Parent-child relationship quality and older parents’ cognitive trajectories

Abstract

This study assesses the bidirectional link between parent-child relationship quality and older parents’ cognitive trajectories over time, with particular attention to potential variations by parents’ gender. Methods: Using a longitudinal, nationally representative sample of parents aged 50 and older from the Health and Retirement Study 2006–2016, the analysis examines three dimensions of the relationship quality between parents and children—contact, support, and strain—in relation to parents’ cognitive function. Parallel latent growth curve models (LGCMs) estimate how baseline relationship quality relates to subsequent change in parents’ cognition and how parents’ baseline cognitive function relates to subsequent change in relationship quality over time. Results: Higher initial levels of relationship strain with children predicts a faster cognitive decline for older parents. Neither initial contact frequency nor initial relationship support is associated with subsequent change in a parent’s cognition, however, good cognitive function predicts a more rapid increase or a slower decline in contact frequency and relationship support over time. No gender difference was found. Conclusion: The results highlight the importance of the “linked lives” of aging parents and their children as well as the potential for bidirectional associations between cognition and family relationships.

Keywords

Intergenerational relations health family gender relationship quality aging

Introduction

One of the most significant factors influencing parents’ health outcomes and quality of life is their relationships with their children (Koropeckyj-Cox, 2002; Reczek & Zhang, 2016). Further, the bond between parents and children is becoming even more important as life expectancy increases in the United States and parents are increasingly likely to experience a significant period of lifespan overlap with their children (Fingerman et al., 2020; Silverstein et al., 2006; Swartz, 2009). As parents age, adult children are typically their primary caregivers as well as essential sources of social connection and support (Carr & Utz, 2020; Umberson et al., 2010). An extensive body of research has found that both closeness to children and support provided by children enhance parents’ psychological well-being, whereas conflict and strain with children can damage health (Koropeckyj-Cox, 2002; Reczek & Zhang, 2016). These broad-brush findings highlight the need for further research on the association between parent-child relationships and parents’ cognitive function.

Cognitive decline has emerged as a burgeoning public health concern in the United States and worldwide (Alzheimer’s Association, 2020). As the U.S. population ages, cognitive decline will continue to pose a significant challenge, placing a greater burden on both the families of older adults and the long-term care system (Crimmins et al., 2016). Cognitive decline influences the lives of not only patients but also their caregivers—most often their adult children—in a significant way. Parents’ memory loss, reduced language ability, and/or psychiatric disturbances (which can accompany the progression of dementia) can threaten the intimacy between parents and children (Alzheimer’s Association, 2020). Further, adult children who are caregivers for aging parents suffer emotional and physical stress that may change the parent-child relationship (Chirico et al., 2021). Simultaneously, the quality and nature of the parent-child relationship can have a profound impact on the life context of parents, thus potentially influencing their cognitive functioning as they age.

Guided by a life course perspective, this study considers the potentially bidirectional link between the parent-child relationship and parents’ cognitive function. Drawing on longitudinal data collected from a nationally representative sample of older adults as part of the Health and Retirement Study (2006–2016), the analysis examines how multiple dimensions of parent-child relationships (contact, support, and strain) shape older parents’ trajectories of cognitive decline as well as how parents’ cognitive function shapes their relationships with children. Given the longstanding literature on gender differences in parent-child relationships (Milkie et al., 2008; Pillemer & Suitor, 2002), the study also examines whether these patterns differ between fathers and mothers. The findings extend the scholarly understanding of parent-child relationships in the context of a rapidly aging U.S. population and can inform medical practitioners and policy makers as they design intervention strategies to protect the cognitive health of older adults.

Parent-child relationships and parents’ cognitive function: Previous empirical evidence

A handful of studies have examined the associations between the quality of intergenerational relationships and the health of older parents, including cognitive health, however, these studies have produced mixed results (e.g., Li et al., 2021; Thomas & Umberson, 2018; Zahodne et al., 2019). For example, a Japanese survey revealed that children’s support was associated with a lower risk of cognitive impairment among parents who did not have spouses (Okabayashi et al., 2004). Two studies of Chinese older adults suggested that both the absence of intergenerational solidarity and the presence of relationship conflict were associated with poorer cognitive performance while visits from children were associated with a lower incidence of cognitive impairment (Li et al., 2021; Yin et al., 2020). Further, a U.S. study suggested that having more frequent contact with children and reporting less strain in relationships with children were associated with better initial memory levels (Zahodne et al., 2019). In contrast, another U.S. study found no evidence that frequency of contact with family members, including children, was associated with episodic memory in later life (Sharifian et al., 2020). Similarly, two studies of Chinese older adults found that support or strain from family members (the study did not distinguish between children, parents, siblings, and relatives) had no statistically significant association with cognitive function (Ge et al., 2017; Li et al., 2019). These mixed finding may be due to various factors, such as different dimensions of relationship quality measured or various social contexts. Therefore, current evidence is far from conclusive regarding the linkage between parent-child relationship and cognitive function.

A thorough review of the literature indicates that Thomas and Umberson (2018) is the only published study of the association between parent-child relationship quality and parental cognitive health in the United States with a gendered perspective. This study (Thomas & Umberson, 2018) found that for mothers, a higher level of strain with children was associated with greater initial cognitive limitations, while for fathers, more strain was associated with slower cognitive decline. Although informative, the study (Thomas & Umberson, 2018) has at least two important limitations that are addressed in the current study. First, the earlier study did not examine the potential bidirectionality of link between parent-child relationships and cognitive health. Second, the study examined parent-child relationships as a static status, and did not consider cognitive function in relation to changes in the relationship between parents and children over time. As discussed in the following section, the life course perspective clearly posits the importance of considering changes in linked lives over time as well as the potential for bidirectionality in the processes linking parent-child relationships and parents’ cognitive function.

Life course perspective: Bidirectional link between “linked lives” and cognitive trajectories

The life course perspective asserts that family members have “linked lives,” suggesting that parents’ well-being is highly contextualized and extensively shaped by their interactions with their children over time (Elder, 1998). The parent-child relationship is dynamic and its impact on parental well-being is usually cumulative across the life course (Umberson et al., 2010). Because older adults are living longer today, the linked lives of parents and children are becoming more important than ever. Moreover, the current generation of adult children have experienced multiple obstacles to achieving socioeconomic stability, including a high demand for education, serious economic downturns, and an unstable job market, all of which have extended the duration of close interaction with their parents and created more complex intergenerational bonds (Kornrich & Furstenberg, 2013). Indeed, many of today’s aging parents provide financial resources to their adult children while receiving care from these children. Given these dynamics, the quality of their “linked lives” may fluctuate over time, with consequences for parents’ health trajectories.

