huge impact on their social environment. Families and intimate partners Following a cancer diagnosis, they are usually expected to autono-
in particular experience levels of psychological burden that are com- mously perform necessary organizational and formal tasks as well as
parable to or even exceed that of patients, making cancer a “we-disease” offering significant levels of emotional support to their ill partner, a
(Carlson et al., 2008; Kayser et al., 2007; Stenberg et al., 2010). dynamic which often results in them suppressing their own feelings for
⁎
Corresponding author at: Department of Medical Psychology and Medical Sociology, University of Leipzig, Ph.-Rosenthal-Str. 55, D-04103 Leipzig, Germany.
E-mail address: Jochen.Ernst@medizin.uni-leipzig.de (J. Ernst).
1
Co-first author; The authors wish it to be known that the first two authors should be regarded as joint first authors.
https://doi.org/10.1016/j.critrevonc.2019.05.008
Received 10 July 2018; Received in revised form 25 April 2019; Accepted 10 May 2019
1040-8428/ © 2019 Elsevier B.V. All rights reserved.
A.-K. Kleine, et al. Critical Reviews in Oncology / Hematology 140 (2019) 52–66
long periods of time (Stenberg et al., 2010). This burdensome situation (3) Intervention effects
frequently leads to a high vulnerability to psychological stress or even a) Which intervention effects were reported for caretaking part-
mental illness: research suggests that between 10 and 50% of caregivers ners?
suffer from psychiatric disorders, whereby higher rates are seen among b) Which (if any) intervention effects were reported for cancer
caregivers of patients with advanced or palliative stage cancer patients?
(Pitceathly and Maguire, 2003). More specifically, caring for a cancer (4) Acceptability of the intervention
patient increases the likelihood of suffering from depression, anxiety, a) How many of the participants dropped out from the interven-
hopelessness, social isolation, various somatic symptoms, and work- tion?
related and financial difficulties (Arno et al., 1999; Bevans and b) What were common reasons for dropping out?
Sternberg, 2012; Braun et al., 2007; Girgis et al., 2013; Grunfeld, 2004; c) How compliant were participants with the intervention (i.e.
Pitceathly and Maguire, 2003). treatment adherence rates)?
However, despite these circumstances, partners’ or spouses’ specific (5) Study quality
burdens often remain invisible. Northouse et al. (2010) suggest this How appropriate were the studies (in terms of risk of bias) for
may be due to the fact that caregiving partners often do not make their drawing conclusions about the efficacy of the interventions (ac-
needs for support known. Moreover, others (including health care cording to the Quality Assessment Tool for Quantitative Studies
professionals) perceive them as caregivers rather than care recipients. developed by the Effective Public Health Practice Project (EPHPP;
For these reasons, caregiving partners currently receive less social, Effective Public Health Practice Project (1998))?
health care-related, and psychological support than patients (Blum and
Sherman, 2010; Given et al., 2004; Jensen and Given, 1991; Schubart 2. Methods
et al., 2008). Because the majority of caring partners’ unmet needs are
related to coping with psychological or emotional distress (Lambert We formulated a review protocol on the basis of the PRISMA
et al., 2012) it is imperative that this particular population be provided statement in accordance with the PICOS elements for a review protocol
appropriate psychosocial interventions. (Liberati et al., 2009). We pre-registered the review in August 2017 at
During the past few years, an increasing number of psychosocial “PROSPERO International prospective register of systematic reviews”
interventions for cancer patients and their caregivers have been de- (CRD42017073481).
veloped and evaluated (see Carlson et al., 2000; Ferrell and Wittenberg,
2017; Frambes et al., 2017; O’Toole et al., 2017 for current reviews).
2.1. Eligibility criteria
However, as far as we know, only one review of interventions solely
aimed at improving caretaking partners’ outcomes has been published
Eligibility criteria are detailed according to the PICOS framework
to date. Carlson et al. (2000) identified seven studies published between
(Liberati et al., 2009) as follows.
1986 and 2000 that evaluated counseling or therapy sessions offered to
partners of cancer patients. Only two of the five studies that included a
comparison group yielded positive intervention effects. The provision 2.1.1. Participants
of psychosocial support to caregivers of cancer patients has gained The participants were spouses or partners (married or unmarried)2
some attention over the past decades and recent research findings have of cancer patients aged 18 years or older. The patients had to have been
shown beneficial effects of psychological interventions for partners of formally diagnosed with a cancer of any type (solid or hematologic), at
cancer patients, for example, with regard to perceived and received any tumor stage, and at any time point since diagnosis. We excluded
social support (Senneseth et al., 2017) or emotional well-being (Lewis studies with mixed chronic disease samples (incl. cancer) and post-be-
et al., 2008). Thus, we deemed it necessary to provide a comprehensive reavement interventions.
overview of the literature published on this topic so far. The main focus
of our review lies on the intervention characteristics and effects on 2.1.2. Interventions
caring spouses. However, researchers have argued that emotionally For inclusion in this review, the psychological interventions had to
supportive care delivered to partners may improve both the partners’ be aimed at improving the well-being of cancer patients’ intimate
and the patients’ emotional well-being (e.g., Carlson et al., 2017). partners, and they had to be offered either solely or predominantly to
Moreover, some interventions offered to spouses of cancer patients the partners. That means we included studies in case other people (e.g.,
focus on increasing their self-efficacy and caregiver skills, thus helping the patient, friends, or family members) took part in the intervention
them to become better caregivers, which likely alleviates the patients’ sessions, but the main aim of the intervention was to improve the
emotional distress (e.g., Duggleby et al., 2017; Jones et al., 2013). partner's well-being and the partners took part in every session that was
Accordingly, we provide information about the effectiveness of dif- offered. On the other hand, we excluded studies that focused on the
ferent interventions for both participating partners and not-partici- outcome of the cancer patients, as well as couple-based interventions.
pating patients. Moreover, since this review is intended to serve as The interventions had to have made use of psychological techniques
practical guidance for the development of future intervention studies, such as: education, coping skills training, psychotherapy, discussion
we provide an overview of variables and instruments used as well as the groups, and relaxation or mindfulness training, either singly or in
acceptability of each intervention conducted (i.e., dropout and com- combination. We included all individual or group interventions pro-
pliance rates and reasons for dropout). Furthermore, by providing in- vided or supervised by health-care professionals as well as self-ad-
formation about the studies’ methodological quality we enable re- ministered online interventions. We included interventions irrespective
searchers and practitioners to critically evaluate the research findings of their setting, duration, and frequency.
of the studies included in our review. With these aims in mind we de-
rived the following research questions:
2.1.3. Outcomes
We took all self-reported psychosocial outcome concerning partners’
(1) Assessed variables and instruments
wellbeing or distress into account. We excluded interventions which
How were the intervention outcomes measured?
assessed purely physical outcomes (i.e. pharmaceutical studies).
(2) Intervention approaches
Which kinds of (evaluated) psychological interventions for partners
of cancer patients have been conducted so far (until December 2
In the following, when referring to partners of cancer patients, we mean
2017)? both unmarried and married partners/spouses of cancer patients.
53
Fig. 1. Search terms.
2.1.4. Studies intervention, duration of the intervention, number of sessions and

