See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/299541086

Maladaptive coping in adults who have experienced early parental loss and
grief counseling

Article in Journal of Health Psychology · March 2016

DOI: 10.1177/1359105316638550

CITATIONS READS

42 4,162

10 authors, including:

Beverley Lim Høeg Charlotte Appel

Danish Cancer Society Regional Hospital Silkeborg
28 PUBLICATIONS 279 CITATIONS 22 PUBLICATIONS 349 CITATIONS

SEE PROFILE SEE PROFILE

Annika von Heymann Christoffer Johansen

Rigshospitalet Rigshospitalet
26 PUBLICATIONS 366 CITATIONS 773 PUBLICATIONS 29,072 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Christoffer Johansen on 10 May 2016.

The user has requested enhancement of the downloaded file.

638550
research-article2016
HPQ0010.1177/1359105316638550Journal of Health PsychologyHøeg et al.

Article

Journal of Health Psychology

Maladaptive coping in adults

1­–11
© The Author(s) 2016
Reprints and permissions:
who have experienced early sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/1359105316638550
parental loss and grief counseling hpq.sagepub.com

Beverley Lim Høeg1, Charlotte W Appel1,

Annika B von Heymann-Horan1,
Kirsten Frederiksen1, Christoffer Johansen1,2,
Per Bøge3, Annemarie Dencker3, Atle Dyregrov4,
Birgit B Mathiesen5 and Pernille E Bidstrup1

Abstract
This study compares maladaptive coping, measured as substance use, behavioral disengagement, self-blame,
and emotional eating, among adults (>18 years) who have experienced early parental loss (N = 1465 women,
N = 331 men) with non-bereaved controls (N = 515 women, N = 115 men). We also compared bereaved
adults who received grief counseling (N = 822 women, N = 190 men) with bereaved controls who had
not (N = 233 women, N = 66 men). Bereaved adults reported significantly more substance use, behavioral
disengagement, and emotional eating than non-bereaved adults. Counseling participants reported significantly
more substance use and self-blame than non-participants. Our results suggest that early loss may negatively
impact the development of adulthood coping.

Keywords
children, coping, counseling, grief, parental loss

The death of a parent in childhood is a painful stress (Folkman and Moskowitz, 2004). It is
experience, with potentially long-term conse- mobilized by appraisal, a process whereby an
quences that may impact adult psychological external and/or internal demand is evaluated
health (Bowlby, 1980; Haine et al., 2008). Early against a person’s resources (Lazarus and
parental loss has been linked to increased risks
of depression, anxiety, and substance use in 1Danish Cancer Society Research Center, Denmark
adulthood (e.g. Appel et al., 2013; Hamdan 2Copenhagen University Hospital, Rigshospitalet,
et al., 2013), indicating a lowered ability to Denmark
3Danish Cancer Society, Denmark
cope with the stresses of life (Auerbach et al., 4Center for Crisis Psychology, Norway
2010). However, to our knowledge, no study 5University of Copenhagen, Denmark
has investigated the long-term coping behavior
of people who have lost a parent or who have Corresponding author:
Beverley Lim Høeg, Survivorship Unit, Danish Cancer
received grief counseling. Society Research Center, Strandboulevarden 49, 2100
“Coping” refers to the cognitive and behav- Copenhagen, Denmark.
ioral strategies that individuals use to manage Email: bevlim@cancer.dk

