Abstract
Study Objectives
In an older African-American sample (n = 231) we tested associations of the household environment and in-bed behaviors with sleep duration, efficiency, and wakefulness after sleep onset (WASO).
Methods
Older adult participants completed a household-level sleep environment questionnaire, a sleep questionnaire, and underwent 7-day wrist actigraphy for objective measures of sleep. Perceived household environment (self-reported) was evaluated using questions regarding safety, physical comfort, temperature, noise, and light disturbances. In-bed behaviors included watching television, listening to radio/music, use of computer/tablet/phone, playing video games, reading books, and eating. To estimate the combined effect of the components in each domain (perceived household environment and in-bed behaviors), we calculated and standardized a weighted score per sleep outcome (e.g. duration, efficiency, WASO), with a higher score indicating worse conditions. The weights were derived from the coefficients of each component estimated from linear regression models predicting each sleep outcome while adjusting for covariates.
Results
A standard deviation increase in an adverse household environment score was associated with lower self-reported sleep duration (β = −13.9 min, 95% confidence interval: −26.1, −1.7) and actigraphy-based sleep efficiency (β = −0.7%, −1.4, 0.0). A standard deviation increase in the in-bed behaviors score was associated with lower actigraphy-based sleep duration (β = −9.7 min, −18.0, −1.3), sleep efficiency (β = −1.2%, −1.9, −0.6), and higher WASO (5.3 min, 2.1, 8.6).
Conclusion
Intervening on the sleep environment, including healthy sleep practices, may improve sleep duration and continuity among African-Americans.
Keywords: household, environment, sleep, African-American, Jackson Heart Study
Statement of Significance.
African-Americans are disproportionately affected by short sleep and poor sleep continuity. However, factors such as household environment are understudied in regards to sleep, despite prior studies demonstrating the unique importance to this population. Understanding the indoor environmental determinants of poor sleep among African-Americans, may provide an intervention target for improving sleep in this population. We provide new evidence demonstrating that the household sleep environment perceptions (e.g. noise and light disruptions, thermal and physical discomfort, lack of safety) are associated with self-reported short sleep duration, and in-bed behaviors (e.g. watching television, use of radio/computer/phone/tablet, eating, and reading) were associated with actigraphy-measured poor sleep continuity. The household environment may be a point of intervention to improve sleep among African-Americans.
Introduction
African-Americans have a high prevalence of short sleep duration and poor sleep quality [1–3]. Numerous research studies have shown inadequate sleep contributes to adverse health and well-being [4]. It is hypothesized that poor sleep may contribute to disparities in cardiovascular health [5, 6]; therefore, it is important to identify the determinants of inadequate sleep, particularly among African-Americans who are most at risk for cardiovascular disease (CVD) [7].
The household environment is an important target of healthy sleep recommendations, which include a dark, safe, and comfortable environment. Environmental factors such as inopportune light, extreme temperatures, and traffic-related noise can disturb sleep [8–10], and are more common in lower socioeconomic status neighborhoods where African-Americans are more likely to reside [11]. Therefore, the physical environment may be a barrier for experiencing healthy sleep recommendations among African-Americans [12, 13]. As a result, neighborhood characteristics have recently emerged as factors that may affect sleep and contribute to disparities in sleep [3, 8, 10]. This limited research has examined the association between the physical environment and sleep, demonstrating an association of increased neighborhood-level noise (most commonly studied) with poor sleep [8, 9, 14]. Data from the Jackson Heart Study (JHS) have also demonstrated that individuals who live in neighborhoods with a high level of disorder [15] characterized as noisy, inopportune light exposure, and poor aesthetics have shorter self-reported sleep duration and a poorer sleep quality than residents in neighborhoods with low measured disorder [16]. Although investigating neighborhood-level factors are important to understand determinants of insufficient sleep, this research does not consider the individual or household-level variation of the environmental exposures within the neighborhood. Individual and household-level factors may both affect sleep independent of neighborhood characteristics as well as provide a more accurate assessment of the influence of even more direct/proximate social and environmental factors on sleep [12].
The limited literature examining the association of the household environment with sleep suggests that poor housing conditions (e.g. crowding, quality of housing structure) are associated with shorter sleep duration and poorer sleep quality [12, 17, 18]. However, specific data regarding the bedroom environment and sleep among adults are lacking. An assessment of household factors could provide important insights into how the sleep environment and ambient conditions influence sleep patterns and disruptions. Only a few prior studies have examined these factors, suggesting the potential importance of housing structure [17–19], traffic-related noise near home environments [20, 21], household crowding [22], inopportune light exposure [5], and ambient temperature [23] on sleep. In particular, the existing literature does not assess other important dimensions of the home environment or the combination of various dimensions, and is limited in the main assessment of self-reported sleep, and, lastly, lacks inclusion of African-American adult populations. This literature is also limited in assessing in-bed behaviors among adults using a multi-level framework. In-bed behaviors such as social media and technology use are associated with shorter sleep duration and poorer sleep continuity [24–26]. Studies have shown these behaviors to be more common among African-Americans [27, 28]. Although these are individual behaviors, they may be partially shaped by the bedroom environment (presence/absence of electronics), as well as reflect the individual’s relationship to their environment (e.g. eating in bed) and thus serve as proxy for a suboptimal environment. In addition, in-bed behaviors may reflect a response to the environment. For example, qualitative data among African-Americans have shown that in-bed behaviors such as watching television or use of electronics are often used as a strategy to feel safer [29]. However, these behaviors also promote light and noise, which can disrupt sleep. To improve sleep among African-Americans, it is important to understand the association between key components of healthy sleep recommendations—the physical environment and bedroom behaviors, on sleep, which is understudied in real-world settings.
Among a sample of African-Americans, we tested the association between perceived household environment including temperature, noise, light exposure, physical comfort and safety as well as in-bed behaviors (watching television, listening to radio/music, use of computer/tablet/phone, playing video games, reading books, eating) in relation to wrist actigraphy-measured sleep duration and continuity as well as self-reported sleep duration. Although perceived safety is not a specific component of healthy sleep recommendations, qualitative data among African-Americans suggests that safety is important [29]. We hypothesized that perceptions of an adverse household environment characterized by extreme temperatures (e.g. too cold or hot), noise disturbances, inopportune light exposure, feeling unsafe/uncomfortable as well as in-bed behaviors that promote light/noise and other behaviors during wake will be associated with a short sleep duration and disturbed sleep.