We considered all quantitative and mixed methods studies. In order outcomes (including instruments and variables used to assess the in-
to be able to interpret the results with regard to change over time we tervention effect), time points of outcome assessment, and narrative
only considered studies which included at least two measurement reports. Any disagreements between the two reviewers were resolved
points in order to detect changes over time. We excluded comments, by discussion.
editorials, case studies, purely qualitative research, posters, dissertation
abstracts, and reviews, as well as unpublished study results. 2.4. Assessment of risk of bias
2.2. Information sources For assessing the internal validity of the studies, we used the EPHPP
developed by the Effective Public Health Project Canada (Armijo-Olivo
A systematic computer-based literature search was conducted in et al., 2012). The EPHPP is a well-established instrument for the eva-
accordance with the PRISMA-Statement (Liberati et al., 2009). We luation of the research quality and internal validity of different study
searched the electronic databases Pubmed, Web of Science, and Psy- designs such as observational, cross sectional, before-and-after studies,
cINFO. We searched in all fields and did not restrict the period of and randomized controlled trials (RCTs). It has been reported to have a
publishing dates and thus searched for any literature that had been good test-retest reliability as well as content and construct validity
published up until August 2017. We searched for studies which were (Jackson and Waters, 2005; Thomas et al., 2004). The quality of a study
written in either English or German. We additionally screened reference is classified as “strong”, “moderate”, or “weak” based on evaluations of
lists of included studies and review articles on related topics. The search the following six domains: (1) selection bias (representativeness and
was re-run prior to final analysis in December 2017. In the second participation rate3); (2) study design; (3) confounders; (4) blindings; (5)
search, we found one further study that met our inclusion criteria and data collection method; and (6) withdrawals/dropouts. An overall assess-
therefore included it in the analysis as well. The search terms used are ment score was formed on the basis of the evaluation scores for these
shown in Fig. 1. domains. Additional elements of the EPHPP that were used for the
evaluation of the studies' methodological quality, but that were not
taken into account for forming the overall score, include adherence
2.3. Study selection and data extraction
rate, possibility of contamination, appropriateness of statistical
methods, and whether the analyses were performed based on inter-
We used the reference management program CITAVI (Swiss
vention allocation (Armijo-Olivo et al., 2012). Two reviewers (NH and
Academic Software GmbH) for data management. After eliminating
AK) independently evaluated the methodological quality of each study
duplicates, the first authors (NH and AK) independently screened all of
on the basis of the EPHPP. Any disagreement was resolved by discus-
the titles based on the eligibility criteria. If it was initially unclear
sion.
whether a selected article might be relevant, it was kept on the list for
closer review. Next, the abstracts of potentially relevant articles were
3. Results
analyzed and the selection was narrowed. If all criteria were met after
the full article had been read, two reviewers (NH and AK) decided
3.1. Study selection
whether or not to include the study in the review.
Information extracted from the study included publication and
8455 records were identified through database searches. After
study details (i.e. year of publication, country of origin, number of
participants), sample characteristics (i.e. age range and mean age, sex,
type of tumor, time passed since diagnosis, tumor stage), intervention 3
Participation rate refers to the percentage of partners that met the inclusion
details (i.e. type of intervention, format of intervention [group or in- criteria and agreed to participate before group assignment. Participation rate is
dividual]), characteristics of the professional who administered the not applicable if partners were self-recruited.
54
Fig. 2. Flow chart of study selection process.
duplicates were eliminated, the titles and abstracts of 6061 articles with regard to the phase of the cancer trajectory: Partners of patients
were scanned. Another 42 potentially relevant articles were identified suffering from stage 0 to stage IV cancer were included and time since
by examining the reference lists of existing reviews and related articles. diagnosis ranged from 4 to 29 months, with an average of 11 months.
After the second search in December 2017, 108 additional articles were Some studies reached a higher degree of homogeneity with regard to
identified as potentially relevant. Thus, 6211 articles were screened on the cancer trajectory by including partners of patients suffering ex-
the basis of their title and abstract. Of those, 6144 articles were ex- clusively from cancer stages I-II (Bultz et al., 2000) or I-III (Duggleby
cluded. Next, the full-texts of 67 articles were assessed for eligibility. 57 et al., 2017), as well as patients diagnosed within six (Lewis et al.,
publications did not meet the inclusion criteria and were excluded due 2008) or 18 months before recruitment and non-metastatic cancer
to the reasons presented in Fig. 2. Ultimately, nine studies were chosen (Carlson et al., 2017). To assess intervention dosage, we calculated the
for inclusion. total number of hours the intervention was provided (mean 7.1; range
3–10.5), the number of sessions (mean 5.6; range 1–10), and the
3.2. Study and sample characteristics duration of the intervention (mean 46.5 days; range 1–70 days).
Duggleby et al. (2017), who provided an online intervention, did not
Table 1 presents a summary of those nine publications, all of which specify the number of hours the intervention was provided nor the
were published between 1986 and 2017. Most of the studies were number of sessions conducted. Therefore, we calculated the average
conducted in the United States (k = 5), in Canada (k = 3), and Norway values using the data that was available. Control group participants
(k = 1). Six of the studies were RCTs, three were pretest-posttest stu- received treatment as usual. In two studies information about the care
dies. Two of the pretest–posttest studies and one of the RCTs were pilot for control group participants was not available (Blanchard et al., 1996;
studies. The studies’ samples comprised a total of 391 participants at Senneseth et al., 2017)
first post-intervention measurement. The average sample size was 43.
Participants were between 32 and 67 years old, the mean age was 54 3.3. Assessed variables and instruments
years. The largest number of studies included interventions for partners
of patients with breast cancer (k = 5), followed by those of patients For a better overview, we classified groups of outcome variables
with prostate cancer (k = 2). The other studies either included patients with the respective measurement instruments as depicted in Table 2.
with a variety of tumor entities (k = 1) or they did not identify the Different kinds of general or cancer-related emotional distress (e.g.,
tumor entities (k = 1). The selection of certain cancer types in a large depression, anxiety) were measured in all of the studies. The assessment
proportion of the included studies led to gender homogenous samples of skills ranged from caregiving skills (i.e., how to be a better caregiver
for most of the trials. For the rest of the trials (k = 2), about equal for the patient) to communication, coping, and self-care skills. Seven
numbers of male and female caregivers were included. In sum, 44% of studies analyzed the effect of the intervention on the caregiver's re-
all of the partners were female. Most of the samples were heterogenous lationships with others (i.e., social support and marital satisfaction).
55
Table 1
Summary of included studies.
Randomized controlled trials

A.-K. Kleine, et al.
Study Country Type of Intervention Measurement Number of Mean age Percent Inclusion Treatment Measurements and instruments Results
intervention period, points, period of participants (SD or female in criteria control
duration of follow-up in recruited/at range) IG group
each session brackets post-
intervention/at
follow-up
Partners Patients Partners Patients
Blanchard USA Individual Six sessions, on t0: baseline 86/78/66 55b 60%b - Partners of Not Depression: CES-D - no quantitative effects - reduced
et al. counseling a weekly basis, t1: post- cancer patients indicated Functioning: SF-20 found depression
(1996) 60 min intervention (various types) Social support: HDL (adapted) - better communicationa compared to CG*
t2: follow-up (six - cancer Marital satisfaction: DAS - appreciation of
months) diagnosed support for spousea
more than Anxiety: STAI Perceived
three months Burden: ZBI, List of health status:
before Pressing Problems FLIC
recruitment Coping: CRS
- patient not
eligible for
hospice
program
Bultz et al. Canada Group Six sessions, on t0: baseline 34/34/32 51 0% - Partners of Treatment General emotional distress:POMS - reduced emotional - improved
(2000) intervention a weekly basis, t1: post- (32–67)b breast cancer as usual, Marital satisfaction: IMS distress compared to communicationa
90–120 min intervention patients wait list Social support: FSSS CG at t2 (p=0.07)
t2: follow-up - patient - improved
56
(three months) diagnosed communication,
within one Mental experience of
year before adjustment to normalizationa
recruitment cancer: MAC
- cancer stages
I-II
Carlson Canada Group six sessions, on t0: baseline 76/63/56/60 61.7 (8.7) 100% - Partners of Treatment General emotional - no quantitative effects
et al. intervention a weekly basis, t1: post- prostate cancer as usual, distress: POMS found
(2017) 90 min intervention patients wait list Anxiety: STAI
t2: follow-up - patient Social support: SSS
(three months) diagnosed Marital satisfaction:
t3: second within 18 IMS
follow-up (six months before
months) recruitment
- non-
metastatic
cancer
Duggleby Canada Web-based six sections, t0: baseline 57/47/44/40 53.7b 0% - Partners of Treatment Caregiver guilt: Quality of - no quantitative effects - no quantitative
et al. intervention four weeks, t1: during breast cancer as usual CGQ life: FACT-B found effects found
(2017) self- intervention patients Hope: - half of the participants
administered t2: post- - cancer stages HHI agreed or strongly agreed
intervention I-III Quality of life: that the intervention
t3: follow-up CQOL-C increased their ability to
(two months) Self-efficacy: deal with changes caused
GSES by the cancera
(continued on next page)
Table 1 (continued)

intervention/at
follow-up
Manne et al. USA Group six sessions, on t0: baseline 68/60 59.6 (9.3) 100% - Partners of Treatment Emotional distress: Not assessed - improved coping on Not assessed
(2004) intervention a weekly basis, t1: post- prostate cancer as usual psychological the subscales positive
60 min intervention patients distress subscale of reappraisal and
(one month) MHI growth*, and denial**
Cancer-specific compared to CG
distress: IES - improved post-
Coping: COPE, traumatic growth*
EACS compared to CG
Post-traumatic
growth: PTGI
Communication: 3
subscales from CPQ
(adapted)
Senneseth Norway Group One session, t0: baseline 35/24 44.3b 53%b - Partners of Not General emotional Not assessed - improved received Not assessed
et al. intervention 180 min t1: post- (7.4) cancer patients indicated distress: GHQ-12 social support*
(2017) intervention - living with Quality of life: - improved perceived
(three months) the patient QOLS-N social support**
- patient Social Support: CSS,
57
diagnosed MSPSS
within 5 years
before
recruitment
- parenting
children under
the age of 18
Pretest-posttest trials
Jones et al. USA Group Five sessions t0: baseline 54/42/41 53 0% - Partners of Depression: - improved support and - improved
(2013) intervention on a weekly t1: post- breast cancer CES-D self-care skills at t1*** despression at t2**
basis, 90 min intervention patients Marital satisfaction: and t2*** - improved
t2: follow-up (3 currently DAS-R, MIS - better communication, communicationa
months) receiving better understanding of
treatment wife, mutual support in
groupa
Support and self-
care skills:
SSC, CASE-S
(continued on next page)
Table 1 (continued)

