Social Relationships and Mental Health Disparities by Race/Ethnicity in Late-middle-age and Older Adulthood

Abstract

Social strain and lack of social support are associated with depression, but it remains unclear whether racial/ethnic differences in social relationships shape racial/ethnic disparities in late-middle age and late-life mental health. Using data from Health and Retirement Study 2006–2016 (N = 5,139), this study estimated negative binomial regressions to examine whether differential exposure to and differential effect of social support and strain with spouses, children, family members, and friends explain Black-White and Latinx-White disparities in depressive symptoms. The findings indicate that minority older adults, especially Latinx, have more depressive symptoms than Whites. Although social support compensates for some mental health gaps, mediation and moderation analyses reveal that such disparities are attributable to Black and Latinx late-middle-aged and older adults’ greater exposure to social strain and diminished returns of social support from spouses and children as compared with Whites. This study offers insights regarding how mental health inequality is impacted by racial/ethnic differences in stress exposure and the protective effect of social relationships.

Keywords

depression disparities race/ethnicity social relationship older adults

Introduction

Depression in older adults is associated with declining physical and cognitive functions, regardless of race/ethnicity (González et al. 2010). In late-middle-age and late adulthood, more people start to experience transitions that shape their social networks, and become more family-focused due to declining health, retirement, divorce, and bereavement (Carr and Moorman 2011). These changes can alter supportive social ties, affecting mental well-being in late-middle-age and late life (Antonucci, Ajrouch, and Birditt 2014). With the aging population in the United States projected to see a rise in racial and ethnic minorities aged 65 and older from 29 percent in 2016 to 44 percent in 2060, understanding persistent health disparities by race/ethnicity, including mental well-being in late-middle-age and late adulthood is a public health imperative (Vespa, Medina, and Armstrong 2020). However, many comparative studies lack representation of individuals aged 50 and older, leaving gaps in knowledge regarding how Black late-middle-aged and older adults fare compared with their White peers (Breslau et al. 2006; Kiecolt, Hughes, and Keith 2008; Wang et al. 2021). In addition, research on mental health disparities between White and Latinx adults, the largest minority group in the United States, remains limited and shows inconsistent patterns (Muruthi, Zalla, and Lewis 2020). While some studies indicate comparable or lower rates of depressive symptoms and psychiatric disorders among Latinx adults relative to their White peers (Breslau et al. 2006; Yang and Park 2019), others report higher rates of depressive symptoms and psychological distress among Latinx adults—including late-middle-aged and older adults—than their White peers (O. P. Almeida et al. 2012; González et al. 2010; Jang et al. 2008; Liang et al. 2011; Sternthal, Slopen, and Williams 2011).

The stress process model (SPM) proposes that mental health disparities stem from two underlying pathways: (1) differential exposure to stressors and access to protective resources, and (2) the differential effect of such factors across social groups (Pearlin 1999; Turner and Avison 2003). Social relationships, servicing as either protective resources (i.e., social support) or stressors (i.e., social strain), directly influence mental health outcomes, such as symptoms of depression, with potential differences in their respective magnitudes based on race/ethnicity, or age group (e.g., young adults or older adults) (Lincoln, Chatters, and Taylor 2003; Shiovitz-Ezra and Leitsch 2010; Stafford et al. 2011; Teo, Choi, and Valenstein 2013). In addition, the convoy model of social relationships underscores the dynamacy in social relations as individuals age, highlighting variations in network composition, quality, and sources of support across personal characteristics and contexts, such as race/ethnicity and life stages (Katz et al. 2020). Accordingly, the distribution and salience of social support and strain in SPM may also vary by race/ethnicity and/or relationship type (e.g., spouse/partner, children, family members, and friends) at various life stages (Antonucci et al. 2014). However, a comprehensive examination of the racial/ethnic patterning of social relations within a diverse aging population, including Latinx, is lacking. Thus, it remains unclear how mental health differs by race/ethnicity, and whether racial/ethnic differences in social relationships explain these disparities (Lincoln et al. 2003; Taylor and Chatters 2020).

Using data from the Health and Retirement Study (HRS) of respondents aged 50 and above, this study addresses three questions: (1) Are depressive symptoms more or less prevalent among Black and Latinx than White older adults? (2) Can differential access to social support (i.e., from spouses, children, extended family members, or friends) and social strain (i.e., with spouses, children, extended family members, or friends) explain the racial/ethnic disparities in depressive symptoms? (3) Does the impact of social support for and/or social strain on depressive symptoms vary by race/ethnicity? By incorporating the SPM and the convoy model, I provide novel evidence on the racial/ethnic patterning of social support and social strain, and mental health disparities in late-middle-age and later life, shedding light on how these facets may be linked to racial/ethnic differences in mental well-being. The findings will enhance understanding of racial/ethnic disparities in mental health and inform future interventions to mitigate the mental health disparities that Black, Latinx late-middle-age, and older adults experience.

Background

Social Support, Social Strain, and Mental Health

The SPM provides a framework for elucidating how structural inequality and social stratification (e.g., by socioeconomic status and race) and social changes affect mental health disparities among disadvantaged groups by shaping their exposure to stressors, vulnerability, and access to protective resources over the life course, considering race/ethnicity, age, and socioeconomic status (Moen 2022; Pearlin 1999). As such, the differential exposure hypothesis ascribes Black and Latinx mental health disadvantages to their greater exposure to stressors and less access to protective resources that could buffer these effects (Assari 2018; Phelan and Link 2015). Moreover, racial/ethnic disparities in mental health may also arise from various effects of stressors and protective resources, driven by two mechanisms: (1) the “diminished return” hypothesis posits that Black and Latinx individuals derive less mental health benefit from protective resources than White individuals (Turner and Avison 2003); and (2) the “higher vulnerability” hypothesis which proposes that Black and Latinx individuals may be more susceptible than Whites to the adverse effects of social stressors (Pearlin 1999; Ulbrich, Warheit, and Zimmerman 1989).

According to the SPM, Social Support and Stressors Shape Mental Health Disparities (Pearlin 1999). Social support encompasses emotional, informative, or instrumental assistance from significant others (Thoits 1995). Emotional support (measured through positive social interactions) has consistently been shown to be related to positive health outcomes (J. Almeida et al. 2009; Burleson 2003). In this study, social support is defined as late-middle-aged and older adults’ perceived emotional support, including care, trust, and understanding from their spouses, children, family members, and friends (Antonucci et al. 2014; Shiovitz-Ezra and Leitsch 2010; Thomas 2016). Conversely, social strain refers to negative relationships involving excessive demands, conflicts, and criticism that undermine mental health (Kasper et al. 2008; Lee and Szinovacz 2016; Shiovitz-Ezra and Leitsch 2010). Social strain in this study comprises stressors characterized by negative interaction, where individuals perceive their interaction with social network members as critical, irritating, and unreliable. While related, social support and strain possess distinct qualitative features, and the absence of support does not necessarily imply the presence of strain, or vice versa (Pierce and Quiroz 2019). Therefore, research on psychological well-being suggests examining support and strain independently and simultaneously, as they represent unique dimensions coexisting within the same relationships (Chen and Feeley 2014; Shiovitz-Ezra and Leitsch 2010).

The convoy model of social relationships posits that individuals are embedded within a dynamic convoy of support, wherein network members and their influences vary over time (Antonucci et al. 2014). The convoy model enhances the SPM by recognizing varying levels of closeness among network members, thereby explaining why certain types of social support may affect health and well-being in later life more than others. Within a relatively complete network, the innermost circle typically comprises spouses and children, followed by family members and friends, while other social connections occupy more peripheral positions (Antonucci et al. 2014). Accordingly, social support and strain from inner circle members yield a greater impact on mental health than that from outer circles. Studies of late-middle-aged and older adults have consistently shown that support and strain derived from spouses/partners have the most pronounced effects on mental health (i.e., depressive symptoms and major depressive disorder) and well-being, followed by that from adult children (Chen and Feeley 2014; Stafford et al. 2011; Teo et al. 2013). However, findings regarding family members and friends have been inconsistent; while some studies suggest associations between support and strain from family members and friends that are similar to those from adult children (Lee and Szinovacz 2016), others indicate that negative interactions with family and friends, but not positive ones, are related to depression (Stafford et al. 2011).

