Associations Between Everyday Discrimination and Sleep: Tests of Moderation by Ethnicity and Sense of Purpose

Abstract

Background

Everyday discrimination holds pernicious effects across most aspects of health, including a pronounced stress response. However, work is needed on when discrimination predicts sleep outcomes, with respect to potential moderators of these associations.

Purpose

The current study sought to advance the past literature by examining the associations between everyday discrimination and sleep outcomes in an ethnically diverse sample, allowing tests of moderation by ethnic group. We also examined the role of sense of purpose, a potential resilience factor, as another moderator.

Methods

Participants in the Hawaii Longitudinal Study of Personality and Health (n = 758; 52.8% female; m_age: 60 years, sd = 2.03) completed assessments for everyday discrimination, sleep duration, daytime dysfunction due to sleep, sleep quality, and sense of purpose.

Results

In the full sample, everyday discrimination was negatively associated with sleep duration, sleep quality, and sense of purpose, while positively associated with daytime dysfunction due to sleep. The associations were similar in magnitude across ethnic groups (Native Hawaiian, White/Caucasian, Japanese/Japanese-American), and were not moderated by sense of purpose, a potential resilience factor.

Conclusions

The ill-effects on health due to everyday discrimination may operate in part on its role in disrupting sleep, an issue that appears to similarly impact several groups. The current research extends these findings to underrepresented groups in the discrimination and sleep literature. Future research is needed to better disentangle the day-to-day associations between sleep and discrimination, and identify which sources of discrimination may be most problematic.

Introduction

The deleterious effects of everyday discrimination on health and well-being are well-documented [1, 2]. An increasing body of work has suggested that one underlying mechanism may be that greater exposure to discrimination is associated with worse sleep [3]. For instance, longitudinal research has noted that global sleep quality, and particularly daytime dysfunction, mediate the associations between everyday discrimination and later physical health outcomes among midlife adults [4]. Given these findings, the researchers noted the need to consider potential moderators of this association, in order to better understand potential resilience factors and to develop more individualized intervention programs.

Regarding resilience factors, one likely candidate is the extent to which individuals feel a sense of purpose in life, defined as the perception that one has clear goals and a life direction that guides them to engage in daily activities in pursuit of those life goals [5, 6]. Sense of purpose consistently predicts positive health outcomes across different cultures, including mortality risk [7] and better sleep quality [8, 9]. Furthermore, sense of purpose may hold unique value for mitigating the ill-effects associated with discrimination. In general, individuals with a stronger sense of purpose appear to show less affective reactivity to daily stressors [10]. Research also shows that purposeful participants report less negative mood during real-world diversity experiences [11]. Therefore, sense of purpose appears to mitigate the impact of daily stressors in general, and reduces reactivity to daily diversity experiences. One explanation for these findings is that purposeful individuals are less impacted by daily events, because they are able to focus on their broader life direction.

Building from the work on majority versus minority status, another important moderator of interest is ethnicity. One could anticipate effects to be moderated based on the individuals’ majority (or privileged) versus minority (less privileged) status in the given context. For individuals of a minority status in society, everyday discrimination is more common and may present a consistent stressor that causes compounding distress [12, 13], and in turn holds a stronger impact on sleep. A handful of studies have examined moderation by race and ethnicity with conflicting results, showing some evidence for moderation in the expected direction [14], evidence for moderation in the opposite direction for some outcomes [15], or no evidence for moderation [16]. Given this mixed picture, it is valuable to extend this literature by considering additional cultural groups, and contexts with nuanced majority versus minority classifications.