intervention/at
follow-up
***
Lewis et al. USA Individual Five sessions t0: baseline 26/20 53 0% - Partners of Depression: Not assessed - improved anxiety - better relationship
(2008) counseling on a biweekly t1: post- breast cancer CES-D and depression** strength,
basis, 60 min intervention patients Anxiety: - improved support and communication and
- married or co- STAI-Y self-care skills*** support from the
habiting Support and self- - improved marital partnera
- local or care skills: SSC, satisfaction on
regional breast CASE-S affectional expression
cancer Marital satisfaction: subscale*
- cancer stages DAS, MIS - better relationship
0-III strength, communication,
- patient understanding of wifea
diagnosed
within 6
months before
recruitment
Sabo et al. USA Group Ten sessions on t0: baseline 24/23 Not 0% - Partners of Emotional distress: Not assessed - improved Not assessed
(1986) intervention a weekly basis, t1: post- reported breast cancer Study-specific scale communication about
60 min intervention patients Communication mastectomy**
58
about mastectomy: - improved verbal
Study-specific scale communicationa
Support skills:
Study-specific scale
Self-esteem: Study-
specific scale
Gender role
expectations: AMR
Marital satisfaction:
LMAT
Notes. AMR, The attitudes toward the male role Scale; CAS, Caregiving Appraisal Scale; CASE-S, Cancer Self-Efficacy Scale - Spouse; CES-D, Center for Epidemiologic Studies-Depression Scale; CG, control group; CGQ,
Caregiver Quilt Questionnaire; CPP, Cancer-PEPSONE program; CPQ, Communication Patterns Questionnaire; CQOL-C, Caregiver Quality of Life Index – Cancer; CRS, Coping Response Scale; CSS, Crisis Support Scale;
CWC, Coping With Cancer; DAS, Dyadic Adjustment Scale; DAS-R, Revised Dyadic Adjustment Scale; EACS, Emotional approach coping scale; FACT-B, Functional Assessment of Cancer Breast; FLIC, Functional Living
Index-Cancer; FSSS, Functional Social Support Scale; GHQ-12, General Health Questionnaire 12-item version; GSES, General Self-Efficacy Scale; HDL, Health and Daily Living Scale; HHH, Helping Her Heal; HHH-G,
Helping Her Heal – group approach; HHI, Herth Hope Index; HSCI, Help Seeking Coping Index; ICR, Index of Coping Responses; IES, Impact of Events Scale; IG, intervention group; IMS, Index of Marital Satisfaction;
LMAT, Locke's Marital Adjustment Test; MAC, Mental Adjustment to Cancer Scale; MaTT, Male Transition Toolkit; MBBS, Montgomery and Borgotta Burden Scale; MBSR, Mindfulness-Based Stress Reduction; MHI, Mental
Health Inventory; MIS, Mutuality and Interpersonal Sensitivity Scale; MSPSS, Multidimensional Scale of Perceived Social Support; PCS, Personal Change Scale; POMS, Profile of Mood States; PTGI, Post-traumatic Growth
Inventory; QOLS-N, Norwegian version of the Quality of Life Scale (QOL); SD, Standard Deviation; SF-20, 20-Item Short Form Health Survey; SET, supportive expressive therapy; SSC, Spouse Skills Checklist; SSS, Social
Support Survey; STAI, State-Trait Anxiety Inventory; STAI-Y, State-Trait Anxiety Inventory – state anxiety subscale; TMS, Total Mood Disturbance Score; ZBI, Zarit Burden Inventory.
a
Qualitative information.
b
information referring to intervention group.
* p < .05.
** p < .01.
*** p < .001.
Table 2
Applied outcome variables and effects of the interventions on the partners.
Outcome Blanchard et al. Bultz et al. Carlson et al. Duggleby Jones et al. Lewis et al. Manne et al. Sabo et al. Senneseth et al.
et al.
General emotional Total mood: Total mood: Psychological distress: Psychological