Furthermore, the convoy model complements the SPM by emphasizing how situational and cultural contexts, such as race/ethnicity, shape the structure and quality of one’s social convoy (Fuller, Ajrouch, and Antonucci 2020). For example, unmarried or childless individuals’ friends may rely more on friends within their convoys (Fuller et al. 2020). Black Americans often have more family-centered convoys, marked by more family obligations and involvement to pool social resources (Ajrouch, Antonucci, and Janevic 2001). Similarly, Latinx supportive convoys typically comprise immediate and extended family members bound by a strong sense of family obligation and respect (Flores et al. 2020). Hence, support and strain from specific relationships may contribute to racial/ethnic mental health disparities, given their unique roles in shaping the association between depression and social relationships. However, there are research gaps in understanding the diverse effects of social support and strain across different relationship types and their links to race/ethnicity, as many studies have focused solely on specific social supports (e.g., spousal relationships).

Differential Exposure to Social Strain and Access to Social Support

The differential exposure hypothesis posits that Black and Latinx older adults may face limited access to social support, and greater exposure to social strain, compared with their White counterparts, stemming from pervasive social and structural inequalities rooted in racism. First, chronic exposure to economic hardship (such as financial strain) can lead to conflict and diminish relationship quality, resulting in reduced social support and increased strain, particularly among Black adults (Kasper et al. 2008; Lincoln 2007). Moreover, racial residential segregation exposes Black and Latinx individuals to higher levels of community-based stressors due to unsafe and undesirable living conditions. These circumstances can limit marriage opportunities and hinder the development of relationships within their communities (Williams and Mohammed 2013). Furthermore, elevated incarceration rates among Black and Latinx males contribute to financial hardship, reduce marriage opportunities, and increase social strain with spouses, partners, children, family members, and communities (Travis and Waul 2003). In addition, the immigration experience poses unique mental health challenges for older Latinx immigrants. They often encounter acculturative stress, including social and linguistic isolation, discrimination based on nativity status, and conflict arising from acculturation gaps within families (e.g., between spouses, and between generations) and outside of family contexts (Calzada et al. 2020; Muruthi et al. 2020). Limited English proficiency and lack of health insurance may also lead to discrimination during medical visits for Latinx older adults, and thereby prompt reliance on informal support networks such as family and culturally preferred coping methods which may more readily available, to address their struggles (Bolger and Prickett 2021).

Evidence regarding the racial/ethnic patterning of social support is mixed. Black adults often report lower spousal support and marital quality than their White counterparts due to the structural inequality and economic disadvantages which particularly affect Black men (Broman 2005). Some research suggests higher levels of social support for Black and Latinx individuals from children, extended family, friends, and “fictive kin,” and they tend to rely on familial support to compensate for social and economic challenges (Ajrouch et al. 2001; Landale, Oropesa, and Bradatan 2006; Mouzon 2013; Staples and Johnson 1993). However, other studies report more similarities than differences across race/ethnicity in social support, with White and Latinx adults having comparable network structures, although White adults report slightly more friends than Black adults (J. Almeida et al. 2009; Flores et al. 2020; Kiecolt et al. 2008; Mouzon 2014). Overall, research has suggested that Black and Latinx people obtain no less, or even more, social support from close kin, with the exception that Black individuals tend to have lower spousal support than their White peers (J. Almeida et al. 2009; Broman 2005; Kiecolt et al. 2008).

Research on racial/ethnic difference in negative aspects of relationship quality, such as social strain, has been limited and inconclusive (Kiecolt et al. 2008). Black individuals may face greater challenges in maintaining relationship quality, such as marital harmony, due to greater exposure to social and economic adversities (Broman 2005; Williams and Mohammed 2009). Moreover, the extensive family networks among Black and Latinx communities may lead to higher levels of social strain than their White counterparties, as the family network could propagate stress and exacerbate its effects, with other family members experiencing their own strain (Cichy, Stawski, and Almeida 2012; Wang et al. 2021). Several studies have indicated that Black adults experience more social strain with their spouses and kin (Broman 2005; Bulanda and Brown 2007; Mouzon 2013), but others have shown no Black-White differences in social strain with spouses, children, or kin (Kiecolt et al. 2008; Kim et al. 2021). Although research on Latinx population has been limited, one study found that U.S.-born Latinx had less social strain with kin than foreign-born Latinx individuals, but there was no significant difference between foreign-born Latinx and White adults (Brown, Mitchell, and Ailshire 2020). Generally speaking, Black individuals report experiencing more spousal strain than White individuals. However, Black and Latinx adults seem to experience similar levels of strain with children and kin, compared with their White peers. Yet, there is limited evidence regarding racial/ethnic differences in relationship quality with children, family members and friends in late-middle age and late adulthood.

Few studies have explored how social support and social strain mediate racial/ethnic disparities in mental health (J. Almeida et al. 2009; Yang and Park 2019). Yang and Park (2019) found that spousal support explains one-fourth of Black-White differences in depression, but not the Latinx-White differences (J. Almeida et al. 2011; Yang and Park 2019) showed that familial support, but not friend support, protects foreign-born Mexicans and Black older adults from depression. Other studies have found no mediation effects explaining the mental health gaps between Black and White adults (Kiecolt et al. 2008; Mouzon 2014). However, the inconsistent findings may stem from different conceptual strategies and measurements of social support (e.g., contact frequency, instrumental support, or perceived support) and the specific types of social support investigated (Mouzon 2013).

The Differential Effects of Social Support and Strain

Older Black and Latinx adults have larger family networks and are more family-oriented than White adults (Landale et al. 2006; Staples and Johnson 1993). Yet, it remains unclear whether some racial/ethnic groups benefit less from social support, and/or experience stronger negative effects from social strain than others (i.e., the moderation effect). According to the “diminished return hypothesis” within the SPM framework, Black and Latinx individuals may experience fewer health benefits from psychosocial resources than White individuals due to structural racism and social inequality (Assari 2018; Turner and Avison 2003). While diminished returns of social support on physical health (e.g., chronic disease and all-cause mortality) have been observed among Black individuals (Assari 2018), their impact on mental health in Latinx aging population is underexplored. Conversely, empirical evidence supporting “the stress-buffering hypothesis” has shown that social support may be more beneficial for those experiencing excessive stress (Cohen and Wills 1985). Black and Latinx adults may derive more mental health benefits from social support than White adults because they develop stronger support systems to counteract the deleterious effects of living in a racist society (Louie and Wheaton 2019). For example, research has shown that social support from spouses and kin benefits the mental health (e.g., depression, psychological distress) of Black and Latinx adults more than White adults (Kiecolt, Hughes, and Keith 2008; Lincoln, Chatters, and Taylor 2003). However, other studies find no moderation effect among Black, Latinx, and White adults (Yang and Park 2019).

Although rarely discussed, the mental health effect of social strain may also differ across racial/ethnic groups. The “higher vulnerability hypothesis” proposes that the emotional toll of social strain could be more detrimental to marginalized groups such as Black and Latinx adults than their White peers, because of their lack of other coping resources (Guo et al. 2015; Ulbrich et al. 1989). Conversely, the “higher resilience hypothesis” posits that Black and Latinx adults may show greater resilience than White adults to social stressors, possibly due to adaptation to life under social and economic adversity (Keyes 2009; Malat, Mayorga-Gallo, and Williams 2018). Additional evidence suggests that social strain might be less problematic for Black and Latinx adults due to contextual and cultural differences, including familial expectations and stress appraisal (Brown et al. 2020; Landale et al. 2006). For instance, older White adults would be more susceptible than older Black adults to family strain’s effects on psychological distress and depressive symptoms (Lincoln et al. 2003; Wang et al. 2021).