Current Study

The current study sought to test these predictions using data from the long-running Hawaii Longitudinal Study of Personality and Health [17]. This dataset provides the opportunity to investigate the role of discrimination on sleep in samples identifying as Native Hawaiian or Japanese/Japanese-American, two underrepresented groups in the literature on discrimination and sleep, in comparison to Whites. First, with respect to the overall sample, we expected reports of everyday discrimination to be negatively associated with subjective sleep quality and sleep duration, and to be positively associated with sleep-related daytime dysfunction. Second, we explored whether the discrimination-sleep association was stronger for participants of less privileged status. Native Hawaiians often are afforded lower status and stigmatized even in Hawaii [18, 19], where most of the current sample still resides; moreover, estimates suggest substantially lower per capita income for this group relative to the White and Japanese/Japanese-American citizens of Hawaii [20]. Although these latter groups hold similar socioeconomic status, census estimates suggest White citizens hold a plurality status. Even though some of the current sample have moved out of Hawaii, it thus may be suggested that White participants hold the most privileged status based on socioeconomic and population statistics, followed by Japanese/Japanese-Americans, with Native Hawaiians holding the least privileged status. Third, we tested sense of purpose as a moderator of the discrimination-sleep association, insofar that individuals with a greater sense of purpose may be less disturbed by daily stressful events and thus have a weaker association. In addition, we explored the association between self-reported discrimination and sense of purpose, given that the past literature has been mixed on this point, finding both positive [21] and negative associations [22] between these constructs.

Method

Participants and Procedure

Participants (n = 758; 52.8% female; mean age: 60 years, range: 56–65, sd = 2.03) for the current study came from the Hawaii Longitudinal Study of Personality and Health [17], for which children who were schooled in Hawai’i were followed up decades later to complete questionnaires as adults. Data came from the seventh questionnaire assessment completed during adulthood, which provided the first and only measurement of everyday discrimination thus far. Participants receive $25 in compensation for completing the questionnaire by mail or electronically. In earlier assessments, participants reported on their educational attainment as adults and ethnicity. Participants reported their highest education level attained on a 9-point scale: only 1.6% of the sample reporting less than a high school degree or equivalent, 16.4% reporting a high school degree or equivalent, 28.0% reporting being a college graduate, and 20.2% reporting some postgraduate or professional degree. In the first adult survey with this sample, participants were asked to report “Which group best describes your cultural identity?” with added instructions noting that if they identified “with more than one group, please choose the one group with which you most strongly associate yourself.” The most common ethnic groups with relevant data reported were Japanese/Japanese-American (n = 276), White/Caucasian (n = 151), and Native Hawaiian/Part-Native Hawaiian (n = 132). As no other group was reported by more than 8.0%, we focus on these three for group analyses below to maximize power for groups comparisons.

Measures

Everyday discrimination.

The nine-item Everyday Discrimination [23] measure was administered, which asked participants to rate how frequently specific acts occurred from 0 (Never) to 5 (Almost Every Day). A sample item is “You are treated with less respect than other people” (α = 0.88). Due to survey limitations, participants were not asked to provide their perceived reason for the discrimination. Past work has shown evidence for the validity of the measure in a Native Hawaiian sample [24].

Sleep variables.

All three sleep measures came from items included in the Pittsburgh Sleep Quality Index [25]. Daytime dysfunction due to sleep was assessed as an average of two items: “During the past month, how often have you had trouble staying awake while driving, eating meals, or engaging in social activity?” and “During the past month, how much of a problem has it been for you to keep up enough enthusiasm to get things done?” The response options were 0 (not at all during the past month), 1 (less than once a week), 2 (once or twice a week), and 3 (three or more times a week) (M = 0.52, sd = 0.588). Sleep quality was assessed using a single-item: “Rate your overall sleep quality” (M = 1.96 on a 0 [Very bad] to 3 [Very good] scale, sd = 0.719). Sleep duration was assessed by asking “How many hours of actual sleep do you get each night?” (M = 6.32, sd = 1.34). Each component has shown predictive value for health and well-being outcomes [26].

Sense of purpose.

Sense of purpose was captured using the Life Engagement Test [6]. Participants are asked to indicate their agreement to six items on a five-point scale from 1 (Strongly Disagree) to 5 (Strongly Agree). A sample item includes “There is not enough purpose in my life” (α = 0.80). Previous research with this dataset has shown the measure’s predictive validity for self-rated health and found associations between purpose and sleep quality [8]; however, that work had not investigated purpose as a moderator.

