Skip to main content
NCBI home page
The Journals of Gerontology Series B: Psychological Sciences and Social Sciences logoLink to The Journals of Gerontology Series B: Psychological Sciences and Social Sciences
. 2024 Aug 12;79(10):gbae137. doi: 10.1093/geronb/gbae137

Associations Between Attachment and Depressive Symptoms Among Older Adults Living With Early-Stage Cognitive Impairment and Their Adult Children

Joan K Monin 1,✉, Maya Shah 2, Wei Chang 3, Brooke C Feeney 4, Kira Birditt 5, Joseph E Gaugler 6, Ana Maria Vranceanu 7,8, Richard Marottoli 9
Editor: Rodlescia S Sneed
PMCID: PMC11407828  PMID: 39129484

Abstract

Objectives

This study examined the extent to which multiple relationship-specific attachment schemas (general, parent-child, and romantic) are interrelated and relate to depressive symptoms among older adults living with early-stage dementia and their adult children, a context likely to activate the attachment system.

Methods

The study consisted of 150 early-stage cognitively impaired parents and one child. Both self-reported general, parent-child, and romantic attachment anxiety as well as avoidance and depressive symptoms. Parents’ cognitive and functional impairment were assessed.

Results

Most attachment variables were interrelated for individuals, except that the child’s parent-child-specific avoidance was not associated with their general or romantic avoidance. The parent’s worse functional, but not cognitive, impairment was associated with the child’s greater parent-child-specific attachment anxiety. Using Actor Partner Interdependence Models, general attachment anxiety was associated with greater depressive symptoms (actor effects). When both dyad members were high in general avoidance, depressive symptoms were greatest (actor × partner effect). A parent had the greatest depressive symptoms when they were avoidantly attached to their child (role × actor effect) and their child was high in anxiety and low in avoidance toward them (role × partner effects). A child had the greatest depressive symptoms when they were low in avoidance toward the parent (role × actor effect) and the parent was low in anxiety and high in avoidance toward the child (role × partner effects). Romantic anxiety was associated with greater depressive symptoms (actor effects).

Discussion

Psychosocial interventions that incorporate attachment theory as a framework may benefit parent-child dyads coping with dementia.

Keywords: Caregiving, Family, Dementia

In the United States, over 6 million adults live with Alzheimer’s disease and related dementias, most of whom are cared for by family members or close others. Caring for family members living with dementia is necessary and sometimes fulfilling, but caregiving can also be stressful and negatively affect both partners’ mental and physical health (Schulz et al., 2020). Our motivation to care for one another as humans, especially as we grow old and face health challenges, is a great asset to public health because family caregivers reduce older adults’ healthcare utilization and allow older adults to age in place (Friedman et al., 2019). Yet, there is a great need to understand how to support families not only with practical, but also the emotional, demands that come with helping a relative.

Much of the research on dementia caregiving focuses on spouses. Less research has examined parent-adult child relationships in the early stages of dementia despite that over half of the people living with dementia have primary caregivers who are adult children (Caregiving, 2020). Research shows that both adult children and spouses experience negative psychological and physical health outcomes as dementia caregivers (Schulz et al., 1995). Although there are mixed findings regarding whether negative mental health outcomes are higher among adult children compared to spousal caregivers (Rigby et al., 2019). Some research suggests that spouses view their caregiving duties as a marital duty, rather than “a large task” or “change in their life” as an adult child may view caregiving (Conde-Sala et al., 2010). Adult child caregivers may experience different interpersonal challenges compared to spousal caregivers. We suggest that looking at these challenges through a developmental lens, drawing from attachment theory, is useful.

Transitioning From Child to Caregiver

Becoming a caregiver can be a demanding and overwhelming change. Before this change, a typical parent-adult child relationship involves the child receiving more emotional, instrumental, and/or financial support from the parent rather than vice versa, especially during the early years. In adulthood, the child may receive parental care similarly to how they have experienced it throughout the duration of their lives, but perhaps with less financial and physical expectations as the child matures (Barca et al., 2014). Scholars have labeled this stage filial maturity in which adult children are “confronted with the aging and increasing vulnerability of the parents and with the possibility that they cease to function as sources of support as they grow older and need the support and comfort of their children” (p. 2) (Cicirelli, 1988; Morais et al., 2024).

During the early stages of dementia, a child may notice the gradual change and deterioration of the parent’s cognitive and functional abilities. These issues can include their parent’s ability to remember to attend scheduled activities, manage finances, respond to social cues accordingly, and more (Barca et al., 2014). The parent-adult child relationship dynamic may adapt to the new situation, in which the child becomes a caregiver providing support for the parent. This role transformation may incite negative emotions in the new caregiver such as grief, worry, aggression, and guilt (Barca et al., 2014). It is important to note that this is not a role reversal that children and parents experience; children are witnessing and supporting a parent through a deteriorating condition, whereas the parent has in the past and continues to support their child through typical development (except in cases where the child also has a deteriorating condition).