Methods
This study was conducted among a sub-sample of participants from the Jackson Heart Sleep Study (JHSS), an ancillary study to the Jackson Heart Study (JHS). The JHS is the largest single-site study of the development and progression of CVD among African-Americans. The details of JHS are previously published [30]. In brief, African-Americans (n = 5,306) from three counties (Hinds, Madison, and Rankin) in Jackson, MS were enrolled and completed a baseline visit between 2000 and 2004. Two follow-up exams were completed after baseline. Participants from the third JHS visit or those who participated in other ancillary studies were recruited and enrolled in the JHSS between 2012 and 2016 [31]. JHSS participants (n = 913) completed a clinic visit when questionnaires on sleep, psychosocial factors, medical history, and socio-demographics were administered and when they were instructed on use of 7-day actigraphy and 1-night of in-home sleep apnea testing, to be initiated the night of the clinic visits. Further details on JHSS were previously published [31]. Data for this study were obtained from a sub-sample of JHSS participants (n = 231) who completed the household-level sleep environment questionnaire conducted during a restricted period of survey administration (2015–2016), underwent actigraphy, and had information on age, body mass index, and educational attainment. Institutional Review Board approval was obtained from the University of Mississippi and Partners Healthcare, and written informed consent was obtained from all participants.
Actigraphy-measured sleep duration (an objective measure of sleep patterns) and continuity as well as self-reported sleep duration were assessed. Participants underwent 7-day actigraphy for measures of sleep duration and sleep continuity (efficiency and wakefulness after sleep onset-WASO). To measure sleep patterns, participants wore a GT3X®+ Activity Monitor on their non-dominant wrist for 7 consecutive days and completed a sleep diary [32]. Using a validated algorithm (Cole–Kripke), actigraphic data during 60-second epochs were scored as sleep or wake by ActiLife version 6.13 analysis software (ActiGraph Corp, Pensicola, FL), and manually edited using the sleep diaries to indicate the sleep period [33]. From valid nocturnal actigraphy data, we computed the average values for the outcome variables: sleep duration, sleep efficiency (percentage of sleep estimated during the nighttime sleep period), and WASO (the number of minutes an individual is awake during the sleep period). In secondary analyses, we tested associations with actigraphy defined sleep midpoint and sleep latency (min) given our hypotheses around sleep timing. Short sleep duration was defined as sleeping less than 6 hours, and low sleep efficiency was defined as <85%. Self-reported sleep duration was defined by participants reported bed and wake times during weekdays and weekends, separately. We then weighted the responses by weekday (5/7) and weekend (2/7) to determine average sleep duration. Because this assessment may include wakefulness before sleep onset (a decision made related to the difficulty individuals have in estimating sleep onset, which may be biased in individuals with perceived sleep difficulties (sleep misperception), we also alternatively analyzed self-reported sleep duration using a question that directly asks the individual to estimate their habitual sleep duration: “How many hours of sleep do you usually get per night (a) on weekdays or workdays and (b) on weekends?.” The same weights were applied to define self-reported “habitual” sleep duration. To be consistent with the literature, self-reported short sleep duration was defined as less than 7 hours [34].
A subset of JHSS participants, those who were enrolled in the last year of the study, completed the household-level sleep environment questionnaire. The questionnaire was administered during the clinic visit. Based on the literature regarding healthy sleep practices and prior research on the home environment [35], we developed a questionnaire (Supplementary File) to assess two broad domains, the perceived household environment and in-bed behaviors. Perceptions of the household environment included safety at night, physical comfort (e.g. mattress, pillows), room temperature, along with noise and light disturbances. Participants reported their level of agreement using a Likert-scale (“strongly agree,” “agree,” “disagree,” “strongly disagree”) to the following questions: “sometimes my sleep is affected because I feel unsafe at night”; “the place where I sleep is physically comfortable (mattress, pillows, etc.)”; and “the place where I sleep is at a comfortable temperature.” To assess noise, participants were asked “is the room where you sleep quiet at night,” and response options included “always,” “sometimes,” or “never.” Participants responded “yes” or “no” to the question “is the room where you sleep dark during the night [or day if you work at night]” to assess light disturbance. For in-bed behaviors, participants reported yes or no to “what behaviors (or things) do you do in the bedroom before sleep.” In-bed behaviors included: watch television, listen to radio/music, computer/tablet/phone use, play video games, read books, or eat meals or snacks.
Covariates included sociodemographic variables and body mass index (BMI). Participants self-reported age in years (date of birth) and gender (male or female). Educational attainment was reported as less than high school, high school or GED, some college, or college degree during the JHS follow-up study (2008–2012). Height and weight were measured during the clinic visit by trained staff following a standardized protocol. BMI was calculated in kg/m2 using measurements of weight and height and analyzed as a continuous variable.
Statistical analyses
The household-level sleep environment questionnaire was administered to 266 JHSS participants enrolled between 2015 and 2016, and of those, 237 (89%) had actigraphy data and completed all 11 questions included in this analysis. We further excluded six participants with missing information on potential confounders. These exclusions resulted in an analytic sample size of 231.
Five of the 11 questions address the household environment, which have different response designs (three questions using a 5-point Likert scale, one question using a 3-point Likert scale, and one binary question). We reverse coded the positively-worded items and linearly transformed the 5-point and 3-point response levels to a scale ranging from 0 to 1 (Supplemental Table 1). The remaining questions are in-bed behaviors, all of which are binary (yes/no) questions. We coded “Yes” as 1 and “No” as 0. The unweighted score of the household environment and in-bed behaviors was the average transformed response, with higher scores indicating a worse sleep environment. We also constructed weighted scores, with weights depending on the associations between the components and the sleep outcome. Each weighted score per domain and sleep outcome was constructed as a weighted average of component items, with weights constructed as follows. A detailed description of the weighted score is in the supplementary file. We regressed each sleep outcome on each sleep environment component while adjusting for age, gender, and BMI (Supplemental Table 2). We assigned zero weights to components with positive coefficients for sleep duration and sleep efficiency and to those with negative coefficients with WASO. Finally, the weight of each component was calculated by dividing its coefficient by the sum of coefficients that are in the expected direction, and the weighted components were summed. Finally, we standardized the unweighted and weighted score as well as the response of each component to a normal distribution of a mean of 0 and a standard deviation of 1.