Downloaded from hpq.sagepub.com by guest on April 12, 2016

2 Journal of Health Psychology 

Folkman, 1984). Appraisal is influenced by adulthood in relation to early parental loss and
multiple factors, including experiences across participation in grief counseling. First, we hypoth-
time (Lazarus, 1999). Within a life-course per- esized that parental loss is associated with disrup-
spective, coping behaviors are seen as driven tion of healthy coping development and that
by developing systems, such as language and bereaved adults would report greater use of mala-
cognition, and shaped by person–environment daptive coping strategies than non-bereaved
interactions (Zimmer-Gembeck and Skinner, adults. Second, we hypothesized that, by facilitat-
2011). ing the bereavement process, receiving counseling
Parents play an essential role for adaptive cop- would be associated with the development of
ing with stress (Gunnar and Cheatham, 2003). A adaptive coping, and that counseling participants
secure parent–child attachment aids the develop- would thus report less use of maladaptive strate-
ment of emotional regulation (Waters et al., gies than non-participants. Additionally, we inves-
2010), while consistent parental support increases tigated whether child gender, which parent was
the likelihood of successful coping experiences, lost, child’s age at the time of loss, and the per-
fostering more flexible coping capabilities as the ceived presence of family support played moder-
child matures (Zimmer-Gembeck and Skinner, ating roles in this association.
2011). The impact of losing a parent may depend
upon the child’s age, which determines the cogni-
tive capability to understand and cope with the Method
death (Webb, 2010).
Grief counseling aims to facilitate the
Procedure
bereavement process, that is, to help the Data were drawn from a larger cross-sectional
bereaved person adapt to the loss and resolve study of early parental loss and grief counseling,
grief (Worden, 2009). While most children which combined register-based information with
adapt without professional help (Akerman and self-reported questionnaire responses (Appel
Statham, 2011), some experience high levels of et al., submitted). In Denmark, all residents have
impairment warranting professional support been recorded in the Central Population Register
(Worden, 2009). Two meta-analyses have sum- (CPR) since 1 April 1968 (Pedersen et al., 2006)
marized the limited studies on effects of grief with unique personal identification numbers,
counseling in bereaved children (age range, containing information on sex, date of birth,
5–18 years) (Currier et al., 2007; Rosner et al., family linkage, migration, and death.
2010). The first meta-analysis, of 13 controlled Three main nationwide organizations have
studies, showed an overall average weighted offered free telephonic, group, and/or individ-
effect size of Cohen’s d = 0.14 (p = 0.08), indi- ual grief counseling to children and young
cating no significant treatment effect. The sec- adults. We identified participants who had
ond meta-analysis, of 13 controlled and 12 received grief counseling at these centers
uncontrolled studies, showed overall effect between 1999 and 2009 and established a sam-
sizes of Hedges’ g = 0.35 (p < 0.01) and 0.49 ple of 1811 participants who met the criteria of
(p < 0.001), respectively, indicating small to being over 18 years of age at 31 December 2010
moderate treatment effects. None of the studies and having lost one or both parents before the
considered adult outcomes. A single study, of age of 30. We identified a second sample of
the Family Bereavement Program (Sandler 1803 bereaved participants from the CPR, who
et al., 2003), included positive coping as an out- met the same conditions but were not on center
come and found effects for girls at the 11-month files, and frequency matched them to the first
follow-up. group on gender, age, and time since parental
In this study, we compared maladaptive coping death. A third sample of 1853 CPR-based non-
strategies (substance use, behavioral disengage- bereaved controls were frequency matched on
ment, self-blame, and emotional eating) in gender and age.

Downloaded from hpq.sagepub.com by guest on April 12, 2016

Høeg et al. 3

In all, 5467 participants (4045 females, 1422 “We would like to know if you participated in
males) were identified and invited to complete professional support at the time your parent died
the questionnaire by mail. Informed consent is and if so, how many times did you participate?”
not required for questionnaire studies in Each support option (“psychologist,” “bereave-
Denmark; however, information on the project ment group,” “internet-based support,” “support
and the contact details of the primary researcher by phone,” “general practitioner,” “priest,”
were provided. Non-responders were reminded “nurse,” or “other”) could have answers ranging
after 3 weeks and again, by telephone, after a fur- from “Never” to “>40 sessions.” A response of
ther 3 weeks. The study protocol was approved “Never” for all the categories was considered
by the Danish Data Protection Agency (Record self-reported No-GC, while all other responses
no. 2009-41-3506). were considered Self-GC.