Hypotheses

Guided by the SPM and the convoy model, I investigate a comprehensive network which includes spouses, children, family members, and friends. I assess how racial/ethnic differences in social support and strain explain disparities in late-life mental health, and whether these effects differ among older Black and Latinx adults. The following hypotheses are tested:

Hypothesis 1: Older Black and Latinx adults experience more depressive symptoms than their White counterparts.

Hypothesis 2a: Older Black and Latinx adults perceive less social support from their spouses, children, family members, and friends than their White peers, increasing their risk of depressive symptoms.

Hypothesis 2b: Racial/ethnic disparities in mental health are partially explained by the differential effects of social support from spouses, children, family members, and friends (i.e., diminished returns).

Hypothesis 3a: Black and Latinx older adults experience more social strain than White older adults, increasing their risk of depressive symptoms.

Hypothesis 3b: Racial/ethnic disparities in mental health are partially explained by differential effects of social strain from spouses, children, family members, and friends (i.e., higher vulnerability).

Methods

Data

Date for this study come from the HRS 2006–2016—a nationally representative biennial survey-based panel study of American adults aged 50 years or older. It oversamples Blacks and Latinx to ensure reliable comparisons regarding health disparities. Social relationships and psychosocial data were collected every four years through the self-administered Psychosocial Leave-Behind Questionnaire (LBQ): a random half of respondents have been asked these questions since 2006 (i.e., 2006, 2010, 2014), while the other half have been asked since 2008 (i.e., 2008, 2012, 2016). Thus, social support and strain were assessed across three waves: 2006/2008, 2010/2012, and 2014/2016, with depressive symptoms measured based on the same three waves. Because the study’s focus was social relationships, respondents who were unmarried (i.e., divorce/separated, widowed, or never married) or reported having no children, no family members, or no friends were excluded at each wave. After excluding any participants who did not complete the Center for Epidemiological Studies Depression Scale (CES-D) (n = 126), those below the age of 50 at the time (n = 375), those who identified their race/ethnicity as “other” (n = 445), and those missing key demographic information (n = 18), the analytic sample contained 5,139 participants aged 50–104 who were married/cohabiting and reported having at least one spouse, one child, one family member, and one friend. Multiple imputations by chained equations were conducted to handle missing values for relationship variables (about 5 percent, M = 20).

Measures

Dependent variable

Depressive symptoms were measured with an abbreviated eight-item version of the CES-D, consisting of dichotomous items asking whether the respondents felt depressed, felt everything was an effort, had restless sleep, felt lonely, felt sad, could not get going, felt happy, enjoyed life, and had a lot of energy during the past week. Positive items were reverse-coded, and scores were summed such that higher values indicated more depressive symptoms (range: 0–8). This abbreviated CES-D has demonstrated comparable internal consistency, reliability, and validity to the original 20-item version (Turvey, Wallace, and Herzog 1999).

Independent variables

Race/ethnicity was self-identified and categorized into three groups: non-Hispanic White (labeled as White [reference]), non-Hispanic Black (labeled Black), and Hispanic/Latino (labeled Latinx).

Social support from spouses, children, family members, or friends was measured separately by three-item scales. Participants rated the extent to which they felt understood, relied upon in times of serious problems, and could open up about worries. Response options ranged from 1 (a lot) to 4 (not at all). Items were reverse-coded and averaged, with higher values indicating greater support. Cronbach’s alpha coefficients of the four social relationship types ranged from .82 to .86, and this HRS measure is commonly used for assessing positive aspects of relationship quality in older adults (Zhang, Hsieh, and Lai 2023).

Social strain within each social relationship type was measured separately by a four-item scale. Participants indicated the frequency of demands, criticism, disappointment, and irritation experienced from each group. Responses ranged from 1 (a lot) to 4 (not at all). All responses were reverse-coded and averaged, with higher scores indicating greater social strain. Cronbach’s alpha coefficient of the social strain variables (a common measure of negative relationship quality) of the four social relationship types ranged from .75 to .78 (Zhang, Hsieh, and Lai 2023).

Control variables

To assess the association between social support and strain with the four social relationship types on depressive symptoms, adjustments were made for structural relationships—including the number of nearby children, family members, and friends. The network size for each social tie was also assessed. All analyses controlled for demographic covariates, including age (in years), gender (0 = male, 1 = female), years of schooling, and nativity status (0 = U.S.-born, 1 = foreign-born). Household income was measured as the total earnings from all sources, including investment returns, pensions, annuities, and welfare payments. Household wealth was assessed as the net value of total assets, including homes, vehicles, bank accounts, and stocks minus debts (e.g., mortgages and other loans). Both income and wealth were measured in dollars and log-transformed to address skewness. I used the RAND version of household income and wealth data in which missing values are imputed (Bugliari et al. 2023). Chronic diseases were assessed based on the presence of seven conditions: hypertension, diabetes mellitus, cancer, chronic lung disease, coronary heart disease, congestive heart failure, and stroke. According to previous research, these variables are correlates of depressive symptoms (Assari 2018). Survey year was controlled due to the rotational design of the LBQ data (0 = starting in 2006, 1 = starting in 2008).

Analytic Plan

The analysis consisted of four steps. First, bivariate analysis (i.e., Pearson’s chi-square tests and t-tests) were conducted to describe sample characteristics at the baseline and to examine racial/ethnic differences in the descriptive statistics across groups (see Table 1). Second, mixed-effects negative binomial regression with a random intercept at the individual level was employed because the dependent variable was count data (i.e., depressive symptoms) and there was overdispersal. All models were adjusted for time-variant covariates including age, household income, and wealth, along with time-invariant controls measured at the baseline—gender, years of schooling, number of children, family members, and friends, nativity status, chronic condition, and survey year (Table 2). Karlson-Holm-Breen (KHB) analyses were conducted to assess any mediation effects, examining the extent to which the association between race/ethnicity and depressive symptoms was mediated through social support or social strain (Table 3) (Breen, Karlson, and Holm 2013; Karlson, Holm, and Breen 2012). Finally, moderation analyses were performed to test the interaction between race/ethnicity and social support and social strain, respectively (Table 4).

Table 1.

Descriptive Statistics of Analytic Variables, HRS, 2006/2008 (N = 5,139).

