Chronic stress exposure, social support, and sleep quality among African Americans: Findings from the National Survey of American Life-Reinterview

Abstract

Objective:

The purpose of this study was to determine whether social support from extended family and church members moderate the association between chronic stress exposure and sleep quality in a nationally representative sample of African American adults.

Design:

Data from African American respondents aged 18 and older were drawn from the National Survey of American Life-Reinterview. The analytic sample for this study included 1,372 African American adults who attended religious services at least a few times a year, as the church-based relationship measures were only assessed for these individuals. Self-reported sleep quality was assessed by sleep satisfaction, trouble falling asleep, and restless sleep. Chronic stress exposure was measured by a nine-item index. OLS and logistic regression were used to estimate the relationship between chronic stress exposure, extended family and church relationships, and sleep quality.

Results:

The data indicated that chronic stress exposure was associated with decreased sleep satisfaction, increased likelihood of trouble falling asleep and restless sleep. Receiving emotional support from family and more frequent contact with church members were associated with decreased restless sleep. Emotional family support moderated the associations between chronic stress exposure and trouble falling asleep and restless sleep. The positive associations between chronic stress exposure and these two sleep quality measures were attenuated among respondents who received high levels of emotional support from their family.

Conclusions:

Together, these findings underscore the detriment of chronic stress exposure to African Americans’ sleep quality and suggest that extended family members are effective stress coping resources and play an important role in this population’s sleep quality.

Introduction

Disparities in sleep health are important but underexplored contributors to racial disparities in health. Sleep is essential for restoration of healthy brain function, the immune system, glucose control and mental health. Poor sleep quality is associated with a wide range of diseases and health problems, including obesity (Buxton et al., 2012), high blood pressure (Knutson et al., 2009), diabetes (Buxton et al., 2010), stroke (Billings et al., 2021), heart disease, and premature mortality (Newman et al., 2000). Several cohort and epidemiological studies conducted in the United States suggest that there are racial differences in sleep quality and duration, with African Americans reporting shorter sleep duration and worse sleep quality than whites. African Americans have a disproportionate share of sleep disturbances, including frequent nocturnal arousals, insomnia, more light and less deep sleep, delayed onset (Turner et al., 2016), greater levels of daytime sleepiness (Hayes et al., 2009), and shortened sleep duration (Hale & Do, 2007). Based on findings from these studies, it is plausible that sleep health disparities may contribute to disparities in physical and mental health. Thus, racial disparities in sleep represent a significant public health problem that deserves immediate attention. A better understanding of modifiable social determinants of sleep quality specifically among African Americans is critical for addressing these disparities. The purpose of this study was to determine whether social support from extended family and church members moderate the association between chronic stress exposure and sleep quality in a nationally representative sample of African American adults.

Chronic stress and sleep

Stress represents the body’s physiological and mental response to a threat. Notably, stress and sleep have a reciprocal relationship (Agorastos & Olff, 2021). Increased stress levels can affect sleep quality, and poor sleep quality is associated with maladaptive changes to the stress response. Chronic stress can result in sleep-wake cycle dysregulation, thus potentially causing malfunction of the internal clock that tells the body when it is time to sleep and when it is time to be awake.

The sleep literature consistently points to the detriments of chronic and acute stress (Johnson et al., 2016; Sims et al., 2017). Research specifically on African Americans shows that chronic stressors are predictive of a range of poor sleep outcomes. General stressors are associated with shorter sleep duration (Bidulescu et al., 2010; Johnson et al., 2016; Sims et al., 2017), poorer sleep quality (Bidulescu et al., 2010; Johnson et al., 2016; Sims et al., 2017), and greater sleep latency (Bidulescu et al., 2010). For example, using data from the Jackson Heart Study, Johnson et al. (2016) found that African American respondents who reported higher levels of chronic and acute stress experienced shorter sleep duration. Specifically, people who reported higher chronic and acute stress had increased odds of very short or short sleep relative to normative sleep. Their study also indicated that African Americans who reported higher levels of chronic and acute stress had poorer sleep quality than their counterparts who reported lower levels of stress. Similarly, Bidulescu et al. (2010) identified associations between higher levels of perceived stress and shorter sleep duration, poorer sleep quality, greater sleep latency, and a higher likelihood of sleep medication use in a population-based sample of African American adults residing in the Atlanta area. More recently, using data from the Health and Retirement Study, Frazier and Brown (2022) found that job stress predicted insomnia symptoms among older Black adults.

Social support and sleep

There is strong evidence indicating that social support has a positive impact on sleep (Kent de Grey et al., 2018). Social support can have a positive impact on sleep for several reasons. Social relationships and support can improve emotion regulation, enhance positive affect, and promote healthy behaviors, all of which contribute to improved sleep outcomes (Berkman & Glass, 2000; Cacioppo et al., 2002; Jackowska et al., 2016; Steptoe et al., 2008). Specifically, studies in this area show that social support is predictive of healthy sleep duration (Patterson et al., 2021), better sleep efficiency (Mousavi et al., 2022; Patterson et al., 2021), higher quality sleep (Chung, 2017), and feeling rested in the morning (Mousavi et al., 2022). Further, social support can protect against short sleep duration (Stafford et al., 2017) and sleep disturbances (Mousavi et al., 2022). For instance, Johnson et al. (2023) found that people who received financial support were less likely to report short sleep duration compared to people who did not receive financial support. Moreover, this study demonstrated that the people with larger social support networks had decreased risk for short sleep duration. In contrast, Frazier and Brown’s (2022) research examined the impact of social support on insomnia symptoms but did not identify a statistically significant relationship between support and insomnia symptoms. Research specific to family relationships finds that people who receive more family support are more likely to have higher sleep quality (Fekete et al., 2014), healthy sleep duration (Patterson et al., 2021), and healthy sleep efficiency (Patterson et al., 2021).

Given its protective qualities for sleep, it makes sense that social support can also buffer the negative effects of chronic stress on sleep. While few studies have investigated the buffering effects of social support, the limited existing evidence indicates that social support can either attenuate or fully mitigate the effects of chronic stressors on sleep (Chen et al., 2021; Matick et al., 2021). For example, Liang et al. (2020), who examined social support and sleep among adult children caregivers of older adults with dementia, found that receiving instrumental support partially buffered the negative effect of role overload (perception of imbalance between role demands and time available to fulfill these demands) on the frequency of sleep disturbances. Similarly, Van Schalkwijk et al.’s (2015) investigation of sleep among Dutch adolescents found that social support buffered the negative effect of academic stress on sleep quality and sleep reduction. In our review of the literature, we were unable to find research in this area on African Americans. Thus, whether social support can moderate the relationship between chronic stress and poor sleep outcomes among African Americans remains unconfirmed.

Family and church-based relationships among African Americans

Among African Americans, extended family and church members are some of the most important support network members (Nguyen, Chatters, & Taylor, 2016; Taylor et al., 2004). They are critical sources of social support and stress coping resources. African American households often include extended family members (Taylor et al., 2003). Moreover, African Americans report that they have frequent contact with their extended family, at least once a week on average, and have very high levels of emotional closeness to extended family (Cross et al., 2018). Moreover, supportive exchanges are very common between extended family members (Cross et al., 2018).