distress POMS ↑ POMS → subscale of MHI → distress:
GHQ-12 →
Depression CES-D → CES-D → CES-D ↑ Study-specific scale →
Anxiety STAI → STAI → STAI-Y ↑
Cancer-related emotional Burden: ZBI, List of Caregiver Cancer-specific distress: Gender role expectations:
distress pressing problems → guilt: IES → AMR →
CGQ →
Quality of life and overall Functioning: Quality of life: Quality of life:
functioning SF-20 → CQOL-C → QOLS-N →
Self-efficacy and skills CRS → Self-efficacy: Self-efficacy: Self-efficacy: Coping: Supportive attitude (study-specific
GSES → CASE-S ↑ CASE-S ↑ COPE scale) →
Support and self- (reappraisal and growth ↑, Frequency of verbal communication
care skills: denial coping ↑, acceptance about mastectomy (study-specific
SSC↑ →, emotional approach) scale) ↑
EACS →
Communication skills:
CPQ →
59
Hope HHI →
Posttraumatic growth PTGI ↑
Self-esteem Study-specific scale →
Social support HDL → FSSS → SSS → Received social
support:
CSS↑
Perceived social
support:
MSPSS ↑
Marital satisfaction Dyadic adjustment: IMS → IMS → Dyadic Dyadic adjustment: Sexual compatibility: LMAT →
DAS → adjustment: DAS (Affectional expression ↑, Marital happiness:
DAS-R → consensus →, cohesion→, LMAT →
Marital quality: satisfaction→)
MIS → Marital quality:
MIS →
Notes. AMR, The attitudes toward the male role Scale; CASE-S, Cancer Self-Efficacy Scale - Spouse; CES-D, Center for Epidemiologic Studies-Depression Scale; CGQ, Caregiver Guilt Questionnaire; CPQ, Communication
Patterns Questionnaire; CQOL-C, Caregiver Quality of Life Index – Cancer; CRS, Coping Response Scale; CSS, Crisis Support Scale; DAS, Dyadic Adjustment Scale; DAS-R, Revised Dyadic Adjustment Scale; EACS,
Emotional approach coping scale; FSSS, Functional Social Support Scale; GHQ-12, General Health Questionnaire 12-item version; GSES, General Self-Efficacy Scale; HDL, Health and Daily Living Scale; HHI, Herth Hope
Index; IES, Impact of Events Scale; IMS, Index of Marital Satisfaction; LMAT, Locke's Marital Adjustment Test; MHI, Mental Health Inventory; MIS, Mutuality and Interpersonal Sensitivity Scale; MSPSS, Multidimensional
Scale of Perceived Social Support; POMS, Profile of Mood States; PTGI, Post-traumatic Growth Inventory; QOLS-N, Norwegian version of the Quality of Life Scale (QOL); SF-20, 20-Item Short Form Health Survey; SSS,
Social Support Survey; STAI, State-Trait Anxiety Inventory; STAI-Y, State-Trait Anxiety Inventory – state anxiety subscale; ZBI, Zarit Burden Inventory.
→ no change after intervention, ↑ improvement after intervention.
3.4. Intervention goals, approaches, and effects everyday life. Lewis et al. (2008) tested its effectiveness in a pilot study,
using the originally intended individual counseling setting. They ob-
We organized our findings according to the topic or focus of each served improvements for most of the assessed variables. Moreover,
intervention. That is, we considered both the main goal of the inter- most of the partners who scored in the clinical range for anxiety and
vention and the intervention approach or strategy used to pursue this marital adjustment before the intervention scored out of the clinical
goal. On this basis, we assigned labels denoting the topic of each in- range after it, supporting the clinical significance of the findings. Exit
tervention. Specifically, we identified six topics (building up social interviews revealed that the program strengthened the husbands’ re-
support, short-term problem solving, improving the marital relationship lationships with their wives. Effects of the intervention on patients was
quality, overcoming role expectations, building up emotional resilience, not measured.
developing coping strategies). However, we want to clarify that these Jones and colleagues’ (2013) primary aim was to test the feasibility
labels only denote the main topic of the intervention. Most of the stu- and acceptability of HHH delivered in a group-format as well as to
dies pursued multiple goals and applied various strategies. All of the obtain a preliminary estimate of its impact on skills, self-confidence,
interventions except for one (Senneseth et al., 2017) were offered to and self-care. Secondary aims were to assess the intervention effects
partners alone. with regard to ratings of marital satisfaction and depression. Similar to
Lewis and colleagues (2008), they expected intervention effects for
3.4.1. Social support patients as well. The results in large reinforced those of Lewis et al.
Senneseth et al. (2017) analyzed the effects of a social network (2008). While they did not observe significantly improved marital
program which was designed with the aim of optimizing social support functioning, they did see positive effects on support and self-care skills.
for families of cancer patients. The program consisted of one session As opposed to Lewis et al. (2008), they measured quantitative outcomes
that was led by three experienced psychologists. In the first part, for patients and found a significant time effect for depression. Exit in-
partners were informed about challenges and coping strategies as well terviews revealed that the program reinforced the husbands’ commu-
as ways to sustain social support. In the second part, they were en- nication skills and recognition of the importance of self-care. The group
couraged to express their needs to their social network members, who format was reported to have been helpful since it offered the oppor-
subsequently stated the types and frequencies of support to which they tunity to identify shared experience and to bond.
could commit. Positive effects were expected with regard to received
and perceived social support, as well as psychological distress and 3.4.4. Role expectations
quality of life. Indeed, they found a significant intervention effect on Sabo et al. (1986) conducted an unstructured multisession men's
both received and perceived social support scales, but not with regard discussion group with the aim of facilitating the husbands’ adjustment
to psychological distress and quality of life. However, descriptive re- by encouraging them to overcome their role expectations, explore the
sults showed that recipients of the intervention reported less distress denial process and to communicate with their wives about mastectomy
and better quality of life at post-intervention while controls reported issues. They did not apply a specific strategy to target these goals. A
small or no changes. significant and favorable difference between the groups was found on
the communication about mastectomy scale. Qualitatively, observations
3.4.2. Short-term problem solving and interviews conducted over the course of the sessions revealed that
Blanchard et al. (1996) used a partly structured short-term problem the support group stimulated verbal communication between the
solving approach which was administered in individual counseling partners and helped the men to reflect on their role as their wife's
sessions by an experienced oncology social worker. The aim of their protector.
study was to help partners develop strategies to manage specific pro-
blems associated with their distress and thus reduce their level of dis- 3.4.5. Emotional resilience
tress. They expected positive effects for the non-participating patients. Bultz et al. (2000) provided a supportive expressive group inter-
In the sessions, partners should learn to identify problems, generate vention with a focus on education and emotional support. The main aim
alternative solutions, and discuss and rehearse an action plan. The of the intervention was to encourage mutual support between the
program did not prove to be differentially effective concerning quan- partners and to help the participants deal with existential concerns
titative outcomes for partners. There was a main effect of time on de- (e.g., fear of recurrence of the cancer illness or patient's death). The
pression and avoidance coping in both groups. There was an inter- expected positive intervention effects for patients due to improved
vention effect for patients. Specifically, patients whose partners were in verbal communication and mutual support. Partners received psy-
the intervention group were significantly less depressed at follow-up. A choeducation by watching a video and talking to a medical oncologist.
number of the participating partners stated that the intervention helped In the group, they explored their feelings, dealt with individual con-
to improve their communication with their partners. Some patients cerns, and confronted fears and anxieties. The Total Mood Disturbance
appreciated the fact that someone was helping their spouse cope with (TMD) score on the POMS was 22.9 scale points lower at the three-
the illness. month follow-up measurement (initial score: 36.6). The control group's
TMD score remained stable at approximately 28. However, this differ-
3.4.3. The marital relationship quality ence failed to reach statistical significance at the three-month follow-up
Jones et al. (2013) and Lewis et al. (2008) both provided a struc- measurement (p = 0.07). On behalf of the patients, total mood dis-
tured skill-building and efficacy-enhancing exercise (Helping Her Heal, turbance decreased (p = 0.19) while confidant support increased
HHH). The HHH program was originally developed as a one-on-one (p = 0.06). Some partners stated that they experienced a beneficial
intervention with the aim of improving marital quality, interpersonal feeling of normalization through the intervention.
communication, and support as well as to improve the partner's emo- Carlson et al. (2017) used the same approach as Bultz et al. (2000).
tional well-being with regard to depressive mood and anxiety. Since the The aim of the intervention was to increase the partner's emotional
intervention aimed at helping partners to be better able to meet their well-being, ameliorate the couple's communication and relationship, as
wives’ changing needs and understand their emotions, positive inter- well as to make the partner a better caregiver. They did not detect
vention effects were expected for non-participating patients. In five significant intervention effects for any of the assessed variables neither
sessions husbands were taught how to (1) become better listeners for for partners nor for patients. There was an effect of time on several
their wife, (2) let go of fixed role models, (3) gain a deeper under- subscales of the POMS for both patients and partners.
standing of their wife, (4) develop strategies to enhance the quality of Manne et al. (2004) provided a closed, structured group interven-