Variables	All		Whites		Blacks		Latinx
	(n = 5,139)		(n = 4,172)		(n = 587)		(n = 380)
	M	SD	M	SD	M	SD	M	SD
Depressive symptoms in Wave 2006/2008	1.03	(1.65)	0.93^{b, c}	(1.56)	1.32^{a, c}	(1.75)	1.71^{a, b}	(2.16)
Depressive symptoms in Wave 2014/2016	1.12	(1.70)	1.01^{b, c}	(1.60)	1.35^{a, c}	(1.81)	1.79^{a, b}	(2.21)
Social support
Spouse	3.52	(0.60)	3.54^{b, c}	(0.58)	3.34^{a, c}	(0.72)	3.44^{a, b}	(0.66)
Children	3.26	(0.70)	3.24^b	(0.70)	3.33^a	(0.69)	3.30	(0.71)
Family members	2.84	(0.85)	2.80^{b, c}	(0.85)	3.07^a	(0.79)	3.02^a	(0.81)
Friends	3.02	(0.74)	3.02^{b, c}	(0.74)	3.09^{a, c}	(0.70)	2.85^{a, b}	(0.76)
Social strain
Spouse	1.92	(0.64)	1.90^{b, c}	(0.63)	2.04^a	(0.69)	2.09^a	(0.71)
Children	1.66	(0.60)	1.65^{b, c}	(0.59)	1.74^a	(0.63)	1.75^a	(0.68)
Family members	1.52	(0.58)	1.50^{b, c}	(0.56)	1.65^a	(0.64)	1.62^a	(0.65)
Friends	1.38	(0.47)	1.36^{b, c}	(0.44)	1.50^a	(0.56)	1.47^a	(0.55)
Number of close children	2.56	(2.87)	2.44^{b, c}	(2.10)	3.21^a	(5.19)	3.14^a	(5.48)
Number of close family members	3.69	(5.64)	3.34^{b, c}	(4.72)	5.85^a	(8.59)	5.10^a	(9.23)
Number of close friends	4.52	(5.90)	4.57^c	(5.75)	4.55	(7.79)	3.84^a	(4.74)
Age	64.82	(7.71)	65.15^{b, c}	(7.76)	63.67^{a, c}	(7.18)	62.47^{a, c}	(7.29)
Chronic medical conditions	0.43	(0.71)	0.42^b	(0.70)	0.48^a	(0.74)	0.47	(0.77)
Year of schooling	13.25	(2.75)	13.59^{b, c}	(2.38)	12.52^{a, c}	(2.72)	10.24^{a, c}	(4.25)
Household income ($1,000)	61.09	(85.46)	71.82^{b, c}	(97.10)	42.10^{a, c}	(44.14)	39.16^{a, c}	(61.81)
Household wealth ($1,000)	370.66	(848.27)	500.42^{b, c}	(996.02)	112.78^{a, c}	(341.36)	148.60^{a, c}	(341.10)
Male	44.86		45.32^b		40.14^a		45.41
Female (%)	55.14		54.68^b		59.86^a		54.59
Nativity status (ref. U.S.-born)
U.S.-born (%)	93.17		96.37^c		95.56^c		54.37^{a, b}
Foreign-born (%)	6.83		3.63^c		4.44^c		45.63^{a, b}

Note. Differences by race/ethnicity were tested using Pearson’s chi-square statistic for categorical variables and t-statistics for continuous variables; M = mean, SD = standard deviation.

Statistically different from White late-middle-aged and older adults at p < .05.

Statistically different from Black late-middle-aged and older adults at p < .05.

Statistically different from Latinx late-middle-aged and older adults at p < .05.

Table 2.

Mixed-Effects Negative Binomial Regression Models of Social Support and Strain on Depressive Symptoms HRS, 2006–2016 (N = 5,139) (IRR).

	(1)	(2)	(3)	(4)
Blacks	1.17**	1.14*	1.11	1.09
	(0.08)	(0.08)	(0.07)	(0.07)
Latinx	1.25***	1.28**	1.19*	1.22*
	(0.11)	(0.11)	(0.10)	(0.10)
Social support
Spouse		0.74***		0.77***
		(0.02)		(0.02)
Children		0.88***		0.92**
		(0.02)		(0.02)
Family		0.97		0.99
		(0.02)		(0.02)
Friends		1.01		1.01
		(0.02)		(0.02)
Social strain
Spouse			1.31***	1.23***
			(0.03)	(0.03)
Children			1.17***	1.13**
			(0.03)	(0.03)
Family			1.10**	1.10**
			(0.03)	(0.03)
Friends			1.08*	1.08*
			(0.03)	(0.03)

Note. All models control for number of children, family members, friends, survey year, and sociodemographic covariates. SEs are in parentheses.

p < .05. **p < .01. ***p < .001.

Table 3.

KHB Analysis on the Mediating Roles of Social Support and Strain on Depressive Symptoms (N = 5,139).

	Black	Latinx
Total effect	0.13**	0.21***
Direct effect	0.04	0.17**
Indirect effect	0.09***	0.03
Proportion Mediated %
Support from spouse	31.65	−1.72
Support from children	−7.36	−2.16
Support from family members	−0.20	−0.11
Support from friends	0.34	−0.42
Strain from spouse	15.80	11.57
Strain from children	5.97	3.24
Strain from family members	11.43	3.61
Strain from friends	7.57	1.87
All factors	65.19	15.89

Note. Estimates are derived from Model 4 in Table 2, adjusting for number of children, family members, friends, survey year, and sociodemographic covariates.

p < .05. **p < .01. ***p < .001.

Table 4.

Moderation Analysis of Race/Ethnicity and Key Variables on Depressive Symptoms (N = 5,139).

Panel 1 Support from spouse			Panel 2 Strain from Spouse
Blacks	−0.41*	(0.16)	Blacks	0.37*	(0.14)
Latinx	−0.16	(0.20)	Latinx	0.39*	(0.17)
Support from spouse	−0.37***	(0.03)	Strain from spouse	0.36***	(0.03)
Blacks*Support from spouse	0.18***	(0.05)	Blacks* strain from spouse	−0.13	(0.07)
Latinx* Support from spouse	0.13*	(0.06)	Latinx* strain from spouse	−0.10	(0.07)
Panel 3 Support from children			Panel 4 Strain from Children
Blacks	0.02	(0.23)	Blacks	0.46**	(0.15)
Latinx	−0.15	(0.25)	Latinx	0.49**	(0.16)
Support from children	−0.20***	(0.02)	Strain from children	0.33***	(0.03)
Blacks* Support from children	0.05	(0.07)	Blacks* Strain from children	−0.18*	(0.08)
Latinx* Support from children	0.12*	(0.07)	Latinx* Strain from children	−0.16*	(0.08)
Panel 5 Support from family members			Panel 6 Strain from family members
Blacks	−0.05	(0.20)	Blacks	0.46**	(0.14)
Latinx	0.30	(0.20)	Latinx	0.34*	(0.17)
Support from family	−0.09***	(0.02)	Strain from family members	0.27***	(0.03)
Blacks* Support from family	0.08	(0.06)	Blacks* Strain from family members	−0.19*	(0.07)
Latinx* Support from family	−0.02	(0.06)	Latinx* Strain from family members	−0.08	(0.08)
Panel 7 Support from friends			Panel 8 Strain from friends
Blacks	0.43*	(0.21)	Blacks	0.20	(0.14)
Latinx	0.03	(0.21)	Latinx	0.24	(0.16)
Support from friends	−0.04	(0.02)	Strain from friends	0.23***	(0.04)
Blacks* Support from friends	−0.09	(0.07)	Blacks* Strain from friends	−0.04	(0.08)
Latinx* Support from friends	0.07	(0.07)	Latinx* Strain from friends	−0.02	(0.09)

Note. Estimates are derived from Model 4 in Table 2—the full model with all focal relationship variables adjusting for numbers of close children, family members, friends, survey year, and sociodemographic covariates.

p < .05. **p < .01. ***p < .001.

Results

Descriptive Findings

Table 1 presents the descriptive statistics for baseline characteristics across race/ethnic groups. Among the 5,139 participants, 81.18 percent were White, 11.42 percent were Black, and 7.40 percent were Latinx late-middle-aged and older adults. On average, participants reported 1.03 depressive symptoms in Wave 1 and 1.12 in Wave 3. However, Black and Latinx participants reported significantly higher depressive symptom scores than their White peers, with Latinx participants having the most symptoms (mean = 1.71), followed by Black adults (mean = 1.32), and this pattern persisted in Wave 3. Despite being younger on average, Black participants reported significantly more chronic diseases than their White peers. Black and Latinx participants also had fewer years of schooling, and lower household income and wealth, compared with Whites. In addition, a higher percentage of Black participants were female (59.86 percent), and 45.63 percent of the Latinx participants were born outside of the United States.

Social support and strain also varied by social ties and race/ethnicity. Blacks and Latinx participants reported less of spousal support than Whites, but they experienced comparable or more social support from children and extended family. Black participants perceived more friend support than Whites, whereas Latinx perceived less. Regarding social strain, Blacks and Latinx participants reported more strain with their spouses, children, family members, and friends than Whites. On average, Blacks and Latinx participants had more close children and family members than their White counterparts, but a similar amount or fewer close friends than them.