Children are often the primary caregivers for older parents as well as major figures in their social networks who can increase parents’ levels of social engagement, emotional support, and social control, all of which can potentially slow the process of cognitive decline (Giles et al., 2012; Umberson, 1987). e.g., parents can connect with their adult child’s family (e.g., their grandchildren) and friends. An extensive literature has shown that social engagement improves individuals’ cognitive reserves by strengthening the ability to cope with neuropathological damage via compensatory cognitive strategies (Giles et al., 2012). Frequent contact with children often serves as a major source of communication and interaction, which may increase older parents’ neural plasticity, thus maintaining brain reserves (Giles et al., 2012). Moreover, children’s emotional support can ease parents’ psychological distress (e.g., depression, loneliness, isolation) and thus reduce parents’ risk of cognitive decline (Costa-Cordella et al., 2021; Shankar et al., 2013). In addition, children can help monitor parents’ health behaviors, and empirical evidence indicates that engaging in a healthy lifestyle, such as reducing smoking and drinking, consuming a healthy diet, and getting regular exercise, is related to a significantly lower risk of cognitive impairment at older ages (Bherer et al., 2013; Hayes et al., 2016; Swan & Lessov-Schlaggar, 2007).

Despite the numerous benefits, the “linked lives” of parents and children may also have a dark side—a strained relationship with children may damage parents’ cognitive health. Negative interactions with children, such as dealing with demanding children, experiencing emotional abuse, and becoming estranged, can consume time and money parents would otherwise use to take care of themselves, and, in turn, prompt the development of sleep problems, chronic diseases, and psychological distress among parents (Milkie et al., 2008; Nomaguchi, 2012). Relationship conflict stimulates parents’ stress hormones (e.g., catecholamines, cortisol), impairs the immune system, and evokes physical responses, all of which may have a long-term negative impact on brain reserve and cognition function (Kuhlmann et al., 2005; Oei et al., 2007). Moreover, epidemiological studies have reported that stress is associated with hypertension and cardiovascular diseases that are, in turn, related to higher risks of cognitive impairment, Alzheimer’s disease, and vascular dementia (Johansson et al., 2010; Sparrenberger et al., 2009). Negative interactions with children can also have a spillover effect or lead to stress proliferation, causing additional or exacerbated life strains for older parents, such as economic hardship, social isolation, and loss of or conflict with the larger family network (e.g., grandchildren, relatives) (Thoits, 2010). Therefore, among older parents, strain in relationships with their children is a potential source of life stress that can be pathogenic and increase vulnerability to cognitive decline (Johansson et al., 2010).

The life course perspective on linked lives also posits that parents’ cognitive function may influence their relationships with their children, especially when children play the role of caregiver. The combination of this association and those described in the previous paragraphs suggest that the link between parent-child relationship quality and parents’ cognitive function is bidirectional. Cognitive impairment can significantly impact patient’s mood, temper, and emotional well-being (Yates et al., 2013). As cognitive function declines, individuals may experience heightened levels of frustration, confusion, anxiety, and depression, often resulting in behavior changes such as becoming more demanding. These changes can have significant impact on their relationships with adult children, adding strain and complexity to family relationships. Moreover, cognitive impairment can diminish patients’ ability to maintain their social contact or interaction with family and other ties because of reduced memory or language abilities (Alzheimer’s Association, 2020). Moreover, taking care of cognitively impaired parents can significantly shape children’s life contexts, leading to changes in relationships between parents and children. e.g., a substantial literature has found that performing intensive care work for parents with cognitive impairment can influence children’s life planning (e.g., with respect to education and career) and increase the likelihood of financial hardship, social isolation, stress, and a variety of other health complications (Brodaty & Donkin, 2022; Chirico et al., 2021). Further, caring for a parent with cognitive decline is extremely stressful and can ignite tensions and conflicts between parents, children, and extended family (Brodaty & Donkin, 2022; Chirico et al., 2021). Given these possible consequences, it is important for researchers to consider the potentially bidirectional links between parent-child relationships and parents’ cognitive function.

Mothers versus fathers

The concept of “linked lives” also points to the role of gender in the life course context, highlighting the gendered nature of the association between intergenerational relationships and parents’ cognitive function. Within a family, mothers typically play the role of “kin-keepers” and feel more obligated than fathers to maintain ties with their children and the extended family network (Lendon et al., 2014; Pillemer & Suitor, 2002; Silverstein & Bengtson, 1997). Indeed, prior research has found a consistently gendered pattern of parental involvement, suggesting that mothers are more likely than fathers to spend social time with their children and are more emotionally involved in children’s lives, while fathers often obtain support from children in part through the kin-keeping activities of mothers (Nomaguchi, 2012). Consequently, in their relationships with their children, mothers tend to experience higher levels of both strain and support than fathers because fulfilling the “kin-keeper” role can amplify both the benefits and costs of interaction (Pillemer & Suitor, 2002). Because women’s relationships are more sensitive to outside influences than men’s relationships are (Liu & Waite, 2014), factors such as cognitive health may influence relationships with children more among mothers than among fathers.

Studies of older parents with children in the United States have consistently shown that the parent-child relationship has a stronger influence on mothers’ psychological well-being than on fathers’ psychological well-being (Milkie et al., 2008; Reczek & Zhang, 2016; Ward et al., 2009). For example, Reczek and Zhang (2016) found that compared to older fathers, older mothers received more support from, had more contact with, and were less dissatisfied with their adult children. Similarly, Ward and coauthors (2009) showed that older mothers experienced more positive relationships, but not more negative relationships, with their adult children than older fathers. Milkie et al. (2008) found that negative treatment by adult children was associated with increased anger for older mothers but not older fathers.

Taken together, the current study considers the potentially bidirectional link between the parent-child relationship and parents’ cognitive function to address three research questions: (1) how does parent-child relationships (contact, support, and strain) shape the subsequent change of older parents’ cognitive function? (2) how does parents’ cognitive function shape their subsequent change of relationships with children? And (3) do these bidirectional processes differ between fathers and mothers? The findings extend the scholarly understanding of parent-child relationships in the context of a rapidly aging U.S. population and can inform medical practitioners and policy makers as they design intervention strategies to protect the cognitive health of older adults.

Methods

Data

Data were drawn from the Health and Retirement Study (HRS, 2006–2016), a nationally representative longitudinal panel survey conducted by the Institute for Social Research at the University of Michigan. Every two years since 1992, HRS data have been collected from samples of Americans over age 50 (Sonnega & Weir, 2014). In 2006, half of the sample members were randomly selected to complete a psychosocial and lifestyle questionnaire; the remaining sample members were asked to complete the questionnaire in 2008, and both groups completed a follow-up every four years (i.e., the 2006 participants were reinterviewed in 2010 and 2014, while the 2008 participants were reinterviewed in 2012 and 2016). This questionnaire asked respondents to assess the quality of their social ties, including support and strain they received from their children as well as the frequency of contact with children (Smith et al., 2017). The current study combines the two non-overlapping samples of respondents who completed the psychosocial questionnaire in either 2006 or 2008 to form an analytical sample containing data on parent-child relationship quality at three time points: Time 1 (baseline) = 2006/08, Time 2 = 2010/12, and Time 3 = 2014/16. The cognition data were collected biennially, at six time points: Time 1 = 2006, Time 2 = 2008, Time 3 = 2010, Time 4 = 2012, Time 5 = 2014, and Time 6 = 2016.

