ArticlePDF Available

Interpersonal Vulnerability to Depression in High-Risk Children: The Role of Insecure Attachment and Reassurance Seeking

Authors:

Abstract and Figures

This study examined the relation between insecure attachment and depression in a sample of 140 children (69 boys and 71 girls; ages 6 to 14) whose parents have a history of major depressive episodes. In addition, we examined whether this relation was moderated by excessive reassurance seeking. Children completed measures assessing insecure attachment to parents, excessive reassurance seeking, and current depressive symptoms. In addition, children and their parents participated in a semistructured clinical interview assessing children's current and past history of depressive symptoms and episodes. In line with hypotheses, children who exhibited high levels of both insecure attachment and excessive reassurance seeking experienced higher levels of current depressive symptoms than children who possessed only one or neither of these interpersonal risk factors. Furthermore, the interaction of insecure attachment with excessive reassurance was associated with a past history and greater severity of depressive episodes even after controlling for current depressive symptoms.
Content may be subject to copyright.
Interpersonal Vulnerability to Depression in High-Risk Children:
The Role of Insecure Attachment and Reassurance Seeking
John R. Z. Abela
Department of Psychology, McGill University
Benjamin L. Hankin
Department of Psychology, University of Illinois at Chicago
Emily A. P. Haigh, Philippe Adams, Theresa Vinokuroff, and Lisa Trayhern
Department of Psychology, McGill University
This study examined the relation between insecure attachment and depression in a
sample of 140 children (69 boys and 71 girls; ages 6 to 14) whose parents have a his
-
tory of major depressive episodes. In addition, we examined whether this relation was
moderated by excessive reassurance seeking. Children completed measures assessing
insecure attachment to parents, excessive reassurance seeking, and current depres
-
sive symptoms. In addition, children and their parents participated in a semi
-
structured clinical interview assessing children’s current and past history of depres
-
sive symptoms and episodes. In line with hypotheses, children who exhibited high
levels of both insecure attachment and excessive reassurance seeking experienced
higher levels of current depressive symptoms than children who possessed only one or
neither of these interpersonal risk factors. Furthermore, the interaction of insecure
attachment with excessive reassurance was associated with a past history and great-
er severity of depressive episodes even after controlling for current depressive
symptoms.
A number of theories have been proposed to explain
the etiology of depression. One such theory is Bowl-
by’s (1969, 1980) attachment theory. According to
Bowlby, early attachment patterns between children
and their caregivers play a vital role in both normal
and abnormal development. Attachment patterns are
thought to derive primarily from the quality and the
quantity of contact that the child has with his or her
caregivers (Ainsworth, Blehar, Waters, & Wall, 1978).
Parents who are sensitive in their caregiving, alert to
their infant’s needs, and who react quickly and ap-
propriately to such needs are likely to have infants who
develop a secure attachment (Wenar & Kerig, 2000;
West, Spreng, Rose, & Adam, 1999). Attachment theo
-
rists hypothesize that the formation of secure attach
-
ment facilitates the subsequent development of trust
-
ing and dependable relationships with others and has
important consequences for the child’s sense of secu
-
rity, adjustment, and emotions.
Not all children, however, are securely attached
to their caregivers. When normal developmental pro
-
cesses are disrupted in some way, a number of types
of insecure attachment patterns have been hypoth
-
esized to result. Insecure attachment patterns have
been posited to serve as risk factors for a diversity of
psychological problems, including depression (e.g.,
Bowlby, 1980; Cummings & Cicchetti, 1990). In line
with this hypothesis, numerous studies with both ado
-
lescents (e.g., Armsden & Greenberg, 1987; Armsden,
McCauley, Greenberg, Burke, & Mitchell, 1990; West
et al., 1999) and adults (e.g., Haaga et al., 2002; Ham
-
men et al., 1995; see Blatt & Homann, 1992) have
demonstrated that an insecure attachment style is asso
-
ciated with a greater likelihood of experiencing depres
-
sion. Similarly, research conducted with children has
found that higher levels of insecure attachment are sig
-
nificantly associated with higher levels of current de
-
Journal of Clinical Child and Adolescent Psychology
2005, Vol. 34, No. 1, 182–192
Copyright © 2005 by
Lawrence Erlbaum Associates, Inc.
182
The research reported in this article was supported, in part, by a
Young Investigator Award from the National Alliance for Research
on Schizophrenia and Depression awarded to John R. Z. Abela.
We would like to thank Martin E. P. Seligman and David Zuroff
for serving as mentors for the National Alliance for Research on
Schizophrenia and Depression grant. We would also like to thank
Michael Birnbaum, Karen Brozina, Chris Bryan, Geneviève Dumas,
Sabrina Drudi, Dean Elterman, Melanie Ewing, Hugo Gagnon, Pau
-
line Gregoire, Rachel Horton, Chris Knundsen, Anais Lavarenne,
Nathali Lefebvre, Geneviève LePage, Alexandra McIntyre-Smith,
Norma Moussaly, Zohreen Murad, Christine Ngo, David Paul, An
-
drew Payne, Jacqueline Poitras, Jay Poitras, Anna Radzioch, Maya
Sakellaropoulo, Raffi Schieir, and Sharmeen Shah for conducting as
-
sessments of the children and their parents.
Requests for reprints should be sent to John R. Z. Abela, Depart
-
ment of Psychology, McGill University, Stewart Biological Sciences
Building, 1205 Dr. Penfield Avenue, Montreal (Quebec) Canada,
H3A 1B1. E-mail:abela@ego.psych.mcgill.ca
pressive symptoms (e.g., Muris, Mayer, & Meesters,
2000; Graham & Easterbrooks, 2000).
Although preliminary evidence suggests that inse
-
cure attachment is associated with depression, several
limitations of past research should be noted. First, the
majority of past research has simply examined whether
insecure attachment is a correlate of current depressive
symptoms as assessed by self-report measures. It is im
-
portant to examine whether insecure attachment pre
-
dicts either past history or future increases in depres
-
sion above and beyond current depressive symptoms.
Furthermore, it is essential to utilize diagnostic inter
-
views, in addition to self-report questionnaires, to as
-
sess depression to be able to examine the full range of
depression from milder levels of dysphoria to more se
-
vere levels of clinical depression (Coyne, 1994). Sec
-
ond, the majority of past research has utilized ado
-
lescent or adult samples. Findings from studies
examining the etiology of depression in adolescents
and adults cannot automatically be extended to chil
-
dren. Rather, age-related differences in cognition,
emotion, and behavior must be taken into account, and
the theories must be tested using child samples. Last,
few studies have examined the factors that moderate
the relation between insecure attachment and depres-
sion. Clearly not all individuals who exhibit insecure
attachment develop depression. Thus, it is important to
examine the factors that might enhance the likelihood
of developing depression among insecurely attached
individuals.
The first goal of this study was to examine whether
insecure attachment to parents serves as a risk factor
for depression, both in symptom severity and clinical
episodes, in children. It is worth noting that the latent
structure of depression has been found to be dimen
-
sional among youth (Hankin, Fraley, Lahey, &
Waldman, in press), indicating that there is continuity
between depressive symptoms and disorder (Flett,
Vredenburg, & Krames, 1997). Consequently, we fo
-
cused primarily on severity of depressive symptoms, as
assessed by both youth self-report questionnaire and
number of symptoms met in a diagnostic interview. At
the same time, given past debate regarding the continu
-
ity of depression (Coyne, 1994), we also assessed de
-
pression in terms of a diagnosis of a depressive epi
-
sode. To test whether insecure attachment predicts
depression, we utilized a retrospective design. As out
-
lined by Alloy et al. (2000), the logic of such a design is
based on two assumptions. First, children who exhibit
the hypothesized vulnerability factor to depression
(i.e., an insecure attachment style) developed this fac
-
tor sometime in the past, thereby increasing their risk
for depression whenever this factor was present. Con
-
sistent with this argument, attachment styles have been
found to exhibit stability throughout childhood and ad
-
olescence (e.g., Ainsworth et al., 1978; Crowell,
Fraley, & Shaver, 1999). A meta-analytic review of the
consistency of attachment styles showed relative sta
-
bility over the first 19 years of life with stronger
stability in childhood (Fraley, 2002). Second, given
that a past history of childhood depression is one of the
best predictors of future elevations in depression (e.g.,
Harrington, Rutter, & Fombonne, 1996), finding an as
-
sociation between increased rate of past depression
and the hypothesized vulnerability factor (i.e., insecure
attachment style) in children not currently experienc
-
ing elevated depression provides support for the vul
-
nerability status of this factor.
The second goal was to examine whether the rela
-
tion between insecure attachment and depression in
children is moderated by excessive reassurance seek
-
ing. Excessive reassurance seeking is defined as “a rel
-
atively stable tendency to excessively and persistently
seek assurances from others that one is loveable and
worthy, regardless of whether such assurance has al
-
ready been provided” (Joiner, Metalsky, Katz, &
Beach, 1999, p. 270). According to Coyne’s (1976)
interactional theory, depressed individuals perpetuate a
cycle of negative interpersonal exchanges that triggers
increases in their depressive symptoms. Applying
Coyne’s model to understanding the etiology of de-
pression in children, we would predict the following: A
mildly depressed child seeks reassurance from parents.
