Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Aug 25.
Published in final edited form as: Res Aging. 2021 Aug 10;44(5-6):405–413. doi: 10.1177/01640275211026919

Interpersonal Correlates of Dementia Caregivers’ Emotional Support Networks: Considering Family History

Jennifer L Cleary 1,*, Jasmine A Manalel 2,*, Sato Ashida 3, Christopher Steven Marcum 2, Jeffrey Rewley 4, Laura Koehly 2
PMCID: PMC12376061  NIHMSID: NIHMS2101661  PMID: 34372731

Abstract

Caregiving networks of individuals with Alzheimer’s disease and related dementias (ADRD) are comprised of family and friends directly involved in caregiving activities and those supporting these activities. The purpose of this study was to investigate whether caregiving-related behaviors and interactions (i.e., uplift, malfeasance, and nonfeasance), kinship tie (i.e., friend, family), and family history of ADRD were associated with caregivers’ emotional support networks. Seventy-one caregivers across 30 families provided information about 776 network members. Perceived emotional support and interactions representing uplift, malfeasance, and nonfeasance were assessed. Results indicated that uplift and friendship were associated with increased, whereas nonfeasance was associated with decreased, likelihood of perceived emotional support. Caregivers with a family history of ADRD were particularly more likely to report emotional support from friends and uplifting network members. Findings suggest the need for differential strategies based on families’ prior caregiving experience to facilitate positive and minimize negative interactions within caregiving networks.

Keywords: Alzheimer’s disease, caregiving, social networks, social support

Family caregivers of individuals with chronic health conditions are at risk for many health concerns, and report heightened levels of physical and psychological distress as compared to their non-caregiver peers (Bom et al., 2018; Pinquart & Sörensen, 2003). Specifically, dementia caregiving has been associated with increased rates of morbidity and mortality (Schulz & Beach, 1999). Rates of familial caregiving are expected to rise as the number of individuals living with Alzheimer’s disease and related dementia (ADRD) increases, exposing a large population of individuals to a serious stressor with documented risks to physical and psychological health (AARP and National Alliance for Caregiving, 2020; Cheng, 2017).

Despite the focus on individual care providers and people with ADRD receiving care, caregiving is an inherently social process (Koehly et al., 2015; Marcum et al., 2020). Multiple people, including family, friends, and formal care providers, may be involved in direct care activities and care decisions. Moreover, these caregivers are embedded in their own support networks. Thus, one can consider caregiving to be a network, where the caregiving network is composed of those directly involved in caregiving activities, as well as those providing indirect care by supporting the caregivers themselves (Carpentier & Ducharme, 2003). In this study, we identified relational and contextual factors that shape the structure and composition of caregivers’ emotional support networks.

Theoretical Framework

The present study considers whether caregiver contextual factors and interpersonal stressors arising from caregiving shape coping resources, guided by two theoretical frameworks: the stress process model, and the support provision model. The stress process model has guided a large body of research aimed at identifying factors associated with adaptation to caregiving stress (Lazarus & Folkman, 1984; Pearlin et al., 1990). The model considers multiple factors in the adaptation process, including background characteristics (i.e., individual and social factors), the appraisal of stressors, and the availability of coping resources. The stress of caregiving can arise from the physical and emotional demands associated with providing care. However, additional stressors indirectly related to or stemming from caregiving stress and burden may also emerge. For example, stressful family relationships may result from negotiating individual members’ caregiving roles or perceptions that family members are not engaging in supportive caregiving activities (Ashida et al., 2018). Such interpersonal stressors may have important implications for the adaptation process, both directly and indirectly through the availability of coping resources.

One coping resource is the emotional support that caregivers receive, which, according to the stress process model, can buffer the stress of caregiving and improve caregiver health and well-being. A number of studies have identified emotional support as especially beneficial to caregivers of those with ADRD (Gilhooly et al., 2016; Wilks & Croom, 2008; Wittenberg-Lyles et al., 2014); however, the factors associated with receipt of support from network members remain unclear. Caregiver background factors and caregiving-related behaviors among family members can impact the adaptation process extending from the stress process model. Further, such factors may shape the composition and structure of emotional support resources available to the caregivers- a premise consistent with the support provision process model, which suggests that interpersonal and contextual factors can determine if and when someone provides support (Dunkel-Schetter & Skokan, 1990). As applied to caregiving, negotiating caregiving activities and kinship ties - that is, whether the provider of support is a family member or a friend- may shape the structure of emotional support networks. Further, the contextual factor of family history of dementia and associated experience with ADRD caregiving may shape how care is provided to the person with ADRD, which in turn can impact the interpersonal dynamics and support exchanges within the family (Rolland & Williams, 2005).