Attachment Theory

Attachment theory may be a useful framework to draw from in thinking about how parents and children view their roles as caregivers and care recipients due to the parent’s dementia. Attachment theory, as described by John Bowlby, postulates that the emotional bond created during the early stages of life between a child and their adult caregiver (often their parent) affects the motivational-behavioral control system of the child throughout the rest of the child’s life (Bowlby, 1988). One’s attachment style influences how one perceives oneself and others, shaping one’s ability to regulate their emotions and navigate close relationships (Karreman & Vingerhoets, 2012). This becomes very relevant for later life caregiving relationships that stem from illness. In particular, attachment styles are activated in situations that introduce a potential or actual threat to one’s state of homeostasis, such as an anxiety-inducing experience, increased distress, or possible loss (Magai & Cohen, 1998). Dementia or cognitive impairment may be one of these situations.

Attachment styles can be measured in multiple ways; however, the most common approach is to measure two dimensions of attachment—anxiety and avoidance. Those who score high on the anxiety dimension want to feel close to others, but often think that their emotions are not reciprocated to the same degree that they feel toward their partner (Brennan et al., 1998). Those who score high on avoidant attachment have difficulty trusting others, being intimate, and depending on others (Brennan et al., 1998). People who score low on both anxiety and avoidance are considered securely attached, and this is thought to be most adaptive. Securely attached individuals find ease in close relationships and rarely feel worried about being abandoned or having another feel close to them. In this paper, we refer to being low in either dimension as higher attachment security.

Attachment Theory and Being a Dementia Caregiver

Attachment theory predicts that caregivers who are high in attachment anxiety will demonstrate a tendency to over-involve themselves to the point that they become intrusive with care; research with romantic couples in which a caregiving role is manipulated support this hypothesis (Collins & Feeney, 2010). Anxious caregivers have also been shown to experience high levels of emotional empathy and distress when witnessing another person suffering (Monin et al., 2013). Further, anxious individuals provide less instrumental support or are less responsive to what care receivers actually desire (Collins & Feeney, 2000). In contrast, caregivers high in avoidance are less prosocial and less likely to show empathic concern compared to less avoidant caregivers (Feeney et al., 2016); their caregiving quality is rated quite low especially in distressing situations (Monin et al., 2012).

More recently, a study of middle-aged adults who had at least one living community-dwelling parent aged 65 and older revealed that securely attached children, using the Portuguese version of the Adult Attachment Scale (AAS; Canavarro et al., 2006) had a greater ability to see their parents objectively, recognizing their parents’ strengths and weaknesses. This led to them seeing their parents as deserving of care. Another recent study of middle-aged children and their older adult parents without cognitive impairment showed that the dyadic combination of the attachment orientations (measured with AAS) was a better predictor of caregiver burden than either dyad member’s orientation. In particular, anxious-avoidant dyads reported the highest child-reported caregiving burden (Romano et al., 2020). These studies highlight the unique ways attachment may play out in parents and children coping with dementia.

Attachment Theory and Being an Older Adult Living With Dementia

The role of attachment becomes more relevant as dependency needs increase and functional and cognitive abilities diminish. Living with dementia can activate the attachment system and lead to behaviors like shadowing caregivers or depending intensely on one’s caregiver’s actions and decisions (Nelis et al., 2012). In addition, insecure individuals living with dementia tend to have more neuropsychiatric symptoms, which may complicate the support processes in the relationship further (Perren et al., 2007). Support-seeking can also be affected. In general, avoidant care receivers tend to use ineffective support-seeking (Collins & Feeney, 2000), which could cause communication problems in a close relationship, affecting both partners’ psychological health. Furthermore, recent research shows that romantic attachment anxiety and avoidance are each related to cognitive decline in older adults (Weidmann & Chopik, 2022), suggesting that attachment insecurity and cognitive functioning may co-vary in their effects on caregivers’ well-being. This prior work also suggests that certain types of relationship-specific attachment may be more impactful for functioning in older adulthood.

Examining Multiple Attachment Schemas

Individuals’ attachment orientations vary by context and relationship, and relationship-specific and general attachment orientations can influence well-being differently (Klohnen et al., 2005). We were interested in whether general, parent-child, and romantic attachment orientations were correlated within individuals in the context of parents and children coping with early-stage dementia. It is likely that the adult child’s general and romantic attachment orientations would be similar to the attachment orientation that was developed with their parent (Magai, 2008). However, it may be that the parent-child attachment relationship is uniquely affected by dementia symptoms (e.g., increased apathy) or the fear of losing the parent to dementia. There is some malleability in attachment orientations; research shows that romantic attachment anxiety and avoidance increase from serious life events, including finding out a loved one or family member is ill (Fraley et al., 2021). However, it is unclear if this applies to general and parent-child attachment in the same way among middle-aged adult children of people living with early-stage dementia. It could be that the parent-child attachment orientation is activated and influences children’s depressive symptoms more so than general and romantic attachment orientations. In terms of the parents’ general and relationship-specific attachment orientations, it is likely that their general, parent-child, and romantic orientations are similar to each other, as their attachment to their child is likely not uniquely affected by the onset of a degenerative condition.

Hypotheses

  1. General, parent-child, and romantic attachment anxiety will be positively significantly correlated within individuals (1a). The same will be true for attachment avoidance (1b).

  2. Greater cognitive and functional impairment in the parent will be related to greater attachment anxiety and avoidance among adult children.

  3. General, parent-child, and romantic attachment anxiety will be significantly associated with one’s own greater depressive symptoms (3a). The same will be true for attachment avoidance (3b; actor effects).

  4. General, parent-child, and romantic attachment anxiety will be significantly associated with the other dyad member’s greater depressive symptoms (4a). The same will be true for attachment avoidance (4b; partner effects).