Parents initially respond with genuine concern and
support. The child, however, perceives this initial sup-
port as inadequate and consequently escalates symp-
toms in an attempt to secure more reassurance. Al-
though parents continue to provide support, they begin
to experience feelings of irritation and guilt, leading to
a separation between the content and affective quality
of their reassuring statements. This discrepancy in
-
creases the child’s fear of rejection, which in turn leads
to a further escalation of symptoms to restore the feel
-
ing of security.
Research with adults has supported the interactional
model of depression (Coyne, 1976). For example, indi
-
viduals with high levels of reassurance seeking exhibit
higher levels of depressive symptoms than individuals
with low levels of reassurance seeking (e.g., Joiner,
Alfano, & Metalsky, 1992; Joiner & Metalsky, 1995).
In addition, excessive reassurance seeking predicts in
-
creases in depression over time both on its own and in
interaction with negative events (e.g., Davilla, 2001;
Joiner & Metalsky, 2001; Joiner & Schmidt, 1998). Far
less research has examined the relation between exces
-
sive reassurance seeking and depressive symptoms in
children. The only available studies, to date, have
shown that higher levels of reassurance seeking were
associated with elevated levels of depressive symp
-
toms (Joiner, 1999) and a diagnosis of a depressive, but
not externalizing, disorder (Joiner, Metalsky, Gencoz,
& Gencoz, 2001). Both of these studies were con
-
ducted with youth in-patient psychiatric samples, and,
although providing preliminary support for children,
183
INTERPERSONAL VULNERABILITY TO DEPRESSION
exclusive use of in-patient samples may be a limitation
because of both the lack of generalizability to commu
-
nity samples and the greater depressive severity and
comorbidity levels (Newman, Moffitt, Caspi, & Silva,
1998).
Given that children who exhibit high levels of reas
-
surance seeking are likely to seek confirmations of
their value and worth from key attachment figures such
as their parents, we hypothesized that children who ex
-
hibit a combination of excessive reassurance seeking
and insecure attachment would be particularly vul
-
nerable to developing depression. In other words, we
posited that the interaction of excessive reassurance
seeking with an insecure attachment style would be as
-
sociated with depression. From the perspective of
Coyne’s (1976) interactional model of depression, this
hypothesis follows from the following line of reason
-
ing: Children who are insecurely attached to their care
-
givers and who consistently seek reassurance from
these caregivers will be particularly likely to have their
requests for reassurance not met as a result of their de
-
velopmental history of unresponsive caregiving. Thus,
although the extant literature shows that both insecure
attachment and reassurance seeking are independent
predictors of depression, we hypothesized that the in-
teraction of these two independent risk factors would
be a particularly potent risk for depression.
To test our hypothesis that the interaction of inse-
cure attachment with excessive reassurance seeking
would be associated with (a) higher levels of current
depressive symptoms and (b) a past history and greater
severity of clinically significant depression, even after
controlling for current depressive symptoms, we as-
sessed attachment, reassurance seeking, and depres
-
sion (both symptoms and episodes) in a high-risk sam
-
ple of 140 children (ages 6 to 14) of depressed parents.
We believe it is particularly interesting and advanta
-
geous to examine this hypothesis in a high-risk sample
for several reasons. First, offspring of depressed par
-
ents are at increased risk for developing depression
(Goodman & Gotlib, 2002). Thus, the use of a high-
risk design maximizes the number of children who
have experienced a past history of clinically significant
depression, thereby increasing power to predict de
-
pression in a sample of younger children expected to
exhibit a relatively low incidence of depression (i.e.,
less than 3% of the general population under 15 years
of age; Hankin et al., 1998). Second, offspring of de
-
pressed parents may be particularly likely to exhibit in
-
secure attachment and excessive reassurance seeking.
Many studies have shown that children of depressed
parents are more likely to have an insecure attachment
than other children, and, although the evidence is not
entirely consistent, this association seems generally
stronger when investigated among children whose par
-
ents show clinical levels of depression (Lyons-Ruth,
Lyubchik, Wolfe, & Bronfman, 2002). Parents’depres
-
sion may contribute to youths’insecure attachment and
excessive reassurance seeking through various mecha
-
nisms (see Goodman & Gotlib, 1999, 2002, for theo
-
retical and empirical reviews), although the precise
processes are not known. It may be that the child devel
-
ops a more insecure attachment because the depressed
parent is less emotionally and physically available.
Also, the child may display more excessive reassur
-
ance seeking because the depressed parent is less re
-
sponsive and available to provide adequate reas
-
surance. The available research shows that parents’
relational difficulties predict their children’s relational
problems, and this transaction increases the risk for
both the child’s depression and his or her insecure at
-
tachment (Lyons-Ruth et al., 2002). Depressed parents
have been found to display less optimal caregiving be
-
haviors and to engage in fewer positive interactions
and more negative interchanges with the child; these
problematic parenting behaviors persist even after de
-
pression remission (Lyons-Ruth et al., 2002). Children
of depressed mothers are less responsive and active,
show flattened affect, exhibit elevated emotionality in
the face of conflict and distress, and display less posi
-
tive and more negative affect in interactions with par-
ents (Goodman & Gotlib, 1999). Thus, the evidence is
consistent with the view that children of depressed par-
ents are more likely to be insecurely attached and dis-
play excessive reassurance seeking.
Method
Participants
Participants were recruited in two phases. In the
first phase, participants were recruited through ads
placed in local newspapers as well as through posters
placed throughout the greater Montreal area. All ads
and posters were in English and specifically advertised
for parents with a history of depressive episodes. One
hundred ninety-five people responded to these ads. Re
-
spondents were invited to participate in a telephone in
-
terview during which a diagnostician administered the
affective disorders module of the Structured Clinical
Interview for the Diagnostic and Statistical Manual of
Mental Disorders (4th ed. [DSM–IV]; American Psy
-
chiatric Association, 1984) Axis I (SCID–I; First, Gib
-
bon, Spitzer, & Williams, 1995). One hundred seven
-
teen parents met criteria for either a current or past
major depressive episode and were invited to partici
-
pate in the study. Eighty-six parents, with 122 children
in the appropriate age range, decided to participate.
In the second phase, diagnosticians called the par
-
ents of children who participated in a previous study in
our lab and invited them to participate in this study.
The only selection criterion for the past study was hav
-
ing a child between the ages of 7 and 13. All 55 parents
184
ABELA ET AL.
agreed to complete the telephone interview. Sixteen of
these parents met criteria for either a current or past
major depressive episode and were invited to partici
-
pate in the study. All 16 parents, with 18 children in the
appropriate age range, decided to participate.
The final sample consisted of 140 children (69 boys
and 71 girls) and one of their parents (88 mothers and
14 fathers). In all cases, the participating parent was
the parent who had completed the phone interview and
met criteria for either a current or past major depressive
episode. Children’s ages ranged from 6 to 14 (M = 9.8,
SD = 2.3, median = 10). Parents’ ages ranged from 27
to 53 (M = 40.3, SD = 6.4, median = 41). The sample
was 84.3% White (n = 118), 4.9% Asian (n = 7), 2.9%
Hispanic (n = 4), 1.9% African American (n = 1), 1.1%
Native American (n = 1), and 4.9% of other descent (n
= 7). The mother tongue of participants included Eng
-
lish (68.7%; n = 70), French (9.8%; n = 10), Spanish
(2.9%; n = 3), and other languages (18.6%; n = 19; e.g.,
German, n = 3, and Portuguese, n = 3). Importantly, all
parents reported that they and their child were fluent in
English. Of the parents, 14.7% were single (n = 21),
43.1% were married (n = 60), 9.8% were separated (n =
14), 27.5% were divorced (n = 38), 1.0% was widowed
(n = 1), and 3.9% were none of the above (n = 6).
The median family income ranged from $30,000 to
$45,000. Of the children, 34 met criteria for a current
(6%; n = 8) or past (21%; n = 26) affective disorder, and
106 had no history of affective disorders.
Procedure
Two research assistants met with one parent–child
pair at a time. At the start of the assessment, parents
completed consent and demographics forms. The chil
-
dren were told that their participation was voluntary
and they could choose not to participate. All children
decided to participate in the study. For the first half of
the assessment, a research assistant read each question
-
naire aloud in English to the child while the child fol
-
lowed along and responded to questions using his or
her own copy. During this time, a diagnostician ob
-
tained information regarding the children’s current and
past depressive symptoms from the parent using the
Schedule for Affective Disorders and Schizophrenia
for School-Age Children, Present Version (K–SADS;
Kaufman, Birmaher, Brent, Rao, & Ryan, 1996). In the
second half of the assessment, a diagnostician obtained
information regarding the child’s current and past de
-
pressive symptoms from the child using the K–SADS.
Families were compensated for their time at the rate of
$60.
Diagnostic interviewers completed an intensive
training program for administering the K–SADS and
SCID–I interviews and for assigning DSM–IV and
RDC diagnoses. The training program consisted of at
-
tending approximately 80 hr of didactic instruction,
listening to audiotaped interviews, conducting practice
interviews, and passing regular exams (85% or above).
The principal investigator held weekly supervision ses
-
sions for the interviewers. The principal investigator
also reviewed interviewers’ notes and tapes to confirm
the presence or absence of a diagnosis. Discrepancies
were resolved through consensus meetings and best-
estimate procedures.
Measures
SCID–I (First et al., 1995). The SCID–I is a sem
-
istructured clinical interview designed to arrive at cur
-
rent and lifetime DSM–IV diagnoses. We employed the
mood disorder module and the psychotic screen for the
DSM–IV to allow for the diagnosis of current and past
history of DSM–IV mood disorders among the parents.