Church members are another unique source of support for African Americans. Congregational relationships are particularly meaningful for African Americans, given the prominence of religion and the church in African American communities (Frazier & Lincoln, 1974; Lincoln & Mamiya, 1990). Church relationships are social relationships that are embedded within a religious and spiritual community that share similar beliefs and values. These shared beliefs and values give added meaning to congregational relationships. On average, African American adults report relatively frequent contact with church members and relatively high levels of emotional support from congregants (Nguyen, Taylor, et al., 2019). Additionally, African Americans report receiving moderate to high levels of social support from fellow congregants (Nguyen, Taylor, et al., 2019; Taylor et al., 2017).

Purpose of the present study

While previous studies have extensively documented the impact of chronic stress on sleep, less is known regarding the role of social support in this relationship. Of the studies that have investigated the role of social support in the relationship between chronic stress and sleep, very few have focused specifically on minoritized populations. A better understanding of factors that might mitigate the effects of chronic stress on sleep in this population is particularly important, as African Americans are disproportionately exposed to chronic stressors (e.g., financial strain, discrimination) (Brown et al., 2020). An additional knowledge gap in this area is the lack of focus on church-based social support. To our knowledge, the impact of social support from church members on the relationship between chronic stress and sleep has yet to be examined. Church members are meaningful social partners for African Americans and social support from church members may play a unique role in the connection between chronic stress and sleep. Adjacently, research on religious participation and sleep demonstrates that religious service attendance and consumption of religious television programs are related to restless sleep and sleep satisfaction among older African American adults and are important coping resources for this population (Nguyen et al., 2022). Additionally, empirical evidence indicates that religious service attendance can mitigate the harmful effects of work-related stress on insomnia symptoms among older Black adults (Frazier & Brown, 2022). Thus, it is possible that church members can buffer against the detriment of chronic stress exposure.

To address these knowledge gaps, this study examined whether social support from extended family and church members moderated the association between chronic stress exposure and subjective sleep quality in a nationally representative sample of African American adults. Specifically, we examined subjective ratings of sleep satisfaction, trouble falling asleep, and restless sleep.

As demonstrated in our review of the extant literature, sleep quality is assessed with a broad range of indicators, including the ones this study utilizes as well as sleep latency, sleep duration, sleep efficiency, and daytime dysfunction, to name a few. We elected to use subjective ratings of sleep satisfaction, trouble falling asleep, and restless sleep to assess subjective sleep quality because these three sleep quality indicators are included in some of the most commonly employed and psychometrically sound scales of sleep quality in survey research, such as the Pittsburgh Sleep Quality Index and the Jenkins Sleep Scale (Buysse et al., 1989; Fabbri et al., 2021; Jenkins et al., 1988). Moreover, because sleep quality was not a focus of the National Survey of American Life-Reinterview, sleep quality measures were limited in this data set. Thus, given data availability and precedence in prior research of sleep quality, this study focused on subjective ratings of sleep satisfaction, trouble falling asleep, and restless sleep as indicators of subjective sleep quality.

Methods

Sample

The African American sample for the current analysis was drawn from the National Survey of American Life: Coping with Stress in the 21st Century-Reinterview (NSAL-RIW). The original NSAL and NSAL-RIW were collected by the Program for Research on Black Americans (PRBA) at the University of Michigan’s Institute for Social Research using a national multistage probability design. The African American sample is a nationally representative sample of households located in the 48 coterminous states with at least one Black adult aged 18 years or older who did not identify ancestral ties in the Caribbean. The data collection was conducted from February 2001 to June 2003. See Jackson et al., 2004 for a more detailed discussion of the NSAL sample.

At the completion of the initial NSAL interview, respondents were given a paper, self-administered questionnaire; this is referred to as the NSAL-RIW. The NSAL-RIW included measures that were not in the original survey, such as several sleep measures. Respondents were instructed to complete the self-administered questionnaire and mail it back to the PRBA. Of the 3,570 African American respondents who completed the NSAL, a total of 2,137 African American respondents completed the NSAL-RIW. Respondents who completed the NSAL-RIW differed from respondents who did not complete the NSAL-RIW on several sociodemographic characteristics. Respondents who completed the NSAL-RIW were more likely to be women, be older, and have lower family incomes. Like the NSAL sample, the NSAL-RIW sample is a nationally representative sample. The analytic sample for this study included 1,372 African American adults aged 18 and older who attended religious services at least a few times a year, as the church-based relationship measures were only assessed for these individuals. We used listwise deletion to handle missing data. After listwise deletion, 1299 cases remained. The NSAL and NSAL-RIW data collections were approved by the University of Michigan Institutional Review Board.

Measures

Independent Variables

Chronic stress exposure was assessed with a 9-item index that asked respondents whether, in the past month, they had experienced problems related to their health, finances, job, family or marriage, love life, gambling, or the police, been treated badly because of their race, or they or their family had been the victim of a crime. A summary score was derived by adding together all events that were endorsed by respondents. Higher scores indicated higher levels of chronic stress exposure.

Extended family relationship variables included frequency of contact with family members, emotional support from family, and overall social support from family. Frequency of contact was assessed with the question, “How often do you see, write or talk on the telephone with family or relatives who do not live with you?” Possible response values ranged from 1 (never) to 7 (everyday). Emotional support from family was assessed with a series of three questions, “Other than your spouse/partner, how often do your family members a) make you feel loved and cared for; b) listen to you talk about your private problems and concerns; and c) express interest and concern in your well-being?” Possible response values ranged from 1 (never) to 4 (very often). A mean family emotional support score was derived; higher scores indicated higher levels of family emotional support. The scale reliability for this measure was α=.73. Overall social support from family was assessed by the question, “How often do people in your family—including children, grandparents, aunts, uncles, in-laws and so on—help you out? Would you say very often (4), fairly often (3), not too often (2), or never (1)?” Respondents who indicated that they had no family or never needed help were omitted from the analysis.

Church attendance was assessed with two questions. A variable that assessed whether or not respondents attended religious services – “Other than for weddings or funerals, have you attended services at a church or other place of worship since you were 18 years old?” – was combined with a variable that assessed frequency of church attendance. The resulting church attendance variable had six categories: never, less than once a year, a few times a year, a few times a month, at least once a week, and nearly everyday. Church-based relationship variables, which we will describe next, were only assessed for respondents who indicated that they attended church at least a few times a year; thus, respondents who reported that they never go to church or attended less than once a year were not asked these church-based relationship questions. This is common practice in survey research on church-based relationships. Consequently, the sample for the current analysis was restricted to only those who reported that they attended church a few times a year or more often than that. Church-based relationship variables included frequency of contact with church members, emotional support from church members, and overall social support from church members. These variables were measured by questions like the ones that assessed family relationships. The scale reliability for the church-based emotional support measure was α=.71.