Note: We did not hypothesize differences in actor effects by parent or child role.

Method

Participants

Data were obtained from the Families Coping Together with Alzheimer’s Disease (FACT-AD) study, which was designed to examine how the mutual support behaviors between parents living with memory loss and their caregiving children affect stress, health, and relational functioning of both dyad members. FACT-AD is a study of parents who are starting to notice changes in their cognition; not all have been diagnosed with Alzheimer’s disease or related dementias to understand the parent-child relationship in the pre-caregiving phase. Dyads were recruited through the Alzheimer’s Disease Research Center at Yale, the Adler Center, Geriatric Psychiatry, The University of Pittsburgh Center for Social and Urban Research through word of mouth and through the use of physical flyers and online social media posts. Parent-oriented flyers asked individuals if they were experiencing memory problems and if they had a child that was at least 18 years old. Adult child flyers asked individuals if they were 18 years old or older and if they had a parent who believed they had memory problems.

Dyads consisted of a parent 55 years or older with either one biological or adopted child, a sibling perceived by the participant to have a significant age gap, a niece or nephew, a grandchild, or a child-in-law at least 18 years or older who reported providing care to the parent. Children were determined to be caregivers if they provided care to their parents with at least one instrumental activity of daily living. Parents were eligible if they were diagnosed with memory loss and/or cognitive impairment of any kind, or if they had self-reported memory loss. Those in the Yale New Haven Health system had their diagnoses verified using electronic health records. The first 21 dyads participated in the study in-person, in which the cognitive assessment took place using the MMSE. After the pandemic began, parents’ cognition was assessed through the Telephone Interview for Cognitive Status (TICS) and converted into a MMSE score (Brandt et al., 1988). Participants scoring above a 12 on the MMSE without distress during the exam were considered eligible. Dyads in which other caregivers were present, such as a spouse or other children, were not excluded.

Procedure

Each participant completed an interview visit in which a trained research staff member helped to administer the self-report baseline interview within a 2-hr timeframe. Participants were also given the option to complete this survey on their own on paper or online.

Measures

General attachment style

The general attachment patterns of both parents and adult children were assessed using the Experiences in Close Relationship scale - Short Form (ECR-RS), which contains nine items (e.g., an avoidance item such as “I don’t feel comfortable opening up to people” or an anxiety item such as “I often worry that people do not really care for me”) on a scale ranging from 1 (strongly disagree) to 7 (strongly agree). Cronbach’s alpha for the children’s anxiety and avoidance scales were 0.92 and 0.86, respectively. For parents’ they were 0.86 and 0.81. Children’s mean anxiety and avoidance scores were 2.54 (SD = 1.81) and 3.11 (SD = 1.41), and parents’ mean scores were 2.28 (SD = 1.60) and 3.21 (SD = 1.50), respectively.

Parent/child attachment style

The same measure as above was used, but the items were reframed to target the parent/child in the study. Because we ask the adult child about attachment to both their mother and their father (or mother and father-like figures), to calculate the scores related to the parent who was in the study with the child, we first calculated their anxiety and avoidance scores pertaining to both their mother and their father. Then we computed new variables selecting the scores for the mother or the father depending on the gender of the parent in the study (e.g., the parent in the study). The Cronbach’s alpha for the scales where children reported on their anxiety and avoidance of their mothers were 0.83 and 0.85, and for fathers they were 0.86 and 0.88. After selecting the focal parent scores in the study, the means were 1.63 (SD = 1.34) for anxiety and 3.69 (SD = 1.56) for avoidance. The Cronbach’s alphas for the scales for parent’s attachment to the child in the study were 0.80 for anxiety and 0.81 for avoidance. For parent’s scores of attachment to their child, they were 1.70 (SD = 1.27) for anxiety and 2.51 (SD = 2.51) for avoidance.

Romantic partner attachment style

The same measure was used as above, but the items were reframed to target one’s romantic partner or imagined romantic partner if they were not currently in a relationship. Eighty-seven children and 67 parents had a current romantic partner (see Table 1 for marital/relationship status). Cronbach’s alphas for the children’s anxiety and avoidance scales were 0.88 and 0.89, respectively. For parents they were 0.86 and 0.88. Children’s mean romantic partner anxiety and avoidance scores were 2.10 (SD = 1.63) and 2.18 (SD = 1.35), and parents’ mean romantic partner anxiety and avoidance scores were 2.17 (SD = 1.78) and 2.59 (SD = 1.70), respectively. Note that the romantic partners were not in the study. No reports come from the romantic partners themselves.

Table 1.