The analysis excludes partners who were under age 50 at baseline as well as those who did not have children in each wave. In addition, respondents were excluded if they had missing values on key covariates at baseline including education, race/ethnicity, marital status, age, number of children, gender of children, and health-related variables. The final analytic sample included 13,169 respondents (5,369 fathers and 7,800 mothers) who had at least one child¹. A flowchart in Figure 1 shows how we reached the final sample step-by-step. Besides, cognition has 3.11% missing at baseline, and 41.54% missing at Time 6; relationship quality variables have around 8%–10% missing at baseline, and 46%–47% missing at Time 3. Missing values for relationship quality and cognitive scores at certain time point were addressed via a full information maximum likelihood (FIML) estimation approach in Mplus (Muthén & Muthén, 1998-2015). Compared to the raw sample, the retained sample tends to be slightly older and more educated, with fewer chronic conditions and slightly more children. They exhibit less depression and are more likely to be drinkers. Additionally, respondents in the retained sample are less likely to be racial minorities, unmarried, or smokers. They also have higher relationship support scores, lower strain scores, and more frequent contact with children.

Figure 1.

Flowchart of sample selection process. Note: (1) Missing dropped in covariates of education, race/ethnicity, marital status, age, number of children, and health-related variables; (2) Refresh sample refers to those whose baseline interview was not in 2006 or 2008; (3) In our final sample (N = 13,169), there are respondents who had missing in RQ or cognition variable at certain time point (but not all time points). These missing were handled by FIML in Mplus.

Measures

Parents’ cognitive function

HRS assesses cognition via the modified version of the Telephone Interview for Cognitive Status (TICS), which includes the following items: immediate and delayed recall of a list of 10 words (1 point for each), five trials of serial 7s (i.e., subtract 7 from 100, and continue subtracting 7 from each subsequent number for a total of five trials, 1 point for each trial), and backward counting (2 points). Final scores range from 0 to 27, with a higher score indicating better cognitive function (Crimmins et al., 2016).

Parent-child relationship quality

We measured three dimensions of parent-child relationship quality, encompassing both quantitative and qualitative aspects: (1) contact frequency, (2) relationship support, and (3) relationship strain. Contact frequency with children was assessed using data from two items (Cronbach’s alpha = 0.66): (1) “How often do you meet up (include both arranged and chance meetings) with your children?” and (2) “How often do you speak on the phone with your children?” Participants responded on a 6-point scale ranging from 1 = three or more times a week to 6 = less than once a year or never (Smith et al., 2017). Both items were reverse coded so higher scores correspond to more frequent contact. The final score is the mean of the two items.

Relationship support was assessed via three items (Cronbach’s alpha = 0.81): (1) “How much do your children really understand the way you feel about things?” (2) “How much can you rely on your children if you have a serious problem?” and (3) “How much can you open up to your children if you need to talk about your worries?” Responses were based on a 4-point scale ranging from 1 = a lot to 4 = not at all (Smith et al., 2017). All three items were reverse coded so higher scores correspond to greater relationship support. The final support score is the mean of the three responses (Zahodne et al., 2019).

Relationship strain was assessed based on four items (Cronbach’s alpha = 0.78): (1) “How often do your children make too many demands on you?” (2) “How often do your children criticize you?” (3) “How much do your children let you down when you are counting on them?” and (4) “How much do your children get on your nerves?” Respondents answered on a 4-point scale ranging from 1 = a lot to 4 = not at all (Smith et al., 2017). All four items were reverse coded so higher scores correspond to greater relationship strain. The final strain score is the mean of the four responses (Zahodne et al., 2019).

Covariates

The analyses control for sociodemographic covariates at the baseline wave (2006) that are related to both parent-child relationship and cognition. Age (in years) is centered at the mean. Race/ethnicity includes four categories: non-Hispanic White (reference), non-Hispanic Black, Hispanic, and non-Hispanic other. Education includes four categories: less than high school (reference), high school graduate, some college, and college graduate or above. Parents’ marital status is a dichotomous variable: 0 = married/partnered and 1 = unmarried (including divorced, widowed, and never married). Number of children indicates the total number of biological children, stepchildren, and other types of children. Parents’ gender is a dichotomous variable: 0 = men; 1 = women. Children’s gender includes three categories: 0 = parents with all sons but no daughters, 1 = parents with all daughters but no sons, and 2 = parents with both sons and daughters. Smoking status is a dichotomous variable: 0 = non-smoker and 1 = current smoker. Drinking status is a dichotomous variable: 0 = not a current drinker and 1 = current drinker. Chronic condition is measured by a comorbidity index ranging from 0 to 4, summarizing the presence of four major chronic conditions: diabetes, stroke, heart disease, and high blood pressure. Depression is measured by a subset of the Center for Epidemiological Studies Depression Scale (CESD). Participants were asked how often during the past week they felt depressed, lonely, sad, and happy, as well as how often they enjoyed life, had a restless sleep, felt everything was an effort, and could not get going. Scores were summed and the values for the final variable ranged from 0 to 8, with higher scores indicating more depressive symptoms. Analyses also include an indicator for the baseline parent-child assessment wave, representing the year the respondent first completed the psychosocial questionnaire interview: 0 = 2006 and 1 = 2008.

Analytical strategy

The study used parallel latent growth curve models (LGCMs) with maximum likelihood estimation to estimate reciprocal relationships between parent-child relationship quality and cognition (see more about the approach in Umberson et al., 2006). Growth curve models analyze changes over time in consideration of variations between and within individuals (Singer et al., 2003). Because there is between-individual variation in initial levels of parent-child relationship quality and cognition as well as subsequent changes in these variables over time, the models estimated the initial level (latent intercept) and the subsequent rate of change (latent slope) for both parent-child relationship quality scores and cognitive scores over the study period. Within construct relationships (cognition intercepts predicting cognition slopes; relationship quality intercepts predicting relationship quality slopes) were included in the models. Moreover, to take advantage of the longitudinal data in modeling the bidirectional relationships, the study used LGCMs to estimate how the intercept of parent-child relationship quality predicts the slope (i.e., subsequent change) of parents’ cognitive trajectories and how the intercept of parents’ cognition predicts the slope of parent-child relationship quality trajectories. The linear equations for the parallel LGCM can be specified as:

C_{it} = π_{0 i} + π_{1 i} T_{it} + ε_{it}

π_{0 i} = β_{1} + X ’ A_{1} + δ_{1 i}

π_{1 i} = β_{2} + X ’ A_{2} + λ_{1} θ_{0 i} + λ_{2} π_{0 i} + δ_{2 i}

Q_{ij} = θ_{0 i} + θ_{1 i} T_{ij} + ν_{ij}

θ_{0 i} = β_{3} + X ’ A_{3} + δ_{3 i}

θ_{1 i} = β_{4} + X ’ A_{4} + λ_{3} π_{0 i} + λ_{4} θ_{0 i} + δ_{4 i}

where C_it denotes the ith individual’s cognitive function at time t. π_0i and π_1i are the latent intercept and latent slope of cognitive trajectory for the ith individual across waves. Q_ij is the ith individual’s parent-child relationship quality score at time j. θ_0i and θ_1i are the latent intercept and slope of parent-child relationship quality trajectory for the ith individual across waves. T_it and T_ij denote analysis time (number of years since baseline) for the trajectories of cognition and parent-child relationship quality. Note that in the cognitive trajectory, t represents five waves while in the parent-child relationship quality trajectory, j represents three waves. X is the vector of covariates, and A ₁ -A ₄ are vectors of the corresponding coefficients. ε_it, ν_ij, and δ_1i-δ_4i represent residual terms. The parameters λ₁ and λ₃ are the focus of the interpretation because they describe the bidirectional relationships between parent-child relationship quality and parents’ cognition. The parameters λ₂ and λ₄ are within construct relationships (λ₂ = the effect of cognition intercepts on cognition slopes; λ₄ = the effect of relationship quality intercepts on relationship quality slopes). Figure 2 depicts the structural path diagram of the parallel LGCMs. The model was estimated separately for each dimension of relationship quality−−Model A: frequency of contact, Model B: relationship support, and Model C: relationship strain. Gender interaction terms were added in the models to determine whether the associations between parent-child relationships and cognition vary by gender. Model-fit was evaluated using AIC and BIC. All models were estimated using Mplus 8.3 (Muthén & Muthén, 1998-2015). To ease the comparison of effect sizes, key continuous variables were standardized in the analysis, including cognition, relationship quality, education, chronic conditions, and depression. For cognition, we standardized the scores for subsequent waves using the mean and standard deviation from the 2006 cognition (baseline). Similarly, for relationship quality variables, we used the 2006/08 baseline to standardize the scores for other waves. Because the HRS dataset includes couples living in the same household, making it likely that both parents were interviewed and provided assessments of their relationship with the same child. Therefore, we reported the clustered standard errors in the analysis of the total sample to adjust for the clustered data structure within household.

Figure 2.

Structural path diagram of the parallel latent growth curve models (LGCMs).

Results

Table 1 displays descriptive statistics for all analytical variables for the full sample and by gender. In general, respondents’ cognitive function declined over the study period. At each time point, women exhibited significantly better cognitive function than men. For both men and women, the frequency of contact with children declined over time, while relationship support from children increased slightly, and relationship strain decreased slightly. Compared to men, women contacted their children more frequently and reported both greater relationship support and greater relationship strain.

Table 1.

Descriptive statistics of focal variables, HRS.

Variables	Mean (SD) or %
Variables	Total (N = 13,169)	Men (n = 5,369)	Women (n = 7,800)
Cognitive function 2006	15.38 (4.41)	15.05 (4.18)	15.59 (4.55)	*
Cognitive function 2008	15.16 (4.46)	14.76 (4.25)	15.42 (4.58)	*
Cognitive function 2010	14.79 (4.49)	14.46 (4.32)	15.01 (4.59)	*
Cognitive function 2012	14.57 (4.52)	14.22 (4.30)	14.79 (4.65)	*
Cognitive function 2014	14.61 (4.66)	14.37 (4.44)	14.75 (4.78)	*
Cognitive function 2016	14.42 (4.65)	14.24 (4.47)	14.53 (4.75)	*
Contact frequency 2006/08	4.59 (1.09)	4.39 (1.16)	4.73 (1.02)	*
Contact frequency 2010/12	4.54 (1.11)	4.33 (1.16)	4.69 (1.04)	*
Contact frequency 2014/16	4.44 (1.12)	4.23 (1.16)	4.58 (1.07)	*
Relationship support 2006/08	3.28 (0.71)	3.17 (0.75)	3.36 (0.67)	*
Relationship support 2010/12	3.29 (0.70)	3.17 (0.75)	3.37 (0.65)	*
Relationship support 2014/16	3.31 (0.70)	3.21 (0.73)	3.38 (0.67)	*
Relationship strain 2006/08	1.70 (0.63)	1.67 (0.62)	1.71 (0.64)	*
Relationship strain 2010/12	1.65 (0.62)	1.63 (0.61)	1.67 (0.63)	*
Relationship strain 2014/16	1.64 (0.63)	1.62 (0.62)	1.65 (0.64)
Age (years)	68.00 (9.74)	68.22 (9.32)	67.84 (10.02)	*
Parent-child gender composition
All sons	15.54	15.66	15.46
All daughters	14.59	14.32	14.77
Both sons and daughters	69.87	70.01	69.77
Race (%)
Non-Hispanic White	76.64	78.77	75.18	*
Non-Hispanic Black	12.6	10.77	13.86	*
Hispanic	6.28	6	6.47
Other	4.48	4.47	4.49
Education (0–17)	12.50 (3.13)	12.71 (3.34)	12.36 (2.97)	*
Marital status (%)
Married	69.1	84.52	58.49	*
Unmarried	30.9	15.48	41.51	*
Number of children (1–20)	3.52 (2.04)	3.50 (1.99)	3.54 (2.07)
Smoking status (%)
Not current smoker	87.1	86.4	87.58	*
Current smoker	12.9	13.6	12.42	*
Drinking status
Not current drinker	66.89	56.73	73.88	*
Current drinker	33.11	43.27	26.12	*
Chronic conditions (0–4)	1.06 (0.96)	1.13 (0.99)	1.01 (0.94)	*
Depression (0–8)	1.43 (1.91)	1.16 (1.66)	1.61 (2.04)	*
Survey starting year (%)
Since 2006	54.03	54.76	53.53
Since 2008	45.97	45.24	46.47

Note. *Statistically significant difference by gender at the p < .05 level. SD: Standard Deviation.

Table 2 shows the results of LGCMs. Models A, B, and C present estimates for contact frequency, relationship support, and relationship strain, respectively. Each model included (1) initial levels of relationship quality (latent intercept, θ_0i) predicting the rate of change in both relationship quality (latent slope, θ_1i) and cognition (latent slope, π_1i); (2) initial levels of cognition (latent intercept, π_0i) predicting the rate of change in both relationship quality (latent slope, θ_1i) and cognition (latent slope, π_1i); (3) covariances between latent slopes of relationship quality and cognition. In Model A, the results suggest that after adjusting for the effects of all covariates, initial level of contact frequency with children was not associated with the rate of change in parents’ cognition (λ₁ = .001, p > .05) while it predicts a faster decrease or a slower increase in the rate of change of contact frequency (λ₄ = −0.031, p < .001). Initial level of cognition predicts a faster increase in cognition (λ₂ = .016, p < .001) and a faster increase or a slower decrease in contact frequency with children (λ₃ = .013, p < .001). Specifically, λ₃ = .013 indicates that, with one standard deviation increase in initial cognition, the slope of contact frequency is expected to increase by 0.013 standard deviation. This effect is much larger than those of standardized education (−0.003), depression (−0.001), and chronic conditions (0.000). Similarly, Model B suggests that initial level of relationship support from children was not associated with the rate of change in parents’ cognition (λ₁ = .000, p > .05) while it predicts a faster decrease or a slower increase in relationship support (λ₄ = −0.018, p < .001). Initial level of cognition predicts a faster increase in cognition (λ₂ = .016, p < .001) and a faster increase or a slower decrease in relationship support from children (λ₃ = .007, p < .05). Specifically, λ₃ = .007 indicates that, with one standard deviation increase in initial cognition, the slope of relationship support is expected to increase by 0.007 standard deviation. This effect is much larger than those of standardized education (−0.001), depression (0.001), and chronic conditions (0.000). Model C suggests that initial level of relationship strain from children predicts a faster decline in parents’ cognition (λ₁ = −0.004, p < .05) and it predicts a faster decrease or a slower increase in relationship strain (λ₄ = −0.024, p < .001). Here, λ₁ = −0.004 indicates that with one standard deviation increase in initial relationship strain, the slope of cognition is expected to decrease by 0.004 standard deviation. This effect size of the negative impact of relationship strain is comparable to the standardized effects of education (−0.005) and depression (0.002). Initial level of cognition was positively associated with its rate of change (λ₂ = .016, p < .001) while it was not associated with the rate of change in relationship strain (λ₃ = −0.005, p > .05). There is no covariance between latent slopes of relationship quality and cognition [Cov (θ_1i, π_1i) = 0] across models. The model-fit indices (shown at the bottom of each panel) suggest the models fit the data very well (i.e., CFI > .95, TLI > .95, RMSEA < .05) (Hu & Bentler, 1999).