Dependent Variables

Sleep quality measures included sleep satisfaction, trouble falling asleep, and restless sleep. Sleep satisfaction was measured with the question, “How satisfied are you with your sleep?” Possible response values ranged from 1 (very dissatisfied) to 4 (very satisfied). Restless sleep was measured by presenting the respondents with the sentence, “My sleep was restless” and asking respondents to indicate the frequency with which they experienced the condition within the past week. Response values ranged from 0 (rarely or none of the time) to 3 (most or all of the time). Trouble falling asleep was measured with the question, “Do you usually have trouble falling asleep?” The response format for this question was yes/no.

Covariates

The multivariable analyses accounted for age (in years), education (in years), family income (in dollars and log transformed), gender (male or female), marital status (married/cohabiting or separated/divorced/widowed/never married), region (Northeast, North Central, South, West), urbanicity (urban, suburban, rural) chronic physical health conditions (count of self-reported, physician-diagnosed physical health conditions), body mass index (BMI), physical activity (score derived from a 3-item index), and self-rated mental health (range: 1–5).

Analysis strategy

OLS regression analyses were performed to determine the associations between chronic stress exposure, social relationships with family and church members, and continuous sleep quality variables (i.e., sleep satisfaction, restless sleep). Logistic regression analyses were performed to determine the associations between chronic stress exposure, social relationships, and trouble falling asleep. To determine the moderating effect of the social relationship variables, interaction terms between chronic stress exposure and each of the four relationship variables were created. Only significant interactions were retained for the final models. For ease of interpretation, instead of centering or standardizing variables, significant interactions were plotted using the margins and marginsplot commands in Stata. Emotional support from family was depicted as low and high in the interaction plots. Low and high emotional support were represented by respondents with emotional support scores of 1 standard deviation below and above the mean, respectively. For each sleep outcome variable, the first model included only the main effects, and subsequent models included the addition of interaction terms. All multivariate analyses took into account the complex multistage clustered design of the NSAL-RIW sample, unequal probabilities of selection, nonresponse, and poststratification. All analyses were conducted in Stata 15.1.

Supplemental Table 1 presents intercorrelations between the study variables. Given the presence of several relatively large correlations in this table, we checked for multicollinearity by computing variance inflation factors (VIF). All VIF values were equal to 1.64 or lower, indicating the absence of multicollinearity.

Results

Table 1 presents descriptive statistics for the sample. Over half of the sample were women (60%), and the mean age was 44 years. On average, respondents had a little more than 12 years of formal education (a little over high school), and the mean family income was $30,460. Close to two out of five respondents were either married or cohabiting. More than half of respondents resided in the South (59%), and over four in five respondents resided in an urban area (86%). Regarding health measures, respondents reported a mean of 1.5 chronic health conditions (range 0–13). The mean BMI was 29.5, and respondents reported a moderate level of physical activity (2.7; range 1–4). The mean self-rated mental health score was moderately high (3.8; range 1–5). The mean church attendance rate was also relatively high (4.3; range 3–6). Frequency of contact with and overall support from church members were moderate, and emotional support from church members was relatively high. With respect to extended family relationships, respondents indicated high levels of contact with and emotional and overall support from family members. Respondents reported an average of 1.7 chronic stress events (range 0–8). Regarding the sleep quality measures, respondents indicated relatively high satisfaction with their sleep. The majority of the sample (71%) reported that they did not have trouble falling asleep, and the frequency of restless sleep was very low.

Table 1.

Demographic Characteristics of the Sample and Distribution of Study Variables

	N (%)	Mean (S.D.)	Min	Max
Gender
Men	406 (39.98)
Women	966 (60.02)
Age		43.86 (15.90)	18	90
Marital status
Separated/divorced/widowed/never married	905 (57.94)
Married/cohabiting	464 (42.06)
Region
Northeast	150 (14.78)
Midwest	230 (17.56)
South	909 (59.42)
West	83 (8.25)
Urbanicity
Urban	1,163 (86.25)
Suburban	91 (5.68)
Rural	118 (8.07)
Education		12.46 (2.51)	0	17
Family income		30,460.10 (31,313.26)	0	520,000
Chronic health conditions		1.49 (1.72)	0	13
BMI		29.48 (6.84)	16.83	61.40
Physical activity		2.67 (0.77)	1	4
Self-rated mental health		3.79 (1.04)	1	5
Church attendance		4.32 (0.91)	3	6
Frequency of contact with church members		3.83 (1.82)	1	6
Emotional support from church members		3.01 (0.81)	1	4
Overall support from church members		2.41 (1.09)	1	4
Frequency of contact with family		6.19 (1.21)	1	7
Emotional support from family		3.27 (0.71)	1	4
Overall support from family		2.82 (0.98)	1	4
Chronic stress exposure		1.70 (1.41)	0	8
Restless sleep		0.77 (1.01)	0	3
Sleep satisfaction		3.05 (0.87)	1	4
Trouble falling asleep
No	914 (71.36)
Yes	442 (28.64)

Percents and N are presented for categorical variables and Means and Standard Deviations are presented for continuous variables. Percentages are weighted and frequencies are un-weighted.

Table 2 presents the multivariable OLS regressions for sleep satisfaction. The baseline model for sleep satisfaction (Model 1) indicated that chronic stress exposure was associated with sleep satisfaction. That is, for every 1 unit increase in chronic stress exposure, sleep satisfaction decreased by .08. Additionally, church attendance was positively associated with sleep satisfaction; for every 1 unit increase in church attendance, sleep satisfaction increased by .12. We did not identify significant interactions between the family and church variables and stress exposure.

Table 2.