Analytic Plan

All measures were averaged across items, and higher scores reflect greater discrimination, daytime dysfunction, sleep quality, sleep duration, and sense of purpose. First, we present associations between discrimination and the primary variables of interest, in the overall sample and then separately by ethnic group. Second, to test ethnicity as a moderator, we conducted multiple regression analyses that first examined discrimination, the covariates, and dummy coded variables to reflect Native Hawaiian and Japanese/Japanese-American identifications. Then, we included interaction terms for the dummy coded variables by discrimination. Third, to test sense of purpose as a moderator, multiple regression analyses were conducted using everyday discrimination, purpose, and their interaction as predictors of sleep measures. Regressions included sex, education level, and birth year as covariates, and discrimination and purpose were centered before inclusion as predictors.

Results

Correlational Analyses and Mean Differences

In line with expectations, everyday discrimination was associated with greater reports of daytime dysfunction (r(754) = 0.28, p < .001), worse sleep quality (r(754) = –0.16, p < .001), shorter sleep duration (r(758) = –0.15, p < .001), and lower sense of purpose (r(754) = –0.25, p < .001). Also as expected, White participants reported the lowest levels of discrimination (M = 0.64, s.e. = 0.04), followed by Japanese/Japanese-Americans (M = 0.73, s.e. = 0.04), with Native Hawaiians reporting the highest levels (M = 0.88, s.e. = 0.07), F(2, 556) = 4.224, p = .015.

Regarding the breakdown by ethnicity, for Whites, discrimination was associated with lower sense of purpose (r(150) = –0.22, p = .007), greater daytime dysfunction (r(151) = 0.18, p = .028), but was unassociated with sleep quality (r(150) = –0.04, p = .658) or sleep duration (r(151) = –0.16, p = .054). For Native Hawaiians, discrimination was associated with lower sense of purpose (r(131) = –0.37, p < .001), greater daytime dysfunction (r(132) = 0.29, p = .001), worse sleep quality (r(131) = –0.23, p = .008), and less sleep duration (r(132) = –0.21, p = .017). For Japanese/Japanese-Americans, discrimination was associated with lower sense of purpose (r(275) = –0.21, p < .001), greater daytime dysfunction (r(276) = 0.34, p < .001), worse sleep quality (r(275) = –0.15, p = .011), but it was not associated with sleep duration (r(276) = –0.06, p = .365).

Moderation by Ethnicity

Next, Table 1 presents the multiple regression tests for examining moderation by ethnicity. In Step 1, discrimination was included along with ethnicity and the covariates. In all models, discrimination remained a significant predictor of the three sleep outcomes. Results also found significant differences by ethnicity for sleep duration such that Native Hawaiians and Japanese Americans had shorter self-reported sleep duration than White participants, but no differences for other sleep variables. In Step 2, interaction terms were added to the models. The interaction effects were never significant, and added less than 1% additional variance.

Moderation by Sense of Purpose

Finally, we examined sense of purpose as a moderator. Results for the three separate multiple regressions are included in Table 2. Everyday discrimination remained a significant predictor of all sleep outcomes, even with the inclusion of the additional predictors. As shown in past work with this sample [8], sense of purpose was associated with better sleep quality, as well as less daytime dysfunction, but it did not uniquely predict sleep duration. Sense of purpose did not moderate the associations between everyday discrimination and sleep.

Discussion

The current study added to the literature on discrimination and sleep in three important ways. First, these findings provide further evidence for the ill-effects of everyday discrimination, insofar that higher levels of discrimination were associated with worse sleep quality, greater disturbances during the day due to sleep, and fewer hours of sleep. Moreover, discrimination was negatively associated for all groups with sense of purpose. Second, although having a sense of purpose may protect against sleep issues [9], it failed to serve as a resilience factor against the ill-effects associated with discrimination. Third, the current study also found little difference in the magnitude of the associations between discrimination and sleep outcomes depending on ethnicity. We discuss each of these findings below, with respect to directions for future research.