Maladaptive coping. Coping was measured

Study sample using the Brief COPE inventory (Carver, 1997),
A total of 2574 people completed the question- which was translated into Danish by two
naire (response rate, 47%). We excluded 12 research psychologists and, subsequently, back-
bereaved respondents who had lost a parent translated into English by a third bilingual psy-
after the age of 29 years, 67 whose self- chologist. The back-translation corresponded
reported parental loss was missing or incon- semantically to the original version. It consists
gruent with register information, and 36 of 14 subscales of two items each that are
participants from center files who self-reported answered on a 4-point Likert scale, ranging
no counseling, obtaining a group of confirmed from 1 (“I haven’t been doing this at all”) to 4
recipients of grief counseling (Conf-GC). We (“I’ve been doing this a lot”). We focused on the
found that 449 (60%) of bereaved participants subscales of substance use, behavioral disen-
in the CPR sample self-reported having gagement, and self-blame. The range of scores
received some kind of counseling (Self-GC), for these subscales was 2–8. One additional
while 299 (40%) confirmed no counseling item (“I have eaten for comfort”) was added in
(No-GC). Finally, we excluded 69 respondents the questionnaire, giving a one-item emotional-
who did not complete the questionnaire and eating subscale, with a score range of 1–4. The
obtained a final study sample of 2390 partici- Brief COPE scales have good internal consist-
pants, as shown in Figure 1. ency, construct validity, and adequate test–
retest reliability (Cooper et al., 2008), although
the Danish version has yet to be validated.
Measures
Early parental loss (loss). Loss was defined as Covariates. Participant’s gender, age at time of
losing one or both parents before the age of 30. loss (0–5, 6–12, 13–18, ⩾19 years), education
We included individuals up to this age because (basic school, high school or vocational, higher
previous studies suggest that loss during young education, unknown), parent lost (father,
adulthood continues to influence later psycho- mother, both), and perceived family support
logical functioning (Appel et al., in press; Nick- (yes, no) were identified as covariates. Gender
erson et al., 2013). Loss was confirmed through and educational level have been related to cop-
the CPR and cross-checked with the question- ing styles (Christensen et al., 2006; Matud,
naire items, “Did your father/mother die before 2004), and child’s educational level has been
you were 30 years of age?” linked to parental educational level (Dubow
et al., 2009), which is in turn associated with
Grief counseling (counseling). Participation in mortality rates (Montez et al., 2012). The parent
counseling was determined from center files and lost, participant age at time of loss, and family
checked from replies to the questionnaire item, support have been shown to affect bereavement

Downloaded from hpq.sagepub.com by guest on April 12, 2016

4 Journal of Health Psychology 

Figure 1. Flowchart of study participants.

Final sample, n = 2426.

outcomes (Raveis et al., 1999; Stroebe et al., could be answered with “high,” “moderate,”
2006). “low,” “none,” “don’t know,” or “have none.”
Participant gender, birth date, parent lost, Responses were dichotomized into a Yes/No
and date of death were obtained by linkage perceived family support variable, where high
with the CPR, while participants self-reported and moderate reported levels of support were
level of education and family support at the coded as “Yes” and the remaining responses as
time of the death. Perceived family support “No.”
was measured from answers to the item, “What
level of support did you experience from the
Statistical analyses
persons around you when your parent died?”
Each person option (“parent,” “siblings,” Descriptive statistics. Descriptive comparisons
“grandparents,” and “other family members”) of bereaved and non-bereaved participants were

