Ghosts from the Past?

The Association between Childhood interpersonal Trauma, Attachment

and Anxiety and Depression in Late Life

Lies Van Asschea, Luc Van de Vena, Mathieu Vandenbulckea, Patrick Luytenb,c

a
Section of Geriatric Psychiatry, Department of Psychiatry, University Hospitals Leuven,
KUL, Belgium
b
Faculty of Psychology and Educational Sciences, University of Leuven, Belgium
c
Research Department of Clinical, Educational and Health Psychology, University College
London, UK

Word count: 4,644

Corresponding author:
Lies Van Assche
University Hospitals Leuven
Herestraat 49
3000 Leuven
+3216341300
[email protected]

1
Abstract

Objectives. Research suggests that vulnerability for anxiety and depression in late life results

from a complex interaction between (neuro)biological and environmental factors. In this

context, there is growing evidence for the role of childhood trauma on vulnerability for both

anxiety and depression throughout the course of life, mainly through its effects on attachment

as a biologically based neurodevelopmental stress regulation system. Yet, the impact of

childhood trauma on depression and anxiety in late life specifically remains unclear. The

current study therefore aims to investigate the association between retrospectively reported

childhood interpersonal trauma, attachment dimensions and levels of anxiety and depression

in late life. Method. A sample of 81 community dwelling older adults completed measures of

early and current adversity, attachment dimensions, and levels of anxiety and depression.

Results. The occurrence and frequency of childhood trauma, but not later negative adult life

events, was associated with late life anxiety and depression. Both attachment anxiety and

avoidance were related to anxiety and depression. Only attachment anxiety affected the

association between childhood trauma, and emotional neglect in particular, and late life

anxiety and depression. Conclusion. Childhood trauma may be associated with anxiety and

depression in late life. Part of this association is probably indirect, via the effect of insecure

attachment and high levels of attachment anxiety in particular.

Keywords: aging, trauma, attachment, anxiety, depression

Disclosure statement

The authors report no conflicts of interest.

2
Introduction

Depression and anxiety are highly prevalent in late life. Indeed, estimates suggest that

depression occurs in approximately 25% of the elderly (Forlani et al., 2014), whereas the

prevalence of anxiety disorders ranges from 10 to 20% in older adults (Cassidy & Rector,

2008; Kirmizioglu, Dogan, Kugu, & Akyüz, 2009). Moreover, comorbidity between

depression and anxiety in late life is substantial (Beekman et al., 2002) and is related to poor

clinical outcome (Beekman et al., 2002; Flint, 1994; Cassidy & Rector, 2008). Studies have

consistently indicated that exposure to childhood adversity increases susceptibility to both

depression and anxiety in early adulthood (Anderson, Fields, & Dobb, 2011; Vink, Aartsen,

& Schoevers, 2008) and many older adults have been confronted with childhood abuse or

neglect. Specifically, estimated prevalence rates of childhood trauma in American adults older

than 55 years were 13.5% for verbal abuse, 9.6% for physical abuse and 9.3% for sexual

abuse (Bynum et al., 2010). Yet, there is still a paucity of studies in this area in older adults,

and relatively little is understood about the mechanisms explaining the relationship between

childhood trauma and anxiety and depression in late life.

The Life Cycle Model of Stress and Wellbeing in Late Life

The life cycle model of stress provides a comprehensive theoretical framework to

understand putative links between early adversity and vulnerability for anxiety and depression

in late life. This model essentially suggests that disorders that have their onset in late life

might reflect ‘delayed’ manifestation effects of exposure to early life stressors (Lupien,

McEwen, Gunnar, & Heim, 2009). Specifically, repeated exposure to severe distress has been

suggested to produce enduring effects on the brain through activation of the hypothalamus-

pituitary-adrenal (HPA) axis which leads to the production of glucocorticoids that exert

harmful effects on brain tissue, particularly during so-called ‘critical time windows’ of

3
increased neuronal plasticity in brain areas involved in the regulation of stress, such as the

hippocampus, the amygdala and the prefrontal cortex (Giedd et al., 1996; Pruessner et al.,

2010). Whilst these ‘programming’ effects may manifest in early childhood or adulthood, in a

substantial proportion of individuals these effects may not manifest until late life (Clark,

Caldwell, Power, & Stansfield, 2010). Both age-related brain changes and age-specific

challenges (and particularly experiences of loss of attachment figures) have been argued to

play a key role in this context. Indeed, old age is associated with several challenges

concerning physical health, sensory loss, retirement with associated changes in income,

narrowing social networks, cognitive declines, and increasing awareness of mortality (Charles

& Carstensen, 2010) that increase frailty which may lead to the onset of psychopathology in

late life in individuals who have been exposed to early life trauma but have shown a resilience

to the development of mental health issues until then. For instance, in a group of 567

community dwelling adults older than 60 years, an increased frailty as assessed using the

Fried biological syndrome model was associated with more anxiety and depression compared

with robust older adults (Ni Mhaolain et al., 2012). More broadly, in social sciences, there has

been increasing interest in the concept of ‘cumulative’ advantage or disadvantage in several

domains in explaining the impact of aging on wellbeing (Ferraro & Shippee, 2009).

Consistent with these concepts and assumptions, a number of studies have indeed found an

association between childhood adversity and increased vulnerability for psychopathology in

late life, even after controlling for early adulthood psychiatric symptoms (Clark, Caldwell,

Power, & Stansfield, 2010; Falk, Hersen, & van Hasselt, 1994; Gershon, Sudheimer,

Tirouvanziam, Williams, & O'Hara, 2013). Yet, more research in this area is clearly needed,

particularly on the role of attachment in these relationships as the attachment system has been

shown to be a key modulator of stress and adversity throughout life (Nolte et al., 2011).