OLS regression analyses of sleep satisfaction among African Americans

	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8
	B (SE)	B (SE)	B (SE)	B (SE)	B (SE)	B (SE)	B (SE)	B (SE)
Chronic stress exposure	−0.08 (0.02)***	−0.24 (0.13)	−0.21 (0.10)*	−0.05 (0.07)	−0.08 (0.05)	−0.18 (0.07)*	−0.11 (0.05)	−0.01 (0.11)
Frequency of contact with family	−0.00 (0.03)	−0.05 (0.05)	−0.00 (0.03)	−0.00 (0.03)	−0.00 (0.03)	−0.00 (0.03)	−0.00 (0.03)	−0.00 (0.03)
Emotional support from family	0.06 (0.04)	0.06 (0.04)	−0.01 (0.07)	0.06 (0.04)	0.06 (0.04)	0.06 (0.04)	0.06 (0.04)	0.06 (0.04)
Overall support from family	0.03(0.03)	0.03 (0.03)	0.03 (0.03)	0.04 (0.05)	0.03 (0.03)	0.03 (0.03)	0.03 (0.03)	0.03 (0.03)
Frequency of contact with church members	−0.00 (0.02)	−0.00 (0.02)	−0.00 (0.02)	−0.00 (0.02)	−0.00 (0.03)	−0.00 (0.02)	−0.00 (0.02)	−0.00 (0.02)
Emotional support from church members	0.04 (0.06)	0.04 (0.05)	0.04 (0.05)	0.04 (0.06)	0.04 (0.06)	−0.02 (0.08)	0.04 (0.05)	0.04 (0.05)
Overall support from church members	−0.01 (0.03)	−0.01 (0.03)	−0.01(0.03)	−0.01 (0.03)	−0.01 (0.03)	−0.01 (0.03)	−0.03 (0.04)	−0.01 (0.03)
Church attendance	0.12 (0.04)**	0.13 (0.04)**	0.12 (0.04)**	0.12 (0.04)**	0.12 (0.04)**	0.12 (0.04)**	0.12 (0.04)**	0.15 (0.06)*
Gender
Mena	0	0	0	0	0	0	0	0
Women	−0.02 (0.06)	−0.02 (0.06)	−0.01(0.06)	−0.02 (0.06)	−0.02 (0.06)	−0.02 (0.06)	−0.02 (0.06)	−0.02 (0.06)
Age	0.00 (0.00)	0.00 (0.00)	0.00 (0.00)	0.00 (0.00)	0.00 (0.00)	0.00 (0.00)	0.00 (0.00)	0.00 (0.00)
Marital status
Separated/divorced/widowed/never marrieda	0	0	0	0	0	0	0	0
Married/cohabiting	0.07 (0.06)	0.07 (0.06)	0.07 (0.06)	0.07 (0.06)	0.07 (0.06)	0.07 (0.06)	0.07 (0.06)	0.07 (0.06)
Region
Southa	0	0	0	0	0	0	0	0
Northeast	−0.07 (0.08)	−0.08 (0.08)	−0.07 (0.08)	−0.07 (0.08)	−0.07 (0.08)	−0.07 (0.08)	−0.07 (0.08)	−0.07 (0.09)
North Central	−0.02 (0.09)	−0.03 (0.09)	−0.03 (0.09)	−0.02 (0.09)	−0.02 (0.09)	−0.03 (0.09)	−0.02 (0.09)	−0.02 (0.09)
West	−0.13 (0.16)	−0.14 (0.16)	−0.13 (0.16)	−0.14 (0.16)	−0.13 (0.16)	−0.14 (0.16)	−0.14 (0.16)	−0.13 (0.16)
Urbanicity
Urbana	0	0	0	0	0	0	0	0
Suburban	0.08 (0.10)	0.08 (0.10)	0.08 (0.10)	0.08 (0.10)	0.08 (0.10)	0.08 (0.10)	0.08 (0.10)	0.08 (0.10)
Rural	0.19(0.10)	0.18 (0.10)	0.19 (0.10)	0.19 (0.10)	0.19 (0.10)	0.18 (0.10)	0.18 (0.10)	0.19 (0.10)
Education	−0.01 (0.01)	−0.01 (0.01)	−0.01 (0.01)	−0.01 (0.10)	−0.01 (0.10)	−0.01 (0.01)	−0.01 (0.01)	−0.01 (0.01)
Family income	−0.05 (0.04)	−0.05 (0.04)	−0.05 (0.04)	−0.05 (0.04)	−0.05 (0.04)	−0.05 (0.04)	−0.05 (0.04)	−0.05 (0.04)
Chronic health conditions	−0.06 (0.02)**	−0.06 (0.02)**	−0.06 (0.02)**	−0.06 (0.02)**	−0.06 (0.02)**	−0.06 (0.02)**	−0.06 (0.02)**	−0.06 (0.02)*
BMI	0.00 (0.00)	0.00 (0.00)	0.00 (0.00)	0.00 (0.00)	0.00 (0.00)	0.00 (0.00)	0.00 (0.00)	0.00 (0.00)
Physical activity	−0.00 (0.03)	−0.00 (0.03)	−0.00 (0.03)	−0.00 (0.03)	−0.00 (0.03)	0.00 (0.03)	−0.00 (0.03)	−0.00 (0.03)
Self-rated mental health	0.16 (0.03)***	0.15 (0.03)***	0.15 (0.03)***	0.16 (0.03)***	0.16 (0.03)***	0.15 (0.03)***	0.16 (0.03)***	0.16 (0.03)***
Frequency of contact with family*chronic stress exposure	--	0.03 (0.02)	--	--	--	--	--	--
Emotional support from family*chronic stress exposure	--	--	0.04 (0.03)	--	--	--	--	--
Overall support from family*chronic stress exposure	--	--	--	−0.01 (0.02)	--	--	--	--
Frequency of contact with church members*chronic stress exposure	--	--	--	--	0.00 (0.01)	--	--	--
Emotional support from church members*chronic stress exposure	--	--	--	--	--	0.03 (0.02)	--	--
Overall support from church members*chronic stress exposure	--	--	--	--	--	--	0.01 (0.02)	--
Church attendance*chronic stress exposure	--	--	--	--	--	--	--	−0.02 (0.02)
Intercept	2.11 (0.47)***	2.46 (0.61)***	2.34 (0.51)***	2.06 (0.50)***	2.11 (0.48)***	2.31 (0.49)***	2.16 (0.48)***	1.98 (0.49)***
N	1,229	1,229	1,229	1,229	1,229	1,229	1,229	1,229

B=regression coefficient; SE=standard error

^aReference category.

^*p < .05;

^**p< .01;

^***p< .001

Table 3 presents the multivariable logistic regressions for trouble falling asleep. The baseline model (Model 1) demonstrated an association between chronic stress exposure and trouble falling asleep, with every 1 unit increase in chronic stress exposure accompanied by 25% increase in the odds of experiencing trouble falling asleep. A significant interaction between emotional support from family and chronic stress exposure (Model 3) indicated that emotional support from family moderated the relationship between stress exposure and trouble falling asleep (Figure 1). Specifically, the positive association between chronic stress exposure and trouble falling asleep was attenuated among respondents who received high levels of emotional support from their family. Thus, the probability of having trouble falling asleep was similar between respondents with high and low levels of family support at low levels of stress exposure. However, at high levels of stress exposure, respondents who received high levels of family support had substantially lower probabilities of experiencing trouble falling asleep compared to respondents who received low levels of family support. We used the Johnson-Neyman (J-N) technique to further explore the point at which the moderating effect of family emotional support became significant. The J-N interaction plot for this significant interaction effect is depicted in Figure 2. The blue line represents the regression slope estimate, and the grey lines represent lower- and upper-level 95% confidence intervals. Confidence interval areas above and below 0 represent regions of significance. The region of significance in Figure 2 indicates that if the family emotional support score is below approximately 3.85, then the effect of chronic stress exposure on trouble falling asleep can be significantly attenuated by family emotional support. No other interaction terms achieved statistical significance in the analyses for trouble falling asleep.

Table 3.