Participant Characteristics

Characteristics Child Parent
Age in years, Mean (SD); Range 47.74 (12.08); 18–81 76.24 (9.78); 49–96
Gender
 Women 118 (78.67%) 106 (70.67%)
 Men 32 (21.33%) 44 (29.33%)
Education
 Did not complete high school 3 (2.00%) 16 (10.67%)
 Completed high school 13 (8.67%) 37 (24.67%)
 College credit 22 (14.67%) 23 (15.33%)
 Associates degree 13 (8.67%) 7 (4.67%)
 Bachelor’s degree 33 (22.00%) 17 (11.33%)
 Some graduate school 28 (18.67%) 15 (10.00%)
 Professional degree 36 (24.00%) 28 (18.67%)
 Missing 2 (1.33%) 7 (4.67%)
Race/ethnicity
 No primary group 3 (2.00%) 5 (3.33%)
 White 110 (73.33%) 110 (73.33%)
 Black/African American 17 (11.33%) 14 (9.33%)
 Asian 7 (4.67%) 7 (4.67%)
 Other 8 (5.33%) 2 (1.33%)
 American Indian or Alaska native 0 (0%) 1 (0.67%)
 Missing 5(3.33%) 11 (7.33%)
Latino/a
 Yes 12 (8.00%) 8 (5.33%)
 Missing 4 (2.67%) 11 (7.33%)
Marital status
 Married, living together 72 (48.00%) 63 (2.00%)
 Married/living apart 5 (3.33%) 3 (2.00%)
 Single 38 (25.33%) 15 (10.00%)
 Divorced 21 (14.00%) 20 (13.33%)
 Bereaved 3 (2.00%) 43 (28.67%)
 Cohabiting with romantic partner 10 (6.67%) 1 (0.67%)
 Missing 1 (0.66%) 5 (3.33%)
Sexual orientation
 Heterosexual/straight 139 (92.67%) 141 (94.00%)
 Gay/lesbian 1 (0.67%) 1 (0.67%)
 Bisexual 7 (4.67%) 0 (0%)
 Missing 3 (2.00%) 8 (5.33%)
Relationship type
 Daughter/mother 82 (55.67%)
 Daughter/father 29 (19.33%)
 Son/mother 17(11.33%)
 Son/father 14 (9.33%)
 Other parent-child like relationship 8 (5.33%)
MMSE score, Mean (SD) 23.39 (5.30)
Open in a new tab

Notes: MMSE = Mini-Mental State Examination; SD = standard deviation.

Depressive symptoms

The participants’ depressive symptoms were measured with a 10-item CESD in which responses to each item (e.g., “I felt depressed”) were given on a scale from 0 (rarely or none of the time) to 3 (most or all of the time). For each item, participants were asked to consider the past week as a timeframe for their feelings. The sum of the responses was used as a depression index. Cronbach’s alpha for the children’s depressive symptoms was 0.85 and for parents it was 0.75. Mean scores were 7.97 (SD = 6.15, range = 0–25) for children and 7.58 (SD = 5.52, range = 0–24) for parents.

Severity of cognitive impairment was measured using the Mini-Mental State Examination (MMSE) for the first 33 participants before the pandemic shutdown (Folstein et al., 1975). Scores on the MMSE range from 0 to 30, with lower scores indicating greater degrees of cognitive impairment. See Table 1 for MMSE data. For 114 of the dyads, we used the TICS (Brandt et al., 1988) because of the pandemic shutdown, and we converted the TICS score into the MMSE score for the entire sample using a crosswalk algorithm (Fong et al., 2009).

Severity of functional impairment

Parents were asked, “At the present time, do you need help from another person with: bathing, walking around home, dressing, getting in and out of a chair, walking a quarter of a mile, walking up a flight of stairs, lifting or carrying 10 pounds, shopping, housework, preparing a meal, taking medications, managing finances, driving.” Responses were: “no help” (0), “help” (1), or “unable to do” (1). Sum scores were calculated.

Potential covariates

Age, gender identity (woman = 0, man = 1), education (did not complete high school; completed high school; some college credit; associate’s degree; bachelor’s degree; some graduate school; professional degree), marital status (married, living together; married, living apart; cohabiting with romantic partner; single; divorced; bereaved), and race (White; Black; Asian; American Indian or Alaska native; Asian; Native Hawaiian or Pacific Islander; Other) and Hispanic heritage (yes = 1; no = 0) were assessed in both dyad members.

Analysis

See Table 1 for participant characteristics and Supplementary Table 1 for the diagnoses of the parents. First, we tested for potential covariates of attachment dimensions and depressive symptoms. See Supplementary Table 3 for interrelations between all study variables. These analyses address hypotheses 1 and 2. We conducted ANOVAs to examine associations between race and education and each of the hypothesized variables (see Supplementary Table 2). There were no significant associations with any of the covariates.

Next, we utilized mixed models with the Actor Partner Interdependence Model (Kenny et al., 2020) to test associations between each individual’s attachment dimensions and their outcome variable (depressive symptoms) for actor effects, as well as each individual’s own attachment dimensions and their partner’s outcome variable (partner effects). We also explored whether there were interactions for actor and partner effects with child role (child = 1, parent = 0) or parent role (parent = 1, child = 0) and between actor and partner effects for each attachment dimension. We ran separate models for general, parent/child-specific, and romantic other attachment variables as the sample size was small and the attachment variables were only moderately correlated.

Results

As shown in Supplementary Table 2, partially supporting our first hypothesis 1a, all types of attachment anxiety were significantly interrelated for both parents and children. For hypotheses 1b for parents, we found that general, parent-child, and romantic avoidant attachment were all significantly positively correlated with each other. However, we found that children’s avoidant attachment to their parent was not significantly correlated with either their general or romantic avoidant attachment; their general and romantic avoidant attachment were positively correlated.

Also, as shown in Supplementary Table 2, not supporting hypothesis 2, greater cognitive impairment in the parent was not significantly correlated with general, parent-child, and relationship-specific anxiety and avoidance in the child. Partially supporting hypothesis 2, only the child’s parent-child attachment anxiety was significantly associated with greater functional impairment in the parent.