4
The Role of Attachment

The detrimental effect of childhood interpersonal trauma may be better understood if we

take into account that the development of stress and affect regulation largely occurs in the

context of attachment relationships (Mikulincer & Shaver, 2007). Contemporary attachment

formulations indeed converge to suggest that the attachment system may be thought of as a

behavioural system that is activated when faced with threat. It involves the coordination of

different subsystems aimed at reducing distress through seeking actual or imagined proximity

to an attachment figure (Fonagy & Luyten, 2009; Mikulincer & Shaver, 2007; Sbarra &

Hazan, 2008).

It is assumed that proximity to a sensitive caregiver who is responsive to the needs of a

child fosters the development of adaptive affect regulation strategies typical of secure

attachment in adulthood. Individuals who are securely attached generally have a strong sense

of self-efficacy. Moreover, they perceive others as trustworthy, accessible and well-

intentioned, and thus have the ability to turn to others in times of stress and adversity.

By contrast, when attachment figures have been unavailable, nonresponsive or abusing,

secondary attachment strategies develop that function as habitual responses to stress in later

life. The first of these secondary or compensatory strategies is typically observed in

individuals with anxious attachment, and consists of a hyperactivation of the attachment

system. Individuals with high levels of attachment anxiety often view themselves as

incompetent in dealing with life’s challenges and they are hypervigilant for signs of rejection

and abandonment by significant others. Difficulties in regulating stress in adults who

primarily use hyperactivating strategies are therefore typically associated with an excessive

need for others (Mikulincer & Shaver, 2007), but paradoxically they are not easily comforted

in times of distress because of their underlying belief that others will not be there for them.

5
 A second compensatory attachment strategy that is observed in individuals with avoidant

attachment features, entails attempts to deactivate the attachment system and avoid contact

with attachment figures during stressful times. When distressed, these individuals have a

tendency towards compulsive autonomy (i.e., the conviction that they have to be able to deal

with distress without the help of others), and consequently they tend to suppress negative

emotions. As there is a relative stability in the quality of attachment throughout life, early

attachment experiences may have a continuing impact on interpersonal functioning

throughout the lifespan (Van Assche et al., 2013).

Indeed, there is considerable research that has found a relationship between adult

attachment insecurity and vulnerability for psychopathology in adulthood and late life

(Mikulicer & Shaver, 2012; Van Assche et al., 2013). Additionally, childhood trauma as

opposed to adulthood trauma has been linked to the development of insecure attachment

patterns (Ogle, Rubin & Siegler, 2015), without consistent evidence for the differential impact

of different types of trauma on attachment (e.g. Erozkan, 2016). Insecure attachment patterns,

in their turn, affect the quality of relationships and psychological function in an ongoing

fashion (Fowler, Allen, Oldham, & Frueh, 2013). Coincidently, studies also converge to

suggest that higher levels of attachment anxiety are negatively related to self-reported

wellbeing in older adults (Andersson & Stevens, 1993; Bodner & Cohen-Fridel, 2010;

Cicirelli, 1989; Cookman, 2005; Kafetsios & Sideridis, 2006; Park & Vandenberg, 1994;

Webster, 1997, 1998; Wensauer & Grossmann, 1995). By contrast, the use of attachment

deactivating strategies has been found to be positively related to greater self-reported

psychological and physical wellbeing in old age in some studies (Jain & Labouvie-Vief,

2010). Although this positive association seems counter-intuitive, it may be explained by a

reporting bias typical of individuals that primarily rely on attachment deactivating strategies.

Jain and Labouvie-Vief (2010), for instance, found that attachment avoidance in community

6
dwelling older adults was at the same time associated with relatively high self-reported

wellbeing, but also with elevated heart rate and blood pressure which suggests that there still

are increased levels of distress on a subconscious level.

The Present Study

The present study aims to further explore the nature of the relationship between childhood

interpersonal trauma and late life anxiety or depression. Specifically, we hypothesized that

attachment anxiety and avoidance in adulthood would mediate the association between

childhood trauma and anxiety or depression in older adults. Based on the findings reviewed

above, we hypothesized that childhood trauma in the elderly will be associated with higher

levels of anxiety and depression. We also predicted a dose-response effect, i.e., that levels of

self-reported anxiety and depression would increase as the number of trauma’s increases.

Finally, we expected that the relationship between childhood trauma and late life anxiety and

depression would be indirect, through the effect of associations with attachment dimensions.

Specifically, we expected that findings for attachment anxiety would be strongest, as

attachment avoidance has been shown to be associated with a tendency to suppress negative

emotions and underreport distressing events, which can be expected to attenuate the

relationship between childhood trauma and feelings of anxiety and depression in late life.

Method

Participants and Procedure

A sample of community dwelling older adults were recruited from organisations for seniors.

The main researcher visited several meetings from senior citizen organisations, provided

information about the current study and prepared leaflets with practical information about

participation in the study (timing, fee, …). Older adults who were present at the time of the

7
visits received a leaflet with information. Those willing to participate filled in a form

containing demographic information such as date of birth, formal education, marital status

and number of children. The candidates for participation in the study also provided a

telephone number or email address. The main researcher collected all the leaflets and

participants were contacted by telephone or email within two months after recruitment. Extra

leaflets were given to the persons responsible for the organisations and older adults who were

not present during the visits but who read the leaflet and were interested in participation,

could send an email or call the main researcher. The study was approved by the ethics

committee of the University Hospitals of Leuven and all subjects signed an informed consent

to the inclusion of material pertaining to themselves. Participation was anonymous. In total 81

older adults aged 74.90 (SD: 6.64, range: 62-90) agreed to participate in the study. There were

29 male (36%) and 52 female volunteers (64%). Most were married (58%) or widowed

(27%). A minority was single (9%) or divorced (6%). The average number of children was 2

(SD = 1.34; range 0-5). The mean number of years in formal education was 12.85 (SD = 3.66,

range 6-16). A brief screening using the Mini Mental State Examination (MMSE) (Folstein,

Folstein, & McHugh, 1975) showed intact cognitive functioning in all of the participants with

a mean total score of 28.99 (SD: 0.13, range 26-30).