Table 2.

Estimated effects of relationship quality (RQ) and cognition from Parallel Growth Curve Models, HRS 2006–2016, Total sample (N = 13,169).

	A. Contact frequency				B. Relationship support				C. Relationship strain
	Cognition slope (π_1i)		RQ slope (θ_1i)		Cognition slope (π_1i)		RQ slope (θ_1i)		Cognition slope (π_1i)		RQ slope (θ_1i)
	Estimate	S.E.	Estimate	S.E.	Estimate	S.E.	Estimate	S.E.	Estimate	S.E.	Estimate	S.E.
RQ intercept (θ_0i)	0.001	0.001	−0.031***	0.004	0.000	0.002	−0.018***	0.005	−0.004*	0.002	−0.024***	0.005
Cognition intercept (π_0i)	0.016***	0.003	0.013***	0.003	0.016***	0.003	0.007*	0.003	0.016***	0.003	−0.005	0.003
Cov (θ_1i, π_1i)	0.000		0.000		0.000		0.000		0.000		0.000
Female (parents)	−0.013***	0.002	0.011**	0.003	−0.013***	0.002	0.002	0.003	−0.013***	0.002	−0.004	0.003
All daughters	−0.002	0.004	0.002	0.005	−0.002	0.004	−0.006	0.004	−0.002	0.004	0.006	0.005
Both daughters and sons	−0.005	0.003	−0.003	0.004	−0.005	0.003	−0.006	0.004	−0.005	0.003	0.006	0.004
Age (years, centered)	−0.003***	0.000	0.001***	0.000	−0.003***	0.000	0.001**	0.000	−0.003***	0.000	0.000	0.000
Non-Hispanic Black	0.008	0.004	0.011*	0.005	0.008	0.004	0.009*	0.004	0.008*	0.004	−0.008	0.005
Hispanic	0.012**	0.004	0.001	0.006	0.012**	0.005	0.000	0.006	0.012*	0.005	−0.012	0.006
Other	0.003	0.006	0.007	0.007	0.003	0.006	0.012	0.006	0.003	0.006	0.003	0.007
Year of schooling (std)	−0.005**	0.002	−0.003	0.002	−0.005**	0.002	−0.001	0.002	−0.005**	0.002	0.002	0.002
Unmarried (ref: married)	−0.005*	0.002	−0.012***	0.003	−0.005*	0.003	−0.004	0.003	−0.005*	0.003	0.009**	0.003
Number of children (1–20)	0.000	0.001	0.002*	0.001	0.000	0.001	0.000	0.001	0.000	0.001	−0.001	0.001
Current smoker	−0.007	0.003	−0.005	0.004	−0.007	0.003	0.000	0.004	−0.007	0.003	0.008	0.004
Current drinker	0.001	0.002	0.004	0.003	0.001	0.002	0.006*	0.003	0.001	0.002	−0.003	0.003
Chronic conditions (std)	−0.001	0.001	0.000	0.001	−0.001	0.001	0.000	0.001	0.000	0.001	0.000	0.002
Depression (std)	0.002	0.001	−0.001	0.001	0.002	0.001	0.001	0.002	0.002	0.001	0.000	0.002
Survey year (since 2008)	−0.001	0.002	0.003	0.003	−0.001	0.002	−0.002	0.002	−0.001	0.002	0.013***	0.003
Means of growth parameters	−0.043***	0.004	−0.029***	0.005	−0.043***	0.004	0.009*	0.004	−0.043***	0.004	−0.020***	0.005
Variances in growth parameters	0.001**	0.000	0.006***	0.001	0.001**	0.000	0.003***	0.001	0.001**	0.000	0.005***	0.001
Model fit index	CFI = 0.991, TLI = 0.984, RMSEA = 0.018				CFI = 0.992, TLI = 0.986, RMSEA = 0.018				CFI = 0.991, TLI = 0.984, RMSEA = 0.018

Note. Some variables are standardized (cognition, relationship quality, education, depression, and chronic diseases). Clustered standard errors in parentheses. ***p < .001, **p < .01, *p < .05.

Table 3 shows the results of LGCMs adding gender interaction with latent intercepts of relationship quality. We found no significant gender differences between men and women.

Table 3.

Estimated effects of relationship quality (RQ) and cognition from Parallel Growth Curve Models, adding gender interaction, HRS 2006–2016, Total sample (N = 13,169).

	A. Contact frequency				B. Relationship support				C. Relationship strain
	Cognition slope (π_1i)		RQ slope (θ_1i)		Cognition slope (π_1i)		RQ slope (θ_1i)		Cognition slope (π_1i)		RQ slope (θ_1i)
	Estimate	S.E.	Estimate	S.E.	Estimate	S.E.	Estimate	S.E.	Estimate	S.E.	Estimate	S.E.
RQ intercept (q_0i)	0.001	0.002	−0.031***	0.004	0.003	0.002	−0.018***	0.005	−0.002	0.002	−0.024***	0.005
Cognition intercept (π_0i)	0.016***	0.003	0.017***	0.004	0.016***	0.003	0.010**	0.004	0.016***	0.003	−0.009*	0.004
RQ intercept X Female	0.000	0.003			−0.005	0.003			−0.003	0.003
Cog intercept X Female			−0.006	0.004		−0.005	0.004			0.007	0.004
Cov (θ_1i, π_1i)	0.000	0.000	0.000	0.000	0.000	0.000
Female (parents)	−0.013***	0.002	0.011**	0.003	−0.013***	0.002	0.003	0.003	−0.013***	0.002	−0.004	0.003
All daughters	−0.002	0.004	0.002	0.005	−0.002	0.004	−0.006	0.004	−0.002	0.004	0.006	0.005
Both daughters and sons	−0.005	0.003	−0.003	0.004	−0.005	0.003	−0.006	0.004	−0.005	0.003	0.006	0.004
Age (years, centered)	−0.003***	0.000	0.001***	0.000	−0.003***	0.000	0.001**	0.000	−0.003***	0.000	0.000	0.000
Non-Hispanic Black	0.008	0.004	0.011*	0.005	0.008	0.004	0.009*	0.004	0.008*	0.004	−0.008	0.005
Hispanic	0.012*	0.005	0.000	0.007	0.012*	0.005	0.000	0.006	0.012*	0.004	−0.011	0.006
Other	0.003	0.005	0.007	0.008	0.003	0.005	0.012	0.006	0.003	0.005	0.003	0.007
Year of schooling (std)	−0.005**	0.002	−0.003	0	−0.005**	0.002	−0.002	0.002	−0.005*	0.002	0.003	0.002
Unmarried (ref: married)	−0.005*	0.003	−0.012***	0.003	−0.005*	0.003	−0.005	0.003	−0.005*	0.003	0.010**	0.003
Number of children (1–20)	0.000	0.001	0.002*	0.001	0.000	0.001	0.000	0.001	0.000	0.001	−0.001	0.001
Current smoker	−0.007	0.003	−0.005	0.005	−0.007	0.003	0.000	0.004	−0.007	0.003	0.007	0.004
Current drinker	0.001	0.002	0.004	0.003	0.001	0.003	0.006*	0.003	0.001	0.002	−0.003	0.003
Chronic conditions (std)	−0.001	0.001	0.000	0.002	−0.001	0.001	0.000	0.001	0.000	0.001	0.000	0.002
Depression (std)	0.002	0.001	−0.001	0.002	0.001	0.001	0.001	0.002	0.002	0.001	0.000	0.002
Survey year (since 2008)	−0.001	0.002	0.003	0.003	−0.001	0.002	−0.002	0.002	−0.001	0.002	0.013***	0.003
Means of growth parameters	−0.043***	0.004	−0.029***	0.005	−0.042***	0.004	0.009*	0.004	−0.043***	0.004	−0.020***	0.005
Variances in growth parameters	0.001**	0.000	0.006***	0.000	0.001***	0.000	0.003***	0.000	0.001**	0.000	0.005***	0.001
Model fit index	AIC = 180485, BIC = 181107				AIC = 179313, BIC = 179935				AIC = 179816, BIC = 180437