Logistic regression analyses of trouble falling asleep among African Americans

	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8
	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)
Chronic stress exposure	1.25*** (1.13, 1.40)	1.46 (0.78, 2.72)	2.21** (1.37, 3.57)	1.45 (0.92, 2.30)	1.10 (0.86, 1.41)	1.21 (0.82, 1.80)	1.07 (0.81, 1.41)	1.08 (0.60, 1.95)
Frequency of contact with family	0.96 (0.83, 1.11)	1.01 (0.82, 1.24)	0.97 (0.84, 1.12)	0.96 (0.84, 1.11)	0.96 (0.83, 1.11)	0.96 (0.83, 1.11)	0.96 (0.83, 1.11)	0.96 (0.83, 1.11)
Emotional support from family	0.97 (0.78, 1.20)	0.97 (0.78, 1.20)	1.34 (0.91, 1.96)	0.97 (0.78, 1.19)	0.97 (0.78, 1.20)	0.97 (0.78, 1.20)	0.96 (0.78, 1.19)	0.97 (0.78, 1.20)
Overall support from family	0.91 (0.76, 1.08)	0.91 (0.76, 1.08)	0.90 (0.76, 1.07)	0.99 (0.74, 1.34)	0.91 (0.77, 1.09)	0.91 (0.76, 1.08)	0.91 (0.77, 1.09)	0.91 (0.77, 1.08)
Frequency of contact with church members	1.00 (0.90, 1.12)	1.00 (0.90, 1.12)	1.00 (0.90, 1.12)	1.00 (0.90, 1.12)	0.93 (0.80, 1.10)	1.00 (0.90, 1.12)	1.00 (0.90, 1.12)	1.00 (0.90, 1.12)
Emotional support from church members	0.87 (0.69, 1.09)	0.87 (0.69, 1.09)	0.86 (0.68, 1.09)	0.86 (0.69, 1.09)	0.86 (0.68, 1.09)	0.85 (0.59, 1.22)	0.86 (0.68, 1.09)	0.87 (0.69, 1.09)
Overall support from church members	1.07 (0.89, 1.29)	1.07 (0.89, 1.28)	1.07 (0.90, 1.28)	1.07 (0.89, 1.28)	1.08 (0.90, 1.29)	1.07 (0.89, 1.29)	0.96 (0.71, 1.28)	1.07 (0.89, 1.29)
Church attendance	0.94 (0.72, 1.22)	0.94 (0.71, 1.22)	0.94 (0.72, 1.22)	0.93 (0.71, 1.22)	0.94 (0.72, 1.23)	0.94 (0.72, 1.22)	0.94 (0.72, 1.22)	0.88 (0.61, 1.26)
Gender
Mena	1	1	1	1	1	1	1	1
Women	1.83** (1.23, 2.71)	1.82** (1.22, 2.71)	1.77** (1.18, 2.66)	1.82** (1.23, 2.71)	1.84** (1.23, 2.75)	1.82** (1.23, 2.71)	1.83** (1.23, 2.72)	1.83** (1.23, 2.72)
Age	0.99 (0.98, 1.01)	0.99 (0.98, 1.01)	0.99 (0.98, 1.01)	0.99 (0.98, 1.01)	0.99 (0.98, 1.01)	0.99 (0.98, 1.01)	0.99 (0.98, 1.01)	0.99 (0.98, 1.01)
Marital status
Separated/divorced/widowed/never marrieda	1	1	1	1	1	1	1	1
Married/cohabiting	1.00 (0.65, 1.53)	1.00 (0.65, 1.53)	0.99 (0.65, 1.52)	0.99 (0.65, 1.52)	1.00 (0.65, 1.52)	1.00 (0.65, 1.52)	0.99 (0.65, 1.51)	1.00 (0.65, 1.53)
Region
Southa	1	1	1	1	1	1	1	1
Northeast	0.74 (0.48, 1.14)	0.75 (0.49, 1.14)	0.76 (0.50, 1.14)	0.75 (0.49, 1.15)	0.73 (0.48, 1.13)	0.74 (0.48, 1.14)	0.74 (0.48, 1.14)	0.74 (0.48, 1.15)
North Central	0.90 (0.60, 1.35)	0.91 (0.61, 1.35)	0.92 (0.61, 1.38)	0.90 (0.60, 1.35)	0.88 (0.59, 1.33)	0.90 (0.60, 1.35)	0.90 (0.60, 1.35)	0.91 (0.60, 1.36)
West	1.61 (0.57, 4.55)	1.62 (0.57, 4.59)	1.62 (0.58, 4.55)	1.61 (0.57, 4.52)	1.56 (0.56, 4.31)	1.61 (0.57, 4.53)	1.59 (0.56, 4.49)	1.61 (0.58, 4.50)
Urbanicity
Urbana	1	1	1	1	1	1	1	1
Suburban	1.30 (0.87, 1.93)	1.30 (0.88, 1.92)	1.27 (0.85, 1.90)	1.30 (0.88, 1.93)	1.28 (0.86, 1.92)	1.30 (0.88, 1.93)	1.32 (0.89, 1.96)	1.29 (0.87, 1.93)
Rural	0.91 (0.52, 1.60)	0.92 (0.53, 1.60)	0.91 (0.51, 1.60)	0.93 (0.54, 1.61)	0.91 (0.51, 1.61)	0.91 (0.52, 1.61)	0.89 (0.51, 1.56)	0.91 (0.52, 1.58)
Education	0.98 (0.91, 1.05)	0.98 (0.91, 1.05)	0.98 (0.91, 1.05)	0.98 (0.91, 1.05)	0.98 (0.91, 1.05)	0.98 (0.91, 1.05)	0.98 (0.91, 1.05)	0.98 (0.91, 1.05)
Family income	0.85* (0.74, 0.98)	0.85* (0.73, 0.98)	0.85* (0.73, 0.98)	0.85* (0.74, 0.98)	0.85* (0.74, 0.98)	0.85* (0.74, 0.98)	0.85* (0.74, 0.98)	0.85* (0.74, 0.98)
Chronic health conditions	1.16* (1.04, 1.29)	1.16** (1.04, 1.29)	1.16** (1.04, 1.28)	1.16** (1.04, 1.29)	1.15* (1.03, 1.29)	1.16* (1.04, 1.29)	1.16* (1.04, 1.29)	1.15* (1.03, 1.29)
BMI	1.00 (0.98, 1.03)	1.00 (0.98, 1.03)	1.00 (0.98, 1.02)	1.00 (0.98, 1.03)	1.00 (0.98, 1.03)	1.00 (0.98, 1.03)	1.00 (0.98, 1.03)	1.00 (0.98, 1.03)
Physical activity	1.03 (0.85, 1.25)	1.03 (0.85, 1.25)	1.03 (0.85, 1.25)	1.03 (0.85, 1.26)	1.03 (0.85, 1.25)	1.03 (0.85, 1.25)	1.04 (0.85, 1.26)	1.03 (0.85, 1.25)
Self-rated mental health	0.77** (0.66, 0.90)	0.77** (0.66, 0.90)	0.78** (0.67, 0.91)	0.78** (0.66, 0.91)	0.77** (0.65, 0.90)	0.77** (0.66, 0.90)	0.77** (0.65, 0.90)	0.77** (0.65, 0.90)
Frequency of contact with family*chronic stress exposure	--	0.98 (0.88, 1.07)	--	--	--	--	--	--
Emotional support from family*chronic stress exposure	--	--	0.84* (0.73, 0.97)	--	--	--	--	--
Overall support from family*chronic stress exposure	--	--	--	0.95 (0.82, 1.10)	--	--	--	--
Frequency of contact with church members*chronic stress exposure	--	--	--	--	1.04 (0.98, 1.11)	--	--	--
Emotional support from church members*chronic stress exposure	--	--	--	--	--	1.01 (0.90, 1.13)	--	--
Overall support from church members*chronic stress exposure	--	--	--	--	--	--	1.07 (0.96, 1.19)	--
Church attendance*chronic stress exposure	--	--	--	--	--	--	--	1.04 (0.91, 1.18)
Intercept	6.20 (0.81, 47.62)	4.47 (0.68, 29.28)	2.07 (0.19, 22.79)	4.44 (0.49, 40.43)	7.86* (1.01, 61.02)	6.62 (0.70, 62.41)	7.93* (1.02, 61.70)	8.41 (0.60, 118.43)
N	1,229	1,229	1,229	1,229	1,229	1,229	1,229	1,229