Of primary importance is that the current study demonstrated the potential for perceived everyday discrimination to impact sleep, in line with previous research [3, 14, 15]. Such findings are important given the research linking these self-reported sleep measures to risk for major health ailments [26]. Although the current study was limited by its use of self-reports, past work has shown that everyday discrimination holds negative associations across multiple self-report and objective measures of sleep [15]. The central advance of the current study was in its ability to expand the literature by investigating these associations in an ethnically diverse sample that included substantial representation of Native Hawaiians and Japanese/Japanese-Americans, underrepresented groups in sleep research. Though the current findings are cross-sectional in nature, the causal direction from discrimination to worse sleep would appear more theoretically motivated than the reverse, given the known impact of daily racial discrimination on stress responses [27]. However, research on reports of daily stressors also has shown that poorer sleep does predict a greater likelihood to report stressors the following day [28]. Future work on discrimination and sleep is needed that better captures the daily events in question, in order to understand the bidirectionality of effects, and disentangle the influence of chronic versus daily discrimination on sleep (see Ref. [29]).

Neither purpose nor ethnicity served to moderate the discrimination-sleep associations. Although the lack of moderation by either sense of purpose was surprising, this finding aligns with a broader literature showing mixed findings for other psychosocial constructs as protective factors [30]. Indeed, past work shows only modest evidence that factors such as racial identity and social support helps to mitigate the impact of racism. Regarding the current findings, everyday discrimination may be more likely to occur in the context of purpose-driven activities (e.g., at work or school), impeding goal pursuit; thus, collecting information on the context may help understand whether having a strong sense of purpose mitigates or perhaps even accentuates the ill-effects of discrimination. With respect to the lack of moderation by ethnicity, the current findings add to the equivocal literature on this topic, by finding no evidence across group comparisons, which is aligned with some of the past studies [16]. A primary limitation was the lack of information on participants’ perceived source or reason for discrimination, although past work also has commonly avoided this topic. Future research with this information would provide greater insight into whether ethnicity-based discrimination is likely to hold broader impact on more marginalized groups.

In addition to the cross-sectional nature of the study, and lack of information on the discriminatory acts, further limitations should be noted. First, participants were asked to choose a single ethnic identification, and this information was assessed only at the first questionnaire, which occurred roughly 16 years prior to the current survey assessment. As such, it would be valuable both to have updated information, and to investigate the role of multi-ethnic status on the impact of discrimination. Second, participants for this study were all recruited on the basis of being originally in the same school cohort [17], and thus there was limited variability with respect to age. Accordingly, work is needed to better investigate the role of intersectionality between ethnic status and age. Past work has suggested that older adults of racial minority status may report less discrimination than younger adults [31], which may explain some of the low base-rates for discrimination reports in the current sample. Furthermore, calls have been made for research on the intersectional interactions between discrimination based on age and race/ethnicity [32], for which it again would be valuable to have greater information on the discriminatory acts experienced.

These caveats aside, the current findings advance our understanding of the role everyday discrimination plays on health and well-being. Specifically, they call for consideration into the impact of discriminatory experiences on sleep, and these findings demonstrate that the potential ill-effects of discrimination appear to persist across different ethnic groups. As such, the current findings call for greater intervention into how to address sleep issues among groups facing discrimination, particularly given that associations were similar in magnitude to past work on the downstream consequences for later mental and physical well-being [4]. Moreover, longitudinal work is needed to investigate mediators linking discrimination and sleep; for instance, although it failed to serve as a moderator, sense of purpose could help explain associations over time.

Acknowledgments This research was supported by a grant from the National Institute on Aging (R01-AG020048).

Compliance with Ethical Standards

Authors’ Statement of Conflict of Interest and Adherence to Ethical Standards Authors Patrick L. Hill, Nancy L. Sin, Grant W. Edmonds, and Anthony L. Burrow all declare that they have no conflict of interest. All participants provided written informed consent. The content is solely the responsibility of the authors and does not represent the official views of the National Institutes of Health.

Authors’ Contributions

PLH was the primary author of the manuscript, and analyzed the data with support from NLS and ALB. All four authors provided feedback on the conceptualization of the project, and critically revised the article across drafts. GWE was central to the acquisition of the data and design of the overall study. All authors provided approval before submission of the manuscript.

Ethical Approval All procedures followed were in accordance with the American Psychological Association ethical standards and with the Helsinki Declaration. This study was approved by the Institutional Review Board at Oregon Research Institute.

References