Measures

Childhood Trauma and Adult Adversity

Information about childhood adverse life events was acquired using the Childhood Trauma

Questionnaire – Short Form (CTQ-SF) (Bernstein et al., 2003). The CTQ-SF contains 25

questions that need to be scored on a five point Likert scale ranging from 1 = ‘Never true’ to 5

= ‘Very often true’, pertaining to five subscales: emotional neglect, physical neglect,

emotional abuse, physical abuse, sexual abuse. Besides dimensional scores, cut-off scores

8
based on the CTQ manual were used to assess the presence or absence of different types of

trauma (emotional abuse cut-off 12, physical abuse cut-off 9, sexual abuse cut-off 7, physical

neglect cut-off 9, emotional neglect cut-off 14) (Bernstein & Fink, 1998). We also divided

participants into two groups based on the presence of any type of trauma as opposed to none.

Finally, we calculated a score based on the total number of traumatic events, resulting in a

score ranging from 0 ( = no trauma) to 5 ( = 5 different types of trauma). The CTQ-SF has

shown good reliability and validity as a measure for retrospective report of childhood trauma

(Bernstein et al., 2003). Specifically, there was measurement invariance across different

samples of individuals with substance abuse, adolescent psychiatric inpatients and a

normative community sample (combined n= 1978). Confirmatory factor analysis showed

significant factor loadings of the individual items and the scales with alpha coefficients

ranging from .84 to .89 across groups on emotional abuse, .81 to .86 for physical abuse, .92 to

.95 for sexual abuse, .85 to .91 for emotional neglect and slightly lower factor loadings

ranging from .61 to .78 for physical neglect. Furthermore, the questionnaire has shown good

criterion related validity in a subsample of 179 adolescents for whom corroborative data on

trauma were available. Specifically, the CTQ latent constructs that were obtained through

confirmatory factor analysis significantly predicted therapists’ observations of abuse and

neglect using the child maltreatment ascertainment interview (CMAI), with an excellent fit in

a path model that allowed covariance among predictor variables χ2 (361, n = 179) = 550.08; p

< .001; comparative fit index (CFI) = .93; root mean square error of approximation (RMSEA)

= .05 (Bernstein et al., 2003). A validation study of the Dutch CTQ-SF showed good

reliability for physical abuse (α = .91), emotional abuse (α = .89), sexual abuse (α = .95) and

emotional neglect (α = .91). There was a slightly lower reliability for physical neglect (α =

.63) (Thombs, Brett, Bernstein, Lobbestael & Arntz, 2009). In the current sample of older

community dwelling adults, internal consistency reliability of the CTQ subscales was good in

9
the subscales emotional abuse (λ-2 = .83; α = .83), sexual abuse (λ-2 = .92; α = .91), physical

abuse (λ-2 = .85; α = .80) and emotional neglect (λ-2 = .81; α = .80). In line with prior

research, reliability was slightly lower in the physical neglect subscale (λ-2 = .62; α = .59).

Negative life events in adulthood (from age 18 onwards) were assessed using the

Psychiatric Epidemiology Research Interview (PERI) (Dohrenwend & Dohrenwend, 1982).

The participants were first asked whether they had experienced specific events pertaining to

several domains (work, family, friends, romantic relationships, health, crime, finance). Next,

they had to indicate how distressing each event was on a Likert scale ranging from 1 to 7. The

total number of events was used as a measure of adulthood adversity and the total sum of

subjective distress associated with the events was used as a measure for the distress associated

with adulthood adversity.

Attachment Dimensions

The Experiences in Close Relationships – Revised (ECR-R) self-report questionnaire (Fraley,

Waller, & Brennan, 2000) was used to assess levels of attachment avoidance and attachment

anxiety in adulthood. The ECR-R consists of 36 items which have to be scored on a 6-point

Likert type scale, with 18 items measuring the level of attachment anxiety and the other 18

items assessing the level of attachment avoidance. Participants were invited to respond to the

questions keeping in mind their relationship with one important attachment figure: partner,

(adult) child, sibling, friend or other. In this study, scores for attachment anxiety and

avoidance were averaged. Higher scores are suggestive of more attachment anxiety or

avoidance. The ECR-R has demonstrated good psychometric qualities (Sibley, Fischer, & Liu,

2005) with a Cronbach’s alfa of .91 for attachment avoidance and .94 for attachment anxiety.

Additionally, confirmatory factor analysis showed that a two factor model exhibited an

excellent fit to the data, with χ2(53, n = 478) = 142.26, CFI = .98, RMSEA = .06. A Dutch

10
version of the ECR-R demonstrated good reliability for attachment avoidance (α = .94) and

attachment anxiety (α = .92). Moreover, confirmatory factor analysis showed that the

hypothesized two-factor model of the ECR-R, showed a good fit with the data, χ2(53, n =

262) = 169.650, CFI = .95, RMSEA = .092 (Kooiman, Klaassens, van Heloma Lugt &

Kamperman, 2013). Internal consistency reliability was also good in our sample for

attachment anxiety (λ-2 = .90; α = .90) and avoidance (λ-2 = .86; α = .85).

Anxiety and Depression

The anxiety subscale of the Hospital Anxiety and Depression Scale (HADS-A) (Zigmond &

Snaith, 1983) and the Geriatric Depression Scale (GDS) (Yesavage et al., 1982) were used to

assess current levels of anxiety and depression, respectively. Both instruments are brief, easy

to administer and have been shown to have good psychometric properties with a Cronbach α

of .83 for the HADS-A (Herrmann, 1997; Bjelland, Dahl, Haug & Neckelmann, 2002) and .82

for the Dutch translation in a group of adults aged more than 65 years (Spinhoven, Ormel,

Sloekers, Kempen, Speckens & van Hemert, 1997). There was a Cronbach α of .94 for the

English version of the GDS (Yesavage et al., 1982) and a Cronbach α of .87 for the Dutch

version of the GDS (van Marwijk, Wallace, de Bock, Hermans, Kaptein & Mulder, 1985).