The psychotic screen was used to exclude any parents
with current or past psychotic disorders. Two parents
were excluded as a result of having experienced a past
psychotic disorder. The SCID–I has been shown to yield
reliable diagnoses of depressive disorders (Zanarini et
al., 2000) and is frequently used in clinical studies of de-
pression in adults.
K–SADS (Kaufman et al., 1996). The K–SADS
is a semistructured clinical interview designed to arrive
at DSM–IV and RDC diagnoses. The K–SADS is ad-
ministered separately to the child and the parent. A
summary diagnosis is based on both sets of infor-
mation. The K–SADS yields reliable diagnoses of de-
pressive disorders (Chambers et al., 1985) and is used
frequently in studies of depression in children. We as
-
sessed both current and past history of clinically signif
-
icant depressive episodes. Dichotomous scores, based
on DSM–IV depression diagnostic criteria, indicate the
presence (coded 1) or absence (coded 0) of current or
past depressive episodes. In addition, we utilized the
K–SADS to assess severity of both current and past
peak levels of depressive symptoms. Severity scores
range from 0 (no symptoms)to9(all depressive symp
-
toms) and indicate the number of DSM–IV depressive
symptom criteria met.
Child Depression Inventory (CDI; Kovacs,
1992). The CDI is a 27-item self-report question
-
naire that measures the cognitive, affective, and behav
-
ioral symptoms of depression. For each item, children
were asked whether it described how they were think
-
ing and feeling in the past week. Items are scored from
0 to 2, with higher score indicating greater symptom
severity. Total CDI scores range from 0 to 52. The CDI
is reliable, valid, and can distinguish children with ma
-
jor depressive disorders from nondepressed children
(Kovacs, 1992). Coefficient alpha was .84 in this study.
185
INTERPERSONAL VULNERABILITY TO DEPRESSION
Inventory of Parent and Peer Attachment (IPPA;
Armsden & Greenberg, 1987). The IPPA is an 18-
item self-report questionnaire that assesses positive and
negative cognitive–affective dimensions (e.g., “internal
working models”) of children’s relationships with their
parents and close friends. Children rate items focusing
on how well these attachment figures serve as sources
of psychological security using a 5-point Likert-scale
ranging from XX (XXXXX)toXX(XXXXX). Sample
items include “I can count on my parents when I need to
get something offmy chest”; “My parents trust my judg
-
ment”; “My parents don’t understand what I’m going
through these days. We used the 12 items assessing at
-
tachmentto parents. Thetotal scoreis equal tothe sum of
all items, and higher scores indicate higher levels of in
-
secure attachment.
Research utilizing the IPPA with adolescent (ages
10 to 20) samples shows the IPPA is internally consis
-
tent and has strong test–retest reliability and good va
-
lidity (Armsden & Greenberg, 1987, 1988; Crowell et
al., 1999). Although far less research has utilized the
IPPA with children (i.e., 6 to 10), preliminary data sup
-
port the reliability and validity of IPPA scores in this
age group. For example, IPPA scores possess moderate
internal consistency and test–retest reliability over a
1-year interval (r = .63; Abela, Adams, & Hankin,
2003) in younger children. In addition, higher scores
on the parental attachment subscale have been found to
be associated with lower scores on the warmth, respon-
siveness, and consistency subscales of the Parenting
Dimensions Inventory (Abela et al., 2003) in both
younger and older children. In this study, coefficient
alpha was .78.
The Reassurance-Seeking Scale for Children
(RSSC; Joiner & Metalsky, 1995). The RSSC is a
modified version of the Reassurance-Seeking Scale
(Joiner & Metalsky, 1995), which was reworded for
use with children. The revised scale consists of four
statements (e.g., “I always need to ask my parents and
friends if they like me”; “I always need to ask my par
-
ents and friends if they really care about me”; “Some
-
times when I ask people if they like me, they tell me to
stop asking”; “Sometimes when I ask people if they
like me, they get mad”). Children rate each statement
on a 3-point scale. Total scores range from 0 to 8, with
higher scores corresponding to higher levels of reas
-
surance seeking. RSSC scores exhibit moderate to high
levels of reliability (Joiner, 1999; Joiner et al., 2001).
Regarding validity, in line with Coyne’s (1976) theory,
higher RSSC scores have been found to be associated
with higher levels of both interpersonal rejection (Join
-
er et al., 2001) and dependency (Abela et al., 2003).
Regarding reliability, RSSC scores have been found to
be relatively stable in children (ages 6 to 14) over a
1-year period (rs = .61; Abela et al., 2003). In this
study, coefficient alpha was .78.
Results
Examining Nonindependence in Data
Given that 38 pairs of siblings participated in this
study, we conducted preliminary analyses examining
whether nonindependence in our data impacted our
findings. To do so, we first ran all analyses including
only (a) the 64 children who did not have a sibling par
-
ticipating in this study and (b) the first child from each
sibling pair to complete the assessment (n = 38). Next,
we ran all analyses including only (a) the 64 children
who did not have a sibling participating in this study
and (b) the second child from each sibling pair to com
-
plete the assessment (n = 38). Last, we ran all analyses
including all 140 children.
1
The direction and magni
-
tude of effects were similar in all three sets of analyses,
suggesting that the inclusion of siblings in this study
did not have a significant impact on the pattern of find
-
ings obtained. In addition, all significant relations re
-
ported were also obtained excluding either the first or
second sibling. Thus, we used the entire sample (n =
140) for the analyses presented next.
Regression Analyses: Current Level
of Depressive Symptoms
Table 1 shows descriptive statistics and associations
(Pearson correlations) for measures.
To examine whether the interaction of excessive re-
assurance seeking with insecure attachment was asso-
ciated with children’s self-reported current level of de-
pressive symptoms (CDI), we conducted hierarchical
multiple regression analyses.
2
First, main effect vari-
ables (i.e., RSSC and IPPA) were entered into the
equation. Second, the RSSC × IPPA interaction was
entered.
3
Results pertaining to children’s self report
(CDI) of current level of depressive symptoms are pre
-
sented in the top panel of Table 2. Higher RSSC scores
were associated with higher levels of depressive symp
-
toms after controlling for IPPA scores, and higher
IPPA scores were associated with higher levels of de
-
186
ABELA ET AL.
1
Details on these specific analyses are available from the first
author.
2
In addition to the moderational model we present in the text, we
also explored a mediational model in which youths’ insecure at
-
tachment would be associated with excessive reassurance seeking,
which, in turn, would be related to depression. However, this medi
-
ational model was not supported because excessive reassurance
seeking was not associated significantly with insecure attachment
(see Table 1). Additional details on the specific mediational analyses
are available from the first author.
3
Given findings showing age and gender effects with depression
(Hankin & Abramson, 2001), we conducted exploratory analyses in
-
cluding age and gender in the regressions as main effects and moder
-
ators. None of the age or gender interactions was significant at each
step as they were entered, nor in the final step, in the hierarchical re
-
gression. Therefore, for the sake of simplicity, results are presented
for the entire sample.
pressive symptoms after controlling for RSSC scores.
Further, in line with hypotheses, the RSSC × IPPA in
-
teraction was associated with depressive symptoms.
To examine the form of the RSSC × IPPA interac
-
tion, predicted CDI scores were calculated by inserting
specific values for RSSC and IPPA scores (i.e., 1 SD
above and below this sample’s mean) into the regres
-
sion equation summarized in the top of Table 2 (see
Cohen & Cohen, 1983, pp. 323, 419 for further de
-
tails). Low levels of depression were found among (a)
children who exhibited low levels of both reassurance
seeking and insecure attachment (CDI = 7.01), (b) chil
-
dren who exhibited high levels of reassurance seeking
but low levels of insecure attachment (CDI = 5.88), and
(c) children who exhibited high levels of insecure at
-
tachment but low levels of reassurance seeking (CDI =
9.68). Children who exhibited high levels of both reas
-
surance seeking and insecure attachment reported the
highest level of depressive symptoms (CDI = 17.04).
Similar regression analyses were conducted to ex
-
amine whether the interaction of excessive reassurance
seeking with insecure attachment was associated with
diagnosticians’ ratings (K–SADS–C-SEV) of chil
-
dren’s current level of depressive symptoms. As the
K–SADS relies, in part, on parental report, BDI scores
were entered into the model first to control for the po
-
187
INTERPERSONAL VULNERABILITY TO DEPRESSION
Table 1. Means, Standard Deviations, and Intercorrelations Among All Measures
1234567891011
1. Children’s
Depression Inventory
2. K–SADS–Current
Diagnosis
.39***
3. K–SADS–Current
Severity
.44*** .85***
4. K–SADS–Past
Diagnosis
.14 .18* .08
5. K–SADS–Past
Severity
.14 .06 .04 .87***
6. Reassurance Seeking
Scale for Children
.31*** .31*** .27** .21* .16
7. Inventory for Parent
and Peer Attachment
.45*** .29*** .27** .08 .01 .08
8. Beck Depression
Inventory
.23** .27** .24** .06 .07 .08 .17
9. SCID–Current
Severity
.14 .31*** .29** .06 .07 .08 .19* .62***
10. Child’s Age .08 –.05 –.05 .15 .15 –.45*** .11 .01 –.09
11. Child’s Gender –.06 .12 .06 .04 .00 .15 .03 .07 .18* –.13
M 10.09 .06 .19 .21 .94 1.52 27.82 19.16 2.61 9.80 .51
SD 6.76 .23 .92 .41 2.11 2.11 8.38 12.41 3.41 2.37 .50
Note: K–SADS = Schedule for Affected Disorders and Schizophrenia; SCID = Structural Clinical Interview for the DSM–IV.