To describe participants’ characteristics in the overall sample and by the binary sleep duration category, we used mean and standard deviations for age, BMI, frequency and percentage for educational attainment, and sleep environment scores (based on the original scale). Chi-square or Wilcoxon rank-sum tests were conducted to compare group differences for each characteristic. Pearson correlations were assessed among the sleep outcomes, and Spearman correlations were assessed among sleep environment components. We performed multivariable linear regression analyses for the associations between each sleep environment component and each sleep outcome (sleep duration, sleep efficiency, and WASO) while adjusting for age, gender, educational attainment, and BMI. Similar models were computed with sleep midpoint and latency. All tests were two-sided and an alpha < 0.05 was used for statistical significance. Analyses were conducted using SAS 9.4 (SAS Institute, Cary, NC).
Results
Selected demographics for the total sample and by actigraphy-measured sleep duration are shown in Table 1. The mean age of the sample was 66.3 years (standard deviation, SD = 10.8), and was mostly female (69.7%), employed (61.4%) and obese (average BMI 31.3 kg/m2, SD = 6.7). On average, demographics were similar between short sleepers and individuals who slept ≥6 hours. However, a higher percent of short sleepers had a college degree or higher (52.1% vs. 42.6%). Thirty-seven percent of participants reported a light disturbance. In general, the household environment scores were not different by actigraphy-measured sleep duration. Watching television was the most common in-bed behavior. On average, short sleepers were more likely to report the following behaviors in bed: listen to radio or music, read books, eat meals; but were less likely to watch television, use a computer/tablet/phone or play a video game compared to those who slept ≥6 hours.
Table 1.
Characteristics of the sub-sample of Jackson Heart Study Sleep participants and by categories of actigraphy-measured sleep duration, the Jackson Heart Sleep Study N = 231
| Characteristics | Total N = 231 (100%) |
< 6 hours N = 48 (20.8%) |
≥ 6 hours N = 183 (79.2%) |
|---|---|---|---|
| Age (years), mean (SD) | 66.3 (10.8) | 65.5 (10.2) | 66.5 (11.0) |
| Male, (%) | 30.3% | 33.3% | 29.5% |
| Educational attainment, (%) | |||
| < High school | 15.1% | 18.8% | 14.2% |
| High school or GED | 19.0% | 14.6% | 20.2% |
| Some college/training | 21.2% | 14.6% | 23.0% |
| Bachelor degree or higher | 44.6% | 52.1% | 42.6% |
| BMI (kg/m2), mean (SD) | 31.3 (6.7) | 31.5 (7.7) | 31.2 (6.4) |
| Household environment | |||
| Feel unsafe at night (range: 1–5), mean (SD) | 1.9 (0.9) | 2.0 (1.0) | 1.9 (0.9) |
| Physically uncomfortable (range: 1–5), mean (SD) | 1.9 (0.8) | 1.9 (0.9) | 1.9 (0.7) |
| Uncomfortable temperature (range: 1–5), mean (SD) | 1.8 (0.6) | 1.8 (0.7) | 1.8 (0.5) |
| Noise disturbance (range: 1–3), mean (SD) | 1.5 (0.7) | 1.6 (0.7) | 1.4 (0.7) |
| Light disturbance, (%) | 36.8% | 39.6% | 36.1% |
| Unweighted score (range: 0–5), mean (SD) | 1.5 (0.4) | 1.5 (0.5) | 1.5 (0.4) |
| Weighted score by self-reported sleep duration (range: 0–5), mean (SD) | 1.5 (0.4) | 1.6 (0.5) | 1.5 (0.4) |
| Weighted score by actigraphy-measured sleep duration (range: 0–5), mean (SD) | 1.2 (0.5) | 1.3 (0.5) | 1.2 (0.5) |
| Weighted score by sleep efficiency (range: 0–5), mean (SD) | 1.7 (0.4) | 1.8 (0.5) | 1.7 (0.4) |
| Weighted score by WASO (range: 0–5), mean (SD) | 1.8 (0.5) | 1.8 (0.6) | 1.7 (0.4) |
| In-bed behaviors | |||
| Watch television, (%) | 77.9% | 70.8% | 79.8% |
| Listen to radio or music, (%) | 27.3% | 33.3% | 25.7% |
| Computer/tablet/phone use, (%) | 44.2% | 39.6% | 45.4% |
| Play video game, (%) | 7.4% | 6.3% | 7.7% |
| Read books, (%) | 56.7% | 60.4% | 55.7% |
| Eat meals or snack, (%) | 36.8% | 47.9% | 33.9% |
| Unweighted score (range: 0–1), mean (SD) | 0.4 (0.2) | 0.4 (0.3) | 0.4 (0.2) |
| Weighted score by self-reported sleep duration (range: 0–1), mean (SD) | 0.3 (0.3) | 0.4 (0.3) | 0.3 (0.3) |
| Weighted score by actigraphy-measured sleep duration (range: 0–1), mean (SD) | 0.4 (0.3) | 0.5 (0.4) | 0.4 (0.3) |
| Weighted score by sleep efficiency (range: 0–1), mean (SD) | 0.4 (0.2) | 0.4 (0.3) | 0.4 (0.2) |
| Weighted score by WASO (range: 0–1), mean (SD) | 0.4 (0.2) | 0.4 (0.3) | 0.4 (0.2) |
SD = standard deviation; BMI = body mass index; WASO = wake after sleep onset.