Interpersonal Factors: Caregiving-related behaviors and interactions and kinship tie

Stressful family relationships can arise as families negotiate caregiving responsibilities. These interpersonal stressors often occur when caregivers’ expectations of network members’ involvement in care are not met. Recent research has characterized the nature of caregivers’ met and unmet expectations in their network members’ caregiving-related activities as examples of “uplift”, “nonfeasance”, and “malfeasance” (Ashida et al., 2018; Brody et al., 1989). These constructs are characterized by observed or perceived caregiving-related behaviors and interactions that occur, or do not occur, towards caregivers and care recipients within a family. Originally, these concepts arose out of qualitative interviews, representing interactions between siblings providing care to aging parents (Brody et al., 1989), and were more recently used in relation to ADRD caregiving networks (Ashida et al., 2018). Uplift refers to behaviors and interactions perceived positively, including helping with caregiving and showing appreciation for caregivers’ efforts. Thus, those network members who are perceived to be engaged in uplifting activities may also be a primary source of emotional support for both caregivers and affected relatives alike. Nonfeasance refers to disengagement from caregiving activities, including not providing care to or spending enough time with the person with ADRD, as well as helping caregivers. By contrast, malfeasance refers to active negative caregiving activities. Network members perceived as malfeasant are viewed as critical toward and angry with those providing care to the person with ADRD or the caregiving situation. Empirically, perceived nonfeasance and malfeasance have been found to be associated with greater caregiving strain (Brody et al., 1989) and perceived nonfeasance was associated with perceptions of under-contribution to caregiving activities by family members (Ashida et al, 2018).

Composition of caregiving networks are often heterogeneous with a mixture of the person with ADRD’s and caregiver’s family members, friends, and professional caregivers. Although the specific caregiving-related roles may vary, both family and friends can be important emotional support sources for caregivers. Indeed, the types and levels of support exchanged between caregivers and their network members are determined by kinship tie (i.e., family or friends) and the norms and expectations accompanying those roles (Antonucci et al., 2014; Antonucci & Sherman, 2019). There is a greater expectation that family members be more involved in caregiving and caregiving decisions, and thus, their caregiving-related behaviors and interactions are more likely to be perceived as nonfeasance and malfeasance than non-family members. A recent study demonstrated that family members, particularly step-family, were more likely to be included in the negative care-related interaction network (i.e., absent, interfering, critical), whereas friends and other non-kin were more likely to be included in the positive care-related support network (i.e., emotional and instrumental support) (Antonucci & Sherman, 2019; Sherman, 2012). This latter finding is consistent with previous qualitative research, in which caregivers of those with ADRD commonly reported friendships to be important sources of emotional support (Lilly et al., 2003).

Contextual Factor: Family history of ADRD

In the context of ADRD, family history of dementia reflects an important contextual factor that shapes caregivers’ support networks. Individuals from families affected by inherited conditions and those with strong genetic bases, including AD, must often balance disease management in their affected relatives as well as anticipation of future caregiving for themselves or others within the family (Koehly, 2017; Rolland, 2006). These intertwined trajectories shape family support systems over time. Families with multiple relatives diagnosed with AD may have developed a schema for disease progression and how caregiving responsibilities will unfold (Alzheimer Association, 2020; Rolland & Williams, 2005). These schemas are based on an individual’s previous experience in providing direct care for a family member affected by the condition, and the family dynamics resulting from that prior experience. Alternatively, for those who were not directly involved in caregiving activities, perhaps due to generational effects, observing how their family negotiated care for a relative can also potentially shape current caregiving roles and expectations (Koehly, 2017). As such, this history of care can potentially exacerbate the salience of interpersonal stressors that may arise through the current caregiving experience. Moreover, if such prior experiences involved supportive caregiving-related interactions and behaviors, resulting in strengthened relationships, one might expect the association between uplifting caregiving-related interactions and emotional support to be stronger for those with a prior family history of ADRD.

Present Study

Guided by the stress process and support provision models, in the current report, we aim to evaluate whether prior family history of ADRD, caregiving-related interpersonal stressors, and kinship tie (i.e., family and friends) shape caregivers’ emotional support networks. We hypothesize that caregivers are less likely to perceive emotional support from network members who demonstrate nonfeasance and malfeasance caregiving behaviors and more likely to perceive emotional support from those who demonstrate uplifting behaviors. Further, we hypothesize that these associations will be stronger in families with a prior family history of ADRD. Given that these prior care experiences may have altered family relationships due to perceived nonfeasant or malfeasant caregiving activities, we hypothesize that non-family social ties, or friends, will be more prominent as sources of emotional support for caregivers with a prior family history of ADRD.

Method

Participants

Participants were recruited from four care facilities, assisted living and day services, in a metropolitan area of the United States via posted flyers and researcher presentations. Eligible participants were at least 18 years of age and caring for and/or a relative of an individual with ADRD. Thirty index participants (i.e., first person to participate from each family) were successfully recruited into study. Following completion of a survey and interview, index participants referred other adult family members of the person with ADRD into study. Referred individuals who were eligible to participate in the study and for whom we had contact information were recruited into study. Enrollment continued until all eligible family members had enrolled in the study, declined, or were not reachable, resulting in an additional 42 participants. For additional details, including sample selection process, see Marcum et al., 2020, Ashida et al., 2018, and Koehly et al., 2015. This study was approved by the Institutional Review Boards at the National Human Genome Research Institute and University of Memphis.

The sample included 71 participants from 30 families who completed a social network assessment, during which they enumerated a list of individuals whom they considered close and important, and family members and others important to the person with ADRD, including professional caregivers. For each enumerated network member, participants provided demographic information, kinship tie (i.e., family, friend, formal caregiver), information regarding support resources received from the network member, and caregiving-related activities and interactions involving the network member. Participants enumerated and reported on 1,248 network members. For the current report, we excluded network members who were formal caregivers and network members who were missing information about key demographic characteristics (e.g., gender, kinship tie) or caregiving-related support and interactions, resulting in a final analytic sample of 776 interpersonal ties.