OR=odds ratio; 95% CI=95% confidence interval

^aReference category.

^*p < .05;

^**p< .01;

^***p< .001

Figure 1.

Predicted probability of trouble falling asleep by chronic stress exposure and emotional support from family among African Americans.

Figure 2.

Johnson-Neyman interaction plot for emotional family support*chronic stress exposure predicting trouble falling asleep.

Table 4 presents the OLS regressions for restless sleep. The baseline model (Model 1) revealed that respondents who reported higher levels of chronic stress exposure were more likely to experience restless sleep relative to respondents who reported lower levels of chronic stress exposure. For every 1 unit increase in chronic stress exposure, restless sleep increased by .15. Additionally, church attendance was associated with decreased likelihood of restless sleep; for every 1 unit increase in church attendance, restless sleep decreased by .08. A significant interaction between emotional support from family and chronic stress exposure (Model 3) indicated that receiving emotional support from family members moderated the relationship between stress exposure and restless sleep (Figure 3). The magnitude of the positive association between chronic stress exposure and restless sleep was weaker among respondents with high family emotional support compared to respondents with low family emotional support. Consequently, at the highest levels of stress exposure, people who had high family emotional support reported substantially less frequent restless sleep relative to people who had low family emotional support. Figure 4 presents the J-N interaction plot for this statistically significant interaction. The J-N interaction plot indicates that the region of significance spans the value of family emotional support, from 1 to 4, indicating that the moderating impact of family emotional support on the effect of chronic stress exposure on restless sleep was significant, regardless of family emotional support score. No other interaction terms achieved statistical significance in the analyses for restless sleep.

Table 4.

OLS regression analyses of restless sleep among African Americans

	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8
	B (SE)	B (SE)	B (SE)	B (SE)	B (SE)	B (SE)	B (SE)	B (SE)
Chronic stress exposure	0.15 (0.03)***	0.15 (0.09)	0.42 (0.09)***	0.16 (0.06)*	0.17 (0.05)**	0.13 (0.07)	0.15 (0.05)**	0.20 (0.11)
Frequency of contact with family	0.03 (0.03)	0.03 (0.03)	0.04 (0.03)	0.03 (0.03)	0.03 (0.03)	0.03 (0.03)	0.03 (0.03)	0.03 (0.03)
Emotional support from family	−0.07 (0.04)	−0.07 (0.04)	0.07 (0.06)	−0.07 (0.04)	−0.07 (0.04)	−0.07 (0.04)	−0.07 (0.04)	−0.07 (0.04)
Overall support from family	−0.02 (0.03)	−0.02 (0.03)	−0.03 (0.03)	−0.02 (0.03)	−0.02 (0.03)	−0.02 (0.03)	−0.02 (0.03)	−0.02 (0.03)
Frequency of contact with church members	−0.00 (0.02)	−0.00 (0.02)	−0.00 (0.02)	−0.00 (0.02)	0.00 (0.02)	−0.00 (0.02)	−0.00 (0.02)	−0.00 (0.02)
Emotional support from church members	0.02 (0.04)	0.02 (0.04)	0.02 (0.04)	0.02 (0.04)	0.02 (0.04)	0.01 (0.05)	0.02 (0.05)	0.02 (0.04)
Overall support from church members	0.03 (0.03)	0.03 (0.03)	0.03 (0.03)	0.03 (0.03)	0.03 (0.03)	0.03 (0.03)	0.03 (0.03)	0.03 (0.03)
Church attendance	−0.08 (0.03)*	−0.08 (0.03)*	−0.08 (0.03)*	−0.08 (0.03)*	−0.08 (0.03)*	−0.08 (0.03)*	−0.08 (0.03)*	−0.06 (0.05)
Gender
Mena	0	0	0	0	0	0	0	0
Women	0.10 (0.07)	0.10 (0.07)	0.08 (0.07)	0.10 (0.07)	0.10 (0.07)	0.10 (0.07)	0.10 (0.07)	−0.10 (0.07)
Age	−0.00 (0.00)	−0.00 (0.00)	−0.00 (0.00)	−0.00 (0.00)	−0.00 (0.00)	−0.00 (0.00)	−0.00 (0.00)	−0.00 (0.00)
Marital status
Separated/divorced/widowed/never marrieda	0	0	0	0	0	0	0	0
Married/cohabiting	−0.09 (0.05)	−0.09 (0.05)	−0.10 (0.05)	−0.09 (0.05)	−0.09 (0.05)	−0.09 (0.05)	−0.09 (0.05)	−0.09 (0.05)
Region
Southa	0	0	0	0	0	0	0	0
Northeast	−0.00 (0.13)	−0.00 (0.13)	0.01 (0.13)	−0.00 (0.13)	−0.00 (0.13)	−0.00 (0.13)	−0.00 (0.13)	−0.00 (0.13)
North Central	−0.03(0.08)	−0.03 (0.08)	−0.03 (0.08)	−0.03 (0.08)	−0.03 (0.08)	−0.03 (0.08)	−0.03 (0.08)	−0.03 (0.08)
West	0.05 (0.11)	0.05 (0.11)	0.05 (0.11)	0.05 (0.11)	0.06 (0.11)	0.05 (0.11)	0.05 (0.11)	0.05 (0.11)
Urbanicity
Urbana	0	0	0	0	0	0	0	0
Suburban	0.08 (0.09)	0.08 (0.09)	0.07 (0.09)	0.08 (0.09)	0.08 (0.09)	0.08 (0.09)	0.08 (0.09)	0.08 (0.09)
Rural	−0.06 (0.11)	−0.06 (0.11)	−0.06 (0.11)	−0.06 (0.11)	−0.06 (0.11)	−0.06 (0.11)	−0.06 (0.11)	−0.06 (0.11)
Education	−0.01 (0.01)	−0.01 (0.01)	−0.01 (0.01)	−0.01 (0.01)	−0.01 (0.01)	−0.01 (0.01)	−0.01 (0.01)	−0.01 (0.01)
Family income	−0.03 (0.05)	−0.03 (0.05)	−0.03 (0.05)	−0.03 (0.05)	−0.03 (0.05)	−0.03 (0.05)	−0.03 (0.05)	−0.03 (0.05)
Chronic health conditions	0.02 (0.02)	0.02 (0.02)	0.02 (0.02)	0.02 (0.02)	0.02 (0.02)	0.02 (0.02)	0.02 (0.02)	0.02 (0.02)
BMI	0.00 (0.01)	0.00 (0.01)	0.00 (0.01)	0.00 (0.01)	0.00 (0.01)	0.00 (0.01)	0.00 (0.01)	0.00 (0.01)
Physical activity	0.00 (0.04)	0.00 (0.04)	0.00 (0.04)	0.00 (0.04)	0.00 (0.04)	0.00 (0.04)	0.00 (0.04)	0.00 (0.04)
Self-rated mental health	−0.16 (0.03)***	−0.16 (0.03)***	−0.16 (0.04)***	−0.16 (0.04)***	−0.16 (0.04)***	−0.16 (0.04)***	−0.16 (0.04)***	−0.16 (0.04)***
Frequency of contact with family*chronic stress exposure	--	−0.00 (0.01)	--	--	--	--	--	--
Emotional support from family*chronic stress exposure	--	--	−0.08 (0.03)**	--	--	--	--	--
Overall support from family*chronic stress exposure	--	--	--	−0.00 (0.02)	--	--	--	--
Frequency of contact with church members*chronic stress exposure	--	--	--	--	−0.00 (0.01)	--	--	--
Emotional support from church members*chronic stress exposure	--	--	--	--	--	0.01 (0.02)	--	--
Overall support from church members*chronic stress exposure	--	--	--	--	--	--	−0.00 (0.01)	--
Church attendance*chronic stress exposure	--	--	--	--	--	--	--	−0.01 (0.02)
Intercept	1.84 (0.57)**	1.84 (0.59)**	1.37 (0.60)*	1.84 (0.59)**	1.82 (0.58)**	1.88 (0.58)**	1.84 (0.58)**	1.75 (0.63)**
N	1,229	1,229	1,229	1,229	1,229	1,229	1,229	1,229