Note. Some variables are standardized (cognition, relationship quality, education, depression, and chronic diseases). Clustered standard errors in parentheses. ***p < .001, **p < .01, *p < .05.

Discussion

The “linked lives” of parents and their children significantly shape parents’ life contexts and thus influence parents’ health outcomes, including cognitive health (Carr & Utz, 2020; Thomas & Umberson, 2018; Zahodne et al., 2019). However, few studies have considered the potentially bidirectional links between parent-child relationship quality and parents’ cognitive function. Using longitudinal data from a nationally representative sample of older adults in the United States, this study estimated parallel growth curve models to examine the bidirectional associations between parent-child relationships and older parents’ cognitive trajectories, with a focus on differences by parents’ gender. The findings advance the limited empirical research on the link between intergenerational relationship quality and cognitive health in important ways.

First, the findings suggest that strain in relationships with children predicted faster cognitive decline for older parents, but neither contact frequency with children nor receiving support from children was associated with changes in older parents’ cognition over time. Importantly, the negative impact of relationship strain on parents’ cognition is notably comparable to the adverse effects of some well-documented traditional risk factors such as age, education, and depression. These results echo the results of several prior studies that found relationship strain had a greater impact on health than relationship support (e.g., Liu & Waite, 2014; Thomas & Umberson, 2018). Further, these findings align with a growing literature on the connection between stress and the risk of cognitive impairment and dementia (Li et al., 2021; Yin et al., 2020; Zahodne et al., 2019). Having a strained relationship with children can negatively influence older parents’ cognitive ability by stimulating stress hormones (e.g., catecholamines, cortisol), impairing the immune system, and evoking a physiological response, all of which may diminish parents’ brain reserve and cognitive ability (Kuhlmann et al., 2005; Oei et al., 2007). Stress caused by relationship conflict is also associated with hypertension and other vascular factors that can increase the risk of cognitive impairment, Alzheimer’s disease, and related dementia (Johansson et al., 2010; Sparrenberger et al., 2009). Moreover, negative interactions with children may significantly increase older parents’ vulnerability to cognitive decline because aging parents often depend on their adult children to act as caregivers, providing support and assistance. Further, relationship conflict with a child is often accompanied by a loss of contact with the child and even the child’s family (e.g., grandchildren), which may lead to feelings of loneliness, isolation, and psychological distress for aging parents, and in turn, speed cognitive decline (Shankar et al., 2013).

As noted, neither contact frequency nor relationship support from children had a statistically significant impact on the subsequent cognitive trajectories of parents. This result aligns with recent research showing no significant association between quality of social relationships and subsequent rates of cognitive function (e.g., Norton et al., 2009; Sharifian et al., 2020; Thomas & Umberson, 2018). For example, in a longitudinal analysis, Sharifian et al. (2020) showed that more frequent contact with family was not a statistically significant predictor of better cognitive function for either men or women in the United States. Norton et al. (2009) examined how caregiver–care recipient closeness influences care recipients’ Alzheimer’s disease progression and found that closeness with adult caregivers have smaller effects of slowing down the Alzheimer’s disease progression than closeness with spouse caregivers.

Second, the results move beyond previous literature by examining the bidirectional nature of the links between parent-child relationship quality and parents’ cognition. The results suggest that while neither contact frequency nor relationship support from children predict parents’ cognitive trajectories, baseline cognitive ability does predict the trajectories of both contact frequency and support from children—for parents who have good cognitive function initially, they tend to experience a more rapid increase in contact frequency with children and relationship support from children over time, however, for those with lower cognitive function, they may experience a decline in contact frequency or support from children. Indeed, previous studies suggest that cognitive limitations may damage networking ability and reduce social support (Alzheimer’s Association, 2020), which may result in major challenges for older parents as they seek to maintain relationships with their children. Parents’ poor cognition may also affect their assessment of receiving support and contact frequency with children because parents with a higher need for care may expect more from their children. This finding addresses an important knowledge gap in that previous studies of the social determinants of cognitive health have largely neglected the possibility of reverse causality or selection effects. Thus, the current results highlight the need for caution when interpreting causality based on previous evidence.

Finally, given the long-documented gendered nature of intergenerational relationships, we had expected that parent-child relationship quality might work differently for mothers and fathers in relation to cognitive trajectories. However, our results suggest no gender difference in our key findings. One possibility is that parent-child relationship quality is equally significant for both mothers and fathers in later life due to reduced social networks outside the family. It is also plausible that cognition hold equal importance for both older fathers and mothers in maintaining their relationship with their children. Given the importance of gender in family relationships, future studies should continue to investigate this topic.

This study has some notable limitations. First, the analysis was constrained by HRS data collection procedures regarding parent-child relationship items—HRS has collected information on relationship quality in three of its waves over a period of 8 years (2006/2010/1014 or 2008/2012/2016), which is a relatively short period for capturing changes in relationship quality. Future studies should use follow-up observations over a longer period to better understand the longitudinal nature of the associations. Second, given the small number of non-biological children in the sample, the study did not distinguish types of children. However, both parent-child relationship quality and its impact on parents’ cognitive health likely vary by the type of child (Carr & Utz, 2020). Third, parents’ marital status and parental status may change over time and interact with their gender, potentially leading to various effects on relationship quality and its link to cognitive health. In our sample, men were more likely than women to remain married and older mothers are slightly more likely than fathers to have stepchildren (results not shown but available). The interplay between gender, marital status, and parental status may contribute to explaining the impact of family relationship on cognition, highlighting the need for further research using larger sample sizes and more robust measures. Fourth, while the cognitive measure employed in this study has been widely used in previous literature (Langa et al., 2023; Ryan et al., 2022), it was heavily loaded in the domain of memory, and may not comprehensively assess other domains of intellectual or cognitive status (e.g., across areas such as reasoning, knowledge, etc.). As more cognitive data is being collected in HRS, future studies should incorporate more diverse measures of cognition to provide a comprehensive understanding of cognitive function among older adults. Moreover, the measure of contact frequency did not capture interaction via newer communication technologies, such as text messages and social media. Future research should consider these modes of communication, especially in analyses of younger cohorts. Lastly, the effect sizes for some key findings are relatively small although significant. Readers should interpret our results with caution, considering the small effect sizes within such a large sample. We encourage future studies to test the robustness of these findings and explore potential gender differences using other datasets.