***p < .001. **p < .01. *p < .05.
Table 2. Reassurance Seeking and Insecure Attachment Predicting Current Level of Depressive Symptoms
Order of Entry Predictors Cumulative R
2
F for Increment
in R
2
df
t for Within-
Set Predictors
Standardized
β Coefficient
Children’s Self Report (CDI)
1. Main Effect Variables .29 27.01*** 2,131
RSSC 131 4.01*** .30
IPPA 131 5.79*** .43
2. RSSC × IPPA .36 14.17*** 1,130 .28
Diagnosticians’ Ratings (K–SADS–C–SEV)
1. BDI .06 7.63** 1,128 .24
2. Main Effect Variables .17 8.82*** 2,126
RSSC 126 3.02** .25
IPPA 126 2.69** .22
3. RSSC × IPPA .30 21.83*** 1,125 .37
Note: CDI = Children’s Depression Inventory; RSSC = Reassurance Seeking Scale for Children; IPPA = Inventory for Parent and Peer Attach
-
ment, Parent subscale; K–SADS–C–SEV = Schedule for Affective Disorders and Schizohrenia for School-Age Children, Current Severity of
Symptoms; BDI = Beck Depression Inventory.
***p < .001. **p < .01. *p < .05.
tential impact of parents’ current levels of depressive
symptoms on their reporting of their child’s current
level of depressive symptoms. Second, main effect
variables (i.e., RSSC and IPPA) were entered into the
equation. Last, the RSSC × IPPA interaction was
entered.
3
Results pertaining to diagnosticians’ ratings
(K–SADS–C-SEV) of children’s current level of de
-
pressive symptoms are presented in the bottom panel
of Table 2. Higher RSSC scores were associated with
higher K–SADS–C-SEV scores after controlling for
IPPA scores, and higher IPPA scores were associated
with higher K–SADS–C-SEV scores after controlling
for RSSC scores. Last, in line with hypotheses, the
RSSC × IPPA interaction was associated with higher
K–SADS–C-SEV scores.
To examine the form of the RSSC × IPPA interac
-
tion, K–SADS–C-SEV scores were calculated by in
-
serting specific values for RSSC and IPPA scores (i.e.,
1 SD above and below this sample’s mean) into the re
-
gression equation summarized in the bottom of Table
2. Relatively low levels of depressive symptoms (i.e.,
<2 depressive symptoms on the K–SADS-C) were
seen among (a) children who exhibited low levels of
both reassurance seeking and insecure attachment
(K–SADS–C-SEV = 0.97), (b) children who exhibited
high levels of reassurance seeking but low levels of in-
secure attachment (K–SADS–C-SEV = 0.17), and (c)
children who exhibited high levels of insecure attach-
ment but low levels of reassurance seeking (K–SADS–
C-SEV = 0.62). Children who exhibited high levels
of both reassurance seeking and insecure attachment
exhibited the highest levels of depressive symptoms
(K–SADS–C-SEV = 2.22).
Regression Analyses: Past History
of Depressive Episodes
Logistic regression analyses were conducted to ex
-
amine whether the interaction of excessive reassurance
seeking with insecure attachment was associated with
a past history of depressive episodes. The dependent
variable was whether the child had a past history of de
-
pressive episodes (coded 1 for presence of past depres
-
sive episode(s) and 0 for none). We first entered current
levels of both parents’ and children’s depressive symp
-
toms to control for the potential impact of current de
-
pressive symptoms on the recall of past depressive epi
-
sodes. To further control for current levels of
depression, children who met diagnostic criteria for a
current depressive episode were removed from the
sample (n = 8). Second, main effect variables were en
-
tered (i.e., IPPA and RSSC scores). Last, we entered
the IPPA × RSSC interaction.
3
Results pertaining to a
past history of depressive episodes are presented at the
top of Table 3. Neither excessive reassurance seeking
nor insecure attachment was associated with a past his
-
tory of depressive episodes after controlling for current
depressive symptoms and the other predictor. Consis
-
tent with hypotheses, the interaction of insecure attach
-
ment and excessive reassurance was associated with a
past history of depressive episodes after controlling for
current depressive symptoms.
To examine the form of this interaction, probabili
-
ties of having experienced a past depressive episode
were calculated by inserting specific values for RSSC
and IPPA scores (i.e., 1 SD above and below this sam
-
ple’s mean) into the regression equation summarized in
the top of Table 3 (see Kleinbaum, 1994, p. 8 for fur
-
ther details). Low probabilities of having experienced a
past depressive episode were obtained for (a) children
who exhibited low levels of both reassurance seeking
and insecure attachment (probability = 6.8%), (b) chil
-
dren who exhibited high levels of reassurance seeking
but low levels of insecure attachment (probability =
5.6%), and (c) children who exhibited high levels of in-
secure attachment but low levels of reassurance seek-
ing (probability = 5.6%). Children who exhibited high
levels of both reassurance seeking and insecure attach-
ment showed the highest probability of experiencing a
past depressive episode (probability = 57.8%).
Regression Analyses: Past Severity
of Depressive Episodes
To examine whether the interaction of excessive re
-
assurance seeking with insecure attachment was asso
-
ciated with diagnosticians’ ratings (K–SADS–P-SEV)
of the severity children’s past depressive episodes, we
conducted hierarchical multiple regression analyses.
Again, both parents’ and children’s depressive symp
-
toms were first entered into the regression equation. In
addition, children who met diagnostic criteria for a cur
-
rent depressive episode were removed from the sample
(n = 8). Second, main effect variables (i.e., RSSC and
IPPA) were entered into the equation. Last, the RSSC ×
IPPA interaction was entered.
3
Results pertaining to
diagnosticians’ ratings (K–SADS–P-SEV) of the se
-
verity children’s past depressive episodes are presented
in the bottom panel of Table 3. The main effect set
was not a significant predictor of K–SADS–P-SEV
scores. At the same time, in line with hypotheses, the
RSSC × IPPA interaction was associated with higher
K–SADS–P-SEV scores.
To examine the form of the RSSC × IPPA interac
-
tion, K–SADS–P-SEV scores were calculated by in
-
serting specific values for RSSC and IPPA scores (i.e.,
188
ABELA ET AL.
3
Given findings showing age and gender effects with depression
(Hankin & Abramson, 2001), we conducted exploratory analyses in
-
cluding age and gender in the regressions as main effects and moder
-
ators. None of the age or gender interactions was significant at each
step as they were entered, nor in the final step, in the hierarchical re
-
gression. Therefore, for the sake of simplicity, results are presented
for the entire sample.
1 SD above and below the sample mean) into the re-
gression equation shown in the bottom of Table 3.
Three of the four groups of children exhibited low lev-
els of depressive symptoms (i.e., K–SADS–P-SEV <
2): (a) children who exhibited low levels of both reas-
surance seeking and insecure attachment (K–SADS–
P-SEV = 1.29), (b) children who exhibited high levels
of reassurance seeking but low levels of insecure at
-
tachment (K–SADS–P-SEV = 0.52), and (c) children
who exhibited high levels of insecure attachment but
low levels of reassurance seeking (K–SADS–P-SEV =
0.19). Children who exhibited high levels of both reas
-
surance seeking and insecure attachment showed the
highest levels of depressive symptoms (K–SADS–P-
SEV = 2.28).
Discussion
The results of this study provide support for our hy
-
pothesis that the interaction of insecure attachment to
parents and excessive reassurance seeking would be
associated with both current depressive symptoms and
past history or severity of depressive episodes. Such
findings suggest that children who both engage in ex
-
cessive reassurance seeking and exhibit insecure at
-
tachment may be particularly likely to develop depres
-
sive symptoms. Thus, the results of this study highlight
the potential importance of integrating Bowlby’s
(1969) attachment theory with Coyne’s (1976) inter
-
actional model to foster a more thorough understand
-
ing of the role of interpersonal factors in the etiology of
depression in children.