Measures

Perceived Emotional Support.

While viewing the list of the enumerated network members, participants were asked “who supports you emotionally?” and indicated the enumerated network members who fulfilled this support function. A binary variable was constructed to reflect a presence (1) or absence of perceived emotional support (0) from a network member.

Caregiving-related behaviors and interactions.

For each enumerated network member, participants indicated who engaged in each of 17 caregiving-related behaviors or interactions. The set of network members’ behaviors or interactions toward the participant or person with ADRD were categorized as uplift, nonfeasance, and malfeasance based on a previous report (Ashida et al., 2017). Uplifting behaviors (7 items) include, for example: “who shows appreciation for your effort in caregiving?” and “who understands what you are going through with caregiving?”. Nonfeasance (4 items) was assessed with such questions as “who doesn’t show appreciation for your effort in caregiving?” and “who doesn’t spend enough time helping your affected relative?”. And, malfeasance (6 items) included interactions like “who doesn’t agree with you about what your relative can do for him/herself?” and “who is overly critical of you about caregiving?”. Responses were summed for each network member within each of the three types of behaviors or interactions. Given the skewed distribution of scores (Supplementary Figure 1), particularly for malfeasance and nonfeasance, we dichotomized these measures to indicate the presence (1 = selected for one or more relevant item) or absence (0 = not selected) of at least one relevant behavior or interaction.

Kinship tie.

Participants reported on how each network member was related to them, as well as to the person with ADRD. Network members who were reported to be social kin (friends, family friends, coworkers, spiritual leaders) of the participant were classified as friends, while biological and non-biological family members (first- and second-degree relatives, including step- and adopted relatives, and spouses) were considered family of the participant. Relationship type was coded dichotomously to indicate friendship (1) or familial tie (0).

Family history of ADRD.

Participants reported on the ADRD diagnosis of their biological relatives. A prior family history of ADRD (1) was indicated for those with at least one biological relative, other than the current person with ADRD, with an Alzheimer’s or related dementia diagnosis, compared to those with no prior family history (0).

Covariates.

Participants self-reported gender and the gender of each of their enumerated network members. Gender homophily was defined at the dyad level and indicated when a participant and enumerated network member were both of the same gender, which has been shown to be associated with perceived social support (Lee et al., 2018). Care needs was measured by assessing respondents’ perceived levels of care needed by person with ADRD for activities of daily living (Linn, 1988). Participants indicated the amount of help the person with ADRD needed for each of 18 behaviors (e.g., mobility, toileting, communication) using the following scale: none (i.e., completely independent or normal behavior; 0); a little (1); a lot (2); total (i.e., the person cannot, will not, or may not perform a behavior or has the most severe form of disability or problem; 3). The family-level score of perceived care needs was computed by summing the item scores for all available reports within a family, including participants and formal care providers, to capitalize on our multi-informant design, improve accuracy, and reduce bias in care needs (Loewenstein et al., 2001).

Analysis Strategy

Descriptive analyses assessed demographic characteristics of participants and their network members, as well as the person with ADRD. In addition, using tests of equal proportions, we describe associations between the contextual factor of prior family health history of ADRD and network characteristics, including emotional support receipt, caregiving-related behaviors and interactions, and kinship composition. As described above, these variables were recoded into binary variables indicating the presence (1) or absence (0) of the construct of interest.

To address our aims, we estimated a series of mixed-effects logistic regression models. The three-level models included random effects for participant and family, which accounted for the clustering of ties (level 1) within each participant’s network (level 2) and the clustering of participants within families (level 3). Including these random effects allowed us to model within-participant and within-family variance (Snijders & Bosker, 1999). In Model 1, binary indicators of caregiving-related behaviors and activities (uplift, nonfeasance, and malfeasance) were entered along with covariates (i.e., participant gender, participant-network member gender homophily, and person with ADRD’s care needs) to assess how caregiving-related behaviors and interactions are associated with received emotional support. In Model 2, kinship tie was added to Model 1 to assess the association between relationship type and received emotional support, controlling for caregiving-related behaviors and interactions as well as covariates. Finally, Model 3 tested whether these associations were moderated by prior family history of ADRD. First the main effect of family history of ADRD (Model 3a) and then interaction terms (Model 3b) were added between family history of ADRD and all three caregiving-related activities and kinship tie. All statistical analyses were performed using R, version 3.3.0 (Carey, 2019; R Core Team, 2018).

Results

Descriptives

The majority of participants were female (71.8%), married (71.8%), and middle-aged, with age ranging from 18 to 96 years (M= 64.8 years)(Table 1). Fifty-eight percent of the sample had earned a Bachelor’s degree or higher, and 79.3% of the sample reported a household income of greater than $50,000 per year. In all, 66.7% of the people with ADRD were female, ranged in age from 63 to 97 years old (M = 81.5 years), and 40% attended a day care program while 60% were in assisted living facilities. Care needs of the person with ADRD ranged from 8.75 to 38 (M = 19.49).