Actigraphy-defined short sleep duration (21%), self-reported short sleep duration (64%), and low (<85%) sleep efficiency (30%) were prevalent. Average actigraphy-defined sleep duration was 6.7 hours (SD = 1.1), whereas average self-reported sleep duration was 6.2 hours (SD = 1.5). Mean sleep efficiency was 87% and WASO was 54.7 min (SD = 25.1). Based on actigraphy, the average sleep midpoint was 3:14 AM (SD = 2:42) and sleep latency was 6.74 min (SD = 1.8). The average self-reported sleep latency was 28.5 min (SD = 35.0).
As observed in prior studies self-reported and actigraphy-measured sleep duration were weakly correlated, r = 0.29. The actigraphy-based measures of sleep continuity, sleep efficiency, and WASO were highly correlated, r = −0.90 (Table 2). The household environment components were generally not highly correlated with the in-bed behaviors (Table 3). Within domains, there were higher correlations for components of the household environment than for the in-bed behaviors domain. Feeling unsafe at night was moderately correlated with physical discomfort and uncomfortable temperature, r = 0.50 for both. Physical discomfort and uncomfortable temperature were highly correlated, r = 0.60. Whereas noise and light disturbances were weakly correlated, r = 0.28. Watching television was weakly correlated with noise and light disturbance, r = 0.15 and r = 0.17, respectively. In general, watching television, computer/tablet/phone use, reading books and eating meals or snacks were weakly correlated, with the highest correlation (r = .32) for watching television and eating meals or snacks (Table 3).
Table 2.
Pearson correlations among six sleep outcomes, the Jackson Heart Sleep Study N = 231
| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | |
|---|---|---|---|---|---|---|---|---|
| 1. Self-reported sleep duration | 1.00 | |||||||
| 2. Self-reported habitual sleep duration | 1.00 ** | 1.00 | ||||||
| 3. Self-reported sleep latency | −0.40 ** | −0.40 ** | 1.00 | |||||
| 4. Actigraphy-measured sleep duration | 0.29 ** | 0.29 ** | −0.01 | 1.00 | ||||
| 5. Actigraphy-measured sleep efficiency | 0.06 | 0.06 | 0.05 | 0.47 ** | 1.00 | |||
| 6. Actigraphy−measured WASO | 0.05 | 0.05 | −0.07 | −0.10 | −0.90 ** | 1.00 | ||
| 7. Actigraphy-measured sleep midpoint | −0.004 | −0.005 | 0.08 | −0.002 | 0.05 | −0.07 | 1.00 | |
| 8. Actigraphy-measured sleep latency | −0.04 | −0.04 | −0.04 | −0.22 ** | −0.36 ** | 0.24 ** | −0.08 | 1.00 |
** p < 0.001.
Table 3.
Spearman correlations among the 11 sleep environment components, the Jackson Heart Sleep Study N = 231
| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1. Feel unsafe at night | 1.00 | ||||||||||
| 2. Physical discomfort | 0.50*** | 1.00 | |||||||||
| 3. Uncomfortable temperature | 0.50*** | 0.60*** | 1.00 | ||||||||
| 4. Noise disturbance | 0.01 | 0.05 | 0.03 | 1.00 | |||||||
| 5. Light disturbance | 0.06 | 0.07 | 0.03 | 0.28*** | 1.00 | ||||||
| 6. Watch television | 0.14* | 0.08 | 0.06 | 0.15* | 0.17* | 1.00 | |||||
| 7. Listen to radio or music | −0.01 | −0.06 | −0.07 | −0.01 | −0.04 | 0.09 | 1.00 | ||||
| 8. Computer/tablet/phone use | 0.07 | 0.00 | 0.06 | 0.11 | 0.12 | 0.26*** | 0.00 | 1.00 | |||
| 9. Play video games | 0.02 | −0.01 | −0.01 | 0.08 | 0.09 | 0.11 | 0.05 | 0.32*** | 1.00 | ||
| 10. Read books | 0.03 | −0.07 | 0.00 | −0.02 | −0.13* | 0.21** | 0.16* | 0.21** | 0.11 | 1.00 | |
| 11. Eat meals or snacks | 0.06 | 0.01 | −0.09 | −0.01 | 0.09 | 0.32*** | 0.22*** | 0.17** | 0.06 | 0.16* | 1.00 |
*** p < 0.001, ** p < 0.01, * p < 0.05.
The results for the linear regression models estimating the association between the household environment or in-bed behaviors are shown in Table 4. Each standard deviation increase in the adverse household environment summary weighted score was associated with 13.9 min (95% confidence interval: −26.1, −1.7) shorter self-reported sleep duration, 13.9 (−26.0, −1.7) min shorter habitual sleep duration, and 0.7% (−1.4, 0.0) lower sleep efficiency on average. There were no associations between the summary score or individual components of the household environment with actigraphy-measured sleep duration or WASO. In secondary analyses, disturbance was associated with a longer sleep latency. Findings were consistent with the unweighted and weighted household environment scores. There were no observed associations between the household environment with self-reported sleep latency or actigraphy-measured sleep midpoint (Supplemental Table 3).
Table 4.