interpersonal connection, and (5) maintain the strategies in their tion in six sessions. The main aim of the intervention was the reduction
60
of psychological distress. Within the sessions, partners received car- participants completing the study as well as treatment adherence rates
egiving information (nutrition and medication) as well as stress man- were high. On the other hand, the representativeness of the target po-
agement and coping skills training. Moreover, relationship (intimacy, pulation was mostly not likely (three studies) or only somewhat likely
communication) and survivorship (post-treatment concerns) issues (six studies) and the participation rate was mostly low. This is due to
were discussed. Each session was facilitated by an expert on the specific the fact that most of the studies used evidence from single-center trials.
topic. Didactic presentations were combined with group contributions. Study participants were not blinded with regard to the experimental
Preliminary analyses revealed that 18% of the sample had elevated condition. Information about blinding of outcome assessors was often
distress scores and 49% scored in the high range for cancer-specific not provided. Six of the trials were RCTs. All of them applied simple
distress at study entry. Significant intervention effects were only seen randomization techniques, which were considered as appropriate.
on four of the five post-traumatic growth scales and two of the five
coping scales. General psychological distress declined over the course of 4. Discussion
the study in both control and intervention group. Wives who attended
group meetings reported gains in terms of greater personal strength, To our knowledge, this is the first systematic review so far that has
spiritual growth, and appreciation of life. examined the structure and effects of psychosocial interventions for
partners of cancer patients. The interventions included covered a broad
3.4.6. Coping strategies range of topics (building up social support, short-term problem solving,
Duggleby et al. (2017) developed an online intervention based on a improving the marital relationship quality, overcoming role expecta-
transition theory framework. The intervention aimed at helping part- tions, building up emotional resilience, developing coping strategies).
ners to become aware of and deal with relevant changes in their si- Four studies reported mixed results (significant intervention effect on
tuation brought about by the cancer illness. In the online program of- one or more, but not all of the variables), and one study reported sig-
fered, partners were encouraged to express themselves (e.g., write nificant improvements for all of the variables assessed. The remaining
down own story) and to develop strategies to care for themselves and studies (k = 4) did not detect any significant intervention effects. The
their partners (e.g., positive self-talk, identify partner's needs, getting studies included in the current review were highly heterogenous, with
help). Moreover, they were provided with relevant information about regard to intervention approaches, goals, sample characteristics, as well
changes to expect, resources (e.g., contact lists), and health informa- as outcomes measured. In the following, we discuss research findings
tion. It was assumed that increased self-efficacy and hope on behalf of with the goal of providing meaningful recommendations for future re-
the patients would positively affect non-participating patients. No sig- search, thereby accounting for the studies’ heterogeneity.
nificant effects were found with respect to relevant outcome variables.
Effects on patients were not measured. According to narrative reports, 4.1. Discussion of results and implications for future research
most participants said that the intervention was convenient and that
they would recommend it to another person. Half of the sample con- Lewis et al. (2008) detected improvements on almost all scales.
sidered the intervention to be effective. Their findings are best comparable to the results of Jones et al. (2013)
who applied the same intervention approach in a group setting. Jones
3.5. Acceptability of the interventions et al. (2013) could not replicate the positive effects for marital func-
tioning. Based on these findings, it could be deduced that the HHH
We considered response rate, dropout, lost-to-follow-up, and treat- approach is – at least concerning marital relationship quality outcomes
ment compliance rates as indicators acceptability. Response rates, i.e. – more effective in the originally intended individual counseling set-
the proportion of partners who took part in the intervention out of all ting. More studies are needed to investigate the conditions that are
partners who had been approached, was comparatively low, ranging critical for the superiority of certain intervention formats.
from 18.2% (Carlson et al., 2017) to 57% (Manne et al., 2004). Dropout Three studies reported positive intervention effects for non-partici-
in the intervention group ranged from 0% (Bultz et al., 2000) to 27.6% pating patients (Blanchard et al., 1996; Bultz et al., 2000; Jones et al.,
(Duggleby et al., 2017). Lost to follow-up rates in the intervention 2013, 2013). Strikingly, intervention effects on patients match the ef-
group ranged from 13.3% (Bultz et al., 2000) to 28.9% (Carlson et al., fects found for partners in three out of four studies in which quantita-
2017). The highest loft to follow-up rate was 31.8% after six months tive outcomes for patients were reported. Specifically, Jones and col-
(Blanchard et al., 1996). Common reasons for dropout and lost to leagues (2008) as well as Bultz and colleagues (2000) observed positive
follow-up were: the patient's death (Blanchard et al., 1996), the dete- effects for both patients and partners while Carlson et al. (2017) failed
rioration of patient's health status (Duggleby et al., 2017; Jones et al., to detect significant intervention effects for both partners and patients.
2013; Senneseth et al., 2017), or time and transportation issues Blanchard et al. (1996) reported decreased depression on the part of the
(Blanchard et al., 1996; Jones et al., 2013). Actual dissatisfaction with patients, but not their participating partners. This indirect effect might
the program was mentioned in several studies, but caused only few be due to positive intervention effects concerning the patient-partner-
dropouts (Blanchard et al., 1996; Carlson et al., 2017; Jones et al., relationship: Exit interviews revealed that the patients felt more sup-
2013; Sabo et al., 1986; Senneseth et al., 2017). ported and that the communication with their partners improved
Treatment compliance was defined on the basis of the number of through the intervention. Furthermore, patients reported that the in-
sessions participants actually attended. It was, when reported, on tervention helped them to worry less about being a burden on the
average high. The lowest reported adherence rate was 86%, that is 5.1 partner – a relief which possibly led to reduced emotional distress
of 6 sessions (Manne et al., 2004). (Blanchard et al., 1996). These findings support the view on coping
with cancer as a dyadic process (Kayser et al., 2007). Future research
3.6. Study quality should find out about the mechanisms that lead to a transfer of inter-
vention effects on non-participating patients. Many cancer patients feel
Table 3 provides a summary of the quality and risk of bias appraisal highly burdened. Finding out about how to alleviate emotional distress
of included studies based on the EPHPP. Five studies received a mod- for both partners and patients without the need for patients to actively
erate overall rating, and the rest were rated as being weak. Thus, many participate in the intervention could be a fruitful path for future re-
of the studies were at risk of bias. The results of the EPHPP revealed a search.
certain pattern of methodological strengths and weaknesses. Specifi- Due to the lack of well-established interventions, several studies
cally, the instruments used were reliable and validated in all of the adapted non-validated programs originally designed for other popula-
studies; all of the studies reported on drop-outs; the percentage of tions (Blanchard et al., 1996; Carlson et al., 2017; Manne et al., 2004).
61
Table 3
Quality ratings accoding to the EPHPP.
Study Representativen- Partcipation Study design Randomization Appropriateness of Differences Percentage of Blinding of Blinding of
ess of target rate randomization procedure between IG and confounders that outcome participants
population CG prior to were controlled assessors
interventiona
Blanchard et al. (1996) Somewhat likely 25% (low) RCT Yes Yes No N.a. No No
information
Bultz et al. (2000) Somewhat likely 31% (low) RCT Yes Yes No information No information No No
information
Carlson et al. (2017) Somewhat likely 18% (low) RCT Yes Yes No information No information No No
information
Duggleby et al. (2017) Not likely No information RCT Yes No information No N.a. No No
information
Jones et al. (2013) No likely No information Cohort No N.a. N.a. N.a. N.a. N.a.
Lewis et al. (2008) Somewhat likely No information Cohort No N.a. N.a. N.a. N.a. N.a.
Manne et al. (2004) Somewhat likely 57% (low) RCT Yes Yes No N.a. No No
information
Sabo et al. (1986) Not likely No information Cohort No N.a. N.a. N.a. N.a. N.a.
Senneseth et al. (2017) Somewhat likely 51.1% (low) RCT Yes Yes No N.a. No No
62
Study Validity of data Reliability of data Reporting of Percentage of Treatment Possibility of contamination Appropriateness of Analyses based on Overall
collection tools collection tools withdrawals and participants adherenceb (unintended intervention)b statistical methodsb intent to treatb rating
drop-outs completing the study
Blanchard et al. (1996) Yes Yes Yes 77% (moderate) 5.5/6 sessions on Yes Yes No Moderate
average (high)
Bultz et al. (2000) Yes Yes Yes 89% (high) 2 missed 1/6, 1 No Yes No Moderate
missed 3/6 sessions
Carlson et al. (2017) Yes Yes Yes 68% (moderate) No information Yes Yes No Moderate
Duggleby et al. (2017) Yes Yes Yes 70% (moderate) No information Yes Yes No Weak
Jones et al. (2013) Yes Yes Yes 87% (high) 91% attended at Yes Yes Yes Weak
least 4/5 sessions
Lewis et al. (2008) Yes Yes Yes 77% (moderate) No information Yes Yes No Weak
Manne et al. (2004) Yes Yes Yes 88% (high) 5.13/6 sessions on Yes Yes No Moderate
average (high)
Sabo et al. (1986) Yes Yes Yes 100% (high) No information Yes Yes No Weak
Senneseth et al. (2017) Yes Yes Yes 93.1% (high) N.a. (only one N.a. Yes No Moderate
session)
Note. IG = intervention group, CG = control group.