It is important to note, however, that because this
study utilized a retrospective design, it is not possible
to ascertain with certainty whether the increased life-
time prevalence of depression among children exhibit-
ing high levels of both insecure attachment and exces-
sive reassurance seeking is due to these interpersonal
vulnerability factors contributing to the cause of past
depression or to past depression leading to the develop
-
ment of these vulnerability factors as a consequence or
a “scar” (Rhode, Lewinsohn, & Seeley, 1990) of the
earlier episode. Considering the early development and
relatively stability of attachment styles (e.g., Ains
-
worth et al., 1978; Fraley, 2002), however, there is ade
-
quate reason to suspect that insecure attachment, as a
vulnerability factor, existed prior to depression eleva
-
tions. At the same time, given that it is unknown how
stable reassurance seeking is over the lifespan, particu
-
larly among youth, caution is needed when interpreting
these retrospective results. Given the uncertain stabil
-
ity of excessive reassurance seeking in childhood, it
may be that an insecure attachment (or other depres
-
sion vulnerability) contributes to decreases in self-es
-
teem or increases in negative life events (see Hankin et
al., 2003, for evidence) that subsequently predict ex
-
cessive reassurance seeking (Joiner et al., 1999). Fur
-
thermore, our results show that excessive reassurance
seeking declines over time with age. This appears con
-
sistent with the perspective that excessive reassurance
seeking may be relatively more developmentally typi
-
189
INTERPERSONAL VULNERABILITY TO DEPRESSION
Table 3. Reassurance Seeking and Insecure Attachment Predicting Past History of Depressive Symptoms
Logistic Regression Analyses: Likelihood of Experiencing Past Depressive Episode (K–SADS–P–DX)
95% CI for Odds Ratio
Order of Entry Predictors Odds Ratio Lower Upper Wald
1. CDI 1.05 0.98 1.13 1.92
BDI 1.00 0.97 1.04 0.02
2. RSSC 1.40 0.87 2.24 1.94
IPPA 1.01 .60 1.69 0.00
3. RSSC × IPPA 3.24 1.24 8.48 5.72*
Hierarchical Multiple Regression Analyses: Severity of Past Depressive Episode (K–SADS–P–SEV)
Order of Entry Predictors Cumulative R
2
F for Increment
in R
2
df
t for Within-
Set Predictors
Standardized
β Coefficient
1. Control Variables .02 1.08 2,118 1.32
118 .12
BDI 118 0.42 .04
2. Main Effect Variables .04 1.28 2,116
RSSC 116 1.16 .11
IPPA 116 –1.01 –.10
3. RSSC × IPPA .10 7.14** 1,115 .28
Note: CI = confidence interval; CDI = Children’s Depression Inventory; BDI = Beck Depression Inventory; RSSC = Reassurance Seeking Scale
for Children; IPPA = Inventory for Parent and Peer Attachment, Parent subscale; K–SADS = Schedule for Affective Disorders and Schizohrenia
for School-Age Children; P–DX = Past Diagnosis; P–SEV = Severity of Past Episode.
*p < .05. **p < .01. ***p < .001.
cal earlier in childhood, but continuing to exhibit high
excessive reassurance seeking that is not normative
may represent more of an emerging depression vul
-
nerability. Thus, it is imperative that future research
examine these issues longitudinally to ascertain more
precisely whether insecure attachment interacts with
reassurance seeking to predict the development of
depression.
Future research is likely to benefit from examining
the factors that mediate the relation between the inter
-
action of these interpersonal vulnerability factors (e.g.,
insecure attachment and excessive reassurance seek
-
ing) and depression in children. Cognitive factors have
been investigated as one mediator of this relation in
adults. For example, Ingram and Ritter (2000) found
that insecurely attached individuals are more likely
than securely attached individuals to attend to negative
stimuli following a negative mood-induction proce
-
dure. In addition, prospective naturalistic studies have
shown that insecure attachment style leads to higher
levels of dysfunctional attitudes, which in turn contrib
-
utes to lower self-esteem, which consequently predicts
future depressive symptoms (Hankin et al., in press;
Roberts, Gotlib, & Kassel, 1996). Thus, preliminary
data provide support for the hypothesis that cognitive
factors mediate the relation between insecure attach-
ment and depression in adults. To our knowledge, how-
ever, no research has examined this mediational com-
ponent of attachment theory in younger populations.
This study advances research on the relation be-
tween insecure attachment and depression for several
reasons. First, most past research has examined wheth-
er insecure attachment is a correlate of current depres-
sive symptoms. This study expanded on such research
by showing that insecure attachment was associated
with a past history of depressive episodes even after
controlling for current level of depressive symptoms.
Second, most past research has examined the relation
between insecure attachment and depressive symp
-
toms. This study expands our present knowledge base
by showing that insecure attachment is not only related
to severity of current depressive symptoms but also to
the past occurrence of clinically significant depressive
episodes. Third, the study utilized a multi-informant,
multimethod design. Both the parent and child were in
-
terviewed to arrive at the child’s depression diagnosis.
Self-report measures and diagnostic interviews of chil
-
dren’s depression were used, so results were not lim
-
ited to only one method (questionnaire or interview).
Last, this study is the first to examine whether exces
-
sive reassurance seeking moderates the relation be
-
tween insecure attachment and depression.
At the same time, we note several limitations. First,
this study used a retrospective design. Although such a
design improves on simple cross-sectional research, it
still does not allow us to determine temporal prece
-
dence or causality. Thus, it is imperative that future re
-
search examine prospectively whether insecure attach
-
ment interacts with reassurance seeking to predict the
development of depression. Second, this study utilized
self-report measures to assess both insecure attach
-
ment and excessive reassurance seeking. Although both
the IPPA and the RSSC exhibit high degrees of reliabil
-
ity and validity, future research should assess these
constructs through other techniques (e.g., observa
-
tion). Third, this study utilized a high-risk design. Al
-
though this design allows for a strong test of vulnera
-
bility theories of depression by maximizing the
number of children who have experienced a depressive
episode (because parental depression is a strong risk
for child depression; Goodman & Gotlib, 2002), re
-
sults cannot be generalized to other populations. Fu
-
ture research should examine whether the interaction
of insecure attachment with excessive reassurance
seeking is associated with depression in a community
sample of children. Fourth, this study examined only
the relations among insecure attachment, excessive re
-
assurance seeking, and depression. Because we did not
assess nondepressive symptoms or disorders, we were
unable to identify whether the combination of these
vulnerability factors is specific to depressive disorders
rather than broadly applicable to other disorders. Fu-
ture research should assess a broader range of symp-
toms (e.g., anxiety) to investigate affective symptom
specificity of these etiological factors. In research with
adults, insecure attachment has been found to predict
both symptoms of depression and anxiety (e.g., Hankin
et al., in press), although a cognitive risk factor path-
way was found to mediate specifically the association
between insecure attachment and later depressive
symptoms, whereas an interpersonal stress generation
mechanism mediated the link between insecure attach
-
ment and both anxiety and depressive symptoms. Re
-
search indicates that excessive reassurance seeking
predicts depression specifically (Joiner & Metalsky,
2001).
In sum, the results of this study show that insecure
attachment to parents is related to depression in chil
-
dren, and excessive reassurance seeking moderates this
relation. This interaction of insecure attachment to par
-
ents with excessive reassurance seeking is associated
with severity of current and past depressive symptoms
as well as increased likelihood of having experienced
past clinically significant depressive episodes. Under
-
standing how such interpersonal factors influence the
etiology of depression may improve identification of
and enhance prevention efforts for children at risk of
experiencing depression.
References
Abela, J. R. Z., Adams, P., & Hankin, B. L. (2003). Interpersonal
vulnerability to depression in children and adolescents: A lon
-
gitudinal study. Manuscript submitted for publication.
190
ABELA ET AL.
Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978).
Patterns of attachment: A psychological study of the strange sit
-
uation. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.
Alloy, L. B., Whitehouse, W. G., Robinson, M. S., Abramson, L. Y.,
Hogan, M. E., Rose, D. T., et al. (2000). The Temple–Wiscon
-
sin Cognitive Vulnerability to Depression Project: Lifetime his
-
tory of Axis I psychopathology in individuals at high and low
cognitive risk for depression. Journal of Abnormal Psychology,
109, 403–418.
American Psychiatric Association. (1994). Diagnostic and statis
-
tical manual of mental disorders 4th ed. Washington, DC:
Author.
Armsden, G. C., & Greenberg, M. T. (1987). The Inventory of Parent
and Peer Attachment: Individual differences and their relation
-
ship to psychological well being in adolescents. Journal of
Youth and Adolescence, 16, 427–454.
Armsden, G. C., McCauley, E., Greenberg, M. T., Burke, P. M.,
& Mitchell, J. R. (1990). Parent and peer attachment in adoles
-
cent depression. Journal of Abnormal Child Psychology, 18,
683–697.
Blatt, S. J., & Homann, E. (1992). Parent–child interaction in the eti
-
ology of dependent and self-critical depression. Clinical Psy
-
chology Review, 12, 47–91.
Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment. New
York: Basic Books.
Bowlby, J. (1980). Attachment and loss: Vol. 3. Loss: Sadness and
depression. New York: Basic Books.
Cohen, J., & Cohen, P. (1983). Applied multiple regression/correla
-
tion analysis for the behavioural sciences (2nd ed.). Hillsdale,
NJ: Lawrence Erlbaum Associates, Inc.
Coyne, J. C. (1976). Toward an interactional description of depres-
sion. Psychiatry, 39, 28–40.
Coyne, J. C. (1994). Self-reported distress: Analog or ersatz depres-
sion? Psychological Bulletin, 116, 29–45.
Crowell, J., Fraley, R. C., & Shaver, P. R. (1999). Measures of indi-
vidual differences in adolescent and adult attachment. In J.
Cassidy & P. R. Shaver (Eds.), Handbook of attachment: The-
ory, research, and clinical applications (pp. 434–465). New
York: Guilford.
Cummings, E. M., & Cicchetti, D. (1990). Toward a transactional
model of relations between attachment and depression. In M. T.
Greenberg, D. Cicchetti, & E. M. Cummings (Eds.), Attach
-
ment in the preschool years: Theory, research and intervention
(339–372). Chicago: University of Chicago Press.
Davila, J. (2001). Refining the association between excessive reas
-
surance seeking and depressive symptoms: The role of related
interpersonal constructs. Journal of Social and Clinical Psy
-
chology, 20, 538–559.
First, M. B., Gibbon, M., Spitzer, R. L., & Williams, J. B. W. (1995).
User’s guide for the Structured Clinical Interview for DSM–IV
Axis I disorders (SCID–I, Version 2.0). Washington, DC:
American Psychiatric Press.
Flett, G. L., Vredenburg, K., & Krames, L. (1997). The continuity of
depression in clinical and nonclinical samples. Psychological
Bulletin, 121, 395–416.