Conclusion

Prior literature has suggested that the quality of relationships between parents and their children is associated with parental well-being (Fingerman et al., 2012; Koropeckyj-Cox, 2002; Reczek & Zhang, 2016), however, studies on intergenerational relationships and older adults’ cognitive health are limited and have often neglected the potential for a bidirectional link between these two factors. This study extends the literature by demonstrating a bidirectional link between intergenerational relationships and parents’ cognition: relationship strain with children may hasten cognitive decline among older parents, while parents’ poor cognition may decrease relationship support and contact with children. These findings contribute to the scholarly understanding of the “linked lives” of aging parents and their children, the impact of this link on parents’ cognitive health, and the gendered nature of these processes. The findings also have implications for social and public policies, highlighting the importance of developing interventions and public programs that reduce relationship strain with children to protect the cognitive functioning of older parents. Future studies should use more detailed measures of parent-child relationship quality and explore the specific mechanisms linking intergenerational relationships and cognitive health.

Footnotes

Author’s Note

This paper was presented at The Gerontological Society of America (2022), Indianapolis, IN, and American Sociological Association (2021) (virtual).

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Institute on Aging (R01AG061118).

Open science statement

As part of IARR’s encouragement of open research practices, the author(s) have provided the following information: This research was not pre-registered. The data used in the research can be publicly posted. The data can be obtained at here .

ORCID iD

Yan Zhang

Note

References

Alzheimer’s Association . (2020). 2020 Alzheimer’s disease facts and figures. Alzheimer’s and Dementia, 14(3), 367–429. https://doi.org/10.1002/alz.1206

Bherer

Erickson

K. I.

Liu-Ambrose

(2013). A review of the effects of physical activity and exercise on cognitive and brain functions in older adults. Journal of Aging Research, 2013(9), Article 657508. https://doi.org/10.1155/2013/657508

Brodaty

Donkin

(2022). Family caregivers of people with dementia. Dialogues in Clinical Neuroscience, 11(2), 217–228. https://doi.org/10.31887/DCNS.2009.11.2/hbrodaty

Carr

Utz

R. L.

(2020). Families in later life: A decade in review. Journal of Marriage and Family, 82(1), 346–363. https://doi.org/10.1111/jomf.12609

Chirico

Ottoboni

Valente

Chattat

(2021). Children and young people’s experience of parental dementia: A systematic review. International Journal of Geriatric Psychiatry, 36(7), 975–992. https://doi.org/10.1002/gps.5542

Costa-Cordella

Arevalo-Romero

Parada

F. J.

Rossi

(2021). Social support and cognition: A systematic review. Frontiers in Psychology, 12, Article 637060. https://doi.org/10.3389/fpsyg.2021.637060

Crimmins

E. M.

Saito

Kim

J. K.

(2016). Change in cognitively healthy and cognitively impaired life expectancy in the United States: 2000–2010. SSM-Population Health, 2, 793–797. https://doi.org/10.1016/j.ssmph.2016.10.007

Elder

G. H.

Jr. (1998). The life course as developmental theory. Child Development, 69(1), 1–12. https://doi.org/10.1111/j.1467-8624.1998.tb06128.x

Fingerman

K. L.

Cheng

Y. P.

Birditt

Zarit

(2012). Only as happy as the least happy child: Multiple grown children’s problems and successes and middle-aged parents’ well-being. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 67(2), 184–193. https://doi.org/10.1093/geronb/gbr086

10.

Fingerman

K. L.

Huo

Birditt

K. S.

(2020). A decade of research on intergenerational ties: Technological, economic, political, and demographic changes. Journal of Marriage and Family, 82(1), 383–403. https://doi.org/10.1111/jomf.12604

11.

Bailey

D. E.

Jr. Dong

(2017). Social support, social strain, and cognitive function among community-dwelling US Chinese older adults. Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences, 72(suppl_1), S16–S21. https://doi.org/10.1093/gerona/glw221

12.

Giles

L. C.

Anstey

K. J.

Walker

R. B.

Luszcz

M. A.

(2012). Social networks and memory over 15 years of followup in a cohort of older Australians: Results from the Australian longitudinal study of ageing. Journal of Aging Research, 2012(1), Article 856048. https://doi.org/10.1155/2012/856048

13.

Hayes

Demirkol

Ridley

Withall

Draper

(2016). Alcohol-related cognitive impairment: Current trends and future perspectives. Neurodegenerative Disease Management, 6(6), 509–523. https://doi.org/10.2217/nmt-2016-0030

14.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1–55. https://doi.org/10.1080/10705519909540118

15.

Johansson

Guo

Waern

Östling

Gustafson

Bengtsson

Skoog

(2010). Midlife psychological stress and risk of dementia: A 35-year longitudinal population study. Brain, 133(Pt 8), 2217–2224. https://doi.org/10.1093/brain/awq116

16.

Kornrich

Furstenberg

(2013). Investing in children: Changes in parental spending on children, 1972–2007. Demography, 50(1), 1–23. https://doi.org/10.1007/s13524-012-0146-4

17.

Koropeckyj-Cox

(2002). Beyond parental status: Psychological well-being in middle and old age. Journal of Marriage and Family, 64(4), 957–971. https://doi.org/10.1111/j.1741-3737.2002.00957.x

18.

Kuhlmann

Piel

Wolf

O. T.

(2005). Impaired memory retrieval after psychosocial stress in healthy young men. Journal of Neuroscience, 25(11), 2977–2982. https://doi.org/10.1523/jneurosci.5139-04.2005

19.

Langa

K. M.

Weir

D. M.

Kabeto

Sonnega

(2023). Langa-weir classification of cognitive function (1995–2020). Survey Research Center Institute for Social Research, University of Michigan. https://hrsdata.isr.umich.edu/sites/default/files/documentation/data-descriptions/1680034270/Data_Description_Langa_Weir_Classifications2020.pdf

20.

Lendon

J. P.

Silverstein

Giarrusso

(2014). Ambivalence in older parent–adult child relationships: Mixed feelings, mixed measures. Journal of Marriage and Family, 76(2), 272–284. https://doi.org/10.1111/jomf.12101

21.

Wang

Lian

Zhu

Liu

(2019). Social networks, community engagement, and cognitive impairment among community-dwelling Chinese older adults. Dementia and Geriatric Cognitive Disorders Extra, 9(3), 330–337. https://doi.org/10.1159/000502090

22.

Guo

Stensland

Dong

(2021). Family relationships and cognitive function among community-dwelling US Chinese older adults. Research on Aging, 43(1), 37–46. https://doi.org/10.1177/0164027520939250

23.

Liu

Waite

(2014). Bad marriage, broken heart? Age and gender differences in the link between marital quality and cardiovascular risks among older adults. Journal of Health and Social Behavior, 55(4), 403–423. https://doi.org/10.1177/0022146514556893

24.