Table 1.

Participant demographics and network member characteristics by AD FH status

No AD FH AD FH Overall
Participant demographics n = 38 n = 33 n = 71
Gender (female) 26 (68.4%) 25 (75.8%) 51 (71.8%)
Race
  Black/African American 13 (34.2%) 7 (21.2%) 20 (28.2%)
  White 25 (65.8%) 26 (78.8%) 51 (71.8%)
Education
  High school/GED 3 (7.9%) 3 (6.4%) 6 (8.4%)
  Associate’s 16 (42.1%) 8 (24.2%) 24 (33.8%)
  Bachelor’s 10 (26.3%) 14 (42.4%) 24 (33.8%)
  Post-graduate 9 (23.7%) 8 (24.2%) 17 (23.9%)
Married 25 (65.8%) 26 (78.8%) 51 (71.8%)
Relationship to affected relative
  Child 22 (57.9%) 20 (60.6%) 42 (59.2%)
  Spouse 8 (21.1%) 1 (3.0%) 9 (12.7%)
  Other relative 8 (21.1%) 11 (33.3%) 19 (26.8 %)
  Friend 0 (0%) 1 (3.0%) 1 (1.4%)
Co-reside with affected relative 11 (28.9%) 1 (3.0%) 12 (16.9%)
Age (years) 65.0 (15.6) 64.7 (13.4) 64.8 (14.5)
Network size 23.8 (7.9) 26.8 (7.1) 25.2 (7.6)
NM Characteristics n = 373 n = 403 n = 776
Emotional Support 171 (45.8%) 159 (39.5%) 330 (42.5%)
Uplift 279 (74.8%) 320 (79.4%) 599 (77.2%)
Malfeasance 58 (15.5%) 63 (15.6%) 121 (15.6%)
Nonfeasance 98 (26.3%) 76 (18.9%) 174 (22.4%)
Kinship Tie (to participant)
 Familial 270 (72.4%) 310 (76.9%) 580 (74.7%)
 Friendship 103 (27.6%) 93 (23.1%) 196 (25.3%)
Gender
 Female 210 (56.3%) 232 (57.6%) 442 (57.0%)
 Male 163 (43.7%) 171 (42.4%) 334 (43.0%)
Age (years) 55.9 (19.2) 52.9 (20.0) 54.4 (19.7)
Open in a new tab

Notes. AD = Alzheimer’s Disease. NM = Network member. M = mean. SD = standard deviation.

Participants enumerated an average of 25 network members, the majority of whom were family (74.7%). Across the 776 dyads, participants reported that 77.2% had at least one caregiving-related behavior or interaction characterized as uplift, 22.4% had at least one characterized as nonfeasance, and 15.6% had at least one characterized as malfeasance. Tests of equal proportions were used to compare prevalence of nonfeasant, malfeasant, and uplifting interactions in the networks between those informants with a prior family history of ADRD and those without. Compared to participants with a family history of ADRD, those without a family history reported a greater proportion of network members engaging in nonfeasance (χ2 = 6.71, p < 0.01) and a lower proportion of network members exhibiting uplift (χ2 = 6.29, p < 0.05). There was no significant difference in the proportion of friends enumerated in the networks of participants with a prior family history of ADRD as compared to those without.

Correlates of Perceived Emotional Support

We tested whether caregiving-related interactions and behaviors and kinship tie were associated with perceived emotional support from network members in a series of nested models (Table 2). Results from Model 1 show that network members who engaged in uplifting caregiving-related behaviors or interactions were more likely to be identified as sources of emotional support (OR = 2.19, p < 0.01). Conversely, network members exhibiting at least one nonfeasant behavior or interaction were less likely to be sources of emotional support (OR = 0.35, p < 0.001). Malfeasance was not significantly associated with perceived emotional support. In model 2, kinship tie was associated with perceived emotional support, with friends perceived to be significantly more likely than family members to provide emotional support (OR = 2.28, p < 01).

Table 2.

Mixed effects logistic regression models estimating emotional support among caregivers (n = 776 NM ties nested within 71 participants across 30 families)

Model 1 Model 2 Model 3a Model 3b
Predictors OR CI OR CI OR CI OR CI
 Intercept 0.41 0.08 – 2.01 0.34 0.07 – 1.66 0.39 0.06 – 2.43 1.56 0.22 – 11.29
 Uplift 2.19 ** 1.24 – 3.87 2.57 ** 1.43 – 4.61 2.58 ** 1.44 – 4.62 0.62 0.25 – 1.52
 Malfeasance 0.86 0.47 – 1.57 0.90 0.49 – 1.64 0.90 0.49 – 1.64 0.53 0.20 – 1.36
 Nonfeasance 0.35 *** 0.20 – 0.61 0.40 ** 0.23 – 0.70 0.40 ** 0.23 – 0.70 0.44 * 0.19 – 0.98
 Kinship (Friendship) 2.28 ** 1.38 – 3.78 2.28 ** 1.37 – 3.78 0.92 0.45 – 1.90
 AD FH 0.85 0.29 – 2.53 0.06 *** 0.01 – 0.31
   x Uplift 13.56 *** 3.87 – 47.57
   x Malfeasance 2.57 0.75 – 8.84
   x Nonfeasance 0.86 0.28 – 2.69
   x Kinship (Friendship) 6.52 *** 2.26 – 18.77
 NM Gender (male) 0.91 0.58 – 1.43 0.86 0.54 – 1.35 0.86 0.54 – 1.35 0.78 0.49 – 1.25
 Gender homophily 1.30 0.82 – 2.05 1.12 0.71 – 1.79 1.12 0.71 – 1.79 1.11 0.69 – 1.78
 Care needs 1.01 0.94 – 1.08 1.00 0.94 – 1.07 1.00 0.93 – 1.07 1.00 0.93 – 1.08
Random Effects (Variance [standard deviation])
 Participant τ00 3.36 (1.83) 3.26 (1.81) 3.29 (1.81) 3.20 (1.79)
 Family τ00 0.44 (0.66) 0.44 (0.67) 0.40 (0.64) 0.55 (0.74)
 Marginal R2 / Conditional R2 0.044 / 0.557 0.060 / 0.558 0.060 / 0.557 0.108 / 0.583
Open in a new tab