Beta Estimates (95% CI) of sleep duration, sleep efficiency, and WASO associated with components of the household environment and in-bed behaviors (per SD increase), the Jackson Heart Sleep Study N = 231
| Self-reported | Actigraphy-measured | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Sleep Duration (min) | Habitual Sleep Duration (min) | Sleep Duration (min) | Sleep Efficiency (%) | WASO (min) | ||||||
| β | 95%CI | β | 95%CI | β | 95%CI | β | 95%CI | |||
| Household environment | ||||||||||
| Feel unsafe at night | −9.1 | −21.3, 3.2 | −9.0 | −21.3, 3.2 | 0.9 | −7.7, 9.4 | −0.3 | −1.0, 0.3 | 2.0 | −1.2, 5.3 |
| Physically uncomfortable | −9.4 | −21.5, 2.7 | −9.3 | −21.4, 2.8 | −1.8 | −10.3, 6.7 | −0.2 | −0.8, 0.5 | 0.3 | −2.9, 3.6 |
| Uncomfortable temperature | −8.7 | −20.8, 3.4 | −8.7 | −20.7, 3.4 | 0.0 | −8.4, 8.5 | −0.4 | −1.0, 0.3 | 1.7 | −1.5, 4.9 |
| Noise disturbance | −6.5 | −18.7, 5.6 | −6.5 | −18.7, 5.7 | −4.8 | −13.2, 3.7 | −0.6 | −1.3, 0.1 | 1.1 | −2.1, 4.3 |
| Light disturbance | −5.1 | −17.5, 7.3 | −5.1 | −17.5, 7.3 | −1.6 | −10.2, 7.1 | 0.3 | −0.3, 1.0 | −2.0 | −5.3, 1.2 |
| Unweighted score | −12.2* | −24.5, 0.2 | −13.8 * | −26.1, −1.7 | −3.2 | −11.8, 5.4 | −0.3 | −1.0, 0.4 | 0.2 | −3.1, 3.5 |
| Weighted score | −13.9* | −26.1, −1.7 | −13.9 * | −26.0, −1.7 | −2.1 | −10.7, 6.4 | −0.7* | −1.4, 0.0 | 2.5 | −0.8, 5.7 |
| In-bed behaviors | ||||||||||
| Watch TV | −1.5 | −13.9, 10.8 | −1.5 | −13.8, 10.8 | 3.7 | −4.9, 12.2 | −0.3 | −1.0, 0.4 | 2.1 | −1.1, 5.4 |
| Listen to radio or music | 7.3 | −4.8, 19.3 | 7.3 | −4.8, 19.4 | −3.9 | −12.3, 4.6 | −0.9** | −1.5, −0.2 | 3.5* | 0.3, 6.7 |
| Computer/tablet/phone use | 3.9 | −9.2, 17.0 | 4.0 | −9.1, 17.1 | 7.0 | −2.1, 16.1 | −0.5 | −1.2, 0.3 | 3.2 | −0.3, 6.6 |
| Play video games | −2.1 | −14.7, 10.4 | −2.1 | −14.6, 10.5 | 0.4 | −8.3, 9.1 | −0.4 | −1.1, 0.3 | 2.7 | −0.6, 6.0 |
| Read books | 4.0 | −8.4, 16.3 | 4.1 | −8.3, 16.4 | −5.8 | −14.4, 2.8 | −0.7* | −1.3, 0.0 | 2.6 | −0.7, 5.8 |
| Eat meals or snacks | −3.8 | −16.0, 8.3 | −3.8 | −16.0, 8.3 | −8.7* | −17.1, −0.4 | −0.9** | −1.5, −0.2 | 2.5 | −0.7, 5.7 |
| Unweighted score | 2.8 | −9.8, 15.4 | 2.9 | −9.6, 15.5 | −3.1 | −11.8, 5.7 | −1.1*** | −1.8, −0.5 | 5.1*** | 1.8, 8.3 |
| Weighted score | −4.2 | −16.7, 8.2 | −4.2 | −16.6, 8.3 | −9.7* | −18.0, −1.3 | −1.2*** | −1.9, −0.6 | 5.3*** | 2.1, 8.6 |
SD = standard deviation.
Adjusted for age, gender, educational attainment, and BMI. *** p < 0.001, ** p < 0.01, * p < 0.05.
For in-bed behaviors, eating meals or snacks in bed was associated with actigraphy-measured sleep, specifically, sleeping 8.7 min (−17.1, −0.4) less on average and a −0.9% (−1.5, −0.2) lower sleep efficiency (Table 4). Overall, a standard deviation increase in the in-bed behavior weighted score was associated with sleeping 9.7 min (−18.0, −1.3) less, 1.2% (−1.9, −0.6) lower sleep efficiency and being awake 5.3 min more (2.1, 8.6) during the sleep period on average. In secondary analyses there were no associations between in-bed behaviors and sleep latency variables or sleep midpoint (Supplemental Table 3).
Discussion
In a sample of African-Americans, we assessed the sleep environment and tested the association between components of the household environment and in-bed behaviors as well as an overall score in relation to self-reported and actigraphy-measured sleep. Adverse perceptions of the household environment were generally associated with shorter self-reported sleep duration, a difference of approximately 14 min per standard deviation change. Whereas, in-bed behaviors, which were adverse healthy sleep practices such as use of electronics and food consumption in bed were related to shorter actigraphy-measured sleep duration, lower sleep efficiency, and longer WASO. These results, adjusted for several important confounders, point to the importance of modifiable factors in the home sleep environment as well as sleep hygiene behaviors on influencing sleep.
This research provides important insights on the perceived home sleep environment in relation to objective and subjective measures of sleep duration and continuity among a cohort of African-Americans. Short sleep duration was common in our sample (63% <7 hours measured by self-report and 21% <6 hours measured by actigraphy). In comparison to the literature on short sleep duration using similar cut-offs per measurement type, our estimates were either higher than some studies using self-report, lower or consistent with other studies using actigraphy [1, 36–39]. African-Americans have a disproportionate burden of suboptimal sleep [1, 3, 40–42]; therefore, it is critical to identify the determinants of poor sleep, which include contextual factors in order to target interventions to improve sleep and subsequent health outcomes in this population [43]. Prior studies have identified that the neighborhood environment, social (safety, disorder, cohesion) and physical (population and intersection density, noise) is associated with short sleep and poor sleep quality, particularly among African-Americans [16, 44–46]. However, these studies did not assess the household environment. There is variation within the neighborhood environment, which can result in different exposures across households within a census tract or block group, which are the typical units of observation used in research studies. Studying the environment at the household-level can provide a more accurate assessment of the proximal environment and a better assessment of the recommended healthy sleep practices. For instance, influences such as intimate partner violence and other safety concerns can be assessed along with household variation in structure like housing insulation, which can contribute to noise or temperature/humidity disturbances [12, 18]. Of note, safety was important to assess given qualitative data among African-Americans suggesting that lack of safety contributes to use of light and noise at night as a safety measure [29].