As shown in Table 2 and Figure 1, in support of hypotheses 3 and 4, when both dyad members were high in general avoidance, depressive symptoms were greatest for both partners (actor × partner effect). Also, attachment anxiety was associated with greater depressive symptoms for both the parent and the child (actor effect). Next, as shown in Supplementary Figures 2–4, a parent had the greatest depressive symptoms when they were avoidantly attached to their child (role × actor effect) and their child was high in anxiety and low in avoidance toward the parent (role × partner effects), whereas a child had the greatest depressive symptoms when they were less avoidantly attached to their parent (role × actor effect) and their parent was low in anxiety and high in avoidance toward them (role × partner effects). Finally, having anxious attachment to a romantic partner was associated with greater depressive symptoms for both parents and children (actor effect).

Table 2.

Simplified Actor Partner Interdependence Model With Attachment Variables Predicting Depressive Symptoms

Variable Estimate SE df t p Value
General attachment
Intercept 4.82 2.03 126.76 2.38 .02
Actor general anxiety 1.53 0.36 178.52 4.23 <.001
Partner general anxiety 0.40 0.35 161.22 1.12 .26
Actor general avoidance −0.38 0.57 149.41 −0.66 .51
Partner general avoidance −0.99 0.58 151.99 −1.71 .09
Actor general anxiety × Partner general anxiety −0.14 0.11 139.16 −1.26 .21
Actor general avoidance × partner general avoidance 0.33 0.15 129.11 2.12 .04
Parent/child attachment
Intercept 10.35 3.24 130.23 3.20 .002
Role −2.60 1.91 128.75 −1.37 .18
Actor parent/child anxiety 1.08 0.99 174.25 1.09 .28
Partner parent/child anxiety −2.45 1.01 169.05 −2.43 .02
Actor parent/child avoidance −1.96 0.82 178.05 −2.39 .02
Partner parent/child avoidance 2.22 0.87 166.04 2.53 .01
Role × Actor parent/child anxiety −0.26 0.60 241.63 −0.43 .67
Role × Actor parent/child avoidance 1.48 0.51 238.65 2.92 .004
Role × partner parent/child anxiety 1.78 0.60 235.97 2.96 .003
Role × partner parent/child avoidance −1.47 0.52 229.19 −2.84 .005
Romantic attachment
Intercept 6.06 1.36 196.53 4.44 <.001
Role −0.67 0.67 119.23 −1.00 .32
Actor romantic anxiety 0.60 0.25 232.90 2.40 .02
Partner romantic anxiety 0.33 0.25 238.79 1.33 .19
Actor romantic avoidance 0.51 0.28 223.64 1.81 .07
Partner romantic avoidance −0.19 0.28 236.03 −0.69 .49
Open in a new tab

Note: SE = standard error.

Figure 1.

Alt Text: Graph showing that when both the actor and the partners’ general avoidant attachment was high, the actor’s depressive symptoms are highest.

Open in a new tab

General actor attachment avoidance associated with depressive symptoms ±1 SD of partner general avoidance above and below the mean. SD = standard deviation.

Discussion

Past research has shown that attachment styles are associated with psychological health in older adult caregiving relationships, especially when dyads are dealing with the challenges of dementia. The present study extends past research by examining associations between attachment styles and psychological health in parent-adult child relationships in the early stages of cognitive change and by examining relationship-specific attachment—general, parent/child, and romantic other. This advance in research is important given that attachment histories between parents and adult children are present throughout the lifespan and can affect caregiving in later life. Adult children have other relationships from which to draw security, so attachment to multiple figures is important to understand further.

Our findings advance past research in the following ways. First, we found that general, parent-child, and romantic attachment anxiety are all interrelated for both parents and children. For avoidant attachment, parent’s different forms of attachment avoidance were also interrelated, whereas the child’s avoidant attachment to their parent was not associated with their general or romantic attachment avoidance. It may be that children become more or less avoidantly attached only to their parent due to changes from dementia, such as increases in the parent’s apathy. It may also be a coping mechanism to deal with anticipatory grief of losing the parent or reaction to fear of loss. Children may be feeling they can no longer rely on the parent to the extent they did before. Alternatively, they may become less avoidant with their parent in particular due to a changing relationship dynamic, where more care is needed.

Second, we found that the child’s parent-child attachment anxiety was uniquely related to greater parent reported functional impairment. Although the parent’s cognitive impairment was significantly associated with the parent’s self-reported functional impairment, it was only the functional impairment that was associated with the child’s attachment anxiety in the relationship with the parent. This may reflect more vigilance demands and caregiving involvement. When a child is more highly involved with helping with their parent’s activities of daily living, they may feel more preoccupied with the parent’s safety. It may be that cognitive impairment that is not causing functional disability may not affect the child’s day-to-day mental health, at least at this time in the pre-caregiving trajectory.

Next, we found that general attachment anxiety was associated with greater depressive symptoms for both dyad members. We also found that when both dyad members were high in general avoidance, depressive symptoms were greatest for both partners. These findings fit with prior attachment findings across age and health contexts (Dagan et al., 2018). They further suggest that the combination of both people being highly avoidant is of high concern for the mental health of both individuals above and beyond the individual effects of avoidance. This combination may be a situation that warrants attachment-based dyadic interventions (Van’t Leven et al., 2013).