Notes. AD FH = Family history of Alzheimer’s disease and related dementia; NM = Network member.

*

p<0.05

**

p<0.01

***

p<0.001

In the interaction model (Model 3b), we observed a main effect of family history such that those with a family history of AD reported significantly smaller emotional support networks than those without a prior family history (OR = 0.06, p < 0.001). Family history of AD also amplified the effects of uplift; those network members involved in at least one uplifting caregiving-related behavior or interaction were significantly more likely to be perceived as providing emotional support by participants with a prior family history of AD compared to those without a prior family health history (OR = 13.56, p < 0.001). In addition, those with a family history of AD were also significantly more likely to nominate friends as emotionally supportive compared to those without family history of AD (OR = 6.52, p < .001). Notably, when moderating effects of prior family history were added to the model, the main effects for uplift and kinship tie were no longer statistically significant (Table 2), indicating that the associations between emotional support and both uplift and kinship tie were observed only among those with a prior family history, but not those without a prior family history. The association between nonfeasance and perceived emotional support remained consistent, regardless of prior family history of ADRD. Additional models that accounted for participants’ relationship to and co-residence with the person with ADRD yielded the same pattern of results.

Discussion

The present study contributes to the literature on caregiving by integrating two theoretical models - the stress process model and the support provision process model - to consider how caregiver contextual factors and interpersonal stressors arising from caregiving shape coping resources. Specifically, we assessed whether family history of ADRD and caregiving-related behaviors and interactions among family members were associated with the composition and functioning of family caregivers’ emotional support networks. We found that caregiving-related behaviors and interactions represent an important interpersonal context underpinning perceived emotional support. Nonfeasance, or disengagement from caregiving, and uplift, or active positive engagement in caregiving, were significantly associated with ADRD caregivers’ perceived emotional support. Moreover, friends played particularly important roles in providing emotional support to caregivers. Finally, prior family history of ADRD was found to be an important contextual factor, moderating some of these associations.

According to the stress process model, nonfeasant and malfeasant caregiving behaviors and interactions reflect indirect caregiving stressors resulting from negotiating family caregiving roles and expectations. Indeed, prior work demonstrates that nonfeasant behaviors were particularly stressful, being associated with caregiver well-being, more so than malfeasant behaviors (Ashida et al., 2018; Brody et al., 1989). Here, we showed that such stressors can also shape the social resources that caregivers call upon to cope with caregiving burden. Given that caregiving interactions characterized by malfeasance, such as arguing, hostility, and disapproval surrounding caregiving decisions and behaviors, have often been linked to poorer well-being and greater caregiver burden (Antonucci & Sherman, 2019), we expected that it would be negatively associated with perceived emotional support. However, this association was not observed. Network members who exhibited malfeasant behaviors, despite reported negative interactions, were still actively involved in the caregiving process. It could be that network members who exhibit malfeasance also demonstrate uplifting behaviors and interactions. Indeed, most close ties are ambivalent in nature, characterized by high positivity and high negativity simultaneously (Fingerman et al., 2004). In contrast, we found that those network members who are disengaged from caregiving activities, or nonfeasant, were unlikely to provide emotional support, regardless of whether the caregiver reported a prior family history of AD. These findings are consistent with previous research in which caregivers have reported feelings of resentment towards family members who fail to show interest (Kutner et al., 2009) .

Family history of ADRD moderated the association between uplift in caregiving and emotional support such that caregivers from families with prior experiences with AD were more likely to report emotional support from network members that demonstrated uplifting caregiving-related behaviors or interactions; there was no such association for caregivers without a family history of ADRD. The effects of uplift and friendship among networks with prior history of ADRD suggest a particularly nuanced and important relationship between uplift and social ties, especially in taxed social systems with prior experience with the illness. Families that have experienced caring for someone with ADRD prior to the current caregiving situation may have existing schemas for the progression of the disease and caregiving needs. The support networks of those caregivers and the interpersonal dynamics of the family that we observe in the current study could be a product of these prior experiences of having a relative with ADRD. Alternatively, families without a family history of ADRD may still be in the process of developing such schema to effectively provide care. Notably, we observed that caregivers with a prior family history of ADRD reported significantly fewer network members exhibiting caregiving-related nonfeasance and significantly more network members exhibiting caregiving-related uplift. These findings suggest the possibility that networks are pruned - that is, network members who stress the system are dropped - based on these prior experiences, reflecting a fracturing of families. Or, alternatively, families grow through the experience, recognizing the importance of expressing praise or gratitude to those providing direct care or making decisions about care. Consequently, when network members interact with them in ways that embody uplift, caregivers may perceive those individuals as especially supportive.