Internal consistency reliability in our sample is good for the HADS-A (λ-2 = .85; α = .84) and

GDS (λ-2 = .82; α = .81). Scores of 7 or higher on the HADS-A are considered clinically

significant. The GDS distinguishes between mild depression (scores between 11 and 13),

moderate depression (scores ranging from 14-20) and severe depression (scores > 20).

Cognition

Finally, a standardized Dutch version of the Mini Mental Status Examination (MMSE)

(Folstein et al., 1975; Kok, Verhey, & Schmand, 2004) was used as a tool for the assessment

11
of cognitive abilities. It measures orientation skills, memory function, concentration, language

and praxis. The maximum score is 30, with 23 considered as an appropriate cut-off value with

a sensitivity of 0.82 (Bour, Rasquin, Boreas, Limburg, & Verhey, 2010). Scores below this

cut-off are suggestive of cognitive dysfunction that is more pronounced than normal.

Data Analysis

Associations between the study variables were explored based on bivariate Pearson

correlations and biserial correlations for the association between continuous and dichotomized

variables. Next, mediation analyses were performed using a bias-corrected bootstrapping

approach, because of the relatively small sample size, to investigate the hypothesized

mediation models involving adult attachment, dimensional measures of childhood trauma and

depression and anxiety in late life (Preacher & Hayes, 2004). Analyses were performed using

SPSS Version 24. Results were considered statistically significant at the p < .05 level.

Results

Almost half of the participants reported having experienced at least one type of childhood

trauma. Emotional and physical neglect were reported most (see Table 1). Participants

reported on average 11 negative life events in adulthood (SD = 3.6; range 3-19), with most

events relating to the family (e.g. a severe illness or death of a family member), children or

friends (e.g. severe illness or death of a friend, falling out with a friend) and health (e.g.

surgery, reduced mobility). Work, financial adversity, crime or issues related to romantic

relationships accounted for only a small proportion of the negative life events reported (Table

1).

There was an average level of anxiety of 4.41, with a range of 0 to 18. Twenty individuals

reported elevated levels of anxiety (cut-off value of 7 or more). There was an average level of

12
5.56 for the GDS, with a range of 0 to 22. There were nine individuals who reported clinically

significant depressive feelings (cut-off value of 11 or more). Anxiety and depression were

strongly correlated, as was expected based on prior research (r = .59, p <.01).

Zero-order correlations between childhood traumatic events, adulthood negative life events

and anxiety and depression are shown in Table 2. First, the presence of any childhood trauma

was positively related to current levels of anxiety and depression. The total number of

childhood traumatic events reported, was significantly positively correlated with anxiety, but

not with depression. The number of negative life events experienced in adulthood was not

significantly correlated with depression or anxiety.

With respect to specific types of trauma, both emotional abuse and neglect showed a

positive association with current levels of anxiety. Emotional neglect was also significantly

and positively related to depression. Physical neglect, physical abuse, and sexual abuse,

however, were not significantly correlated with depression or anxiety (Table 2).

Physical abuse was strongly negatively correlated with attachment avoidance. Emotional

neglect, on the other hand, was significantly positively associated with attachment anxiety.

There were no significant correlations between attachment anxiety or avoidance and physical

neglect, emotional abuse or sexual abuse. Both attachment anxiety and attachment avoidance

showed a significant positive correlation with current levels of anxiety and depression (Table

2).

Hence, only for attachment anxiety and emotional neglect, conditions for mediation were

satisfied and we thus investigated whether emotional neglect was associated with late life

anxiety and depression through the effect of attachment anxiety. Results showed that there

was indeed an indirect effect of emotional neglect on late life anxiety through attachment

anxiety (Table 3). There was no evidence that emotional neglect influenced current levels of

anxiety directly (c’ = 1.85, p = .06). The indirect effect coefficient was .04 (confidence

13
intervals: .00-.15). Additionally, there was an indirect relationship between emotional neglect

and late life depression through the effect of attachment anxiety (Table 4). Again, there was

no evidence that emotional neglect influenced current levels of depression directly (c’ = 1.83,

p = .14). The indirect effect coefficient was .05 (confidence intervals: .00-.17).

Discussion

In this sample of community-dwelling older adults approximately half of all participants

reported having been exposed to at least one type of childhood trauma, which slightly exceeds

prevalence rates usually reported in younger cohorts (Saunders & Adams, 2014). Prior

research indeed suggests that the occurrence of childhood maltreatment has declined in some

regions, particularly in Western countries (Finkelhor & Jones, 2006), leading to lower

prevalence rates in younger cohorts. Moreover, emotional and physical neglect was reported

more often than abuse, which may also reflect cohort effects, as pedagogic changes and

greater improvements in welfare have occurred over the past few decades.

Surprisingly, only childhood trauma, but not negative life events that occurred in

adulthood, were related to late life anxiety and depression. We also found a dose-response

effect as individuals who were exposed to different types of childhood trauma, experienced

higher levels of both anxiety and depression. These findings are consistent with prior research

on the enduring effects of childhood trauma (Lupien et al., 2009). Indeed, other researchers

have also looked into the differential impact of early versus adulthood trauma and they found

a significant interaction between the developmental timing of the trauma and attachment

anxiety, showing that attachment anxiety was stronger in older individuals with Post

Traumatic Stress Disorder (PTSD) symptoms and childhood trauma compared with older

adults with PTSD symptoms and adulthood trauma (Ogle, Rubin & Siegler, 2015).

14
 In line with these findings, attachment anxiety was positively related to emotional neglect

in early youth in our sample of older adults and it was also related to current levels of anxiety

and depression. Moreover, the relationship between childhood emotional neglect and current

levels of anxiety and depression was indirect, via the effect of attachment anxiety. Cognitive-

affective schema’s in individuals with high levels of attachment anxiety are assumed to

encompass the belief that one is incompetent to solve problems oneself. This belief might lead

to higher levels of anxiety and depression in late life, when individuals are exposed to more

age-related losses and social networks typically decrease (Kafetsios & Sideridis, 2006).