Combinations of parent and child orientations appeared to be impactful when it came to parent-child specific attachment. The importance of the combination of anxious and avoidant attachment orientations on depressive symptoms is in line with Romano and colleagues’ findings regarding caregiving burden (Romano et al., 2020). We found a parent had the greatest depressive symptoms when they were avoidantly attached to their child and their child was high in anxiety and low in avoidance toward the parent. This is a pattern that has also been seen in heterosexual romantic relationships in multiple age groups, with the combination of an anxious wife and an avoidant husband leading to the highest physiological stress during conflict (Beck et al., 2013). It also is reminiscent of the demand-withdrawal pattern in close relationships that can be particularly stressful (Heavey et al., 1995). Parents and children who fit this communication style may need special attention in supportive interventions.

A child had the greatest depressive symptoms when they were low in avoidance toward the parent and the parent was low in anxiety and high in avoidance toward the child. It seems that avoidance was more of the issue with the parent and child specific bond just as we found that avoidance was particularly related to this bond in hypothesis 1. In contrast, attachment anxiety tended to be more relevant to romantic relationships for both dyad members. Having an anxious attachment to a romantic partner may put both parents and children at higher risk for depression in life in general (Dagan et al., 2018), something that does not stop in the context of coping with dementia.

Overall, the findings of this study should be interpreted in the context of certain limitations. First, our sample could have been more diverse, as the majority of the participants were both White and female. This makes our findings less generalizable to the overall population, especially to those who are more prone to developing dementia (Mehta & Yeo, 2017). Second, this was a correlational study, making it impossible to show the causality or direction of effects.

Our findings should be considered preliminary evidence for psychological health effects of attachment styles in parent-adult child dyads in the early stages of dementia or at low levels of cognitive impairment. Future research should include larger sample sizes that can examine whether there are unique associations between the relationship-specific and general measures of attachment and other psychological outcomes. Longitudinal data over the lifespan would also allow for examination of changes prior to the onset of the role reversal and as the caregiving relationship continues between children and their parents coping with dementia. In addition, assessing the attachment orientations of multiple children and the parent would provide insight into more family-centered rather than dyad-centered interventions.

Taken together, findings from this study can inform future interventions for parent-adult child dyads coping with dementia. For example, healthcare professionals can better inform dyads of helpful coping strategies to buffer against depressive symptoms based on the attachment styles of each individual as well as combinations of attachment styles of dyad members. This would enable each partner to explore coping strategies they are comfortable with that also respond to their partner’s needs. If parent-child dyads are able to communicate their needs and provide support in the way it is needed and also received well, these dyads can decrease friction in their relationships and provide better care for each other during this early stage, and potentially in the progression of dementia. Existing attachment-based interventions such as Emotion Focused Therapy (Johnson, 2019), Attachment-Based Family Therapy (Diamond et al., 2002), and Circle of Security (Cassidy et al., 2017) may provide insights into therapeutic strategies for helping this population; however, further work is needed to tailor these approaches to the specific needs of adult children and their parents living with early-stage dementia. Another future direction based on our findings is to explore the feasibility and potential impact of integrating attachment assessments into caregiver burden screenings.

Supplementary Material

Supplementary data are available at The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences online.

gbae137_suppl_Supplementary_Material
gbae137_suppl_supplementary_material.docx (224.4KB, docx)

Contributor Information

Joan K Monin, Department of Social and Behavioral Sciences, Yale School of Public Health, New Haven, Connecticut, USA.

Maya Shah, Department of Social and Behavioral Sciences, Yale School of Public Health, New Haven, Connecticut, USA.

Wei Chang, Department of Social and Behavioral Sciences, Yale School of Public Health, New Haven, Connecticut, USA.

Brooke C Feeney, Department of Psychology, Carnegie Mellon University, Pittsburgh, Pennsylvania, USA.

Kira Birditt, Survey Research Center, Institute for Social Research, University of Michigan, Ann Arbor, Michigan, USA.

Joseph E Gaugler, Division of Health Policy and Management, School of Public Health, University of Minnesota, Minneapolis, Minnesota, USA.

Ana Maria Vranceanu, Center for Health Outcomes and Interdisciplinary Research, Massachusetts General Hospital, Boston, Massachusetts, USA; Department of Psychiatry, Massachusetts General Hospital, Boston, Massachusetts, USA.

Richard Marottoli, Division of Geriatrics, Department of Internal Medicine, Yale School of Medicine, New Haven, Connecticut, USA.

Funding

This study was supported by an R01 from the National Institute on Aging (R01AG058565 to J. K. Monin).

Conflict of Interest

None.

Data Availability

Data, analytic methods, and study materials will be made available to other researchers upon request and sent directly to the researchers. This study was not preregistered.