a
Referring to differences with regard to demographic variables and pre-intervention scores on outcome measures.
b
Not taken into account for the overall score.
This might have led to a failure to meet the specific needs of caregiving Recruitment strategies such as the use of multiple sites, the active in-
partners, and hence may also explain some of the null-findings volvement of healthcare providers and peers, as well as alternative
(Blanchard et al., 1996; Carlson et al., 2017). For example, Bultz et al. strategies such as mailing lists need to be explored (Ford et al., 2008;
(2000) and Carlson et al. (2017) utilized the same intervention ap- Jones et al., 2013).
proach. However, while Bultz and colleagues (2000) saw a considerable A related issue concerns the relevance of the findings. Statistically
improvement in the intervention group's total mood disturbance score significant findings do not necessarily indicate meaningful or practi-
(POMS) compared to the control group, Carlson and colleagues (2017) cally relevant intervention effects. Accordingly, researchers and clin-
failed to replicate this effect. They deduced that the null-finding might icians become increasingly interested in receiving information on the
be partly due to the fact that the intervention was originally developed clinical relevance of research findings (Ogles et al., 2001). There was
for partners of breast cancer patients and was thus not suitable for only one study included in the current review that reported on the
addressing the needs of prostate cancer patients. In the future, care clinical relevance of their findings (Lewis et al., 2008). Specifically,
should be taken to develop and deliver programs that are tailored to the Lewis and colleagues (2008) used norm values and cutoff scores to
specific needs of certain subgroups (e.g., male partners of breast cancer evaluate whether participants scored in the normal or clinical range at
patients or female partners of prostate cancer patients). study entry and post-intervention. Manne and colleagues (2004) also
A related issue concerns the gender distribution in the samples. used norm values to classify the participants’ scores at study entry, but
Most of the intervention studies included in the current review target omitted to provide information on how the results of those scoring in
partners of either prostate or breast cancer patients. Previous research the clinical range changed over the course of the study as compared to
has shown that partners of cancer patients adjust differently to the individuals scoring in the normal range at study entry. Reporting on the
patients’ illness depending on the partner's gender with regard to per- clinical relevance of research findings not only enables researchers to
ceived burden, emotional well-being as well as caregiving behavior evaluate results more critically, but also serves as an important mean
(e.g., Glantz et al., 2009; Hagedoorn et al., 2000; Pitceathly and for the derivation of practical recommendations (please see Ogles et al.,
Maguire, 2003). Accordingly, researchers should concentrate on de- 2001 for an overview of methods for defining improvement in clinical
veloping interventions that take the specific needs of female or male trials). Another aspect of clinical relevance concerns the long-term ef-
partners of cancer patients into account. fects of interventions. Five of the nine studies included in the current
Research indicates that the level and form of distress partners ex- review applied follow-up measurements. The duration of follow-up
hibit greatly depends on the phase of the cancer illness trajectory the ranged from two to six months. Two studies reported significant in-
patient is in (Oberst and Scott, 1988; Sales, 1992). As such, null-find- tervention effects at three months follow-up (Bultz et al., 2000; Jones
ings may have been caused by specific needs of the partners having et al., 2013). These comparatively short follow-up periods decrease the
been neglected in the intervention. For example, Manne et al. (2004) meaningfulness of the findings. In order to detect long-term interven-
discussed the lack of impact seen on the emotional approach and action effects and thus contribute to the evaluation of clinical sig-
ceptance coping scales, positing that those items assessed processes nificance, future studies should establish longer follow-up periods.
which were more relevant to adjustments respondents usually make On the basis of dropout, lost to follow-up, and treatment compliance
during the first several months after their partner's diagnosis. As most of rates, it can be deduced that the interventions were generally well ac-
the wives in their sample had been diagnosed more than a few months cepted. However, response rates were low (between 18.1 and 57%)
previous to when they joined the study, missing treatment effects might indicating that researchers should explore more effective ways to re-
be due to these outcomes being less amendable to change. Strikingly, cruit partners of cancer patients for intervention studies. Duggleby et al.
two of the RCTs that detected intervention effects comprised partners of (2017), the only team who conducted an online study, were also the
patients that were relatively homogenous with regard to the cancer only researchers to report a comparatively high number of dropouts.
illness trajectory. Specifically, Bultz et al. (2000) included partners of Dropouts and low adherence rates are common problems among in-
cancer patients suffering from cancer stages I and II and Lewis et al. ternet-based interventions (Cavanagh, 2010; Melville et al., 2010;
(2008) included partners of cancer patients that were diagnosed within Wangberg et al., 2008). The familiarity with computer-based applica-
six months before recruitment and suffered from cancer stages 0 to III. tions will likely increase over the next few years and, consequently,
Both detected significant intervention effects. Possibly, these interven- online interventions may be an accessible and well accepted alternative
tions were specifically suitable in meeting the needs of partners of pa- to face-to-face interventions (Nguyen et al., 2004). Moreover, the re-
tients suffering from early-stage cancer. Accordingly, researchers straining threshold for taking part in online interventions may be lower
should consider the phase of the cancer illness trajectory the patient is for some individuals than participating in face-to-face interventions.
in and the resulting challenges partners have to deal with in order to This could pave the way for including highly burdened individuals or
develop interventions that are effective. those who may not be able to take part in face-to-face interventions
One main reason the studies cited for null-findings was small sample (e.g., due to long distances or time-consuming caregiving tasks). Ac-
size, a factor known to increase the risk of studies being underpowered cordingly, future research should aim at developing effective online
(Carlson et al., 2017; Duggleby et al., 2017). Some authors reported interventions for partners of cancer patients.
having had serious recruitment difficulties that resulted in small and One goal of our review was to evaluate the instruments that were
unrepresentative samples (Blanchard et al., 1996; Bultz et al., 2000; used to detect intervention effects. While psychometric properties (i.e.,
Lewis et al., 2008). Partners of cancer patients are a highly burdened reliability and validity) of the instruments were generally good, not all
population (Pitceathly and Maguire, 2003; Stenberg et al., 2010) and of them were suitable with regard to certain research aims and inter-
thus may consider participation in an intervention study too time- vention foci. For example, Blanchard and colleagues (1996) applied a
consuming or emotionally burdensome. Accordingly, most of the pa- short-term problem-solving approach with the aim of reducing pro-
tients whose partners participated in the intervention studies were blem-related distress. However, they measured a variety of outcomes
suffering from early stage or non-metastatic cancers; their partners (depression, functioning, social support, marital satisfaction, burden,
were often coping relatively well from the beginning and did not ex- coping) that do not seem appropriately targeted by the intervention,
hibit relevant mood disturbance symptoms at baseline. This led to especially taking the focus and the brevity of the program into account.
ceiling effects, i.e. treatment effects were difficult to detect as there was Bultz et al. (2000) measured social support using the functional social
little space for improvement (Lewis et al., 2008; Manne et al., 2004). support scale (Broadhead et al., 1988). However, their intervention
Much remains to be learned about how to effectively meet the challenge aimed at improving the marital relationship quality and not social
of recruiting more (highly burdened) partners in order to solve the support in general. The same applies to Carlson et al. (2017). While
problem of insufficient statistical power and ceiling effects. social support interventions may affect more distal variables, such as
63
quality of life or emotional distress (e.g., Senneseth et al., 2017), it to partners of cancer patients rather than cancer patients alone or both
seems unreasonable to assume that interventions that aim at improving partners and patients together in case they aim at enhancing the pa-
the marital relationship quality or alleviating emotional distress levels tient's and the partner's well-being. Patients who are highly burdened
readily affect general social support outcomes. Instead of measuring the due to the illness may not always be able to take part in face-to-face
entire spectrum of well-being outcomes, researchers are well advised to interventions. Offering the intervention to the partner and thereby
focus on the assessment of variables that suit the content and goals of enhancing the patient's well-being outcomes may be a feasible alter-
the intervention they delivered. In summary, we recommend the ap- native. However, more research has to be conducted in order to find out
plication of appropriate (i.e. well suited to the content of the inter- about the conditions under which treatment effects are transferred to
vention) and sufficient (i.e. covering a broad range of emotional and non-participating patients.
physical health) outcome variables, which measure both positive and Moreover, clinicians are well advised to take the intervention re-