However, at the same time, individuals with high levels of attachment anxiety may be

inclined to ‘over-report’ trauma, and emotional neglect in particular. Indeed, because of their

excessive need of reassurance, individuals with high levels of attachment anxiety might feel

neglected more easily when attachment figures have been unable to meet their emotional

needs (Mikulincer & Florian, 1998).

Attachment avoidance, to the contrary, was associated with a smaller number of reported

childhood trauma’s. This was particularly the case for physical abuse. Although it is difficult

to draw strong conclusions based on the current cross-sectional data, this latter finding is

similar to findings from other studies suggesting that individuals high on attachment

avoidance tend to cope with distress through defensive attempts to deemphasize the

importance of negative emotions related to these events (Shaver, Collins, & Clark, 1996).

This may also result in a tendency to underreport distressing events. However, this

mechanism, which may be adaptive in the short term, might become counterproductive in the

longer term, and especially in late life, as it may increasingly fail, leading to higher levels of

self-reported anxiety and depression in individuals with high levels of attachment avoidance

(Jain & Labouvie-Vief, 2010). Indeed, Gross and Levenson (1993) have reported that

suppression of negative emotions actually puts individuals under constant and considerable

15
strain as it results in greater physiological arousal, which could lead to physical and emotional

‘wear and tear’ and ultimately a collapse of emotion regulation strategies resulting in higher

self-reported anxiety and depression.

In summary, results suggest that childhood trauma may negatively impact on late life

wellbeing, especially in individuals with higher levels of attachment insecurity. Hence, a

systematic screening of early life trauma in older patients seems important and an awareness

of the importance of security in attachment to the therapist as well as for instance family

members or nursing staff. Specifically, attachment based interventions may be beneficial in

treating older patients who present with anxious or depressive symptoms and who have a

history of trauma (Mikulincer & Florian, 1998). Also, the clinician needs to be aware of the

possibility of underreporting distress and distressing events in individuals with avoidant

attachment characteristics.

Further longitudinal research is needed, however, to replicate these findings. Furthermore,

the mechanism underlying this association may be different in individuals that have high

levels of attachment avoidance versus those with high levels of attachment anxiety.

Other limitations of the current study include the relatively small sample size (n=81),

which possibly reduces statistical power. Power analysis for multiple correlations indeed

showed that a sample of 87 individuals would insure a power of .80 to detect correlations with

a coefficient of at least .30 and alpha set at .05 (Gatsonis & Sampson, 1989; Hulley,

Cummings, Browner, Grady & Newman, 2013). Still, the use of bootstrapping in the

mediation analyses mitigates this limitation somewhat (Preacher & Hayes, 2004). However,

because of the small sample size, we were not able to conduct analyses for men and women

separately, although gender has often been shown to play an important role in research on the

impact of childhood trauma (Varese et al., 2012). Second, the cross-sectional nature of the

study design limits the possibilities for causal interpretation. Retrospective reports of

16
childhood trauma may introduce an important bias. It is likely, as we have discussed, that the

quality of current attachment working models has an influence on the reporting and

recollection of childhood traumatic events, with attachment avoidance leading to a tendency

to under-report such events, whereas attachment anxiety, to the contrary, may lead to the

over-reporting of negative affect and/or negative events. Next, an important methodological

limitation is the lack of data on PTSD symptoms in our group. Also, we did not assess

adulthood trauma, but adulthood negative life events that might not necessarily be traumatic

and thus less harsh. This may already explain a difference in the observed associations of

adulthood versus childhood adversity. Indeed, some prior researches have even demonstrated

an inverse relationship (Shrira, Shmotkin & Litwin, 2012). Specifically, when data concerning

potentially traumatic adulthood and childhood events were collected in 1130 older Israeli

participants, and these were correlated with measures of wellbeing, only adulthood adversity

was significantly and negatively related to wellbeing. Importantly, the authors refer to a

cumulative effect, suggesting that at a certain point throughout the lifespan individuals may

have reached a critical threshold and wellbeing starts to reduce with each additional adverse

event. This threshold for experiencing (possibly non-traumatic) adversity may not have been

reached in early youth, explaining the absence of a significant relationship between wellbeing

and childhood adversity in their sample. Finally, we have studied a sample of community

dwelling older adults and hence the current findings might not be generalizable to clinical

populations.

Taking into account these limitations, the current results suggests that early life trauma

shows a positive association with anxiety and depression in late life (Comijs et al., 2013),

especially in individuals with insecure attachment features. Still, even though findings in the

current study are in line with earlier findings from studies on childhood adversity and diverse

17
psychopathology throughout the lifespan, prospective research is needed to replicate and

extend the current findings.

18
References

Anderson, K. A., Fields, N. L., & Dobb, L. A. (2011). Understanding the impact of early-life

trauma in nursing home residents. Journal of Gerontological Social Work, 54(8), 755-

767.

Andersson, L., & Stevens, N. (1993). Associations between early experiences with parents

and well-being in old age. Journal of Gerontology, 48(3), 109-116. doi:

10.1093/geronj/48.3.P109

Beekman, A. T., Penninx, W. J. H., Deeg, D. J. H., de Beurs, E., Geerlings, S. W., & Van

Tilburg, W. (2002). The impact of depression on the well-being, disability and use of

services in older adults: A longitudinal perspective. Acta Psychiatrica Scandinavica,

105(1), 20-27.

Bernstein, D. P., & Fink, L. (1998). Childhood trauma questionnaire: A retrospective self-

report. Manual. Orlando: Psychological Corporation.

Bernstein, D. P., Stein, J. A., Newcomb, M. D., Walker, E., Pogge, D., Ahluvalia, T., . . .

Zule, W. (2003). Development and validation of a brief screening version of the

Childhood Trauma Questionnaire. Child abuse & Neglect, 27(2), 169-190.

Bjelland, I., Dahl, A. A., Haug, T. T., & Neckelmann, D. (2002). The validity of the Hospital

Anxiety and Depression Scale. An updated literature review. Journal of

Psychosomatic Research, 52(2), 69-77.