We investigated the role of family and friends in providing emotional support to caregivers. Given the familial nature of dementia caregiving, friends are typically not involved directly in care, but could still serve as sources of emotional support outside of the caregiving context (Antonucci & Sherman, 2019). However, we were interested in whether or not those relationships even existed in caregiving networks - studies have shown that a pruning occurs during times of caregiving burden such that maintaining social relationships with friends loses priority to caregiving responsibilities and interactions within the caregiving network (Roth, 2020; Wittenberg-Lyles et al., 2014).

We observed a moderation effect of context on kinship tie, such that friendship ties were associated with perceived emotional support among caregivers with a prior family history of ADRD, but not those without a prior family history. The prevalence of friends in caregivers’ emotional support networks could be the result of network selection or selective pruning. Friendships tend to be higher quality relationships because there is less criticism and conflict, and these relationships are largely formed and maintained by choice (Antonucci et al., 2004; Fuller-Iglesias et al., 2013).

However, resentment towards friends can result if they provide unsolicited help (Almberg et al., 2000). Pruning of friendship ties could also occur such that only supportive friendship ties are maintained and included within the caregiving network of those with family history of ADRD, suggesting that the friends retained in the network are especially important. Given that friends are not typically expected to be involved in caregiving activities, inclusion of friendship ties in the caregiving network may indicate network selection resulting in a broadening of support sources. Guiding caregivers, especially those who may be pruning and reducing interactions with friends, to identify potential support sources from social ties outside of their family may be beneficial.

Clinical Implications

This research provides insight into the characteristics of caregivers’ social environment that place caregivers at risk for poor adaptation to caregiver stress due to inadequate support. Although caregiver well-being is also impacted by a wide range of community or societal factors, including availability of healthcare, financial support, and community stigma, caregiver support networks represent a modifiable social resource that can be leveraged to buffer the negative impact of caregiving stress.

Behaviors and interactions on the scale that characterize uplift, such as showing appreciation for the caregiver’s efforts and keeping in touch (i.e., in person, phone, electronically), could be encouraged by family members and friends, while behaviors and interactions reflecting nonfeasance could be avoided or reversed. This is particularly important for caregivers with a prior family history of ADRD. Not only are caregivers with a prior family history impacted from their family’s previous caregiving experience, but they are also affected by their personal risk of ADRD due to its heritability (Alzheimer Association, 2020). Supporting families as they negotiate their personal risk, risk of future generations, and their caregiving roles is a research priority moving forward (Koehly, 2017). On the other hand, for families without prior experiences with ADRD, caregivers could benefit from guidance on regulating their own support networks. Strategies may include maintaining friendships with those who could be providing emotional support even as caregiving demands limit the time caregivers have available to nurture these friendships. Alternatively, caregivers can identify new friends (e.g., through support groups) who share similar experiences with ADRD caregiving.

Limitations

Here, we rely on participant self-report of network members’ caregiving-related behaviors and interactions as well as emotional support receipt. Such self-report data reflect the participants’ unique perspective of the caregiving network system. However, research suggests that the support recipient’s perception of emotional support received may be more reflective of their actual coping resources than measures that factor in other measurement modes, including perceived or observed support from network members’ perspectives (Birditt et al., 2012). Additionally, we did not model the network system as a whole, which might provide insights into exchanges and structure (e.g., reciprocal support exchanges) that are not investigated here. Instead, we modeled ego-centric networks across multiple family members, which limits our ability to make conclusions about whole family network systems from structural perspective (Koehly et al., 2015). Finally, the large odds ratios noted for the interaction between family history and uplift are likely due in part to small cell sizes and a fairly homogeneous distribution of emotional support ties within the subset of network members of respondents who had a prior FH of ADRD and who engaged in uplift, though this pattern remains important to note.

Previous reports have used factor scores to capture malfeasance, nonfeasance, and uplift (Ashida et al., 2018); however, here, we dichotomized these measures to reflect the presence of at least one corresponding behavior. We used dichotomized measures with the intent to inform future intervention aimed at increasing coping resources for caregivers. From this perspective, our findings suggest that targeting at least one positive caregiving-related behavior has the potential to activate emotional support ties. On the other hand, mitigating nonfeasance will be important in facilitating emotional support provision within these networks. Additionally, it is important to recognize that the caregiving-related behaviors and interactions are not mutually exclusive. Although we examined uplift, malfeasance, and nonfeasance as separate behaviors, it is possible, and likely, that network members simultaneously exhibit more than one. Investigating multiplexity among these caregiving-related behaviors might more fully capture the complexity of interpersonal dynamics (Koehly & Marcum, 2018). Finally, all participants are from an urban city in the U.S., limiting the generalizability of the findings to other populations with different cultural and social background.