B=regression coefficient; SE=standard error

^aReference category.

^*p < .05;

^**p< .01;

^***p< .001

Figure 3.

Predicted value of restless sleep by chronic stress exposure and emotional support from family among African Americans.

Figure 4.

Johnson-Neyman interaction plot for emotional family support*chronic stress exposure predicting restless sleep.

The church relationship variables were not associated with the three measures of sleep quality. To determine whether the extended family relationship variables may have explained a large proportion of the variance, rendering the church relationship variables non-significant, we conducted separate regression models for the extended family and church variables (Supplement Tables 2–4). These models indicated that the church-based relationship variables were not associated with sleep quality.

Discussion

This study investigated the interrelations between extended family and church-based relationships, chronic stress exposure, and sleep quality in a nationally representative sample of African American adults. Our findings demonstrated that chronic stress exposure was associated with all three sleep quality variables—sleep satisfaction, trouble falling asleep, and restless sleep. We also found that receiving emotional support from extended family members and frequent contact with church members were linked to decreased frequency of restless sleep. Further, receiving emotional support from family members moderated the associations between chronic stress exposure and two sleep quality variables.

The data indicated that African American adults who reported greater chronic stress exposure experienced lower sleep satisfaction, more trouble falling asleep, and more frequent restless sleep. These findings are in concordance with extant evidence. Studies of chronic stress and sleep among African Americans demonstrate that greater chronic stress exposure and perceived chronic stress lead to compromised sleep (Bidulescu et al., 2010; Johnson et al., 2016; Sims et al., 2017). The fact that our findings demonstrated that chronic stress exposure was related to all three sleep quality variables underscores the consistent deleterious impact of chronic stress exposure on sleep quality.

Additionally, church attendance was associated with sleep satisfaction and restless sleep. African Americans who attended church more frequently were more likely to report higher levels of sleep satisfaction and less likely to report restless sleep compared to African Americans who attended church less frequently. This is consistent with extant research, demonstrating the salubrious effect of church attendance on sleep health among older African Americans (Hill et al., 2006; Hill et al., 2020; Nguyen et al., 2022). For example, Nguyen et al.’s (2022) research showed that relative to older African Americans who never attended church, older African Americans who attended church less than once a year, at least once a week, or nearly everyday were more satisfied with their sleep. These findings suggest that church attendance may protect against sleep problems.

Interestingly, the current analysis indicated that the extended family and church-based relationship variables were not directly associated with any of the three sleep quality outcomes. It is possible that this is due to suppressor variables that were not accounted for in the analyses. Suppressor variables are variables that are not correlated or weekly correlated with the outcome variable, but substantially correlated with predictor variable(s) in the model (Pandey & Elliott, 2010). The inclusion of suppressor variables can increase the overall model predictability and the predictor’s regression weight (Pandey & Elliott, 2010). Thus, it is possible that in our current analyses, we may have unknowingly omitted important suppressor variable(s) that would have removed irrelevant predictive variance from the social relationship variables; the omission of potential suppressor variable(s) in our analyses may have rendered the social relationship variables non-significant. We recommend that future research explore potential suppressor variables (variables that are correlated with social relationships but weakly or not correlated with sleep quality) to fully understand the association between social relationships and sleep quality.

The significant interactions between chronic stress exposure and emotional support from family revealed that emotional family support mitigated the negative impact of chronic stress exposure on trouble falling asleep and the frequency of restless sleep. Specifically, while more chronic stress exposure was associated with greater trouble falling asleep and restless sleep for all respondents, the magnitude of these relationships was weaker among respondents who reported receiving high levels of emotional family support, suggesting that emotional support from family is more protective at higher levels of chronic stress exposure than at low levels of exposure. Overall, these significant interactions underscore the important stress buffering function of family support; they demonstrate that family support can reduce the effects of chronic stress exposure on sleep quality. Our findings are consistent with the limited literature on the moderating effects of social support on the chronic stress-sleep connection (Liang et al., 2020; Van Schalkwijk et al., 2015). Studies in this area find that social support blunted the effects of chronic stressors on a range of sleep outcomes, including short sleep duration (Chen et al., 2021), sleep disturbances (Liang et al., 2020), and sleep quality (Matick et al., 2021; Van Schalkwijk et al., 2015). Our findings demonstrate that emotional support from extended family members may be an effective stress coping resource and could mitigate some of the harm of chronic stress exposure on sleep quality among African Americans.