References

  1. Barca, M. L., Thorsen, K., Engedal, K., Haugen, P. K., & Johannessen, A. (2014). Nobody asked me how I felt: Experiences of adult children of persons with young-onset dementia. International Psychogeriatrics, 26(12), 1935–1944. 10.1017/S1041610213002639 [DOI] [PubMed] [Google Scholar]
  2. Beck, L. A., Pietromonaco, P. R., DeBuse, C. J., Powers, S. I., & Sayer, A. G. (2013). Spouses’ attachment pairings predict neuroendocrine, behavioral, and psychological responses to marital conflict. Journal of Personality and Social Psychology, 105(3), 388–424. 10.1037/a0033056 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bowlby, J. (1988). A secure base: Parent-child attachment and healthy human development. Basic Nooks. [Google Scholar]
  4. Brandt, J., Spencer, M., & Folstein, M. (1988). The Telephone Interview for Cognitive Status. Neuropsychiatry Neuropsychology and Behavioral Neurology, 1(2), 111–117. https://journals.lww.com/cogbehavneurol/abstract/1988/00120/the_telephone_interview_for_cognitive_status.4.aspx [Google Scholar]
  5. Brennan, K. A., Clark, C. L., & Shaver, P. R. (1998). Self-report measurement of adult attachment: An integrative overview. Attachment Theory and Close Relationships, 46, 76. https://psycnet.apa.org/doi/10.1037/t23266-000. [Google Scholar]
  6. Canavarro, M., Dias, P., & Lima, V. (2006). Adult attachment evaluation: A critical review on the application of adult attachment scale-R (AAS-R-r) in Portuguese population. Psicologia, 20, 154–186. 10.17575/rpsicol.v20i1.381 [DOI] [Google Scholar]
  7. Caregiving, A. A. T. N. A. F. (2020). Caregiving in the U.S. https://www.caregiving.org/wp-content/uploads/2021/01/full-report-caregiving-in-the-united-states-01-21.pdf [Google Scholar]
  8. Cassidy, J., Brett, B. E., Gross, J. T., Stern, J. A., Martin, D. R., Mohr, J. J., & Woodhouse, S. S. (2017). Circle of security–parenting: A randomized controlled trial in Head Start. Development and Psychopathology, 29(2), 651–673. 10.1017/S0954579417000244 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cicirelli, V. G. (1988). A measure of filial anxiety regarding anticipated care of elderly parents. Gerontologist, 28(4), 478–482. 10.1093/geront/28.4.478 [DOI] [PubMed] [Google Scholar]
  10. Collins, N., & Feeney, B. C. (2000). A safe haven: An attachment theory perspective on support seeking and caregiving in intimate relationships. Journal of Personality and Social Psychology, 78(6), 1053–1073. 10.1037//0022-3514.78.6.1053 [DOI] [PubMed] [Google Scholar]
  11. Collins, N. L., & Feeney, B. C. (2010). An attachment theoretical perspective on social support dynamics in couples: Normative processes and individual differences. In Sullivan K. T. & Davila J. (Eds.), Support processes in intimate relationships (pp. 89–120). Oxford Academic. 10.1093/acprof:oso/9780195380170.003.0004 [DOI] [Google Scholar]
  12. Conde-Sala, J. L., Garre-Olmo, J., Turró-Garriga, O., Vilalta-Franch, J., & López-Pousa, S. (2010). Differential features of burden between spouse and adult-child caregivers of patients with Alzheimer’s disease: An exploratory comparative design. International Journal of Nursing Studies, 47(10), 1262–1273. 10.1016/j.ijnurstu.2010.03.001 [DOI] [PubMed] [Google Scholar]
  13. Dagan, O., Facompré, C. R., & Bernard, K. (2018). Adult attachment representations and depressive symptoms: A meta-analysis. Journal of Affective Disorders, 236, 274–290. 10.1016/j.jad.2018.04.091 [DOI] [PubMed] [Google Scholar]
  14. Diamond, G. S., Reis, B. F., Diamond, G. M., Siqueland, L., & Isaacs, L. (2002). Attachment-based family therapy for depressed adolescents: A treatment development study. Journal of the American Academy of Child & Adolescent Psychiatry, 41(10), 1190–1196. 10.1097/00004583-200210000-00008 [DOI] [PubMed] [Google Scholar]
  15. Feeney, J., Cassidy, J., & Shaver, P. R. (2016). Handbook of attachment: Theory, research, and clinical applications. The Guilford Press. [Google Scholar]
  16. 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. 10.1016/0022-3956(75)90026-6 [DOI] [PubMed] [Google Scholar]
  17. Fong, T. G., Fearing, M. A., Jones, R. N., Shi, P., Marcantonio, E. R., Rudolph, J. L., Yang, F. M., Kiely, D. K., & Inouye, S. K. (2009). Telephone Interview for Cognitive Status: Creating a crosswalk with the Mini-Mental State Examination. Alzheimer's & Dementia, 5(6), 492–497. 10.1016/j.jalz.2009.02.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Fraley, R. C., Gillath, O., & Deboeck, P. R. (2021). Do life events lead to enduring changes in adult attachment styles? A naturalistic longitudinal investigation. Journal of Personality and Social Psychology, 120(6), 1567–1606. 10.1037/pspi0000326 [DOI] [PubMed] [Google Scholar]
  19. Friedman, E. M., Rodakowski, J., Schulz, R., Beach, S. R., Martsolf, G. R., & James, A. E.III. (2019). Do family caregivers offset healthcare costs for older adults? A mapping review on the costs of care for older adults with versus without caregivers. Gerontologist, 59(5), e535–e551. 10.1093/geront/gny182. [DOI] [PubMed] [Google Scholar]