Fraley, R. C. (2002). Attachment stability from infancy to adulthood:
Meta-analysis and dynamic modeling of developmental mecha
-
nisms. Personality and Social Psychology Review, 6, 123–151.
Goodman, S. H., & Gotlib, I. H. (2002). Children of depressed par
-
ents: Mechanisms of risk and implications for treatment. Wash
-
ington, DC: American Psychiatric Association.
Graham, C. A., & Easterbrooks, M. A. (2000). School-aged chil
-
dren’s vulnerability to depressive symptomatology: The role
of attachment security, maternal depressive symptomatology,
and economic risk. Development and Psychopathology, 12,
201–213.
Hammen, C. L., Burge, D., Daley, S. E., Davila, J., Paley, B., &
Rudolph, K. D. (1995). Interpersonal attachment cognitions
and prediction of symptomatic responses to interpersonal
stress. Journal of Abnormal Psychology, 104, 436–443.
Hankin, B. L., & Abramson, L. Y. (2001). Development of gender
differences in depression: An elaborated cognitive vulnera
-
bility–transactional stress theory. Psychological Bulletin, 127,
773–796.
Hankin, B. L., Abramson, L. Y., Moffitt, T. E., McGee, R., Silva, P.
A., & Angell, K. E. (1998). Development of depression from
preadolescence to young adulthood: Emerging gender differ
-
ences in a 10-year longitudinal study. Journal of Abnormal Psy
-
chology, 107, 128–140.
Hankin, B. L., Fraley, R. C., Lahey, B. B., & Waldman, I. D. (in
press). Is depression best viewed as a continuum or discrete cat
-
egory? A taxometric analysis of childhood and adolescent de
-
pression in a population-based sample. Journal of Abnormal
Psychology.
Hankin, B. L., Kassel, J. D., & Abela, J. R. Z. (in press). Adult at
-
tachment styles and specificity of emotional distress: Prospec
-
tive investigations of cognitive risk and interpersonal stress
generation as mediating mechanisms. Personality and Social
Psychology Bulletin.
Harrington, R., Rutter, M., & Frombonne, E. (1996). Developmental
pathways in depression: Multiple meanings, antecedents, and
endpoints. Development and Psychopathology, 8, 601–616.
Ingram, R. E., & Ritter, J. (2000). Vulnerability to depression: Cog
-
nitive reactivity and parental bonding in high-risk individuals.
Journal of Abnormal Psychology, 109, 588–596.
Joiner, T. E. (1999). A test of interpersonal theory of depression in
youth psychiatric inpatients. Journal of Abnormal Child Psy-
chology, 27, 77–85.
Joiner, T. E., Alfano, M. S., & Metalsky, G. I. (1992). When depres-
sion breeds contempt: Reassurance seeking, self-esteem, and
rejection of depressed college students by their roommates.
Journal of Abnormal Psychology, 101, 165–173.
Joiner, T. E., & Metalsky, G. I. (1995). A prospective test of an inte-
grative theory of depression. Journal of Personality and Social
Psychology, 69, 778–788.
Joiner, T. E., & Metalsky, G. I. (2001). Excessive reassurance seek-
ing: Delineating a risk factor in the development of depressive
symptoms. Psychological Science, 12, 371–378.
Joiner, T. E., Metalsky, G. I., Katz, J., & Beach, S. R. H. (1999). De
-
pression and excessive reassurance-seeking. Psychological In
-
quiry, 10, 269–278.
Joiner, T. E., & Schmidt, N. B. (1998). Excessive reassurance seek
-
ing predicts depressive but not anxious reactions to acute stress.
Journal of Abnormal Psychology, 107, 533–537.
Kaufman, J., Birmaher, B., Brent, B., Rao, U., & Ryan, N. (1996).
The Schedule for Affective Disorders for School-Age Children.
Pittsburgh: University of Pittsburgh Medical Center.
Kleinbaum, D. G. (1994). Logistic regression: A self-learning text.
New York: Springer-Verlag.
Kovacs, M. (1992). The Children’s Depression Inventory manual.
North Tonawanda, NY: Multi-Health Systems.
Muris, P., Mayer, P., & Meesters, C. (2000). Self-reported attach
-
ment style, anxiety, and depression in children. Social Behavior
and Personality, 28, 157–162.
Newman, D. L., Moffitt, T. E., Caspi, A., & Silva, P. A.
(1998). Comorbid mental disorders: Implications for treatment
and sample selection. Journal of Abnormal Psychology, 107,
305–311.
Roberts, J., Gotlib, I. H., & Kassel, J. D. (1996). Adult attachment
security and symptoms of depression: The mediating roles of
dysfunctional attitudes and low self-esteem. Journal of Person
-
ality and Social Psychology, 70, 310–320.
Rohde, P., Lewinsohn, P. M., & Seeley, J. R. (1994). Are adolescents
changed by an episode of major depression? Journal of
the American Academy of Child & Adolescent Psychiatry, 33,
1289–1298.
191
INTERPERSONAL VULNERABILITY TO DEPRESSION
West, M. L., Spreng, S. W., Rose, S. M., & Adam, K. S. (1999). Rela
-
tionship between attachment-felt security and history of sui
-
cidal behaviours in clinical adolescents. Canadian Journal of
Psychiatry, 44, 578–582.
Zanarini, M. C., Skodol, A. E., Bender, D., Dolan, R., Sanislow, C.,
Schaefer, E., et al. (2000). The Collaborative Longitudinal Per
-
sonality Disorders Study: Reliability of Axis I and Axis II diag
-
noses. Journal of Personality Disorders, 14, 291–299.
Received June 3, 2003
Accepted September 17, 2004
192
ABELA ET AL.
... Two studies utilized longitudinal data collected in the National Institute of Child Health and Human Development Study of Early Child Care (NICHD SECC; Milan et al. 2009Milan et al. , 2013. One study included retrospective measures assessing for past depressive symptoms and episodes alongside current depressive symptoms (Abela et al. 2005). Two studies made use of a multi-wave longitudinal design, one over the course of one year (Abela et al. 2009) and the other over the course of 20 years (Bureau et al. 2009). ...
... Two studies made use of data from the NICHD study of Early Child Care, which was sampled from 10 geographic areas across the U.S (Milan et al. 2009(Milan et al. , 2013, another study drew a sample from 7 public high schools in a large metropolitan area (Woodhouse et al. 2009), two studies examined samples from Belgium (Brenning et al. 2012), another two examined families from the Montreal Area (Abela et al. 2005;, one study sampled in southern Californian families (Fox and Borelli 2015), and two studies did not (2009) report what geographic location they drew their sample from (Brenning et al. 2011b;Bureau et al. 2009). ...
... The majority of the studies included in this review examined the mother-adolescent relationship. Only three of the nine studies identified included fathers in their examination of the intergenerational transmission of depressive symptoms (Abela et al. 2005(Abela et al. , 2009Woodhouse et al. 2009) and of these three studies, only Woodhouse et al. (2009) separated fathers in a way that allowed each parent-child relationship to be measured independently. All included studies provided information on parental marital status. ...
Article
Full-text available
John Bowlby’s theory of attachment has inspired numerous studies which have concluded that insecure attachment can contribute to the development of depressive symptoms in adolescents. It has also spurred research confirming that other environmental factors, such as parental depressive symptoms, can contribute to the development of insecure attachment in adolescents. However, few studies have investigated how insecure adolescent attachment works as a mechanism in the relationship between adolescent and parental depressive symptoms. Accordingly, this systematic review was conducted to better ascertain how attachment is related to the development of adolescent depressive symptoms in the context of parental depressive symptoms. Inclusion criteria for this review were met if the child study population had a mean age in our identified “adolescent” range (age 10–19), if depressive symptoms were measured in at least one parent and their adolescent child, and if adolescent attachment was related to (i.e., mediated or moderated) the intergenerational relationship between depressive symptoms. A total of nine studies were identified searching among three databases (APA PsychNET, Academic Search Elite and PubMed) and one search engine (Google Scholar) through March 2020. The overall sample of this review included mothers (N = 1,684), fathers (N = 123), and adolescents (N = 1805). Samples for the original studies were recruited from both clinical and community populations, with clinical samples being determined by both parental and adolescent levels of depressive symptoms in different studies. Attachment was measured as both an internal working model of self or “state of mind” and as a behavior in which a stressed child seeks comfort or support from an attachment figure acting as a “secure base.” Measurements of attachment were taken at varying stages in child development ranging from infancy through adolescence. Additionally, most reviewed studies did not include fathers or much racial or social class diversity. Overall, the review determined that both conceptualizations of attachment (internal working model and support-seeking behavior) successfully mediated or moderated the relationship between maternal and adolescent depressive symptoms.
... The findings reported above, lend support to the hypothesis that, less secure attachment was associated with greater severity of depression as reflected in the child's self-report on the Children's Depression Inventory. This finding is congruent with the results of a number of studies that relate the psychopathology to the quality of intimate relationships with parents and peers (Raja & et al, 1992;Laible & et al, 2000;Muris & et al, 2000;Graham & et al, 200;Buist & et al, 2004;Abela & et al, 2005;Hankin, 2005;Essau, 2004;Zachrisson, 2009). These findings are congruent with Bowlby's hypothesis regarding the relationships between attachment, anxiety and depression. ...