Regression Results about Social Support

Table 2 presents the incidence rate ratios (IRRs) from mixed-effects negative binomial regression analyses of race/ethnicity and depressive symptoms with the consideration of social support and strain with all four social ties. Regarding social support, Model 1 shows that Black and Latinx participants have an increased risk of experiencing more depressive symptoms than their White counterparts, which is consistent with Hypothesis 1 (Black IRR = 1.17, p < .01; Latinx IRR = 1.25, p<.001). In Model 2, which incorporates all sources of social support, higher spousal support was associated with lower rates of depressive symptoms (IRR = 0.74, p < .001). Meanwhile, increase in support from children was associated with lower rates of depressive symptoms (IRR = 0.88, p < .001). Although social support from family and friends did not show a significant association with numbers of depressive symptoms in Model 2, they were independently linked to lower depression rates (IRR of family = 0.92, p < .001; IRR of friends = 0.96, p < .05, results not shown but available upon request).

Table 3 presents the mediation effects of social support on depressive symptoms based on KHB analysis. The results partially support Hypothesis 2a, as differences in social support partially explain the depressive symptoms disparities between Blacks and Whites; Blacks’ disadvantage in spousal support explained 31.65 percent of the Black-White difference in depressive symptoms, while support from children mediated 7.36 percent. Although support from spouses, children, family members, and friends had direct effects on depressive symptoms for Latinx participants, the indirect or mediating effect of this social support was statistically insignificant. This suggests that social support rarely explains Latinx-White depressive symptom disparities.

However, the moderation effects of social support tend to be weaker for Blacks and Latinx than for Whites (see Table 4). As shown in Panel 1 in Table 4, Blacks (interaction b = 0.18, p < .01, Figure 1) and Latinx (interaction b = 0.13, p < .05) derived fewer mental health benefits from spousal support than Whites (see Figure 1). Likewise, Panel 3 shows that Latinx older adults (interaction b = 0.12, p < .05, Figure 2) had diminished returns from support from children, compared with Whites. These findings partially support Hypothesis 2b (i.e., the diminished return hypothesis), which states that social support is less beneficial for the mental health of older Black and Latinx adults than it is for their White counterparts.

Regression Results about Social Strain

In terms of social strain, higher strain with spouses (IRR = 1.31, p < .001), children (IRR = 1.17, p < .001), family members (IRR = 1.10, p < .01), and friends (IRR = 1.08, p < .05) were associated with higher rates of depressive symptoms over time, respectively (see Model 3 in Table 2). Accounting for all relationship covariates, Model 4 in Table 2 shows no significant differences between Blacks and Whites in depressive symptoms. However, Latinx still had 1.22 times greater rates of expecting more depressive symptoms than Whites. This implies that social relationships may explain Black older adults’ depressive symptoms to a greater extent than those of Latinx.

The results from the KHB analysis partially supported Hypothesis 3a, indicating that Black individuals’ greater exposure to social strain partially explains their depressive symptom disadvantages relative to their White counterparts. Spousal Strain with was the most potent mediator, explaining 15.80 percent of the Black-White gap in depressive symptoms. Following this, strain with family members and strain with friends, and with children account for 11.43 and 7.57 percent, 5.97 percent of this gap, respectively. Overall, social strain with a spouse, children, family members, and friends jointly explained 40.77 percent of the association between Black older adults and depressive symptoms (indirect effect = 0.09, p < .001). Social strain with spouse and with family members mediated about 12 and 3.6 percent the relationship between Latinx and depressive symptoms, respectively. As such, social strain made a limited contribution to the depressive symptom disadvantage faced by Latinx older adults.

Furthermore, the findings of moderation analysis show that strain with children is a significant moderator for both Blacks and Latinx (see Table 4). Strain with children was less harmful for Blacks (interaction b = −0.18, p < .05) and for Latinx (interaction b = −0.16, p < .05, Figure 3). Likewise, family strain was less detrimental for Black older adults (interaction b = −0.19, p < .05, Figure 4) than for Whites. However, there were no moderating effects for social strain with spouses or friends. Given that Blacks and Latinx were equally or less vulnerable to social strain than Whites, these results do not support Hypothesis 3b, the higher vulnerability hypothesis.

Discussion

This study provides a nuanced understanding of racial/ethnic disparities in mental health by examining how differential exposure to social strain and access to social support, as well as the magnitude of the effects of social relationships, are associated with mental health disparities among late-middle-aged and older adults across racial/ethnic groups. This study reveals that Black and especially Latinx late-middle-aged and older adults exhibit more depressive symptoms than their White counterparts. In addition, it demonstrates that access to social support and exposure to social strain vary among Black and Latinx older adults and their White peers. Meanwhile, social support and social strain play different roles in contributing to the racial/ethnic disparities in depressive symptoms, with relationships with spouses, children, and close kin being more influential than relationships with friends. Moreover, the findings suggest that racial/ethnic differences in social support’s effect magnitude are also associated with this mental health inequality between Black and Latinx late-middle-aged and older adults, compared with Whites.

This study’s findings indicate that social support is positively associated with mental health among late-middle-aged and older adults, regardless of race/ethnicity. In addition, more support from spouses, children, family members, and friends is correlated with lower rates of depressive symptoms. Aligning with the convoy model, this study links social support from spouses to the lowest rates of depressive symptoms, followed by support from children, family members, and friends (Antonucci et al. 2014; Lee and Szinovacz 2016; Stafford et al. 2011; Teo et al. 2013). Sensitivity analysis of the average marginal effects demonstrated that the associations between social support and depressive symptoms remained robust, despite the modest effect sizes. However, supplementary analyses (results not shown) exploring social relationships’ structural characteristics (e.g., contact frequency and number of close members in focal networks) revealed that these factors are either negligible or insignificant in explaining racial/ethnic gaps in depressive symptoms.

Although the SPM presumes that disadvantaged social groups have fewer protective resources (Pearlin 1989), this study reveals that social supports are equally or more available to Black and Latinx than White older adults. However, there were two exceptions: White individuals exhibited an advantage in spousal support compared with their Blacks and Latinx older adults, and in support from friends relative to Latinx. Moreover, this study indicates that Black older adults’ spousal support disadvantages account for 31.65 percent of the depressive symptom disparity between Black and White older adults. Conversely, Black older adults’ advantages in support from children compensate for 7.36 percent of the Black-White gap. The negative mediation effect suggests that support from children suppressed or reduced the direct effect of race/ethnicity on depression. In other words, the Black-White disparity in depressive symptoms would be even larger without the current level of support from children. These findings lend partial support to Hypothesis 2a, which posits that limited access to social support contributes to racial/ethnic disparities in depressive symptoms. These findings highlight spousal support’s role in late-life mental health among Black adults. One explanation for this finding could be the disproportionate rates of incarceration among Black males, stemming from systemic racism. Higher incarceration rates may challenge spousal relationships, reduce marital quality, and subsequently decrease spousal support (Travis and Waul 2003). The lack of spousal support, which may encompass emotional, social, and potentially financial aspects within the marital relationship, is associated with the overall difference in depressive symptoms between Black and White older adults.

While social support services are a protective factor for late-life mental health, this study highlights an overlooked mechanism of mental health inequality. It reveals that the cognitive health effect of spousal support for Black and Latinx late-middle-aged and older adults, and support from children for Latinx late-middle-aged and older adults, was weaker than among their White peers. This finding aligns with literature indicating that socioeconomically privileged groups benefit more from social resources and psychosocial assets they possess (Assari 2018).