Conclusions

This study expands the stress process model to consider how the background context of a prior family health history of ADRD and interpersonal stressors shape available coping resources. We demonstrated that having had family members, other than the person with ADRD, diagnosed with ADRD strengthens the association between network members’ uplifting caregiving behaviors and emotional support. We also underscore the importance of friends as support providers among caregivers. Our findings provide novel, feasible points of intervention to foster emotionally supportive social environments for caregivers of individuals with dementia. Future work might test the impact of discussing the health implications of uplifting, nonfeasance, and malfeasance behaviors with families on facilitating positive and minimizing negative interactions among them. Interpersonal relationships function both as stressors and as coping resources and their impacts depend on the characteristics of interactions. Our study showed that the way interactional characteristics manifest as stressors or coping resources may differ based on whether the family has prior experience with ADRD, suggesting the need for differential strategies to facilitate positive and minimize negative interactions. By creating more supportive environments for caregivers using the strategies appropriate for the contexts of each family, we may mitigate some of the documented negative psychological and physical health effects of caregiving for a loved one with ADRD.

Supplementary Material

1

Funding:

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Authors supported by funding from the Intramural Research Program of the National Human Genome Research Institute [ZIAHG20335]. J.C. was supported by T32HD007109.

Biography

JLC is a graduate student at the University of Michigan. JAM is a postdoctoral fellow, CSM is a Staff Scientist, and LK is a Senior Investigator at the National Human Genome Research Institute. SA is an Associate Professor at the University of Iowa. JR is a postdoctoral fellow at the University of Pennsylvania and Philadelphia VA Medical Center.