... The findings reported above, lend support to the hypothesis that, less secure attachment was associated with greater severity of depression as reflected in the child's self-report on the Children's Depression Inventory. This finding is congruent with the results of a number of studies that relate the psychopathology to the quality of intimate relationships with parents and peers (Raja & et al, 1992;Laible & et al, 2000;Muris & et al, 2000;Graham & et al, 200;Buist & et al, 2004;Abela & et al, 2005;Hankin, 2005;Essau, 2004;Zachrisson, 2009). These findings are congruent with Bowlby's hypothesis regarding the relationships between attachment, anxiety and depression. ...
Article
Full-text available
social self-efficacy refers to the individual capabilities in dealing with social challenges and the ability to handle interpersonal conflicts. A lot of research has demonstrated that family and the relationships among its members can pave the way to create social self-efficacy beliefs among children especially in dealing with social situations. The aim of this study is to examine the relationship between family communication patterns (Conversation and Conformity) and social self-efficacy in students. The study sample consisted of 496 fourth to sixth grade elementary school students (252 boys and 244 girls), who were administered the revised family communication patterns scale (Fitzpatrick and Ritchie, 1994) and the social selfefficacy questionnaire for students (Wheeler and Led, 1982). The reliability and validity of the instruments have been confirmed. Collected data was analyzed using the simultaneous multiple regression analysis. In addition and in order to compare social efficacy of boys and girls in different educational grades, a two-way ANOVA was used. Results showed that family Conversation orientation is a positive and significant predictor of social self-efficacy. Family Conformity orientation is, on the other hand, a significant negative predictor of social self-efficacy. There was no significant difference among different grades with regard to the total score of social self-efficacy, subscale of conflict situations, and subscale of non-conflict. In addition no significant differences were observed when comparing boys and girls in terms of their mean score of the subscale of non-conflict. However, regarding the conflict subscale, girls had significantly higher scores than the boys. Moreover the mean scores of boys and girls were significantly different with regard to the total social self-efficacy scale score and girls had higher levels of social self-efficacy than boys. As a result, the promotion of Conversation in the family is a predictor of social self-efficacy in children while emphasizing Conformity in the family diminishes the children's social self-efficacy.
... Given the different strategies for managing distress that are characteristic of anxious and avoidant individuals, it would follow that individuals with anxiety are more inclined to engage in ERS (a hyperactivating strategy) than individuals engage in avoidance (who favour deactivating strategies). In line with this, research has repeatedly shown an association between attachment anxiety and higher levels of overall and daily ERS (Abela et al., 2005;Davila, 2001;Katz, Petracca, & Rabinowitz, 2009;Shaver et al., 2005). As individuals with anxious attachment hold positive IWMs of others in addition to low self-esteem and a fear of abandonment, they are more likely to rely on reassurance from others to confirm their self-worth and relationship security (Mikulincer & Shaver, 2016). ...
... Part of the first objective was to replicate the positive association found in the literature between attachment anxiety and overall and daily ERS (Abela et al., 2005;Davila, 2001;Katz et al., 2009;Shaver et al., 2005). In support of prior research and the first hypothesis, higher attachment anxiety was associated with greater levels of daily ERS in both men and women. ...
Article
Full-text available
The current daily diary study examined the moderating impact of attachment style on the association between excessive reassurance seeking (ERS) behavior and trust in romantic dyads. A sample of 110 heterosexual couples completed measures of attachment, ERS, and relationship trust. In line with prior research, an anxious attachment style was associated with higher daily ERS, and an avoidant attachment style with lower daily ERS. Lower levels of trust were also associated with greater daily ERS. Moreover, analyses remained significant while controlling for symptoms of depression. This study extended the literature by demonstrating that for women with an anxious attachment style, and men with an avoidant attachment style, ERS was related to lower next day trust. In contrast, the partners of men with an avoidant attachment style, who also engaged in ERS, reported higher levels of next day trust. This study was also the first to examine how individual attachment styles influenced the perception of, and reactions to, ERS. Women with an anxious attachment style liked when their male partners engaged in ERS, as illustrated by higher levels of reported trust. These results support the idea that attachment styles play an important role in determining whether or not ERS leads to negative interpersonal consequences. They also suggest that it is the combination of relationship insecurities and ERS that leads to negative interpersonal consequences.
... A number of these behaviors have negative effects on the relationships of these individuals with others and are specific characteristics of depression . Therefore, it may be concluded that the higher the level of depression of a person, the greater is the susceptibility of that person to experience relational problems with others (Abela et al., 2005;Prinstein et al., 2005). Besides this, the positive friendship quality will decline, and conflict with peers will increase . ...
... Sejumlah perilaku tersebut berdampak negatif terhadap relasi individu tersebut dengan orang lain dan merupakan ciri perilaku yang khas dari depresi . Maka dari itu, dapat disimpulkan bahwa semakin tinggi tingkat depresi seorang individu, semakin rentan individu tersebut mengalami permasalahan relasional dengan orang lain (Abela et al., 2005;Prinstein et al., 2005). Selain itu, kualitas persahabatan positif akan menurun dan konflik dengan teman akan meningkat . ...
Article
Full-text available
Friendship is one of the forms of interpersonal relationship which may be experienced by people at some point in their lives. Friendships play larger roles when people move into adolescence, when peers begin to play larger roles. However, studies concerning friendship tend greatly to be focused upon mental health benefits for adolescents and their role as factors protective against mental health problems, predominantly internalizing disorder. Nonetheless, several of the latest pieces of research indicate that friendship does not always provide protection against internalizing disorder. Experts in the field of developmental psychopathology have recommended looking at the connection between friendships and internalizing disorder from a different viewpoint. Not just viewing friendships as protection against internalizing disorder, there are guidelines asking about the effects of internalizing disorder influencing friendships. This study attempted to delve deeper into whether the internalizing disorder such as depression and anxiety, regarding two indexes of the features of friendship (the positive features of friendship and the negative features of friendship), with non-clinical adolescents in the South East Asian context, involving 400 Indonesian participants (84% female adolescents; 16% male adolescents; Mage = 16.04) and 269 Malaysian participants (73% female adolescents; 27% male adolescents; Mage = 15.86). The results of hierarchical linear regression analysis indicated that the increase in depression rates predicted lower levels of the positive features of friendships and higher levels of the negative features of friendship. Meanwhile, an increase in anxiety predicted higher levels of the negative features of friendship. Besides this, there were no significant results from regression analysis regarding anxiety and the positive features of friendship. Additional findings and practical implications are discussed comprehensively. Persahabatan merupakan salah satu bentuk relasi interpersonal yang akan dialami seseorang pada satu titik dalam kehidupannya. Persahabatan mengambil peran yang lebih besar ketika individu beranjak ke masa remaja, ketika teman sebaya mulai mengambil peran yang lebih besar. Hanya saja, studi tentang persahabatan cenderung banyak difokuskan pada manfaatnya bagi kesehatan mental remaja serta perannya sebagai faktor protektif terhadap masalah kesehatan mental, terutama masalah internalisasi. Namun, beberapa penelitian terkini menunjukkan bahwa persahabatan tidak selalu memberikan proteksi terhadap masalah internalisasi. Ahli-ahli psikopatologi perkembangan sebenarnya sudah memberi saran untuk melihat hubungan antara persahabatan dan permasalahan internalisasi dari sudut pandang yang berbeda. Alih-alih melihat persahabatan sebagai pelindung dari permasalahan internalisasi, ada arahan untuk bertanya apa efek masalah internalisasi mempengaruhi persahabatan. Studi ini mencoba untuk mendalami lebih jauh mengenai peran masalah internalisasi seperti depresi dan kecemasan kepada dua indeks fitur persahabatan (fitur persahabatan positif dan fitur persahabatan negatif) pada remaja non klinis dalam konteks Asia Tenggara, dengan melibatkan 400 partisipan Indonesia (84% remaja perempuan; 16% remaja laki-laki; Musia = 16,04) dan 269 peserta Malaysia (73% remaja perempuan; 27% remaja laki-laki; Musia = 15,86). Hasil analisis hierarchical linear regression menunjukkan bahwa naiknya depresi memprediksi fitur persahabatan positif yang lebih rendah dan memprediksi fitur persahabatan negatif yang lebih tinggi. Sedangkan naiknya kecemasan memprediksi fitur persahabatan negatif yang lebih tinggi. Selain itu, tidak ditemukan hasil yang signifikan pada hasil analisis regresi untuk kecemasan dan fitur persahabatan positif. Temuan tambahan dan implikasi praktis didiskusikan secara komprehensif.
... The role of insecure and anxious attachment styles have been highlighted as influential in accounting for ERS in depression and health anxiety (e.g. Abela et al., 2005;Rector, Kamkar, Cassin, Ayearst, & Laposa, 2011;Stuart & Noyes, 1999). Studies including samples of adult non-clinical participants have also shown that insecure attachment is associated with obsessive compulsive symptoms via inflated sense of responsibility -where a person experiences a sense of personal responsibility for some potential negative outcome for self and/or others (e.g. ...
... To date, there are no directly comparable studies exploring ERS in the context of OCD among young people. Two studies investigated ERS in the context of depression in young people but did not employ a clinical comparison group (Abela et al., 2005;Abela, Zuroff, Ho, Adams, & Hankin, 2006). Lebowitz and colleagues (2014) compared samples of young people with OCD and anxiety finding no difference between them, in support of the current findings, but this related to parent report of accommodation which included reassurance provision rather than a discrete measure of ERS. ...