We did not find moderating effects for emotional and overall support from church members on the associations between chronic stress exposure and the three sleep quality variables. It is possible that support from family may have had a greater impact on sleep quality. Thus, the interactions between chronic stress exposure and church support did not attain statistical significance. It is important to note that most research documenting the salutary health effects of church support focuses on older adults. As a group, older adults have higher rates of religious involvement, higher levels of social embeddedness within congregational networks, and religious sentiments are more pronounced in this population (Krause & Wulff, 2005; Nguyen, Taylor, et al., 2019; Taylor et al., 2004). Given that this study focused on African American across the adult life span, our analysis might not have detected a significant moderating effect for church support due to age group differences in the centrality of religion. Church members may be less significant as social partners and sources of support and stress coping for the current sample compared to older African Americans. Relatively younger adults may have greater access to other sources of support, such as co-workers and same age friends, than older adults, and these sources of support may be more salient for adults at this developmental stage. It is also important to note that the cumulative effects of chronic stress exposure over the life course would likely be more pronounced among older adults. Thus, the combination of decreased religious salience and decreased life course stress accumulation in the current sample (relative to older African Americans) may have contributed to the null findings.

Additionally, frequency of contact and overall social support did not moderate the relationship between chronic stress exposure and sleep quality. It is plausible that frequency of contact and overall support were not sufficient stress coping resources. That is, emotional support may help a person deal with chronic stress by providing comfort and a sense of being cared for, which could alleviate some of the negative impact of chronic stress exposure on sleep. In contrast, overall social support is a broader measure of social support and can include multiple dimensions of social support other than emotional support, such as tangible support (e.g., help with chores, transportation), which may not directly address chronic stress or feelings of distress that the individual is dealing with. Furthermore, frequency of contact may not reflect supportive exchanges that would help a person deal with chronic stress. That is, frequency of contact does not tap into the quality and function of relationships, which may be more important in the context of dealing with chronic stressors, but rather frequency of contact taps into the quantitative aspects of relationships.

Strengths and limitations

This study has several notable strengths. This is the first analysis of the moderating role of social support from extended family and church members on the relationship between chronic stress exposure and sleep quality specifically in African Americans. Due to social, political, and economic marginalization, many African Americans are exposed to greater levels of chronic stressors than white Americans (Brown et al., 2020). Consequently, it is critical to identify modifiable factors that can mitigate the harm of this unequal stress exposure. Our focus on the role of social support in the connection between chronic stress exposure and sleep quality among African Americans not only contributes to the literature by bridging an important knowledge gap but also informs possible social interventions to address sleep concerns. Another strength of this study is the differentiation between social support from extended family and church members. The distinction between these two sources of social support extends the current literature by allowing us to determine the unique contributions of specific types of social support network members to African Americans’ stress coping process. Additionally, our analysis was based on a nationally representative sample of African Americans. Many studies in this area use regional and local data and/or rely on convenience sampling methods, which limit the generalizability of the study findings. Because our sample is nationally representative, the current findings can be extrapolated to the U.S. population level.

Several study limitations should be noted. The sleep quality variables in the current analysis are self-reported. Self-reported measures are susceptible to social desirability and memory recall biases. Future research should use objective measures of sleep quality obtained through actigraphy and polysomnography to verify the current findings. Because our data were cross-sectional, the current findings do not support causal inferences. For example, it is unclear whether people who attended church more frequently are more likely to have better sleep health or whether people who had better sleep health were physically healthier and have fewer functional limitations, which would have allowed them to attend church more frequently. We recommend that future studies use prospective data and longitudinal analytical strategies to determine the temporal ordering of these variables to allow for causal inferences. Further, given that numerous interaction terms were tested in the current analysis, we acknowledge that there is a possibility that the statistically significant interactions that we identified may be spurious. Finally, the NSAL only sampled community-dwelling individuals; thus, institutionalized and unhoused individuals and individuals living on military bases were excluded from the NSAL. Consequently, the findings from this study are only generalizable to community-dwelling African American adults, and we cannot extrapolate these findings to populations of institutionalized or unhoused African American adults and African American adults living on military bases.

Implications

These findings hold important implications for research, practice, and policy. Regarding research implications, future research should examine other types of non-kin relationships that serve as important stress coping resources for African Americans. For example, friendships and fictive kin relationships are significant sources of support for this group (Nguyen, Chatters, Taylor, Levine, et al., 2016; Nguyen, Chatters, Taylor, & Mouzon, 2016; Nguyen, Taylor, et al., 2016; Nguyen, Walton, et al., 2019; Taylor et al., 2021). Yet, very few studies have focused on friendships and even fewer studies have examined fictive kin relationships specifically among African Americans. We recommend that that future research elucidate the role of these types of non-kin relationships in African Americans’ sleep health to clarify the unique contributions of these social groups. Furthermore, while this study focused on a global measure of chronic stress exposure, we suggest that future research investigate how specific types of stressors may impact sleep health among African Americans. For example, prior research indicates that relational stressors, such as negative interactions, negatively impact the health and well-being of African Americans (Lincoln et al., 2012; Nguyen et al., 2021; Wang et al., 2022). Future research could investigate the effects of negative interactions on sleep health and whether different sources of negative interactions (e.g., spouse/partner, extended family, friends) may differentially impact sleep.

Regarding practice implications, these findings suggest that assessments of African Americans’ family support network is important for determining potential stress coping resources. Assessments can identify people with weak or limited family ties who may be at greater risk for stress-induced sleep problems. Stress management interventions could focus on helping people strengthen their relationships with extended family members; these types of interventions will be particularly important for people with weak or limited family ties. Additionally, behavioral interventions for stress-induced sleep problems could benefit from using a family systems approach. Given that these findings demonstrate that emotional support from extended family may buffer the negative effects of chronic stress exposure on sleep quality, sleep health and hygiene psychoeducation programs may incorporate information on the role of family support in stress management. Also, clinicians and providers should be aware of cultural differences in the structure and function of extended family relationships and how these can influence stress and sleep when working with diverse populations.

Regarding policy implications, policies that prioritize funding for programs that emphasize the role of extended family support in community health will be important for promoting sleep health. Further, directing resources toward the development of family-centered community events and spaces (e.g., family spaces and times at public libraries) can provide opportunities for extended family gathering and interactions and strengthen extended family ties. Health policies that acknowledge the role of social support, especially family support, in health can facilitate the integration of extended family members into broader health care plans. For instance, stress management workshops that recognize and utilize the support of extended family members could be encouraged.

Conclusion

In sum, this study found that emotional support from extended family members attenuated the positive association between chronic stress exposure and trouble falling asleep and restless sleep in African American adults. Furthermore, confirming prior research, the data demonstrated that chronic stress exposure consistently predicted poor sleep quality across all three sleep measures. Together, these findings underscore the detriment of chronic stress exposure to African Americans’ sleep health and suggest that extended family members are effective stress coping resources and play an important role in the sleep quality of this population. Our findings suggest that interventions for sleep difficulties should consider the role of extended family and church members and integrate them into sleep interventions whenever possible.

Data Availability Statement

The National Survey of American Life-Reinterview data set is publicly available and can be accessed at https://www.icpsr.umich.edu/web/ICPSR/studies/27121