The household environment and in-bed behaviors were associated differently with sleep duration and continuity, depending on the sleep measurement. In general, perceptions of the household environment were associated with self-reported sleep but not objectively measured sleep. Whereas the in-bed behaviors were associated with objectively measured sleep. These differences may be attributable to the sleep environment domain. For example, the household environment components assessed perceptions on a Likert scale that may more closely relate to other subjective measures such as self-reported sleep, which correlates poorly with objectively measured sleep [43]. However, some of the questions around perceptions of the household environment, such as noise, reference sleep in the question, which in relation to self-reported sleep duration could result in same source bias, reflecting the possibility that individuals who perceive greater sleep disturbances also self-report shorter sleep duration and those who report shorter sleep duration may perceive greater sleep disturbances.
The findings of our study are consistent with the limited adult literature, which suggests housing is associated with suboptimal sleep [17, 18, 36]. For instance, a study of African-Americans demonstrated that housing conditions and housing quality were associated with shorter actigraphy-measured sleep duration (15 min) and worse continuity [17]. We found 14- (self-reported) and 10- (actigraphy-measured) min differences per standard deviation change in perceptions of the household environment and in-bed behaviors, respectively. We were able to extend beyond the prior study by identifying the particularly influential components of the home environment as well as simultaneously assessing the association of multiple adverse household factors on sleep. This is particularly important because no one component drove the associations, the cumulative presence was associated with poor sleep.
In general, there were no associations between household environmental factors and objective measures of sleep duration or continuity. The null findings may be attributable to misclassification of the household environment based on measures that largely were assessed as perceived disturbances. The results could also be prone to error as a result of social desirability (leading to an overly favorable assessment of the environment). In contrast, associations were observed between self-reported sleep and the environment, possibly reflecting same source bias (or those who perceive environmental disturbances also perceive shorter sleep). It is also plausible that there are other factors that may mitigate the adverse effects of the environment on sleep that were not collected or accounted for in this study.
Unexpectedly light disturbance was not associated with the sleep outcomes despite literature showing that inopportune light exposure can delay the phase of the biological clock and lead to later sleep timing [47]. However, in the current study light disturbance was classified as sleeping in a room that was not dark night, but did not address timing of light nor light brightness. Future studies may benefit from quantitative and objective metrics of the household environment, including the timing of environmental exposures.
There are a number of pathways by which the home environment can directly or indirectly affect sleep. For example, these various environment factors (e.g. inopportune light exposure, temperature) emanating from either outside or inside the home environment can disrupt the circadian rhythms, which can delay sleep timing or disrupt sleep [48]. Other components such as television use or screen use can also delay circadian rhythms and disturb sleep [49]. Environmental factors such as noise or temperature can directly affect sleep, through problems initiating or maintaining sleep. Also, safety could indirectly disturb sleep through a state of vigilance that is associated with poor sleep [50]. These features of the home environment can coexist and accumulate in a manner that can negatively affect sleep.
The household environment can be a targeted point of intervention to improve sleep. Efforts to reduce light exposure include eye masks, amber lenses on electronic devices, light blocking window shades, as well as others [51, 52]. An intervention around healthy sleep practices or sleep hygiene may also be effective in this population, with targeting in-bed behaviors (especially surrounding television watching). Future studies should continue to comprehensively assess the influence of the household environment—ideally with devices to objectively-measure features of the physical environment—and tailor interventions to mitigate the adverse components. Furthermore, a score that summarizes information across the household and bedroom behaviors may also be useful for assessing the sleep environment.
Our study is unique in assessing the sleep environment in relation to subjective and objective sleep among African-Americans. This is particularly important given the environment as a unique barrier to healthy sleep among African-Americans [29, 44, 45]. We conducted the study within a sample of African-Americans with known short sleep duration and poor sleep continuity [53], which is important for identifying drivers and potentially modifiable risk factors for sleep deficiency within this population. However, we cannot determine which sets of risk factors drive between-group differences, which was not the goal of this paper.
This is the first study to assess specific domains of the household environment in relation to objectively measured sleep. We included both self-reported and actigraphy-measured sleep duration and continuity. Two measures of sleep duration were assessed: reported bed and wake times as well as habitual sleep duration. However, self-reported data are limited by recall biases, which differ for each self-reported assessment. Our prior work indicated that compared to actigraphy, self-reported sleep duration based on bed and wake time assessment over-estimates sleep duration, while self-reported habitual sleep duration under-estimates sleep duration [43]. Therefore, we present both, showing that regardless of the self-reported measure, there was an association with the perceived household environment weighted score and self-reported sleep duration. The use of actigraphy provided an objective measure of sleep. Our study is limited by the modest sample size. The sleep environment was based on self-report, which can be prone to measurement error; however, perceptions are important to health and should be considered. The current study focuses on the nocturnal main sleep period; therefore, napping was not evaluated. Environmental sleep disruption can disrupt daytime napping. Future studies should further consider the effects of environmental exposures- assessed across the day and night- on both nighttime and 24-hour sleep patterns. Additional domains of the sleep environment such as crowding, heating and cooling access, and air quality potentially due to poor ventilation systems that were not assessed in the current study may also prove relevant and should be investigated in future studies. Also, the weighted scores have weights based on associations observed in the data between each set of exposures and each sleep outcome, potentially resulting in overfitting of the data. These are good summaries of the evidence observed in this dataset, but additional research is needed to assess their applicability to other datasets. Future studies should assess specific sources of lack of safety such as intimate partner violence that may disturb sleep. Also, our measure of safety is in reference to sleep, and does not capture perceived safety in general, which should be expanded in future studies. Further psychometric assessments of environmental factors that influence sleep may further advance our understanding of the inter-relationships between sleep and the environment. The household questionnaire used in the current study was developed based the features/concepts identified in the literature at the neighborhood-level (e.g. noise, temperature, light, safety) and bedroom activities (e.g. use of electronics) to be associated with sleep [8, 16, 35]. This questionnaire was not designed to be specific to a given race/ethnic group although some components may be particularly relevant to African-Americans or those of lower socioeconomic status who may more frequently reside in low income housing and be exposed to factors such as noise. Some behaviors (e.g. electronic media use) may be used specifically to create light or noise in response to safety concerns [29]. Our questionnaire has not undergone formal assessment of reproducibility or convergent validity, thus further research is needed on its reliability and discriminative ability. Our study was conducted among a mostly college educated sample of African-Americans in Jackson MS; thus, these results cannot be generalized to samples in other geographic areas or other race/ethnic groups. However, JHS recruited individuals from three counties around Jackson, and this cohort is one of the best characterized African-American samples in the U.S.