... Similarly, Beck (1967) suggests the presence of a negative triad that includes negative interpretation about the self, world and future among depressive individuals . These findings also echo interpersonal theories that emphasize the role of one's negative view of relationship with others in the development of depression (Abela et al., 2005) and empirical findings suggesting that individuals with symptoms of or risk for depression are more likely to have low self-worth, pessimistic expectations about interpersonal relations and belief that they are unworthy of healthy social relationships (Rudolph & Clark, 2001). ...
Article
Early maladaptive schemas (EMS) are broad, pervasive themes and patterns of emotions, memories, cognition and physical sensations that impede the goal of individuals. Maladaptive behaviours can occur as a response to maladaptive or negative schemas, often culminating in depression or anxiety. The current meta-analysis integrates the existing literature to estimate the magnitude of effect of association between EMS and depression among adolescents and young adults. A systematic search of seven different databases including Embase, CINAHL, Medline, ASSIA, Psych INFO, Scopus and Web of Science was carried out identifying 24 relevant studies of adolescents (10–18 years) and young adults (19–29 years). The random-effect model estimate for association between overall EMS and depression was r = 0.56 (95% CI 0.49–0.63, Z = 12.88, p ≤ 0.0001), suggesting higher predominant EMS significantly linked to higher levels of depressive symptoms, with a large effect size. Separate meta-analytical results with schema domains indicated moderately stronger associations between schemas of disconnection/rejection, impaired autonomy/performance and other-directedness with depression. Age and gender were not found to have any significant moderating effect on the associations. The findings suggest that it is vital for clinicians to identify specific maladaptive schemas contributing towards depression, to have a better understanding of underlying cognitive processes and in turn promote psychological health, well-being and resilience in adolescents and young adults. Furthermore, findings will also assist clinicians to focus more on the content of three significant schema domains, which emerged as particularly salient factors underlying adolescent depression.
... Studies have persistently reported the association between social support and mental health problems among adolescents [44][45][46]. Scientific evidence revealed that poor relationships with schoolmates, feeling of insecurity, and adverse life events increase the risk of mental disorders in adolescents [47,48]. Moreover, a longitudinal study revealed the feeling of connectedness, absence or low levels of conflict, and social environment, which is encouraging to express once the emotion is protective against the occurrence of emotional and behavioral disorders [49,50]. ...
Article
Full-text available
Objectives To determine the prevalence and socio-demographic and parental-related factors of depression among school adolescents in Jimma town, southwest Ethiopia. Methods Using a cross-sectional survey, 546 school adolescents were screened for depression using the patient health questionnaire (PHQ-9) from five randomly selected public and private schools. Oslo social support scale, adverse childhood experience tool, and socio-demographic questionnaire were used to gather data on risk factors. Linear regression analysis was used, and unstandardized beta (β) coefficients with 95% confidence intervals (CI) were reported to declare statistical significance. Results A total of 546 adolescents participated in the study, with a response rate of 97.3%. The mean (±SD) age of participants was 16.8 ± 1.3 years. The majority (81%) of the adolescents were attending day classes at public schools. The prevalence of depression was found to be 28% using the patient health questionnaire. Based on the PHQ-9 depression severity scale, 18.5% and 8.2% of the adolescent had moderate and moderate to severe depression while 1.3% had severe depression. In the final multivariate linear regression analysis, it was found that sex, rural residence, low social support, being in higher grade level, and adverse childhood experience were found to be independently associated with a higher score of depression. Conclusion One in three adolescents was found to have a depressive syndrome. We recommend schools to integrate school mental health service that contains routine screening and intervention services. Moreover, efforts are needed to sensitize and educate the communities on child protection, social support, and safeguarding to effectively tackle the magnitude of adolescent depression.
Chapter
Affective disorders, including major depressive disorder, persistent depressive disorder, and bipolar spectrum disorders, are common, recurrent, and debilitating disorders. In this review, we present updated research on these mood disorders, including their phenomenology, epidemiology, risk factors, and theory and evidence for the role of behavioral/motivational, cognitive, interpersonal/psychodynamic, and neurobiological mechanisms in their etiology and course. We also present treatments for affective disorders derived from the various theoretical perspectives on their etiology and evidence regarding the efficacy of these interventions as well as directions for future research.
Article
Introduction: Negative feedback seeking and excessive reassurance seeking behaviors in interpersonal relationships have been shown to frequently occur in conjunction with levels of depression. Method: We used meta-analysis to examine 102 studies (134 effects), relating depression with negative feedback seeking (k = 31) and/or excessive reassurance seeking (k = 103). Results: Depression had positive, moderate effect sizes with both negative feedback seeking (r = .26, 95% CI [.21, .32], p < .001, k = 31) and excessive reassurance seeking (r = .33, 95% CI [.31, .36] p < .001, k = 103). Subgroup analysis revealed the effect size for negative feedback seeking was smaller in romantic relationships compared to other relationship types. Effect sizes for excessive reassurance seeking did not differ for romantic and other relationships but were smaller in romantic relationships of a longer duration. Participant gender and symptom severity did not moderate effect sizes. Studies with child and adolescent samples had larger effects for negative feedback seeking and smaller effects for excessive reassurance seeking, relative to adult samples. Discussion: Longer closer relationships may protect against maladaptive interpersonal behaviors in individuals with depression.
Article
Full-text available
The results of two studies are reported. Study I involved the development of the Inventory of Parent and Peer Attachment (IPPA), a self-report instrument for use with adolescents. Subject were 179 college students aged 16-20 years. Item content of the instrument was suggested by attachment theory's formulations concerning the nature of feelings toward attachment figures. In Study II, the convergent validity of the IPPA was examined. Also, a hierarchial regression model was employed to investigate the association between quality of attachment and self-esteem, life-satisfaction, and affective status. Respondents were 86 adolescents from the Study I sample. As hypothesized, perceived quality of both parent and peer attachments was significantly related to psychological well-being. Results of the development of a theoretically focused, exploratory classification scheme indicated that adolescents classified as highly securely attached reported greater satisfaction with themselves, a higher likelihood of seeking social support, and less symptomatic response to stressful life events.
Article
Three studies investigated the relation between adult attachment security and symptoms of depression. Study 1 examined the overall magnitude of the association between adult attachment and depression, and Studies 2 and 3 tested whether this relation was mediated by dysfunctional attitudes and low self-esteem. Results from the three studies were consistent with a mediation model. This model suggests that insecure adult attachment styles are associated with dysfunctional attitudes, which in turn predispose to lower levels of self-esteem. Such depletions in self-esteem are directly associated with increases in depressive symptoms over time. Insecure attachment appears to lead to depressive symptoms in adulthood through its impact on self-worth contingencies and self-esteem.
Article
Although various conceptual proposals have suggested that disruptions in childhood bonding processes may be linked to the origins of these cognitive structures, little research has tested these proposals. This study assessed the information processing of vulnerable individuals and its relationship to childhood bonding. Formerly depressed (vulnerable) and never depressed (nonvulnerable) individuals participated in a mood induction task followed by an attentional allocation task. Results indicated that vulnerable individuals uniquely diverted attention toward negative stimuli when they were in a negative mood. Furthermore. level of maternal caring was found to be associated with performance on this task for vulnerable individuals in this mood state. These data support the idea that cognitive variables form a pathway between troublesome parental-child/adolescent interactions and depression.
Article
Two studies examined the validity of Joiner and colleagues' (Joiner, Alfano, & Metalsky, 1992) measure of excessive reassurance seeking. These studies specifically examined whether the association between excessive reassurance seeking and depressive symptoms was better accounted for by conceptually related interpersonal variables also known to be associated with depression (e.g., sociotropy, attachment insecurity). Results from cross-sectional and longitudinal analyses in college student samples generally supported the validity of excessive reassurance seeking and its association with depressive symptoms. The implications of these findings for the role of excessive reassurance seeking in depressive vulnerability are discussed.
Article
The arrival of a book for review usually gives rise to pleasant anticipation, and whatever criticisms have to be made, it is that almost always possible to find some pleasant things to say. But finding praise for this tome is a problem — it is a volume too far. It is to be hoped that the authors
Article
Ethological attachment theory is a landmark of 20th century social and behavioral sciences theory and research. This new paradigm for understanding primary relationships across the lifespan evolved from John Bowlby's critique of psychoanalytic drive theory and his own clinical observations, supplemented by his knowledge of fields as diverse as primate ethology, control systems theory, and cognitive psychology. By the time he had written the first volume of his classic Attachment and Loss trilogy, Mary D. Salter Ainsworth's naturalistic observations in Uganda and Baltimore, and her theoretical and descriptive insights about maternal care and the secure base phenomenon had become integral to attachment theory. Patterns of Attachment reports the methods and key results of Ainsworth's landmark Baltimore Longitudinal Study. Following upon her naturalistic home observations in Uganda, the Baltimore project yielded a wealth of enduring, benchmark results on the nature of the child's tie to its primary caregiver and the importance of early experience. It also addressed a wide range of conceptual and methodological issues common to many developmental and longitudinal projects, especially issues of age appropriate assessment, quantifying behavior, and comprehending individual differences. In addition, Ainsworth and her students broke new ground, clarifying and defining new concepts, demonstrating the value of the ethological methods and insights about behavior. Today, as we enter the fourth generation of attachment study, we have a rich and growing catalogue of behavioral and narrative approaches to measuring attachment from infancy to adulthood. Each of them has roots in the Strange Situation and the secure base concept presented in Patterns of Attachment. It inclusion in the Psychology Press Classic Editions series reflects Patterns of Attachment's continuing significance and insures its availability to new generations of students, researchers, and clinicians.