Downloaded from hpq.sagepub.com by guest on April 12, 2016

Høeg et al. 5

made according to gender, age, and educational 1993). All analyses were carried out using SAS
level. Bereaved participants were further com- Enterprise Guide 5.1.
pared by the gender of the deceased parent, age
at time of loss, and perceived family support
Results
according to counseling group (Conf-GC, Self-
GC, and No-GC). Chi-squared tests were used Study sample
for categorical variables and t tests for continu-
ous variables. Table 1 shows the characteristics of the full
study sample. There were no significant differ-
Parental loss and coping. Multivariable linear ences by bereavement status, but there were
regression models were used to examine the significant differences by counseling status.
mean differences in substance use, behavioral Those in the Self-GC group tended to be
disengagement, self-blame, and emotional eat- younger, with correspondingly lower levels of
ing according to bereavement status. As histo- attained education, and had experienced loss at
grams indicated that scores on the maladaptive a younger age than the Conf-GC and No-GC
subscales peaked for the response that the groups.
behavior was not used, we also used logistic
regression to estimate differences in behavior
Parental loss and coping
use (substance use, behavioral disengagement,
self-blame >2; emotional eating >1) according Bereaved participants scored significantly
to bereavement status and expressed as odds higher for substance use (adjusted mean dif-
ratios. As log and linear models gave the same ference (MD) = 0.21, 95% confidence interval
conclusions, linear regression was used for the (CI) [0.11, 0.30]), behavioral disengagement
remaining analyses to allow possible compari- (MD = 0.13, 95% CI [0.02, 0.25]), and emo-
sons with the other scales that were more nor- tional eating (MD = 0.15, 95% CI [0.08,
mally distributed. Analyses were carried out 0.23]) but not for self-blame (MD = 0.00,
unadjusted as well as adjusted for age and edu- 95% CI [−0.13, 0.14]). The odds ratios with
cational level. 95 percent CIs gave similar results (Table 2).
To investigate whether the differences Only perceived family support moderated the
between bereaved and non-bereaved partici- association between parental death and
pants were moderated by gender, parent lost, maladaptive coping on all four subscales
age at time of loss, and perceived family sup- (p < 0.0001 for all subscales), while parent
port, we added an interaction term between lost and participant’s age at time of loss had a
bereavement status and each of the four modi- moderating effect only on substance use
fiers in separate models. The mean changes in (p = 0.03 and p < 0.0001, respectively; results
coping scores were estimated, and effect modi- not shown).
fication was tested in F tests.
Grief counseling and coping
Grief counseling and coping. Multivariable linear
regression models were used to compare differ- Grief counseling recipients and non-recipients
ences in mean coping scores between grief did not differ significantly on the scales of
counseling participants (Conf-GC and Self- behavioral disengagement and emotional eat-
GC) and No-GC. Analyses were unadjusted or ing; however, Conf-GC participants reported
adjusted for participant’s gender, parent lost, significantly more substance use (adjusted
age at time of loss, educational level, and per- MD = 0.16, 95% CI [0.01, 0.30], p < 0.05) and
ceived family support, which may be associated self-blame (MD = 0.32, 95% CI [0.13, 0.52],
with both receiving counseling and coping p < 0.05) than No-GC participants. The unad-
(Holahan and Moos, 1987; Vessey and Howard, justed and adjusted results are shown in Table 3.

Downloaded from hpq.sagepub.com by guest on April 12, 2016

6 Journal of Health Psychology 

Table 1. Characteristics of the full sample by bereavement status and of the bereaved sample by grief
counseling status (group percentages in parentheses).

Characteristic Full sample p value Bereaved sample p value

χ2 χ2
Bereaved Non- Conf-GC Self-GC No-GC
(n = 1796) bereaved (n = 1012) (n = 449) (n = 299)
(n = 630)
Gender
Male 331 (18) 115 (18) 0.92 190 (19) 67 (15) 66 (22) 0.04
Female 1465 (82) 515 (82) 822 (81) 382 (85) 233 (78)
Age
19–25 years 717 (40) 243 (39) 0.24 396 (39) 211 (47) 100 (33) <0.0001
26–30 years 535 (30) 179 (28) 298 (29) 136 (30) 85 (28)
31–35 years 434 (24) 155 (25) 257 (25) 88 (20) 82 (27)
>36 years 110 (6) 53 (8) 61 (6) 14 (3) 32 (11)
Educational level
Basic school 108 (6) 23 (4) 47 (5) 41 (9) 18 (6)
 High school or 604 (34) 207 (33) 319 (32) 179 (40) 97 (32)
vocational
Higher education 1078 (60) 399 (63) 643 (64) 226 (50) 184 (62)
Unknown 6 (0) 1 (0) 0.10 3 (0) 3 (1) 0 <0.0001
Age at time of loss
<5 years 23 (1) n/a 9 (1) 6 (1) 7 (2) 0.0014
6–12 years 239 (13) n/a 133 (13) 70 (16) 30 (10)
13–18 years 574 (32) n/a 312 (31) 166 (37) 82 (27)
>18 years 960 (53) n/a 558 (55) 207 (46) 180 (60)
Perceived family support
Yes 1641 (91) n/a 922 (91) 419 (93) 269 (90) 0.22
No 155 (9) n/a 90 (9) 30 (7) 30 (10)
Parent lost
Father 1003 (56) n/a 499 (49) 271 (60) 209 (70) <0.001
Mother 707 (39) n/a 442 (44) 168 (37) 87 (29)
Both 86 (5) n/a 71 (7) 10 (2) 3 (1)

Conf-GC: confirmed grief counseling; Self-GC: self-reported counseling; No-GC: self-reported no counseling; n/a: not
available.
Some percentages may not add up to 100 due to rounding. Age calculated at 1 March 2012 (mid-point of questionnaire
collection period). Participants identified from the counseling center files who self-reported no counseling (n = 36) were
included in the overall bereaved sample but excluded from the bereaved sample by counseling status.