In our sample of African-Americans, we found that the accumulation of adverse components of the sleep environment and in-bed behaviors were associated with poor sleep. These results highlight the importance of aspects of the physical bedroom environment as well as in-bed behaviors in relation to sleep duration and quality. In summary, these findings support the development and testing of interventions that consider both contextual factors and individual behaviors as strategies for improving sleep duration and continuity.
Supplementary Material
Acknowledgement
The authors wish to thank the staffs and participants of the JHS. We also thank Manasvi Sundar for her assistance with the manuscript.
Conflict of interest statement. Dr. Redline reports consulting fees from Eisai Inc and Jazz Inc.
Funding
Research reported in this publication was supported by the National Heart, Lung, and Blood Institute, (NHLBI) K01HL138211 and 3R01HL11006. The Jackson Heart Study is supported and conducted in collaboration with Jackson State University (HHSN268201800013I), Tougaloo College (HHSN268201800014I), the Mississippi State Department of Health (HHSN268201800015I) and the University of Mississippi Medical Center (HHSN268201800010I, HHSN268201800011I and HHSN268201800012I) contracts from the National Heart, Lung, and Blood Institute (NHLBI) and the National Institute for Minority Health and Health Disparities (NIMHD). This work was funded, in part, by the Intramural Program at the National Institutes of Health (NIH), National Institute of Environmental Health Sciences (NIEHS, Z1AES103325-01).
References
- 1. Chen X, et al. . Racial/Ethnic differences in sleep disturbances: The Multi-Ethnic Study of Atherosclerosis (MESA). Sleep. 2015;38(6):877–888. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Hale L, et al. . Racial differences in self-reports of sleep duration in a population-based study. Sleep. 2007;30(9):1096–1103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Johnson DA, et al. . Are sleep patterns influenced by race/ethnicity – a marker of relative advantage or disadvantage? Evidence to date. Nat Sci Sleep. 2019;11:79–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Cappuccio FP, et al. . Sleep duration predicts cardiovascular outcomes: a systematic review and meta-analysis of prospective studies. Eur Heart J. 2011;32(12):1484–1492. [DOI] [PubMed] [Google Scholar]
- 5. Jackson CL, et al. . Sleep as a potential fundamental contributor to disparities in cardiovascular health. Annu Rev Public Health. 2015;36:417–440. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Curtis DS, et al. . Habitual sleep as a contributor to racial differences in cardiometabolic risk. Proc Natl Acad Sci U S A. 2017;114(33):8889–8894. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Carnethon MR, et al. ; American Heart Association Council on Epidemiology and Prevention; Council on Cardiovascular Disease in the Young; Council on Cardiovascular and Stroke Nursing; Council on Clinical Cardiology; Council on Functional Genomics and Translational Biology; and Stroke Council. Cardiovascular Health in African Americans: A scientific statement from the American Heart Association. Circulation. 2017;136(21):e393–e423. [DOI] [PubMed] [Google Scholar]
- 8. Johnson DA, et al. . Environmental determinants of insufficient sleep and sleep disorders: Implications for population health. Curr Epidemiol Rep. 2018;5(2):61–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Nivison ME, et al. . An analysis of relationships among environmental noise, annoyance and sensitivity to noise, and the consequences for health and sleep. J Behav Med. 1993;16(3):257–276. [DOI] [PubMed] [Google Scholar]
- 10. Billings ME, et al. . Physical and social environment relationship with sleep health and disorders. Chest. 2020;157(5):1304–1312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Reardon SF, et al. . Neighborhood Income composition by household race and income, 1990–2009. The Annals of the American Academy of Political and Social Science. 2015;660:78–97. [Google Scholar]
- 12. Jackson CL. Housing conditions as environmental and social determinants of sleep health. In: Duncan DT, Kawachi I, Redline S, eds. The Social Epidemiology of Sleep. Oxford, UK: Oxford University Press; 2019:373–407. [Google Scholar]
- 13. Altevogt BM, et al., eds. Sleep Disorders and Sleep Deprivation: An Unmet Public Health Problem. Washington, DC: The National Academies Collection: Reports funded by National Institutes of Health; 2006. [PubMed] [Google Scholar]
- 14. Muzet A. Environmental noise, sleep and health. Sleep Med Rev. 2007;11(2):135–142. [DOI] [PubMed] [Google Scholar]
- 15. Mujahid MS, et al. . Assessing the measurement properties of neighborhood scales: from psychometrics to ecometrics. Am J Epidemiol. 2007;165(8):858–867. [DOI] [PubMed] [Google Scholar]
- 16. Johnson DA, et al. . The Social patterning of sleep in African Americans: Associations of socioeconomic position and neighborhood characteristics with sleep in the Jackson Heart Study. Sleep. 2016;39(9):1749–1759. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Troxel WM, et al. . Broken Windows, Broken Zzs: Poor housing and neighborhood conditions are associated with objective measures of sleep health. J Urban Health. 2020;97(2):230–238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Chambers EC, et al. . Sleep and the housing and neighborhood environment of Urban Latino Adults living in low-income housing: The AHOME Study. Behav Sleep Med. 2016;14(2):169–184. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Simonelli G, et al. . Sleep and quality of life in urban poverty: the effect of a slum housing upgrading program. Sleep. 2013;36(11):1669–1676. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Griefahn B, et al. . Noise emitted from road, rail and air traffic and their effects on sleep. J Sound Vibr. 2006;295:129–140. [Google Scholar]
- 21. Jakovljević B, et al. . Road traffic noise and sleep disturbances in an urban population: cross-sectional study. Croat Med J. 2006;47(1):125–133. [PMC free article] [PubMed] [Google Scholar]
- 22. Johnson DA, et al. . Influence of neighbourhood-level crowding on sleep-disordered breathing severity: mediation by body size. J Sleep Res. 2015;24(5):559–565. [DOI] [PubMed] [Google Scholar]