  20. Heavey, C. L., Christensen, A., & Malamuth, N. M. (1995). The longitudinal impact of demand and withdrawal during marital conflict. Journal of Consulting and Clinical Psychology, 63(5), 797–801. 10.1037//0022-006x.63.5.797 [DOI] [PubMed] [Google Scholar]
  21. Johnson, S. M. (2019). The practice of emotionally focused couple therapy. Routledge. 10.4324/9781351168366 [DOI] [Google Scholar]
  22. Karreman, A., & Vingerhoets, A. J. (2012). Attachment and well-being: The mediating role of emotion regulation and resilience. Personality and Individual Differences, 53(7), 821–826. 10.1016/j.paid.2012.06.014 [DOI] [Google Scholar]
  23. Kenny, D. A., Kashy, D. A., & Cook, W. L. (2020). Dyadic data analysis. Guilford Publications. [Google Scholar]
  24. Klohnen, E. C., Weller, J. A., Luo, S., & Choe, M. (2005). Organization and predictive power of general and relationship-specific attachment models: One for all, and all for one? Personality and Social Psychology Bulletin, 31(12), 1665–1682. 10.1177/0146167205278307 [DOI] [PubMed] [Google Scholar]
  25. Magai, C. (2008). Attachment in middle and later life. In Cassidy J. & Shaver P. R. (Eds.), Handbook of attachment: Theory, research, and clinical applications (2nd ed., pp. 532–551). The Guilford Press. [Google Scholar]
  26. Magai, C., & Cohen, C. I. (1998). Attachment style and emotion regulation in dementia patients and their relation to caregiver burden. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 53(3), P147– P154. 10.1093/geronb/53b.3.p147 [DOI] [PubMed] [Google Scholar]
  27. Mehta, K. M., & Yeo, G. W. (2017). Systematic review of dementia prevalence and incidence in United States race/ethnic populations. Alzheimer's & Dementia, 13(1), 72–83. 10.1016/j.jalz.2016.06.2360 [DOI] [PubMed] [Google Scholar]
  28. Monin, J. K., Feeney, B. C., & Schulz, R. (2012). Attachment orientation and reactions to anxiety expression in close relationships. Personal Relationships, 19(3), 535–550. 10.1111/j.1475-6811.2011.01376.x [DOI] [Google Scholar]
  29. Monin, J. K., Schulz, R., & Kershaw, T. S. (2013). Caregiving spouses’ attachment orientations and the physical and psychological health of individuals with Alzheimer’s disease. Aging and Mental Health, 17(4), 508–516. 10.1080/13607863.2012.747080 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Morais, D., Faria, C., & Fernandes, L. (2024). Filial maturity and caregiving to aging parents. Geriatrics (Basel, Switzerland), 9(1), 17. 10.3390/geriatrics9010017. https://www.mdpi.com/2308-3417/9/1/17 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Nelis, S. M., Clare, L., & Whitaker, C. J. (2012). Attachment representations in people with dementia and their carers: Implications for well-being within the dyad. Aging & Mental Health, 16(7), 845–854. 10.1080/13607863.2012.667779 [DOI] [PubMed] [Google Scholar]
  32. Perren, S., Schmid, R., Herrmann, S., & Wettstein, A. (2007). The impact of attachment on dementia-related problem behavior and spousal caregivers’ well-being. Attachment & Human Development, 9(2), 163–178. 10.1080/14616730701349630 [DOI] [PubMed] [Google Scholar]
  33. Rigby, T., Ashwill, R. T., Johnson, D. K., & Galvin, J. E. (2019). Differences in the experience of caregiving between spouse and adult child caregivers in dementia with Lewy bodies. Innovation in Aging, 3(3), igz027. 10.1093/geroni/igz027 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Romano, D., Karantzas, G. C., Marshall, E. M., Simpson, J. A., Feeney, J. A., McCabe, M. P., & Mullins, E. R. (2020). Carer burden and dyadic attachment orientations in adult children-older parent dyads. Archives of Gerontology and Geriatrics, 90, 104170. 10.1016/j.archger.2020.104170 [DOI] [PubMed] [Google Scholar]
  35. Schulz, R., Beach, S. R., Czaja, S. J., Martire, L. M., & Monin, J. K. (2020). Family caregiving for older adults. Annual Review of Psychology, 71, 635–659. 10.1146/annurev-psych-010419-050754 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Schulz, R., O’Brien, A. T., Bookwala, J., & Fleissner, K. (1995). Psychiatric and physical morbidity effects of dementia caregiving: Prevalence, correlates, and causes. Gerontologist, 35(6), 771–791. 10.1093/geront/35.6.771 [DOI] [PubMed] [Google Scholar]
  37. Van’t Leven, N., Prick, A. -E. J. C., Groenewoud, J. G., Roelofs, P. D. D. M., de Lange, J., & Pot, A. M. (2013). Dyadic interventions for community-dwelling people with dementia and their family caregivers: A systematic review. International Psychogeriatrics, 25(10), 1581–1603. 10.1017/S1041610213000860 [DOI] [PubMed] [Google Scholar]
  38. Weidmann, R., & Chopik, W. J. (2022). Romantic attachment, stress, and cognitive functioning in a large sample of middle-aged and older couples. Journal of Research in Personality, 98:104233. 10.31234/osf.io/rymhg [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

gbae137_suppl_Supplementary_Material
gbae137_suppl_supplementary_material.docx (224.4KB, docx)

Data Availability Statement

Data, analytic methods, and study materials will be made available to other researchers upon request and sent directly to the researchers. This study was not preregistered.

Articles from The Journals of Gerontology Series B: Psychological Sciences and Social Sciences are provided here courtesy of Oxford University Press

RESOURCES