Discussion (Giordano et al., 2014; Hamdan et al., 2013).

Those who lost both parents and those who expe-
Our results underline the important role of paren- rienced death between the ages of 6 and 18 years
tal loss in coping strategies. In accordance with appeared most vulnerable to substance use as
our first hypothesis, bereaved adults reported adults. For the other subscales, parent lost and
significantly higher substance use, behavioral child age at time of loss did not modify our
disengagement, and emotional eating than non- results, indicating that parental death may have a
bereaved adults. The finding on substance use long-term detrimental effect regardless of parent
bears out several longitudinal studies that found gender and child age.
higher levels of alcohol and substance abuse A new contribution of our study is the finding
among both youth and adults after parental loss that early parental loss is associated with

Downloaded from hpq.sagepub.com by guest on April 12, 2016

Høeg et al. 7

Table 2. Mean score, estimated mean differences, and odds ratio for use of coping behavior with
95 percent confidence intervals by bereavement status.

COPE Bereaved Non-bereaved MDa [95% CI] ORa [95% CI]

subscales
Mean No. with Mean No. with
score (SE) score score (SE) score
>2 (%) >2 (%)
Substance use 2.46 (0.02) 336 (19) 2.24 (0.04) 71 (11) 0.21 [0.11 to 0.30]* 1.79 [1.36 to 2.36]*
Behavioral 2.86 (0.03) 740 (41) 2.71 (0.05) 211 (34) 0.13 [0.02 to 0.24]* 1.35 [1.12 to 1.64]*
disengagement
Self-blame 3.22 (0.04) 968 (54) 3.21 (0.06) 319 (51) 0.00 [−0.13 to 0.14] 1.12 [0.93 to 1.34]
Emotional 1.63 (0.02) 745 (42) 1.47 (0.03) 217 (34) 0.15 [0.08 to 0.23]* 1.34 [1.11 to 1.62]*
eating

SE: standard error; MD: mean difference; OR: odds ratio; CI: confidence interval.
Sample size: n = 1796 bereaved, n = 630 non-bereaved.
aAdjusted for age and educational level.

*p < 0.05.

Table 3. Mean score and estimated mean differences with 95 percent confidence intervals for maladaptive
coping scales by counseling status: confirmed grief counseling (Conf-GC), self-reported counseling (Self-
GC), or self-reported no counseling (No-GC).

COPE scales Conf-GC Self-GC No-GC Conf-GC versus No- Self-GC versus No-GC
mean mean mean GC MD [95% CI] MD [95% CI]
score (SE) score (SE) score (SE)
Substance use
Unadjusted 2.50 (0.04) 2.44 (0.05) 2.33 (0.07) 0.17 [0.03 to 0.32]* 0.11 [−0.06 to 2.27]
Fully adjusteda 3.25 (0.49) 3.16 (0.49) 3.09 (0.49) 0.16 [0.01 to 0.30]* 0.07 [−0.09 to 0.23]
Behavioral disengagement
Unadjusted 2.83 (0.04) 2.92 (0.06) 2.85 (0.07) −0.02 [−0.18 to 0.14] 0.07 [−0.11 to 0.25]
Fully adjusted 3.15 (0.54) 3.19 (0.55) 3.17 (0.55) −0.02 [−0.18 to 0.14] 0.03 [−0.16 to 0.21]
Self-blame
Unadjusted 3.30 (0.05) 3.20 (0.07) 2.98 (0.09) 0.32 [0.13 to 0.51]* 0.22 [0.0081 to 0.44]*
Fully adjusted 4.06 (0.64) 3.92 (0.64) 3.73 (6.65) 0.32 [0.13 to 0.52]* 0.19 [−0.03 to 0.40]
Emotional eating
Unadjusted 1.66 (0.03) 1.61 (0.04) 1.58 (0.05) 0.08 [−0.03 to 0.19] 0.03 [−0.09 to 0.16]
Fully adjusted 1.28 (0.38) 1.20 (0.38) 1.22 (0.38) 0.06 [−0.05 to 0.18] −0.02 [−0.14 to 0.11]

SE: standard error; MD: mean difference; CI: confidence interval.

Sample size: n = 1012 Conf-GC, n = 449 Self-GC, n = 299 No-GC.
aAdjusted for participant gender, gender of deceased parent, participant age at time of loss, education level, and

perceived family support.