This discrepancy can be attributed to several structural factors. Black Americans, for instance, often face systemic challenges such as segregation, which can limit their access to high-quality health care and mental health services. Given their marginalized social positions and stress exposure, Black and Latinx older adults, especially in immigrant communities, may turn to social resources such as family members, friends, and religious institutions for emotional support (Bolger and Prickett 2021; Woodward, Taylor, and Chatters 2011). Yet, racism, discrimination, and collective trauma may create additional stressors that overshadow the benefits of social support (Abrutyn 2023). Moreover, mental illness stigma in Black and Latinx communities may lead to supportive relationships substituting for, or delaying, necessary medical intervention (Misra et al. 2021; Thoits 2011b). It is possible that emotional social support does not necessarily translate into adjuvant mental health support and encouragement to seek professional mental health help for Black and Latinx older adults, particularly when there are unmet needs for professional treatment. However, this finding contradicts claims by Kiecolt et al. (2008) and Lincoln et al. (2003) that social support (i.e., spousal support and average perceived support) has been more beneficial for Black adults than for White adults. Both studies focused on the non-elderly population (aged 15–54 years) and used data from the 1990–1992 wave of the National Comorbidity Survey, where Black adults reported either similar or lower levels of psychological distress than White adults. Future studies should further explore whether Black and Latinx older adults experience diminished returns from psychosocial resources on mental health as an underlying explanation for mental health inequality by race/ethnicity in late-middle age and later life.

Given the limited research on the effects of social strain on the racial/ethnic patterning of mental health, this study explores how differential exposure to, and differential effects of, social strain influence racial/ethnic disparities in depressive symptoms. Consistent with the SPM (Pearlin 1989), Black and Latinx late-middle-aged and older adults generally report more social strain with spouses, children, family members, and friends than their White counterparts. Moreover, aligning with previous research, all types of social strain are associated with more depressive symptoms, with spousal strain having the most pronounced effect, followed by strain with children, family members, and then friends (Chen and Feeley 2014; Stafford et al. 2011). This study reveals that social strain with spouses, family members, and friends collectively mediates nearly one-third of the Black-White disparity in depressive symptoms. Black older adults often have higher proportions of close kin in their social networks, such that they may rely more on family members for assistance and support (Ajrouch, Antonucci, and Janevic 2001). Therefore, the chronic stressors Black individuals experience, including financial hardship, racial discrimination, and residential challenges due to segregation, may increase their likelihood of encountering daily hassles, and experiencing more social strain (Connidis and McMullin 2002; Williams and Mohammed 2009).

This study demonstrates that Black and Latinx older adults exhibit lower susceptibility to social strain on depressive symptoms than their White counterparts, aligning with previous research (Lincoln et al. 2003; Malat et al. 2018; Wang et al. 2021). Despite greater exposure to social strain, the detrimental effects of strain with children on mental health were weaker for Black and Latinx late-middle-aged and older adults. Moreover, strain with family members was less harmful among Black than among White older adults. This finding corroborates recent research on racial/ethnic differences in stress appraisal and stress experience, showing that relationship-based stressors could be less disturbing for Black and Latinx than for White older adults (Brown et al. 2020; Wang et al. 2021). It is conceivable that long-term exposure to stressors may build resilience in Black and Latinx older adults, equipping them with context-specific coping skills for adapting to and managing stress better than their White peers (Malat et al. 2018; Wang et al. 2021). Moreover, historical legacies of slavery, segregation, and systemic racism have shaped cultural norms in Black communities, emphasizing cooperative support, extended familism (e.g., strong mother/child relationships, communal patterns of childcare, mutual support among kinship ties), and religiosity as strategies for navigating adversity (Sarkisian and Gerstel 2004). Higher levels of extended familism and religious involvement foster greater tolerance, and reduce tension among children and family members, thereby mitigating the negative effects of interactions within mother/child relationships and broader kinship networks (Taylor, Chatters, and Levin 2004).

Similarly, Latinx culture emphasizes family solidarity, with familism promoting cohesion, loyalty, and care for older adults in multigenerational households (J. Almeida et al. 2009; Guo et al. 2015). Through frequent interaction and negotiation, these cultural values may mitigate strain with children’s adverse effects on mental well-being. In addition, compared with White individuals, who tend to be more culturally individualistic, Latinx communities boast strong support networks transcending their immediate families (J. Almeida et al. 2009). These networks, which include extended kin, friends, neighbors, and communities, could provide additional support, and buffer them from the negative effects of strain with children. This finding suggests that racial/ethnic patterning of stress exposure in the SPM should consider the differential effect of stressors through which Black and Latinx older adults might be less vulnerable and more resilient to some stress exposure, as previously posited. Future research is needed to test the role of social support and strain in late-life mental health across racial/ethnic groups.

Limitations

This study has several limitations warranting further investigation. First, although mediation analyses were conducted, causality was not addressed. Thus, the relationship between social support and strain and depressive symptoms could be bidirectional. Second, the analytic sample was confined to late-middle-aged and older adults with comprehensive networks (e.g., spouse/partner, children, other family, and friends) over the 10-year study period. Thus, sampling weights were not applied, potentially limiting the study’s generalizability to the broader population. In addition, selection bias and the highly stable social network within the sample may have led to underestimating social relationships’ effect on depressive symptoms. Third, while this study measured the qualitative characteristics of major social relationships in late-middle-aged and later life; unobserved relationship factors such as instrumental assistance (e.g., actual help, caregiving support, or companionship) and congregational support networks may also influence late-life depression, especially for Black older adults and those lacking spouses or close kin (Nguyen 2020; Thoits 2011a). In addition, recent evidence suggests that coping resources (e.g., self-esteem, mastery, and religious attendance) offer more protection for older Black adults than for their White peers for some mental health outcomes (Louie et al. 2021), warranting further investigation into additional psychosocial coping resources beyond those examined in this study. Finally, this research reveals that social support and social strain play minor roles in explaining Latinx-White depressive symptom disparities. Even when considering nativity status’ moderation effect (i.e., immigrants and U.S.-born citizens) in supplementary analyses, Latinx U.S. natives were more depressed than Latinx immigrants, Black natives, and White natives. Although the Latinx population exhibits considerable intra-ethnic heterogeneity (e.g., ethnic and racial identity, migration history), data limitations precluded the examination of these subgroup differences and cultural factors. Future studies could investigate intergroup differences and similarities when data are available to unpack how social relationships and unmeasured cultural factors influence Latinx older adults’ mental health.

Conclusion

This study enriches the literature on the mental health disparities of racial/ethnic older adults by assessing the differential access/exposure to, and differential effects of, social support and social strain with major social ties. The findings reveal that Black and Latinx late-middle-aged and older adults face more mental health disadvantages than their White counterparts. These disparities are linked to heightened exposure to social strain across major social ties and diminished returns on social support from spouses and children, despite generally having more available social support than White peers. The findings emphasize social strain’s importance in explaining mental health gaps in late-middle-age and later life, especially among Blacks. Overall, this study enhances our understanding of the psychosocial mechanism that shapes late-life mental health among Black and Latinx older adults.

Footnotes

ORCID iD

Wenhua Lai

References

Abrutyn

Seth

. 2023. “The Roots of Social Trauma: Collective, Cultural Pain and Its Consequences.” Society and Mental Health. Published electronically November 30. doi: 10.1177/2156869323121308.

Ajrouch

Kristine J.

Antonucci

Toni C.

Janevic

Mary R.

2001. “Social Networks Among Blacks and Whites: The Interaction Between Race and Age.” The Journals of Gerontology: Series B 56:112–18.

Almeida

Joanna

Molnar

Beth E.

Kawachi

Ichiro

Subramanian

S. V.

2009. “Ethnicity and Nativity Status as Determinants of Perceived Social Support: Testing the Concept of Familism.” Social Science and Medicine 68:1852–58.

Almeida

Joanna

Subramanian

S. V.

Kawachi

Molnar

B. E.

2011. “Is Blood Thicker Than Water? Social Support, Depression and the Modifying Role of Ethnicity/Nativity Status.” Journal of Epidemiology and Community Health 65:51–56.

Almeida

Osvaldo P.

Draper

Brian

Snowdon

John

Lautenschlager

Nicola T.