- 23. Sandberg JC, et al. . Association between housing quality and individual health characteristics on sleep quality among Latino farmworkers. J Immigr Minor Health. 2014;16(2):265–272. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Bhat S, et al. . “To sleep, perchance to tweet”: in-bed electronic social media use and its associations with insomnia, daytime sleepiness, mood, and sleep duration in adults. Sleep Health. 2018;4(2):166–173. [DOI] [PubMed] [Google Scholar]
- 25. Exelmans L, et al. . Bedtime mobile phone use and sleep in adults. Soc Sci Med. 2016;148:93–101. [DOI] [PubMed] [Google Scholar]
- 26. Lemola S, et al. . Adolescents’ electronic media use at night, sleep disturbance, and depressive symptoms in the smartphone age. J Youth Adolesc. 2015;44(2):405–418. [DOI] [PubMed] [Google Scholar]
- 27. Gradisar M, et al. . The sleep and technology use of Americans: findings from the National Sleep Foundation’s 2011 Sleep in America poll. J Clin Sleep Med. 2013;9(12):1291–1299. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Fobian AD, et al. . Impact of media use on adolescent sleep efficiency. J Dev Behav Pediatr. 2016;37(1):9–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Rottapel RE, et al. . Adapting sleep hygiene for community interventions: a qualitative investigation of sleep hygiene behaviors among racially/ethnically diverse, low-income adults. Sleep Health. 2020;6(2):205–213. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Fuqua SR, et al. . Recruiting African-American research participation in the Jackson Heart Study: methods, response rates, and sample description. Ethn Dis. 2005;15(4 Suppl 6):S6–18. [PubMed] [Google Scholar]
- 31. Johnson DA, et al. . Prevalence and correlates of obstructive sleep apnea among African Americans: the Jackson Heart Sleep Study. Sleep. 2018;41(10). doi: 10.1093/sleep/zsy154. [DOI] [PMC free article] [PubMed]
- 32. Morgenthaler T, et al. ; Standards of Practice Committee; American Academy of Sleep Medicine. Practice parameters for the use of actigraphy in the assessment of sleep and sleep disorders: an update for 2007. Sleep. 2007;30(4):519–529. [DOI] [PubMed] [Google Scholar]
- 33. Cole RJ, et al. . Automatic sleep/wake identification from wrist activity. Sleep. 1992;15(5):461–469. [DOI] [PubMed] [Google Scholar]
- 34. Watson NF, et al. ; Consensus Conference Panel; Non-Participating Observers; American Academy of Sleep Medicine Staff. Recommended amount of sleep for a healthy adult: A joint consensus statement of the American Academy of Sleep Medicine and Sleep Research Society. J Clin Sleep Med. 2015;11(6):591–592. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Spilsbury JC, et al. . Sleep environments of children in an Urban U.S. Setting exposed to interpersonal violence. Behav Sleep Med. 2016;14(6):585–601. [DOI] [PubMed] [Google Scholar]
- 36. Johnson DA, et al. Black–white differences in housing type and sleep duration as well as sleep difficulties in the United States. Int J Environ Res Public Health. 2018;15(4):564. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Jackson CL, et al. . Racial/ethnic disparities in short sleep duration by occupation: the contribution of immigrant status. Soc Sci Med. 2014;118:71–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Williams NJ, et al. . Social and behavioral predictors of insufficient sleep among African Americans and Caucasians. Sleep Med. 2016;18:103–107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Centers for Disease Control and Prevention (CDC). Effect of short sleep duration on daily activities – United States, 2005–2008. MMWR Morb Mortal Wkly Rep. 2011;60(8):239–242. [PubMed] [Google Scholar]
- 40. Jackson CL, et al. . Racial disparities in short sleep duration by occupation and industry. Am J Epidemiol. 2013;178(9):1442–1451. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Liu Y, et al. . Prevalence of healthy sleep duration among adults–United States, 2014. MMWR Morb Mortal Wkly Rep. 2016;65(6):137–141. [DOI] [PubMed] [Google Scholar]
- 42. Chattu VK, et al. . Do disparities in sleep duration among racial and ethnic minorities contribute to differences in disease prevalence? J Racial Ethn Health Disparities. 2019;6(6):1053–1061. [DOI] [PubMed] [Google Scholar]
- 43. Jackson CL, et al. Concordance between self-reported and actigraphy-assessed sleep duration among African-American adults: findings from the Jackson Heart Sleep Study. Sleep. 2020;43(3). doi: 10.1093/sleep/zsz246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Johnson DA, et al. . Associations between the built environment and objective measures of sleep: the multi-ethnic study of atherosclerosis. Am J Epidemiol. 2018;187(5):941–950. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Johnson DA, et al. . The neighborhood social environment and objective measures of sleep in the multi-ethnic study of atherosclerosis. Sleep. 2017;40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46. Troxel WM, et al. . Neighborhood disadvantage is associated with actigraphy-assessed sleep continuity and short sleep duration. Sleep. 2018;41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47. Chang AM, et al. . Human responses to bright light of different durations. J Physiol. 2012;590(13):3103–3112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48. Vetter C, et al. . Light me up? Why, when, and how much light we need. J Biol Rhythms. 2019;34(6):573–575. [DOI] [PubMed] [Google Scholar]
- 49. Hale L, et al. . Recent updates in the social and environmental determinants of sleep health. Curr Sleep Med Rep. 2015;1(4):212–217. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50. Hicken MT, et al. . “Every shut eye, ain’t sleep”: The role of racism-related vigilance in racial/ethnic disparities in sleep difficulty. Race Soc Probl. 2013;5(2):100–112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51. Hu RF, et al. . Effects of earplugs and eye masks on nocturnal sleep, melatonin and cortisol in a simulated intensive care unit environment. Crit Care. 2010;14(2):R66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52. Burkhart K, et al. . Amber lenses to block blue light and improve sleep: a randomized trial. Chronobiol Int. 2009;26(8):1602–1612. [DOI] [PubMed] [Google Scholar]
- 53. Johnson DA, et al. . Objective measures of sleep apnea and actigraphy-based sleep characteristics as correlates of subjective sleep quality in an epidemiologic study: The Jackson Heart Sleep Study. Psychosom Med. 2020;82(3):324–330. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.