Bodner, E., & Cohen-Fridel, S. (2010). Relations between attachment styles, ageism and

quality of life in late life. International Psychogeriatrics, 22(8), 1353-1361. doi:

10.1017/s1041610210001249

Bour, A., Rasquin, S., Boreas, A., Limburg, M., & Verhey, F. (2010). How predictive is the

MMSE for cognitive performance after stroke? Journal of Neurology, 257(4), 630-

637.

19
Bynum, L., Griffin, T., Riding, D. L., Wynkoop, K. S., Anda, R. F., Edwards, V. J., ... &

Croft, J. B. (2010). Adverse childhood experiences reported by adults-five states,

2009. Morbidity and Mortality Weekly Report, 59(49), 1609-1613.

Cassidy, K., & Rector, N. A. (2008). The silent geriatric giant: Anxiety disorders in late life.

Geriatrics and Aging, 11(3), 150-156.

Charles, S., & Carstensen, L. L. (2010). Social and emotional aging. Annual Reviews in

Psychology, 61, 383-409.

Cicirelli, V. G. (1989). Feelings of attachment to siblings and well-being in later life.

Psychology and Aging, 4(2), 211-216.

Clark, C., Caldwell, T., Power, C., & Stansfield, C. A. (2010). Does the influence of

childhood adversity on psychopathology persist across the lifecourse? A 45-year

prospective epidemiologic study. Annals of Epidemiology, 20, 385-394.

Comijs, H. C., van Exel, E., van der Mast, R. C., Paauw, A., Oude Voshaar, R., & Stek, M. L.

(2013). Childhood abuse in late-life depression. Journal of Affective Disorders,

147(1), 241-246. doi: 10.1016/j.jad.2012.11.010

Cookman, C. (2005). Attachment in older adulthood: Concept clarification. Journal of

Advanced Nursing, 50(5), 528-535. doi: 10.1111/j.1365-2648.2005.03435.x

Dohrenwend, B. P., & Dohrenwend, B. S. (1982). Perspectives on past and future of

psychiatric epidemiology. American Journal of Public Health, 72, 1271-1279.

Erozkan, A. (2016). The link between types of attachment and childhood trauma. Universal

Journal of Educational Research, 4(5), 1071-1079. doi: 10.13189/ujer.2016.040517

Falk, B., Hersen, M., & van Hasselt, V. (1994). Assessment of post-traumatic stress disorder

in older adults: A critical review. Clinical Psychology Review, 14, 383-415.

Ferraro, K. F., & Shippee, T. P. (2009). Aging and cumulative inequality: How does

inequality get under the skin? The Gerontologist, 49(3), 333-343.

20
Finkelhor, D., & Jones, L. (2006). Why have child maltreatment and victimization declined?

Journal of social issues, 62(4), 685-716.

Flint, A. J. (1994). Epidemiology and comorbidity of anxiety disorders in the elderly. The

American Journal of Psychiatry, 151(5), 640-649.

Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). "Mini-mental state". A practical

method for grading the cognitive state of patients for the clinician. Journal of

Psychiatric Research, 12(3), 189-198.

Fonagy, P., & Luyten, P. (2009). A developmental, mentalization-based approach to

understanding and treatment of borderline personality disorder. Development and

Psychopathology, 21(4), 1355-1381.

Forlani, C., Morri, M., Ferrari, B., Dalmonte, E., Menchetti, M., De Ronchi, D., & Atti, A. R.

(2014). Prevalence and gender differences in late-life depression: A population-based

study. The American Journal of Geriatric Psychiatry, 22(4), 370-380. doi:

10.1016/j.jagp.2012.08.015

Fowler, J. C., Allen, J. G., Oldham, J. M., & Frueh, C. B. (2013). Exposure to interpersonal

trauma, attachment insecurity, and depression severity. Journal of Affective Disorders,

149(1-3), 313-318.

Fraley, R. C., Waller, N. G., & Brennan, K. A. (2000). An item response theory analysis of

self-report measures of adult attachment Journal of Personality and Social

Psychology, 78(2), 350-365.

Gatsonis, C., & Sampson, A. R. (1989). Multiple correlation: Exact power and sample size

calculations. Psychopharmacology Bulletin, 106(3), 516-524.

Gershon, A., Sudheimer, K., Tirouvanziam, R., Williams, L. M., & O'Hara, R. (2013). The

long-term impact of early adversity on late-life psychiatric disorders. Current

Psychiatry Reports, 15, 352.

21
Giedd, J. N., Snell, J. W., Lange, N., Rajapakse, J. C., Casey, B. J., Kozuch, P. L., . . .

Rapoport, J. L. (1996). Quantitative magnetic resonance imaging of human brain

development: ages 4-18. Cerebral Cortex, 6(4), 551-560.

Gross, J. J., & Levenson, R. W. (1993). Emotional suppression: Physiology, self-report, and

expressive behavior. Journal of Personality and Social Psychology, 64(6), 970-986.

Herrmann, C. (1997). International experiences with the Hospital Anxiety and Depression

Scale: A review of validation data and clinical results. Journal of Psychosomatic

Research, 42(1), 17-41.

Hulley, S. B., Cummings, S. R., Browner, W. S., Grady, D., & Newman, T. B. (2013).

Designing clinical research: An epidemiological approach. Philadelphia, PA:

Lippincott Williams & Wilkins.

Jain, E., & Labouvie-Vief, G. (2010). Compensatory effects of emotion avoidance in adult

development. Biological Psychology, 84(3), 497-513.

Kafetsios, K., & Sideridis, G. D. (2006). Attachment, social support and well-being in young

and older adults. Journal of Health Psychology, 11(6), 863-875. doi:

10.1177/1359105306069084

Kirmizioglu, Y., Dogan, O., Kugu, N., & Akyüz, G. (2009). Prevalence of anxiety disorders

among elderly people. International Journal of Geriatric Psychiatry, 24(9), 1026-

1033.

