Fig 1.
Study sample derivation flowchart – BRFSS 2016–2023.
This flow diagram illustrates the selection process for the final analytic sample used in the study examining associations between sleep duration and health outcomes. Starting with 3,911,129 adult respondents from the 2016–2023 Behavioral Risk Factor Surveillance System (BRFSS), 1,712,145 participants were excluded due to missing data on sleep duration. An additional 2,788,668 respondents were excluded for missing data on one or more of the four primary outcome variables: depression (n = 1,299,214), poor mental health days (n = 1,190,226), poor physical health days (n = 289,872), and general health status (n = 9,356). The resulting final analytic sample included 318,654 respondents with complete information. These were categorized into: short sleep (≤5 hours, n = 65,579; 19.7%), recommended sleep (6–8 hours, n = 224,164; 70.5%), and long sleep (≥9 hours, n = 31,257; 9.8%).
Table 1.
Sociodemographic and health characteristics of the study population by sleep duration category (BRFSS 2016-2023).
Table 2.
Weighted prevalence of self-reported sleep duration categories across survey years in the United States, BRFSS 2016–2022.
Table 3.
Average treatment effects of sleep duration on incidence of depression.
Table 4.
Average treatment effects of sleep duration on incidence of poor mental health days.
Table 5.
Average treatment effects of sleep duration on incidence of poor physical health days.
Table 6.
Adjusted predicted probabilities of general health categories by sleep duration.