*p < 0.05.

behavioral disengagement and emotional eating experience of ultimate helplessness for a child
as coping strategies in adulthood. Disengagement and this vulnerability may interact with the sur-
refers to giving up efforts to handle a stressor viving parent’s limited resources to model and
and has been linked with low self-efficacy, the foster subsequent mastery experiences needed
lack of belief in one’s abilities (Bandura, 1977; for shaping high self-efficacy (Bandura, 1977;
Carver et al., 1989). The death of a parent is an Fan and Williams, 2009). Our finding on

Downloaded from hpq.sagepub.com by guest on April 12, 2016

8 Journal of Health Psychology 

emotional eating backs one previous study that baseline of maladaptive coping than the non-
found an association between parental bereave- counseling group. Second, we had no detailed
ment and youth obesity (Weinberg et al., 2013). information about the structure or content of the
A possible explanation may be that eating pat- grief counseling, and it is possible that the ser-
terns are influenced by family functioning and vices offered were insufficient for participants
quality of parenting (Rhee, 2008), two factors suffering from more severe and complicated
potentially affected by parental loss. Interestingly, grief reactions. Third, the possibility for grief
emotional eating may predict eating disorders counseling to be harmful has been posited by
(Stice et al., 2002), which in turn have been Jordan and Neimeyer (2003), although the
described as ways of coping with problems of methodology used in that study has since
personal control (Polivy and Herman, 2002). proved questionable (Larson and Hoyt, 2007).
Future research may investigate the possibility However, since children have a “short sadness
that parental death influences maladaptive cop- span” and a need to focus on non-grief related
ing through pathways involving efficacy and activities (Webb, 2010: 17), counseling may
control beliefs. force the child to dwell unnecessarily on aspects
The only factor that modified the association of grief and interfere with the natural oscillation
between loss and all four maladaptive coping toward restoration-oriented tasks, as described
scales in our study was perceived family sup- in the dual-process model of grief (Stroebe and
port. Significantly lower levels of maladaptive Schut, 1999). Screening may therefore be use-
coping according to bereavement status were ful as there is evidence that interventions for
found in adults who reported higher family sup- symptomatic participants produce better out-
port. Emotional support has a positive effect on comes than those with no selection criteria
bereavement outcomes by lowering psycholog- (Rosner et al., 2010).
ical distress (Raveis et al., 1999) and surviving This study has several strengths. This is the
parents, who are able to provide higher levels of first study to examine the role of parental loss in
emotional support to the grieving child, may adult coping. Also, our study is the largest on
also provide the stable attachment and security childhood bereavement and grief counseling. In
needed in the child’s environment post-loss. the latest meta-analysis on childhood grief
This highlights the need to target the family and interventions, the largest study had 230 partici-
“parenting capacity” of the surviving parent pants, while the numbers for the rest ranged
when intervening in bereaved children. from 17 to 87 (Rosner et al., 2010). The identi-
Contrary to our second hypothesis, we found fication of a population-based non-bereaved
no significant differences on behavioral disen- control group is also unique, made possible by
gagement and emotional eating between those national registries that were established years
who received counseling and those who did not. before this study was hypothesized thereby
Furthermore, both groups of participants who avoiding information, recall, and selection bias.
received counseling (Conf-GC and Self-GC) Also, grief counseling was available free of
reported significantly higher usage of substance charge and our sample of counseling recipients
use and self-blame than non-recipients, although were not recruited in an experimental setting,
the difference between Self-GC and non-recipi- hence increasing the ecological validity of the
ents (No-GC) became insignificant in the fully results and decreasing sample selection bias
adjusted model. These findings highlight a num- (Steele et al., 2012). We also matched our
ber of methodological and theoretical issues. groups on important confounders such as age,
First and most important, it is highly proba- gender, age at time of parental death, and time
ble that those who had sought and received since parental death.
grief counseling experienced higher distress However, the cross-sectional design limits
levels compared to those who did not, resulting the understanding of the causal nature of the
in a counseling group with a significantly higher association between early parental death, grief