Milkie

M. A.

Bierman

Schieman

(2008). How adult children influence older parents’ mental health: Integrating stress-process and life-course perspectives. Social Psychology Quarterly, 71(1), 86–105. https://doi.org/10.1177/019027250807100109

25.

Muthén

L. K.

Muthén

(1998-2015). Mplus user’s guide (8th ed.) Muthén & Muthén.

26.

Nomaguchi

K. M.

(2012). Parenthood and psychological well-being: Clarifying the role of child age and parent–child relationship quality. Social Science Research, 41(2), 489–498. https://doi.org/10.1016/j.ssresearch.2011.08.001

27.

Norton

M. C.

Piercy

K. W.

Rabins

P. V.

Green

R. C.

Breitner

J. C.

Ostbye

Corcoran

Welsh-Bohmer

K. A.

Lyketsos

C. G.

Tschanz

J. T.

(2009). Caregiver–recipient closeness and symptom progression in alzheimer disease. The Cache county dementia progression study. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 64(5), 560–568. https://doi.org/10.1093/geronb/gbp052

28.

Oei

N. Y.

Elzinga

B. M.

Wolf

O. T.

de Ruiter

M. B.

Damoiseaux

J. S.

Kuijer

J. P.

Veltman

D. J.

Scheltens

Rombouts

S. A. R. B.

(2007). Glucocorticoids decrease hippocampal and prefrontal activation during declarative memory retrieval in young men. Brain Imaging and Behavior, 1(1–2), 31–41. https://doi.org/10.1007/s11682-007-9003-2

29.

Okabayashi

Liang

Krause

Akiyama

Sugisawa

(2004). Mental health among older adults in Japan: Do sources of social support and negative interaction make a difference? Social Science & Medicine, 59(11), 2259–2270. https://doi.org/10.1016/j.socscimed.2004.02.024

30.

Pillemer

Suitor

J. J.

(2002). Explaining mothers’ ambivalence toward their adult children. Journal of Marriage and Family, 64(3), 602–613. https://doi.org/10.1111/j.1741-3737.2002.00602.x

31.

Reczek

Zhang

(2016). Parent–child relationships and parent psychological distress: How do social support, strain, dissatisfaction, and equity matter? Research on Aging, 38(7), 742–766. https://doi.org/10.1177/0164027515602315

32.

Ryan

J. M.

Fisher

G. G.

Hassan

Faul

J. D.

Rodgers

W. L.

Weir

D. R.

(2022). Health and retirement study imputation of cognitive functioning measures: 1992 – 2018 (version 7.0) data description. Survey Research Center Institute for Social Research, University of Michigan. https://hrsdata.isr.umich.edu/sites/default/files/documentation/data-descriptions/1651088507/COGIMP9218_dd.pdf

33.

Shankar

Hamer

McMunn

Steptoe

(2013). Social isolation and loneliness: Relationships with cognitive function during 4 years of follow-up in the English longitudinal study of ageing. Psychosomatic Medicine, 75(2), 161–170. https://doi.org/10.1097/psy.0b013e31827f09cd

34.

Sharifian

Kraal

A. Z.

Zaheed

A. B.

Sol

Zahodne

L. B.

(2020). Longitudinal associations between contact frequency with friends and with family, activity engagement, and cognitive functioning. Journal of the International Neuropsychological Society, 26(8), 815–824. https://doi.org/10.1017/S1355617720000259

35.

Silverstein

Bengtson

V. L.

(1997). Intergenerational solidarity and the structure of adult child-parent relationships in American families. American Journal of Sociology, 103(2), 429–460. https://doi.org/10.1086/231213

36.

Silverstein

Gans

Yang

F. M.

(2006). Intergenerational support to aging parents: The role of norms and needs. Journal of Family Issues, 27(8), 1068–1084. https://doi.org/10.1177/0192513X06288120

37.

Singer

J. D.

Willett

J. B.

Willett

J. B.

(2003). Applied longitudinal data analysis: Modeling change and event occurrence. Oxford University Press.

38.

Smith

Ryan

Sonnega

Weir

(2017). Psychosocial and lifestyle questionnaire 2006–2016. Survey Research Center, Institute for Social Research.

39.

Sonnega

Weir

(2014). The health and retirement study: A public data resource for research on aging. Open Health Data, 2(1), Article e7. https://doi.org/10.5334/ohd.am

40.

Sparrenberger

Cichelero

F. T.

Ascoli

A. M.

Fonseca

F. P.

Weiss

Berwanger

Fuchs

S. C.

Moreira

L. B.

Fuchs

F. D.

(2009). Does psychosocial stress cause hypertension? A systematic review of observational studies. Journal of Human Hypertension, 23(1), 12–19. https://doi.org/10.1038/jhh.2008.74

41.

Swan

G. E.

Lessov-Schlaggar

C. N.

(2007). The effects of tobacco smoke and nicotine on cognition and the brain. Neuropsychology Review, 17(3), 259–273. https://doi.org/10.1007/s11065-007-9035-9

42.

Swartz

(2009). Intergenerational family relations in adulthood: Patterns, variations, and implications in the contemporary United States. Annual Review of Sociology, 35(1), 191–212. https://doi.org/10.1146/annurev.soc.34.040507.134615

43.

Thoits

P. A.

(2010). Stress and health: Major findings and policy implications. Journal of Health and Social Behavior, 51(Suppl), S41–S53. https://doi.org/10.1177/0022146510383499

44.

Thomas

P. A.

Umberson

(2018). Do older parents’ relationships with their adult children affect cognitive limitations, and does this differ for mothers and fathers? The Journals of Gerontology: Serie Bibliographique, 73(6), 1133–1142. https://doi.org/10.1093/geronb/gbx009

45.

Umberson

(1987). Family status and health behaviors: Social control as a dimension of social integration. Journal of Health and Social Behavior, 28(3), 306–319. https://doi.org/10.2307/2136848

46.

Umberson

Pudrovska

Reczek

(2010). Parenthood, childlessness, and well-being: A life course perspective. Journal of Marriage and Family, 72(3), 612–629. https://doi.org/10.1111/j.1741-3737.2010.00721.x

47.

Umberson

Williams

Powers

D. A.

Liu

Needham

(2006). You make me sick: Marital quality and health over the life course. Journal of Health and Social Behavior, 47(1), 1–16. https://doi.org/10.1177/002214650604700101

48.

Ward

R. A.

Spitze

Deane

(2009). The more the merrier? Multiple parent-adult child relations. Journal of Marriage and Family, 71(1), 161–173. https://doi.org/10.1111/j.1741-3737.2008.00587.x

49.

Yates

J. A.

Clare

Woods

R. T.

(2013). Mild cognitive impairment and mood: A systematic review. Reviews in Clinical Gerontology, 23(4), 317–356. https://doi.org/10.1017/s0959259813000129

50.

Yin

Yang

Xiong

Zhu

(2020). Social support and the incidence of cognitive impairment among older adults in China: Findings from the Chinese longitudinal healthy longevity survey study. Frontiers in Psychiatry, 11, 254. https://doi.org/10.3389/fpsyt.2020.00254

51.

Zahodne

L. B.

Ajrouch

K. J.

Sharifian

Antonucci

T. C.

(2019). Social relations and age-related change in memory. Psychology and Aging, 34(6), 751–765. https://doi.org/10.1037/pag0000369