negative aspects of the caregiving experience. Moreover, outcome cipients’ characteristics into account when selecting specific interven-
variables should account for the processes that are amendable to tion approaches. For example, the needs of female partners of prostate
change in different phases of caregiving with regard to the cancer ill- cancer patients may differ from those of male partners of breast cancer
ness trajectory. Researchers should provide a clear rationale for each patients and the needs of partners of cancer patients who suffer from
outcome measured based on theoretical considerations and related re- early-stage cancer may differ from those of partners of patients suf-
search findings. fering from a metastatic or terminal cancer disease.
4.2. Discussion of methodological weakness 5. Limitations of the review
It was notable that, according to the EPHPP, none of the trials Recent research indicates that the number and quality of partner
achieved a satisfactory methodological quality rating score. Five studies interventions in oncologic settings has improved considerably
received moderate ratings, and the rest were rated as being weak. Thus, (Northouse et al., 2012). Thus, it may seem a contradiction that only
many of the studies were at risk of bias. Strikingly, some aspects of nine publications met the inclusion criteria for the present paper.
methodological quality were rated as being high in all of the included However, it is possible that further partner interventions exist but they
studies, e.g., the reliability and validity of the instruments or the per- lack more comprehensive documentation. On the one hand, considering
centage of participants completing the study. On the other hand, the only published material ensures that the included studies are of a higher
representativeness of the target population was mostly not likely or quality. On the other hand, excluding unpublished studies likely in-
only somewhat likely and participation rate was mostly low indicating troduces an upward bias concerning the estimation of intervention ef-
that recruitment strategies should be expanded and researchers should fects. However, as most of the obtained studies reported statistically
not rely on evidence from single-center trials. While the categories of non-significant effects, we consider the influence of publication bias
the EPHPP are informative in that they provide certain criteria that leading to an overestimation of treatment effects rather negligible.
should be considered when planning intervention studies as well as The validity of a systematic review is strongly dependent on the risk
with regard to the evaluation of research findings, some criteria appear of bias present in the studies it includes (Noordzij et al., 2009). Some
to be less relevant than others. For example, while blinding of outcome researchers have addressed the issue of risk of bias that arises when
assessors should be aimed at, blinding of participants is difficult to there is a high degree of clinical and methodological heterogeneity
apply in case of psychological intervention studies that require in- between studies included in meta-analyses (Melsen et al., 2014;
formed consent. Moreover, recruitment might be more or less difficult Shibata, 2013). The present review comprises highly heterogenous
in different situations and for different target populations. For example, studies with regard to both clinical (e.g., patient population, diagnostic
in the case of recruiting partners of cancer patients for an intervention methods, interventions) and methodological (e.g., study design, defi-
study, a participation rate of over 50% appears to be comparably high nitions of outcomes) heterogeneity. We met the challenge of accounting
(and not low as suggested by the EPHPP). In summary it can be stated for clinical heterogeneity among the trials by providing a detailed re-
that in evaluating the methodological quality of intervention studies, port of the participants’ characteristics, the interventions’ content, as
researchers may rely on instruments such as the EPHPP but are well- well as the assessed variables and outcomes. The EPHPP tool provides a
advised to carefully assess the relevance of the criteria applied. In line rough estimation of the studies’ methodological quality. To assess the
with previous recommendations (see e.g., Guyatt et al., 2011) we full range of risk of bias, it is recommended to choose the domains that
strongly advise against using summary ratings since they entail the risk are to be assessed based on a combination of theoretical and empirical
of overestimating irrelevant aspects of methodological quality while at considerations rather than to estimate risk of bias using overall re-
the same time underestimating the impact of serious methodological sulting scores (Higgins et al., 2011). As we included trials comprising
flaws. That is, weights are arbitrary assigned to different criteria different study designs (such as RCTs and one group pre-post studies), it
(Guyatt et al., 2011). Altogether, more randomized controlled trials is likely that, due to different degrees of bias, the studies do not all
that meet high methodological standards are needed to verify the as- estimate the same effects (Gagnier et al., 2012).
sumptions made in the current review.
6. Conclusion
4.3. Practical implications
The status of research on effective ways of improving cancer pa-
Senneseth et al. (2017) found statistically significant improvements tients’ partners’ psychosocial wellbeing remains significantly limited.
with regard to perceived an received social support and descriptive Studies that address this population's needs are rare and their metho-
findings indicated positive intervention effects with regard to distress dological quality is typically moderate or weak. We were able to derive
and quality of life. Presumably, partners are more likely to receive the several implications for future research and practice. Specifically, in-
support they need when they are actively encouraged to reach out for terventions targeting partners of cancer patients often affect both
help (Gage, 2013; Ryan et al., 2008). Practitioners may consider in- partners and patients. Researchers need to find out about the me-
cluding the social network when offering support to partners of cancer chanisms that support the transfer of intervention effects on non-par-
patients. ticipating patients. Moreover, intervention approaches should match
We found that outcomes of cancer patients often match those of the needs of specific sub-populations (especially with regard to gender
participating partners (Bultz et al., 2000; Carlson et al., 2017; Jones and the cancer phase trajectory) in order to effectively increase the
et al., 2013). Thus, practitioners may consider delivering interventions partners’ well-being. Null-findings were likely due to small sample
64
sizes. Partners were often coping well already at baseline. Accordingly, Gage, E.A., 2013. The dynamics and processes of social support: Families’ experiences
effort should be made to recruit more and especially burdened in- coping with a serious paediatric illness. Sociol. Health Illness 35 (3), 405–418.
Gagnier, J.J., Moher, D., Boon, H., Beyene, J., Bombardier, C., 2012. Investigating clinical
dividuals. Only one of the included studies reported on the clinical heterogeneity in systematic reviews: A methodologic review of guidance in the lit-
relevance of their findings (Lewis et al., 2008). Future studies should erature. BMC Med. Res. Methodol. 12, 111. https://doi.org/10.1186/1471-2288-12-
apply cut-off scores and mean values to evaluate the clinical relevance 111.
Girgis, A., Lambert, S., Johnson, C., Waller, A., Currow, D., 2013. Physical, psychosocial,
(next to the statistical significance) of their findings both with regard to relationship, and economic burden of caring for people with cancer: A review. J.
post-intervention as well as follow-up outcomes. Another focus should Oncol. Pract. 9 (4), 197–202. https://doi.org/10.1200/JOP.2012.000690.
lie on the development of effective internet-based interventions since Given, B., Wyatt, G., Given, C., Sherwood, P., Gift, A., DeVoss, D., Rahbar, M., 2004.
Burden and depression among caregivers of patients with cancer at the end of life.
they may be a feasible alternative for highly burdened partners or in Oncol. Nurs. Forum 31 (6), 1105–1117. https://doi.org/10.1188/04.ONF.1105-
case participation in face-to-face interventions is not feasible. Finally, 1117.
researchers should provide clear rationale for the assessment of certain Glantz, M.J., Chamberlain, M.C., Liu, Q., Hsieh, C.-C., Edwards, K.R., van Horn, A., Recht,
L., 2009. Gender disparity in the rate of partner abandonment in patients with serious
outcomes. They need to carefully consider whether outcome variables
medical illness. Cancer 115 (22), 5237–5242.
account for the processes that are targeted in the intervention. Grunfeld, E., 2004. Family caregiver burden: Results of a longitudinal study of breast
cancer patients and their principal caregivers. Can. Med. Assoc. J. 170 (12),
Conflict of interest 1795–1801. https://doi.org/10.1503/cmaj.1031205.
Guyatt, G.H., Oxman, A.D., Vist, G., Kunz, R., Brozek, J., Alonso-Coello, P., …, Falck-
Ytter, Y., 2011. GRADE guidelines: 4. Rating the quality of evidence—study limita-
The authors certify that there is no conflict of interest regarding the tions (risk of bias). J. Clin. Epidemiol. 64 (4), 407–415.
material discussed in the manuscript. Hagedoorn, M., Buunk, B.P., Kuijer, R.G., Wobbes, T., Sanderman, R., 2000. Couples
dealing with cancer: Role and gender differences regarding psychological distress and
quality of life. Psycho-Oncology 9 (3), 232–242.
Acknowledgement Higgins, J.P.T., Altman, D.G., Gøtzsche, P.C., Jüni, P., Moher, D., Oxman, A.D., …, Sterne,
J.A.C., 2011. The Cochrane Collaboration's tool for assessing risk of bias in rando-
mised trials. BMJ (Clinical Research Ed.) 343, d5928.
The study was supported by a grant from the Deutsche José Carreras Jackson, N., Waters, E., 2005. Criteria for the systematic review of health promotion and
Leukämie-Stiftung (Trial registration: DJCLS 23R/2016). public health interventions. Health Promotion Int. 20 (4), 367–374. https://doi.org/
10.1093/heapro/dai022.
Jensen, S., Given, B.A., 1991. Fatigue affecting family caregivers of cancer patients.
References Cancer Nurs. 14 (4), 181–187.
Jones, J.M., Lewis, F.M., Griffith, K., Cheng, T., Secord, S., Walton, T., …, Catton, P.,
Armijo-Olivo, S., Stiles, C.R., Hagen, N.A., Biondo, P.D., Cummings, G.G., 2012. 2013. Helping Her Heal-Group: A pilot study to evaluate a group delivered educa-