Downloaded from hpq.sagepub.com by guest on April 12, 2016

Høeg et al. 9

counseling, and coping in adulthood. Counseling Appel CW, Frederiksen K, Hjalgrim H, et al.
participants were heterogeneous with regard to (submitted) Depressive symptoms and men-
type and extent of counseling received, thus the tal health related quality of life in adulthood
current findings cannot be generalized to spe- after grief counseling due to early parental
loss.
cific counseling practices. We were also unable
Appel CW, Johansen C, Christensen J, et al.
to include other factors that influence a child’s
(in press) Risk for use of antidepressants among
development, such as family functioning, or children and young adults exposed to the death
adjust for other potentially confounding factors, of a parent. Epidemiology, 27(3).
such as substance use in the parent(s), which Appel CW, Johansen C, Deltour I, et al. (2013) Early
could be associated with both parental death and parental death and risk of hospitalization for
our outcome. affective disorder in adulthood. Epidemiology
This study underlines the negative role of 24: 608–615.
early parental loss on coping behaviors in adult- Auerbach RP, Abela JRZ, Zhu X, et al. (2010)
hood. However, future studies with prospective Understanding the role of coping in the devel-
design measuring coping behaviors over time opment of depressive symptoms: Symptom
specificity, gender differences, and cross-cul-
before and after loss and counseling, respec-
tural applicability. British Journal of Clinical
tively, are needed. Additional research in cop-
Psychology 49: 547–561.
ing using a developmental perspective, and a Bandura A (1977) Self-efficacy: Toward a unify-
further shift in focus to the areas of resilience ing theory of behavioral change. Psychological
and post-traumatic growth after parental death, Review 84: 191–215.
may provide additional theoretical and clinical Bowlby J (1980) Loss: Sadness and Depression.
insights. New York: Basic Books.
Carver CS (1997) You want to measure coping but
Acknowledgements your protocol’s too long: Consider the brief
cope. International Journal of Behavioral
The authors thank the grief counseling organizations
Medicine 4: 92–100.
(The Danish Cancer Society, Children’s Welfare,
Carver CS, Scheier MF and Weintraub JK (1989)
and The Danish Counseling and Research Center for
Assessing coping strategies: A theoretically
Grieving Children and Youth) for their contribution
based approach. Journal of Personality and
to this study. They also thank Visti B. Larsen for his
Social Psychology 56: 267–283.
invaluable assistance in data management.
Christensen U, Schmidt L, Kriegbaum M, et al. (2006)
Coping with unemployment: Does educational
Declaration of Conflicting Interests attainment make any difference? Scandinavian
The author(s) declared no potential conflicts of inter- Journal of Public Health 34: 363–370.
est with respect to the research, authorship, and/or Cooper C, Katona C and Livingston G (2008)
publication of this article. Validity and reliability of the brief COPE in car-
ers of people with dementia: The LASER-AD
Funding Study. Journal of Nervous Mental Disorders
196: 838–843.
The author(s) disclosed receipt of the following
Currier JM, Holland JM and Neimeyer RA (2007)
financial support for the research, authorship, and/or
The effectiveness of bereavement interven-
publication of this article: This research was sup-
tions with children: A meta-analytic review
ported by the Danish foundation TrygFonden (J.nr.
of controlled outcome research. Journal of
7134-08).
Clinical Child & Adolescent Psychology 36:
253–259.
References Dubow EF, Boxer P and Huesmann LR (2009)
Akerman R and Statham J (2011) Childhood Long-term effects of parents’ education on chil-
Bereavement: A Rapid Literature Review. dren’s educational and occupational success:
London: Childhood Wellbeing Research Mediation by family interactions, child aggres-
Centre. sion, and teenage aspirations. Merrill-Palmer