Pirkis

Jane

Byrne

Gerard

Sim

Moira

Stocks

Nigel

Flicker

Leon

Pfaff

Jon J.

2012. “Factors Associated with Suicidal Thoughts in a Large Community Study of Older Adults.” British Journal of Psychiatry 201:466–72.

Antonucci

Toni C.

Ajrouch

Kristine J.

Birditt

Kira S.

2014. “The Convoy Model: Explaining Social Relations from a Multidisciplinary Perspective.” Gerontologist 54:82–92.

Assari

Shervin

. 2018. “Health Disparities Due to Diminished Return among Black Americans: Public Policy Solutions.” Social Issues and Policy Review 12:112–45.

Bolger

Daniel

Prickett

Pamela J.

2021. “Where Would You Go? Race, Religion, and the Limits of Pastor Mental Health Care in Black and Latino Congregations.” Religions 12:1–15.

Breen

Richard

Karlson

Kristian Bernt

Holm

Anders

. 2013. “Total, Direct, and Indirect Effects in Logit and Probit Models.” Sociological Methods and Research 42:164–91.

10.

Breslau

Joshua

Aguilar-Gaxiola

Sergio

Kendler

Kenneth S.

Maxwell

Williams

David

Kessler

Ronald C.

2006. “Specifying Race-ethnic Differences in Risk for Psychiatric Disorder in a USA National Sample.” Psychological Medicine 36:57–68.

11.

Broman

Clifford L.

2005. “Marital Quality in Black and White Marriages.” Journal of Family Issues 26:431–41.

12.

Brown

Lauren L.

Mitchell

Uchechi A.

Ailshire

Jennifer A.

2020. “Disentangling the Stress Process: Race/Ethnic Differences in the Exposure and Appraisal of Chronic Stressors among Older Adults.” Journals of Gerontology, Series B Psychological Sciences and Social Sciences 75:650–60.

13.

Bugliari

Delia

Carroll

Joanna

Hayden

Orla

Hayes

Jessica

Hurd

Michael D

Lee

Stephen

Main

colleen m Regan

Mccullough

Erik Meijer

Pantoja

Philip

Rohwedder.

Susann

2023. “RAND HRS Longitudinal File 2020 (V1) Documentation.” RAND Center for the Study of Aging.

14.

Bulanda

Jennifer Roebuck

Brown

Susan L.

2007. “Race-ethnic Differences in Marital Quality and Divorce.” Social Science Research 36:945–67.

15.

Burleson

Brant R.

2003. “Emotional Support Skills.” Pp. 569–612 in Handbook of communication and social interaction skills. Routledge.

16.

Calzada

Esther J.

Gulbas

Lauren E.

Hausmann-Stabile

Carolina

Kim

Su Yeong

Cardoso

Jodi Berger

. 2020. “Mental Health Issues within Latinx Populations: Evaluating the State of the Field.” Pp. 45–62 in New and Emerging Issues in Latinx Health. Springer.

17.

Carr

Deborah

Moorman

Sara M.

2011. “Social Relations and Aging BT.” Pp. 145–60 in Handbook of Sociology of Aging, edited by Settersten

R. A.

Angel

J. L.

New York: Springer.

18.

Chen

Yixin

Feeley

Thomas Hugh

. 2014. “Social Support, Social Strain, Loneliness, and Well-being among Older Adults: An Analysis of the Health and Retirement Study” Journal of Social and Personal Relationships 31:141–61.

19.

Cichy

Kelly E.

Stawski

Robert S.

Almeida

David M.

2012. “Racial Differences in Exposure and Reactivity to Daily Family Stressors.” Journal of Marriage and Family 74:572–86.

20.

Cohen

Sheldon

Wills

Thomas A.

1985. “Stress, Social Support, and the Buffering Hypothesis.” Psychological Bulletin 98:310.

21.

Connidis

Ingrid Arnet

McMullin

Julie Ann

. 2002. “Ambivalence, Family Ties, and Doing Sociology.” Journal of Marriage and the Family 64:594–601.

22.

Flores

Melissa

Ruiz

John M.

Goans

Christian

Butler

Emily A.

Uchino

Bert N.

Hirai

Michiyo

Tinajero

Ruben

Smith

Timothy W.

2020. “Racial-ethnic Differences in Social Networks and Perceived Support: Measurement Considerations and Implications for Disparities Research.” Cultural Diversity and Ethnic Minority Psychology 26:189–99.

23.

Fuller

Heather R.

Ajrouch

Kristine J.

Antonucci

Toni C.

2020. “The Convoy Model and Later-life Family Relationships.” Journal of Family Theory and Review 12:126–46.

24.

González

Hector M.

Tarraf

Wassim

Whitfield

Keith E.

Vega

William A.

2010. “The Epidemiology of Major Depression and Ethnicity in the United States.” Journal of Psychiatric Research 44:1043–51.

25.

Guo

Man

Shijian

Liu

Jinyu

Sun

Fei

. 2015. “Family Relations, Social Connections, and Mental Health Among Latino and Asian Older Adults.” Research on Aging 37:123–47.

26.

Jang

Yuri

Chiriboga

David A.

Kim

Giyeon

Phillips

Karon

. 2008. “Depressive Symptoms in Four Racial and Ethnic Groups: The Survey of Older Floridians (SOF).” Research on Aging 30:488–502.

27.

Karlson

Kristian Bernt

Holm

Anders

Breen

Richard

. 2012. “Comparing Regression Coefficients Between Same-sample Nested Models Using Logit and Probit: A New Method.” Sociological Methodology 42:286–313.

28.

Kasper

Judith D.

Ensminger

Margaret E.

Green

Kerry M.

Fothergill

Kate E.

Juon

Hee-Soon

Robertson

Judith

Thorpe

Roland J.

2008. “Effects of Poverty and Family Stress Over Three Decades on the Functional Status of Older African American Women.” The Journals of Gerontology Series B: Psychological Sciences and Social Sciences 63:S201–S210.

29.

Katz

Benjamin

Turney

Indira

Lee

Ji Hyun

Amini

Reza

Ajrouch

Kristine J.

Antonucci

Toni C.

2020. “Race/Ethnic Differences in Social Resources as Cognitive Risk and Protective Factors.” Research in Human Development 17:57–77.

30.

Keyes

Corey L. M.

2009. “The Black-White Paradox in Health: Flourishing in the Face of Social Inequality and Discrimination.” Journal of Personality 77:1677–706.

31.

Kiecolt

K. Jill

Hughes

Michael

Keith

Verna M.

2008. “Race, Social Relationships, and Mental Health.” Personal Relationships 15:229–45.

32.

Kim

Yijung K.

Kim

Kyungmin

Fingerman

Karen L.

Umberson

Debra J.

2021. “Racial Differences in Early Parental Death, Midlife Life Problems, and Relationship Strain With Adult Children.” The Journals of Gerontology: Series B 76:1617–28.

33.

Landale

Nancy S.

Oropesa

R. Salvador

Bradatan

Cristina

. 2006. “Hispanic Families in the United States: Family Structure and Process in an Era of Family Change.” Hispanics and the Future of America 5:1–30.

34.

Lee

Hyo Jung

Szinovacz

Maximiliane E.

2016. “Positive, Negative, and Ambivalent Interactions With Family and Friends: Associations With Well-being.” Journal of Marriage and Family 78:660–79.

35.

Liang

Jersey

Xiao

Quiñones

Ana R.

Bennett

Joan M.

Wen

. 2011. “Multiple Trajectories of Depressive Symptoms in Middle and Late Life: Racial/Ethnic Variations.” Psychology and Aging 26:761–77.

36.

Lincoln

Karen D.

2007. “Financial Strain, Negative Interactions, and Mastery: Pathways to Mental Health among Older African Americans.” Journal of Black Psychology 33:439–62.

37.

Lincoln

Karen D.

Chatters

Linda M.