Assessment of study quality for systematic reviews: A comparison of the Cochrane tional intervention for male spouses of women with breast cancer. Psycho-Oncology
Collaboration Risk of Bias Tool and the Effective Public Health Practice Project 22 (9), 2102–2109. https://doi.org/10.1002/pon.3263.
Quality Assessment Tool: methodological research. J. Eval. Clin. Pract. 18 (1), 12–18. Kayser, K., Watson, L.E., Andrade, J.T., 2007. Cancer as a “we-disease”: Examining the
https://doi.org/10.1111/j.1365-2753.2010.01516.x. process of coping from a relational perspective. Families Syst. Health 25 (4),
Arno, P.S., Levine, C., Memmott, M.M., 1999. The economic value of informal caregiving. 404–418. https://doi.org/10.1037/1091-7527.25.4.404.
Health Affairs (Project Hope) 18 (2), 182–188. https://doi.org/10.1377/hlthaff.18.2. Lambert, S.D., Harrison, J.D., Smith, E., Bonevski, B., Carey, M., Lawsin, C., …, Girgis, A.,
182. 2012. The unmet needs of partners and caregivers of adults diagnosed with cancer: A
Bevans, M., Sternberg, E.M., 2012. Caregiving burden, stress, and health effects among systematic review. BMJ Support. Palliative Care. https://doi.org/10.1136/
family caregivers of adult cancer patients. JAMA 307 (4), 398–403. https://doi.org/ bmjspcare-2012-000226. bmjspcare-2012-000226.
10.1001/jama.2012.29. Lewis, F.M., Cochrane, B.B., Fletcher, K.A., Zahlis, E.H., Shands, M.E., Gralow, J.R., …,
Blanchard, C.G., Toseland, R.W., McCallion, P., 1996. The effects of a problem-solving Schmitz, K., 2008. Helping Her Heal: A pilot study of an educational counseling in-
intervention with spouses of cancer patients. J. Psychos. Oncol. 14 (2), 1–21. https:// tervention for spouses of women with breast cancer. Psycho-Oncology 17 (2),
doi.org/10.1300/J077v14n02_01. 131–137. https://doi.org/10.1002/pon.1203.
Blum, K., Sherman, D.W., 2010. Understanding the experience of caregivers: A focus on Liberati, A., Altman, D.G., Tetzlaff, J., Mulrow, C., Gøtzsche, P.C., Ioannidis, J.P.A., …,
transitions. Semin. Oncol. Nurs. 26 (4), 243–258. https://doi.org/10.1016/j.soncn. Moher, D., 2009. The PRISMA statement for reporting systematic reviews and meta-
2010.08.005. analyses of studies that evaluate health care interventions: Explanation and ela-
Braun, M., Mikulincer, M., Rydall, A., Walsh, A., Rodin, G., 2007. Hidden morbidity in boration. PLoS Med. 6 (7), e1000100. https://doi.org/10.1371/journal.pmed.
cancer: Spouse caregivers. J. Clin. Oncol. 25 (30), 4829–4834. https://doi.org/10. 1000100.
1200/JCO.2006.10.0909. Manne, S., Babb, J., Pinover, W., Horwitz, E., Ebbert, J., 2004. Psychoeducational group
Broadhead, W.E., Gehlbach, S.H., Gruy, F.V.de, Kaplan, B.H., 1988. The Duke? UNC intervention for wives of men with prostate cancer. Psycho-Oncology 13 (1), 37–46.
Functional Social Support Questionnaire. Med. Care 26 (7), 709–723. https://doi. https://doi.org/10.1002/pon.724.
org/10.1097/00005650-198807000-00006. McLean, L.M., Jones, J.M., 2007. A review of distress and its management in couples
Bultz, B.D., Speca, M., Brasher, P.M., Geggie, P.H., Page, S.A., 2000. A randomized facing end-of-life cancer. Psycho-Oncology 16 (7), 603–616. https://doi.org/10.
controlled trial of a brief psychoeducational support group for partners of early stage 1002/pon.1196.
breast cancer patients. Psycho-Oncology 9 (4), 303–313. Melsen, W.G., Bootsma, M.C.J., Rovers, M.M., Bonten, M.J.M., 2014. The effects of
Carlson, L.E., Bultz, B.D., Speca, M., St. Pierre, M., 2008. Partners of cancer patients. J. clinical and statistical heterogeneity on the predictive values of results from meta-
Psychosocial Oncol. 18 (2), 39–63. https://doi.org/10.1300/J077v18n02_03. analyses. Clin. Microbiol. Infect. 20 (2), 123–129. https://doi.org/10.1111/1469-
Carlson, L.E., Bultz, B.D., Speca, M., St-Pierre, M., 2000. Partners of cancer patients. J. 0691.12494.
Psychosocial Oncol. 18 (3), 33–43. https://doi.org/10.1300/J077v18n03_03. Melville, K.M., Casey, L.M., Kavanagh, D.J., 2010. Dropout from Internet-based treatment
Carlson, L.E., Rouleau, C.R., Speca, M., Robinson, J., Bultz, B.D., 2017. Brief supportive- for psychological disorders. Br. J. Clin. Psychol. 49 (Pt 4), 455–471. https://doi.org/
expressive group therapy for partners of men with early stage prostate cancer: 10.1348/014466509X472138.
Lessons learned from a negative randomized controlled trial. Support. Care Cancer 25 Nguyen, H.Q., Carrieri-Kohlman, V., Rankin, S.H., Slaughter, R., Stulbarg, M.S., 2004.
(4), 1035–1041. https://doi.org/10.1007/s00520-016-3551-1. Internet-based patient education and support interventions: A review of evaluation
Cavanagh, K., 2010. Turn on, tune in and (don’t) drop out: Engagement, adherence, at- studies and directions for future research. Comput. Biol. Med. 34 (2), 95–112.
trition, and alliance with internet-based interventions. Oxford Guide to Low Intensity https://doi.org/10.1016/S0010-4825(03)00046-5.
CBT Interventions 227, 232. Noordzij, M., Hooft, L., Dekker, F.W., Zoccali, C., Jager, K.J., 2009. Systematic reviews
Duggleby, W., Ghosh, S., Struthers-Montford, K., Nekolaichuk, C., Cumming, C., Thomas, and meta-analyses: When they are useful and when to be careful. Kidney Int. 76 (11),
R., …, Swindle, J., 2017. Feasibility study of an online intervention to support male 1130–1136. https://doi.org/10.1038/ki.2009.339.
spouses of women with breast cancer. Oncol. Nursing Forum 44 (6), 765–775. Northouse, L., Williams, A.-l., Given, B., McCorkle, R., 2012. Psychosocial care for family
https://doi.org/10.1188/17.ONF.765-775. caregivers of patients with cancer. J. Clin. Oncol. 30 (11), 1227–1234. https://doi.
Effective Public Health Practice Project, 1998. Quality Assessment Tool For Quantitative org/10.1200/JCO.2011.39.5798.
Studies. Retrieved from https://merst.ca/ephpp/. Northouse, L.L., Katapodi, M.C., Song, L., Zhang, L., Mood, D.W., 2010. Interventions
Ferrell, B., Wittenberg, E., 2017. A review of family caregiving intervention trials in with family caregivers of cancer patients: Meta-analysis of randomized trials. CA: a
oncology. CA: a Cancer Journal for Clinicians 67 (4), 318–325. https://doi.org/10. Cancer J. Clin. 60 (5), 317–339. https://doi.org/10.3322/caac.20081.
3322/caac.21396. Oberst, M.T., Scott, D.W., 1988. Postdischarge distress in surgically treated cancer pa-
Ford, J.G., Howerton, M.W., Lai, G.Y., Gary, T.L., Bolen, S., Gibbons, M.C., …, Bass, E.B., tients and their spouses. Res. Nurs. Health 11 (4), 223–233. https://doi.org/10.1002/
2008. Barriers to recruiting underrepresented populations to cancer clinical trials: A nur.4770110404.
systematic review. Cancer 112 (2), 228–242. https://doi.org/10.1002/cncr.23157. Ogles, B.M., Lunnen, K.M., Bonesteel, K., 2001. Clinical significance: History, application,
Frambes, D., Given, B., Lehto, R., Sikorskii, A., Wyatt, G., 2017. Informal caregivers of and current practice. Clin. Psychol. Rev. 21 (3), 421–446.
cancer patients: review of interventions, care activities, and outcomes. Western J. O’Toole, M.S., Zachariae, R., Renna, M.E., Mennin, D.S., Applebaum, A., 2017. Cognitive
Nurs. Res. https://doi.org/10.1177/0193945917699364. 193945917699364. behavioral therapies for informal caregivers of patients with cancer and cancer
65
survivors: A systematic review and meta-analysis. Psycho-Oncology 26 (4), 428–437. Senneseth, M., Dyregrov, A., Laberg, J., Matthiesen, S.B., Pereira, M., Hauken, M.A.,
https://doi.org/10.1002/pon.4144. 2017. Facing spousal cancer during child-rearing years: The short-term effects of the
Pitceathly, C., Maguire, P., 2003. The psychological impact of cancer on patients’ partners cancer-pepsone programme—a single-center randomized controlled trial. Psycho-
and other key relatives. Eur. J. Cancer 39 (11), 1517–1524. https://doi.org/10.1016/ Oncology. https://doi.org/10.1002/pon.4329. Advance online publication.
S0959-8049(03)00309-5. Shibata, M.C., 2013. What is wrong with meta-analysis? The importance of clinical het-
Robert Koch-Institut (RKI), 2016. Bericht zum Krebsgeschehen in Deutschland 2016. erogeneity in myocardial regeneration research. Int. J. Clin. Pract. 67 (11),
Ryan, P.J., Howell, V., Jones, J., Hardy, E.J., 2008. Lung cancer, caring for the caregivers. 1081–1085. https://doi.org/10.1111/ijcp.12232.
A qualitative study of providing pro-active social support targeted to the carers of Stenberg, U., Ruland, C.M., Miaskowski, C., 2010. Review of the literature on the effects
patients with lung cancer. Palliative Med. 22 (3), 233–238. https://doi.org/10.1177/ of caring for a patient with cancer. Psycho-Oncology 19 (10), 1013–1025. https://
0269216307087145. doi.org/10.1002/pon.1670.
Sabo, D., Brown, J., Smith, C., 1986. The male role and mastectomy. J. Psychosocial Thomas, B.H., Ciliska, D., Dobbins, M., Micucci, S., 2004. A process for systematically
Oncol. 4 (1–2), 19–31. https://doi.org/10.1300/J077v04n01_02. reviewing the literature: Providing the research evidence for public health nursing
Sales, E., 1992. Psychosocial impact of the phase of cancer on the family. J. Psychosocial interventions. Worldviews Evid.-Based Nurs. 1 (3), 176–184. https://doi.org/10.
Oncol. 9 (4), 1–18. https://doi.org/10.1300/J077v09n04_01. 1111/j.1524-475X.2004.04006.x.
Schubart, J.R., Kinzie, M.B., Farace, E., 2008. Caring for the brain tumor patient: Family Wangberg, S.C., Bergmo, T.S., Johnsen, J.-A.K., 2008. Adherence in Internet-based in-
caregiver burden and unmet needs. Neuro-Oncology 10 (1), 61–72. https://doi.org/ terventions. Pat. Prefer. Adher. 2, 57–65.
10.1215/15228517-2007-040.
66

Interventii PT Apartinatori

Uploaded by

Copyright:

Available Formats

Interventii PT Apartinatori

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Interventii PT Apartinatori

Uploaded by

Copyright:

Available Formats

Critical Reviews in Oncology / Hematology 140 (2019) 52–66

Contents lists available at ScienceDirect

Critical Reviews in Oncology / Hematology

Psychological interventions targeting partners of cancer patients: A T

ARTICLE INFO ABSTRACT

Fig. 1. Search terms.

2.1.4. Studies intervention, duration of the intervention, number of sessions and

Fig. 2. Flow chart of study selection process.

Randomized controlled trials

Randomized controlled trials

intervention period, points, period of participants (SD or female in criteria control

Randomized controlled trials

intervention period, points, period of participants (SD or female in criteria control

General emotional Total mood: Total mood: Psychological distress: Psychological

Note. IG = intervention group, CG = control group.

4.2. Discussion of methodological weakness 5. Limitations of the review

You might also like