Kok, R. M., Verhey, F., & Schmand, B. (2004). Meetinstrumenten bij cognitieve stoornissen.

Tijdschrift voor Psychiatrie, 46(10), 665-669.

Kooiman, C. G., Klaassens, E. R., van Heloma Lugt, J. Q., & Kamperman, A. M. (2013).

Psychometrics and validity of the Dutch Experiences in Close Relationships-Revised

(ECR-R) in an outpatient mental health sample. Journal of Personality Assessment,

95(2), 217-224. doi: 10.1080/00223891.2012.740540

22
Lupien, S. J., McEwen, B. S., Gunnar, M. R., & Heim, C. (2009). Effects of stress throughout

the lifespan on the brain, behaviour and cognition. Nature Reviews Neuroscience,

10(6), 434-445. doi: 10.1038/nrn2639

McEwen, B. S., & Gianaros, P. J. (2010). Central role of the brain in stress and adaptation:

Links to socioeconomic status, health, and disease. Annals of the New York Academy

of Sciences, 1186, 190-222. doi: 10.1111/j.1749-6632.2009.05331.x

Mikulincer, M., & Florian, V. (1998). The relationship between adult attachment styles and

emotional and cognitive reactions to stressful events. In J. Simpson & S. Rholes

(Eds.), Attachment theory and close relationships (pp. 143-165). New York: Guilford.

Mikulincer, M., & Shaver, P. R. (2007). Attachment in Adulthood: Structure, Dynamics, and

Change. New York: Guilford Press.

Mikulincer, M., & Shaver, P. R. (2012). An attachment perspective on psychopathology.

World Psychiatry, 11(1), 11-15.

Ni Mhaolain, A. M., Fan, C. W., Romero-Ortuno, R., Cogan, L., Cunningham, C., Kenny, R.

A., & Lawlor, B. (2012). Frailty, depression, and anxiety in later life. International

Psychogeriatrics, 24(8), 1265-1274. doi: 10.1017/s1041610211002110

Nolte, T., Guiney, J., Fonagy, P., Mayes, L., & Luyten, P. (2011). Interpersonal stress

regulation and the development of anxiety disorders: An attachment-based

developmental framework. Frontiers in Behavioral Neuroscience, 5(55). doi:

10.3389/fnbeh.2011.00055

Ogle, C. M., Rubin, D. C., & Siegler, I. C. (2014). Cumulative exposure to traumatic events in

older adults. Aging & Mental Health, 18(3), 316-325.

Park, D., & Vandenberg, B. (1994). The influence of separation orientation on life satisfaction

in the elderly. The International Journal of Aging and Human Development, 39(2),

177-187. doi: doi:10.2190/A9QW-P21Y-L974-T67J

23
Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect

effects in simple mediation models. Behavior Research Methods, Instruments and

Computers, 36(4), 717-731.

Pruessner, J. C., Li, L. M., Serles, W., Pruessner, M., Collins, D. L., Kabani, N., . . . Evans, A.

C. (2000). Volumetry of hippocampus and amygdala with high-resolution MRI and

three-dimensional analysis software: Minimizing the discrepancies between

laboratories. Cerebral Cortex, 10(4), 433-442.

Pruessner, J. C., Dedovic, K., Pruessner, M., Lord, C., Buss, C., Collins, L., . . . Lupien, S. J.

(2010). Stress regulation in the central nervous system: Evidence from structural and

functional neuroimaging studies in human populations - 2008 Curt Richter Award

Winner. Psychoneuroendocrinology, 35(1), 179-191. doi:

10.1016/j.psyneuen.2009.02.016

Saunders, B. E., & Adams, Z. W. (2014). Epidemiology of traumatic experiences in

childhood. Child and Adolescent Psychiatric Clinics of North America, 23(2), 167-

184. doi: 10.1016/j.chc.2013.12.003

Sbarra, D. A., & Hazan, C. (2008). Coregulation, dysregulation, self-regulation: An

integrative analysis and empirical agenda for understanding adult attachment,

separation, loss, and recovery. Personality and Social Psychology Review, 12(2), 141-

167.

Shaver, P. R., Collins, N. L., & Clark, C. (1996). Attachment styles and internal working

models of self and relationship partners. In G. J. O. Fletcher & J. Fitness (Eds.),

Knowledge structures in close relationships: A social psychological approach (pp. 25-

61). Mahwah, NJ: Lawrence Erlbaum.

24
Shrira, A., Shmotkin, D., & Litwin, H. (2012). Potentially traumatic events at different points

in the life span and mental health: Findings from SHARE-Israel. The American

Journal of Orthopsychiatry, 82(2), 251-259. doi: 10.1111/j.1939-0025.2012.01149.x

Sibley, C. G., Fischer, R., & Liu, J. H. (2005). Reliability and validity of the Revised

Experiences in Close Relationships (ECR-R) self-report measure of adult romantic

attachment. Personality and Social Psychology Bulletin, 31(11), 1524-1536.

Spinhoven, P., Ormel, J., Sloekers, P. P. A., Kempen, G. J. M., Speckens, A. E. M & Van

Hemert, A. M. (1997). A validation study of the Hospital Anxiety and Depression

Scale (HADS) in different groups of Dutch subjects. Psychological Medicine, 27, 363-

370.

Teicher, M. H., & Samson, J. A. (2013). Childhood maltreatment and psychopathology: A

case for ecophenotypic variants as clinically and neurobiologically distinct subtypes.

The American Journal of Psychiatry, 170(10), 1114-1133. doi:

10.1176/appi.ajp.2013.12070957

Thombs, B. D., Bernstein, D. P., Lobbestael, J., & Arntz, A. (2009). A validation study of the

Dutch Childhood Trauma Questionnaire-Short Form: factor structure, reliability, and

known-groups validity. Child Abuse & Neglect, 33(8), 518-523.

Tisserand, D. J., Pruessner, J. C., Sanz Arigita, E. J., van Boxtel, M. P., Evans, A. C., Jolles,

J., & Uylings, H. B. (2002). Regional frontal cortical volumes decrease differentially

in aging: An MRI study to compare volumetric approaches and voxel-based

morphometry. NeuroImage, 17(2), 657-669.