Taylor

Robert Joseph

. 2003. “Psychological Distress among Black and White Americans: Differential Effects of Social Support, Negative Interaction and Personal Control” Journal of Health and Social Behavior 44:390–407.

38.

Louie

Patricia

Upenieks

Laura

Erving

Christy L.

Tobin

Courtney S. Thomas

. 2021. “Do Racial Differences in Coping Resources Explain the Black–White Paradox in Mental Health? A Test of Multiple Mechanisms.” Journal of Health and Social Behavior 63:55–70.

39.

Louie

Patricia

Wheaton

Blair

. 2019. “The Black-White Paradox Revisited: Understanding the Role of Counterbalancing Mechanisms during Adolescence.” Journal of Health and Social Behavior 60:169–87.

40.

Malat

Jennifer

Mayorga-Gallo

Sarah

Williams

David R.

2018. “The Effects of Whiteness on the Health of Whites in the USA.” Social Science and Medicine 199:148–56.

41.

Misra

Supriya

Jackson

Valerie W.

Chong

Jeanette

Choe

Karen

Tay

Charisse

Wong

Jazmine

Yang

Lawrence H.

2021. “Systematic Review of Cultural Aspects of Stigma and Mental Illness among Racial and Ethnic Minority Groups in the United States: Implications for Interventions.” American Journal of Community Psychology 68:486–512.

42.

Moen

Phyllis

. 2022. “The Uneven Stress of Social Change: Disruptions, Disparities, and Mental Health.” Society and Mental Health 12:85–98.

43.

Mouzon

Dawne M.

2013. “Can Family Relationships Explain the Race Paradox in Mental Health?” Journal of Marriage and Family 75:470–85.

44.

Mouzon

Dawne M.

2014. “Relationships of Choice: Can Friendships or Fictive Kinships Explain the Race Paradox in Mental Health?” Social Science Research 44:32–43.

45.

Muruthi

James R.

Zalla

Lauren C.

Lewis

Denise C.

2020. “Depressive Symptoms Among Aging Hispanic Americans: Longitudinal Effects of Positive Spousal Support and Previous Depressive Symptoms.” Journal of Aging and Health 32:481–90.

46.

Nguyen

Ann W.

2020. “Religion and Mental Health in Racial and Ethnic Minority Populations: A Review of the Literature.” Innovation in Aging 4:1–13.

47.

Pearlin

Leonard I

. 1989. “The Sociological Study of Stress.” Journal of Health and Social Behavior 241–56.

48.

Pearlin

Leonard I.

1999. “Stress and Mental Health: A Conceptual Overview.” Pp. 161–75 in A Handbook for the Study of Mental Health: Social Contexts, Theories, and Systems. Cambridge University Press.

49.

Phelan

Link

Bruce

. 2015. “Is Racism a Fundamental Cause of Inequalities in Health?” Annual Review of Sociology 41:311–30.

50.

Pierce

Kayla D. R.

Quiroz

Christopher S.

2019. “Who Matters Most? Social Support, Social Strain, and Emotions.” Journal of Social and Personal Relationships 36(10):3273–92.

51.

Sarkisian

Natalia

Gerstel

Naomi

. 2004. “Kin Support among Blacks and Whites: Race and Family Organization” American Sociological Review 9:812–37.

52.

Shiovitz-Ezra

Sharon

Leitsch

Sara A.

2010. “The Role of Social Relationships in Predicting Loneliness: The National Social Life, Health, and Aging Project.” Social Work Research 34:157–67.

53.

Stafford

Mai

McMunn

Anne

Zaninotto

Paola

Nazroo

James

. 2011. “Positive and Negative Exchanges in Social Relationships as Predictors of Depression: Evidence from the English Longitudinal Study of Aging.” Journal of Aging and Health 23:607–28.

54.

Staples

Robert

Johnson

Leanor Boulin

. 1993. Black Families at the Crossroads: Challenges and Prospects. San Francisco, CA: Jossey-Bass.

55.

Sternthal

Michelle J.

Slopen

Natalie

Williams

David R.

2011. “Racial Disparities in Health: How Much Does Stress Really Matter?” Du Bois Review 8:95–113.

56.

Taylor

Robert Joseph

Chatters

Linda M.

Levin

Jeffrey S.

2004. Religion in the Lives of African Americans: Social, Psychological, and Health Perspectives. Sage.

57.

Taylor

Robert Joseph

Chatters

Linda M.

2020. “Psychiatric Disorders among Older Black Americans: Within-and Between-group Differences.” Innovation in Aging 4:1–16.

58.

Teo

Alan R.

Choi

Hwa Jung

Valenstein

Marcia

. 2013. “Social Relationships and Depression: Ten-year Follow-up from a Nationally Representative Study.” PLoS ONE 8:1–8.

59.

Thoits

Peggy A.

1995. “Stress, Coping, and Social Support Processes: Where Are We? What Next?” Journal of Health and Social Behavior. Retrieved October 15, 2024 (https://www.jstor.org/stable/2626957).

60.

Thoits

Peggy A.

2011. “Mechanisms Linking Social Ties and Support to Physical and Mental Health.” Journal of Health and Social Behavior 52:145–61.

61.

Thomas

Patricia A.

2016. “The Impact of Relationship-specific Support and Strain on Depressive Symptoms Across the Life Course.” Journal of Aging and Health 28:363–82.

62.

Travis

Jeremy

Waul

Michelle

. 2003. Prisoners Once Removed: The Impact of Incarceration and Reentry on Children, Families, and Communities. Washington, DC: The Urban Institute.

63.

Turner

R. Jay

Avison

William R.

2003. “Status Variations in Stress Exposure: Implications for the Interpretation of Research on Race, Socioeconomic Status, and Gender.” Journal of Health and Social Behavior 44:488–505.

64.

Turvey

Carolyn L.

Wallace

Robert B.

Herzog

Regula

. 1999. “A Revised CES-d Measure of Depressive Symptoms and a DSM-based Measure of Major Depressive Episodes in the Elderly.” International Psychogeriatrics 11:139–48.

65.

Ulbrich

Patricia M.

Warheit

George J.

Zimmerman

Rick S.

1989. “Race, Socioeconomic Status, and Psychological Distress: An Examination of Differential Vulnerability.” Journal of Health and Social Behavior 30 (1): 131–46.

66.

Vespa

Jonathan

Medina

Lauren

Armstrong

David

. 2020. “Demographic Turning Points for the United States: Population Projections for 2020 to 2060.” Current Population Reports. Retrieved October 15, 2024 (https://www.census.gov/content/dam/Census/library/publications/2020/demo/p25-1144.pdf).

67.

Wang

Fei

Nguyen

Ann W.

Lincoln

Karen D.

Qin

Weidi

Hamler

Tyrone

. 2021. “The Moderating Role of Race and Ethnicity in the Relationship Between Negative Family Interactions and Mental Health Among Older Adults.” The Gerontologist 62:674–84.

68.

Williams

David R.

Mohammed

Selina A.

2009. “Discrimination and Racial Disparities in Health: Evidence and Needed Research.” Journal of Behavioral Medicine 32:20.

69.

Williams

David R.

Mohammed

Selina A.

2013. “Racism and Health I: Pathways and Scientific Evidence.” American Behavioral Scientist 57:1152–73.

70.

Woodward

Toler Amanda

Taylor

Robert Joseph

Chatters

Linda M.

2011. “Use of Professional and Informal Support by Black Men with Mental Disorders.” Research on Social Work Practice 21:328–36.

71.

Yang

Tse Chuan

Park

Kiwoong

. 2019. “Racial/Ethnic Disparities in Depression: Investigating How Sources of Support and Types of Integration Matter.” Social Science Research 82:59–71.

72.

Zhang

Zhenmei

Hsieh

Ning

Lai

Wen Hua

. 2023. “Social Relationships in Later Life: Does Marital Status Matter?” Journal of Social and Personal Relationships 40:2946–68.