Downloaded from hpq.sagepub.com by guest on April 12, 2016

10 Journal of Health Psychology 

quarterly (Wayne State University Press) 55: Trauma: Theory, Research, Practice, and
224–249. Policy 5: 119–127.
Fan W and Williams CM (2009) The effects of Pedersen CB, Gøtzsche H, Møller JØ, et al. (2006)
parental involvement on students’ academic The Danish Civil Registration System: A
self-efficacy, engagement and intrinsic motiva- cohort of eight million persons. Danish Medical
tion. Educational Psychology 30: 53–74. Bulletin 53: 441–449.
Folkman S and Moskowitz JT (2004) Coping: Pitfalls Polivy J and Herman CP (2002) Causes of eating
and promise. Annual Review of Psychology 55: disorders. Annual Review of Psychology 53:
745–774. 187–213.
Giordano GN, Ohlsson H, Kendler KS, et al. (2014) Raveis V, Siegel K and Karus D (1999) Children’s
Unexpected adverse childhood experiences psychological distress following the death of a
and subsequent drug use disorder: A Swedish parent. Journal of Youth and Adolescence 28:
population study (1995–2011). Addiction 109: 165–180.
1119–1127. Rhee K (2008) Childhood overweight and the rela-
Gunnar MR and Cheatham CL (2003) Brain and tionship between parent behaviors, parenting
behavior interface: Stress and the developing style, and family functioning. The ANNALS of
brain. Infant Mental Health Journal 24: 195–211. the American Academy of Political and Social
Haine RA, Ayers TS, Sandler IN, et al. (2008) Science 615: 11–37.
Evidence-based practices for parentally bereaved Rosner R, Kruse J and Hagl M (2010) A meta-anal-
children and their families. Professional ysis of interventions for bereaved children and
Psychology, Research and Practice 39: 113–121. adolescents. Death Studies 34: 99–136.
Hamdan S, Melhem NM, Porta G, et al. (2013) Sandler IN, Ayers TS, Wolchik SA, et al. (2003) The
Alcohol and substance abuse in parentally family bereavement program: Efficacy evalua-
bereaved youth. Journal of Clinical Psychiatry tion of a theory-based prevention program for
74: 828–833. parentally bereaved children and adolescents.
Holahan CJ and Moos RH (1987) Personal and Journal of Consulting and Clinical Psychology
contextual determinants of coping strategies. 71: 587–600.
Journal of Personality and Social Psychology Steele C, Andrews H and Upton D (2012)
52: 946–955. Statistics in Psychology. Essex: Pearson
Jordan JR and Neimeyer RA (2003) Does grief coun- Education Ltd.
seling work? Death Studies 27: 765–786. Stice E, Presnell K and Spangler D (2002) Risk fac-
Larson DG and Hoyt WT (2007) What has become of tors for binge eating onset in adolescent girls:
grief counseling? An evaluation of the empirical A 2-year prospective investigation. Health
foundations of the new pessimism. Professional Psychology 21: 131–138.
Psychology: Research and Practice 38: 347– Stroebe M and Schut H (1999) The dual pro-
355. cess model of coping with bereavement:
Lazarus RS (1999) Stress and Emotion: A New Rationale and description. Death Studies 23:
Synthesis. London: Free Association Books. 197–224.
Lazarus RS and Folkman S (1984) Stress, Appraisal, Stroebe M, Folkman S, Hansson RO, et al. (2006)
and Coping. New York: Springer. The prediction of bereavement outcome:
Matud MP (2004) Gender differences in stress Development of an integrative risk factor
and coping styles. Personality and Individual framework. Social Science & Medicine 63:
Differences 37: 1401–1415. 2440–2451.
Montez J, Hummer R and Hayward M (2012) Vessey JT and Howard KI (1993) Who seeks psy-
Educational attainment and adult mortality chotherapy? Psychotherapy: Theory, Research,
in the United States: A systematic analysis of Practice, Training 30: 546–553.
functional form. Demography 49: 315–336. Waters SF, Virmani EA, Thompson RA, et al. (2010)
Nickerson A, Bryant RA, Aderka IM, et al. (2013) Emotion regulation and attachment: Unpacking
The impacts of parental loss and adverse parent- two constructs and their association. Journal of
ing on mental health: Findings from the national Psychopathology and Behavioral Assessment
comorbidity survey-replication. Psychological 32: 37–47.

Downloaded from hpq.sagepub.com by guest on April 12, 2016

Høeg et al. 11

Webb NB (2010) The child and death. In: Webb NB Worden JW (2009) Grief Counseling and Grief Therapy:
(ed.) Helping Bereaved Children: A Handbook A Handbook for the Mental Health Practitioner.
for Practitioners (3rd edn). New York: Guilford New York: Springer Publishing Company.
Press, pp. 3–21. Zimmer-Gembeck MJ and Skinner EA (2011)
Weinberg RJ, Dietz LJ, Stoyak S, et al. (2013) A Review: The development of coping across
prospective study of parentally bereaved youth, childhood and adolescence: An integrative
caregiver depression, and body mass index. review and critique of research. International
Journal of Clinical Psychiatry 74: 834–840. Journal of Behavioral Development 35: 1–17.

Downloaded from hpq.sagepub.com by guest on April 12, 2016

View publication stats