Van Assche, L., Luyten, P., Bruffaerts, R., Persoons, P., van de Ven, L., & Vandenbulcke, M.

(2013). Attachment in old age: Theoretical assumptions, empirical findings and

implications for clinical practice. Clinical Psychology Review, 33(1), 67-81. doi:

10.1016/j.cpr.2012.10.003

25
van Marwijk, H. W., Wallace, P., de Bock, G. H., Hermans, J., Kaptein, A. A., & Mulder, J.

D. (1995). Evaluation of the feasibility, reliability and diagnostic value of shortened

versions of the geriatric depression scale. British Journal of General Practice,

45(393), 195-199.

Varese, F., Smeets, F., Drukker, M., Lieverse, R., Lataster, T., Viechtbauer, W., . . . Bentall,

R. P. (2012). Childhood adversities increase the risk of psychosis: A meta-analysis of

patient-control, prospective- and cross-sectional cohort studies. Schizophrenia

Bulletin, 38(4), 661-671. doi: 10.1093/schbul/sbs050

Vink, D., Aartsen, M. J., & Schoevers, R. A. (2008). Risk factors for anxiety and depression

in the erlderly: A review. Journal of Affective Disorders, 106(1-2), 29-44.

Webster, J. D. (1997). Attachment style and well-being in elderly adults: A preliminary

investigation. Canadian Journal on Aging, 16(1), 101-111.

Webster, J. D. (1998). Attachment styles, reminiscence functions and happiness in young and

elderly adults. Journal of Aging Studies, 12(3), 315-330. doi: 10.1016/S0890-

4065(98)90006-8

Wensauer, M., & Grossmann, K. E. (1995). [Quality of attachment representation, social

integration and use of social network resources in advanced age]. Zeitschrift fur

Gerontologie und Geriatrie, 28(6), 444-456.

Yesavage, J. A., Brink, T. L., Rose, T. L., Lum, O., Huang, V., Adey, M., & Leirer, V. O.

(1982). Development and validation of a geriatric depression screening scale: A

preliminary report. Journal of Psychiatric Research, 17(1), 37-49.

Zigmond, A. S., & Snaith, R. P. (1983). The hospital anxiety and depression scale. Acta

Psychiatrica Scandinavica, 67(6), 361-370.

26
Table 1. Self-reported childhood trauma and adulthood negative life events in a sample of

community dwelling older adults (n=81)

Type of childhood trauma Number of individuals with scores above cut-off

(% of individuals)

CTQ emotional abuse 5 (6%)

CTQ physical abuse 4 (5%)

CTQ sexual abuse 12 (15%)

CTQ emotional neglect 16 (20%)

CTQ physical neglect 28 (35%)

CTQ any type of trauma 40 (49%)

Negative life events in adulthood Average number of events (% of events)

Total number 11 (100%)

Work 1 (9.09%)

Relationship 1 (9.09%)

Children 2 (18.18%)

Family 3.5 (31.82%)

Crime 1 (9.09%)

Finance 0 (0%)

Friends 1.25 (11.36%)

Health 1.25 (11.36%)

Self-reported anxiety and depression Mean SD, range

HADS-A 4.41 3.58, 0-18

GDS 5.56 4.36, 0-22

Note. CTQ = Childhood Trauma Questionnaire; HADS-A = Hospital Anxiety and Depression

Scale-Anxiety subscale; GDS = Geriatric Depression Scale.

Table 2. Correlations between childhood trauma, adulthood negative life events, attachment and late life anxiety or depression in community

dwelling older adults (n=81)

Anxiety Depression

CTQ total number .28* .19

CTQ presence .25* .31*

CTQ emotional neglect .35* .29*

CTQ physical neglect .16 .16

CTQ emotional abuse .38* .18

CTQ physical abuse .11 .12

CTQ sexual abuse .06 -.15

PERI total number .01 .17

PERI distress .12 .28*

Attachment anxiety Attachment avoidance

CTQ total number .06 -.15

CTQ presence -.04 -.15

 CTQ emotional neglect .28* .10

CTQ physical neglect .08 .08

CTQ emotional abuse .06 .09

CTQ physical abuse .26 -.43*

CTQ sexual abuse -.10 -.32

PERI total number -.06 -.07

PERI distress .04 -.02

Anxiety .29** .30**

Depression .26* .40***

Note. CTQ = Childhood Trauma Questionnaire; PERI = Psychiatric Epidemiology Research Interview; * = p <0.05, ** = p <0.01, *** = p

<0.001.
Table 3. Model coefficients for the mediation through attachment anxiety of the effect of

childhood emotional neglect on levels of anxiety in late life

Consequent

Attachment anxiety (M) State anxiety (Y)

Antecedent Coeff. SE P Coeff. SE p

Emotional neglect (X) a 0.06 0.03 .07 c’ 1.85 0.10 .06

Attachment anxiety (M) b 0.81 0.36 .03

Constant i1 1.96 0.34 < .01 i2 .42 1.28 .74

R2 = 0.20 R2 = 0.19

F(1, 79) = 3.39, p = .07 F(2, 78) = 5.41, p = .06

Note. X = independent variable, Y = dependent variable, M = mediator

Table 4. Model coefficients for the mediation through attachment anxiety of the effect of

childhood emotional neglect on levels of depression in late life

Consequent

Attachment anxiety (M) Depression (Y)

Antecedent Coeff. SE P Coeff. SE p

Emotional neglect (X) a 0.06 0.03 .07 c’ 1.83 0.12 .14

Attachment anxiety (M) b 0.93 0.44 .04

Constant i1 1.96 0.34 < .01 i2 1.30 1.58 .42

R2 = 0.20 R2 = 0.31

F(1, 79) = 3.39, p = .07 F(2, 78) = 4.13, p = .02

Note. X = independent variable, Y = dependent variable, M = mediator