References

  1. AARP and National Alliance for Caregiving. (2020). Caregiving in the United States 2020 (Issue May). [Google Scholar]
  2. Almberg BE, Grafström M, & Winblad B (2000). Caregivers of relatives with dementia: Experiences encompassing social support and bereavement. Aging and Mental Health, 4(1), 82–89. 10.1080/13607860056026 [DOI] [Google Scholar]
  3. Alzheimer Association. (2020). 2020 Alzheimer’s disease facts and figures. Alzheimer’s Dementia, 16(3), 391–460. 10.1002/alz.12068 [DOI] [Google Scholar]
  4. Antonucci TC, Ajrouch KJ, & Birditt KS (2014). The convoy model: Explaining social relations from a multidisciplinary perspective. The Gerontologist, 54(1), 82–92. 10.1093/geront/gnt118 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Antonucci TC, Akiyama H, & Takahashi K (2004). Attachment and close relationships across the life span. Attachment & Human Development, 6(4), 353–370. 10.1080/1461673042000303136 [DOI] [PubMed] [Google Scholar]
  6. Antonucci TC, & Sherman CW (2019). Applying the convoy model to support in care situations. International Journal of Care and Caring, 3(1), 23–38. 10.1332/239788219x15492857438483 [DOI] [Google Scholar]
  7. Ashida S, Marcum CS, & Koehly LM (2018). Unmet expectations in Alzheimer’s family caregiving: Interactional characteristics associated with perceived under-contribution. Gerontologist, 58(2), e46–e55. 10.1093/geront/gnx141 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Birditt KS, Antonucci TC, & Tighe L (2012). Enacted support during stressful life events in middle and older adulthood: an examination of the interpersonal context. Psychology and Aging, 27(3), 728–741. 10.1037/a0026967 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Bom J, Bakx P, Schut F, & van Doorslaer E (2018). The impact of informal caregiving for older adults on the health of various types of caregivers: A systematic review. The Gerontologist. 10.1093/geront/gny137 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Brody EM, Hoffman C, Kleban MH, & Schoonover CB (1989). Caregiving daughters and their local siblings: Perceptions, strains, and interactions. Gerontologist, 29(4), 529–538. 10.1093/geront/29.4.529 [DOI] [PubMed] [Google Scholar]
  11. Carey Vincent J. Ported to R by Thomas Lumley, Brian Ripley. Files src/dgedi.f and src/dgefa.f are for LINPACK authored by Cleve Moler. Note that maintainers are not available to give advice on using a package they did not author. (2019). gee: Generalized Estimation Equation Solver. R package version 4.13–20. https://CRAN.R-project.org/package=gee [Google Scholar]
  12. Carpentier N, & Ducharme F (2003). Care-giver network transformations: The need for an integrated perspective. Ageing and Society, 23(4), 507–525. 10.1017/S0144686X03001211 [DOI] [Google Scholar]
  13. Cheng S-T (2017). Dementia caregiver burden: A research update and critical analysis. Current Psychiatry Reports, 19(9), 64. 10.1007/s11920-017-0818-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dunkel-Schetter C, & Skokan LA (1990). Determinants of social support provision in personal relationships. Journal of Social and Personal Relationships, 7(4), 437–450. 10.1177/0265407590074002 [DOI] [Google Scholar]
  15. Fingerman KL, Hay EL, Birditt KS, & Hay L (2004). The best of ties, the worst of ties: Close, problematic, and ambivalent social relationships. Journal of Marriage and Family, 66(3), 792–808. [Google Scholar]
  16. Fuller-Iglesias HR, Webster N, & Antonucci TC (2013). Adult family relationships in the context of friendship. Research in Human Development, 10(2), 184–203. 10.1080/15427609.2013.786562 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Gilhooly KJ, Gilhooly MLM, Sullivan MP, McIntyre A, Wilson L, Harding E, Woodbridge R, & Crutch S (2016). A meta-review of stress, coping and interventions in dementia and dementia caregiving. BMC Geriatrics. 10.1186/s12877-016-0280-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Koehly LM (2017). It’s interpersonal: Family relationships, genetic risk, and caregiving. The Gerontologist, 57(1), 32–39. 10.1093/geront/gnw103 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Koehly LM, Ashida S, Schafer EJ, & Ludden A (2015). Caregiving networks — Using a network approach to identify missed opportunities. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 70(1), 143–154. 10.1093/geronb/gbu111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Koehly LM, & Marcum CS (2018). Multi-relational measurement for latent construct networks. Psychological Methods, 23(1), 42–57. 10.1037/met0000110 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kutner J, Kilbourn KM, Costenaro A, Lee CA, Nowels C, Vancura JL, Anderson D, & Keech TE (2009). Support needs of informal hospice caregivers: A qualitative study. Journal of Palliative Medicine, 12(12), 1101–1104. 10.1089/jpm.2009.0178 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lazarus RS, & Folkman S (1984). Cognitive appraisal processes. In Stress, Appraisal, and Coping (pp. 22–54). Springer Publishing Company. [Google Scholar]
  23. Lee S, Chung JE, & Park N (2018). Network environments and well-being: An examination of personal network structure, social capital, and perceived social support. Health Communication, 33(1), 22–31. 10.1080/10410236.2016.1242032 [DOI] [PubMed] [Google Scholar]
  24. Lilly ML, Richards BS, & Buckwalter KC (2003). Friends and social support in dementia caregiving. Assessment and intervention. Journal of Gerontological Nursing, 29(1), 29–36. 10.3928/0098-9134-20030101-10 [DOI] [PubMed] [Google Scholar]
  25. Linn MW (1988). Rapid Disability Rating Scale-2 (RDRS-2). Psychopharmacology Bulletin. [PubMed] [Google Scholar]
  26. Loewenstein DA, Arguelles S, Bravo M, Freeman RQ, Arguelles T, Acevedo A, & Eisdorfer C (2001). Caregivers’ judgments of the functional abilities of the Alzheimer’s Diseasepatient: A comparison of proxy reports and objective measures. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 56(2), P78–P84. 10.1093/geronb/56.2.P78 [DOI] [PubMed] [Google Scholar]
  27. Marcum CS, Ashida S, & Koehly LM (2020). Primary caregivers in a network context. The Journals of Gerontology: Series B, 75(1), 125–136. 10.1093/geronb/gbx165 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Pearlin LI, Mullan JT, Semple SJ, & Skaff MM (1990). Caregiving and the stress process: An overview of concepts and their measures. Gerontologist, 30(5), 583–594. 10.1093/geront/30.5.583 [DOI] [PubMed] [Google Scholar]
  29. Pinquart M, & Sörensen S (2003). Differences between caregivers and noncaregivers in psychological health and physical health: A meta-analysis. Psychology and Aging, 18(2), 250–267. 10.1037/0882-7974.18.2.250 [DOI] [PubMed] [Google Scholar]
  30. R Core Team (2018). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/ [Google Scholar]
  31. Rolland JS (2006). Genetics, family systems, and multicultural influences. Families, Systems, & Health, 24(4), 425–441. 10.1037/1091-7527.24.4.425 [DOI] [Google Scholar]
  32. Rolland JS, & Williams JK (2005). Toward a biopsychosocial model for 21 st-century genetics. Family Process, 44(1), 3–24. 10.1111/j.1545-5300.2005.00039.x [DOI] [PubMed] [Google Scholar]
  33. Roth AR (2020). Informal caregiving and network turnover among older adults. The Journals of Gerontology: Series B, 75(7), 1538–1547. 10.1093/geronb/gby139 [DOI] [PubMed] [Google Scholar]
  34. Schulz R, & Beach SR (1999). Caregiving as a risk factor for mortality. JAMA, 282(23), 2215–2219. 10.1001/jama.282.23.2215 [DOI] [PubMed] [Google Scholar]
  35. Sherman CW (2012). Remarriage as context for dementia caregiving: Implications of positive support and negative interactions for caregiver well-being. Research in Human Development, 9(2), 165–182. 10.1080/15427609.2012.680845 [DOI] [Google Scholar]
  36. Snijders TAB, & Bosker RJ (1999). Multilevel analysis: An introduction to basic and advanced multilevel modeling. Sage. [Google Scholar]
  37. Wilks SE, & Croom B (2008). Perceived stress and resilience in Alzheimer’s disease caregivers: Testing moderation and mediation models of social support. Aging and Mental Health, 12(3), 357–365. 10.1080/13607860801933323 [DOI] [PubMed] [Google Scholar]
  38. Wittenberg-Lyles E, Washington K, Demiris G, Oliver DP, & Shaunfield S (2014). Understanding social support burden among family caregivers. Health Communication, 29(9), 901–910. 10.1080/10410236.2013.815111 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

1
NIHMS2101661-supplement-1.pdf (324.8KB, pdf)

RESOURCES