Abstract P270: Neighborhood Socioeconomic Status and Sleep Duration Among U.S. Adolescents: Does the Association Differ by Race?

Background: Short sleep duration is linked to adverse cardiovascular risk factors among adolescents and racial disparities in sleep among this group is well documented. Thus, identifying determinants and its differential impact on sleep duration by race may be important for informing sleep interventions and reducing sleep disparities. Evidence suggests neighborhood socioeconomic status (NSES) may play a role in shorter sleep duration among adolescents, but few studies have examined whether this relationship differs by race. Furthermore, previous studies were limited by sample size, geographic representation to specific U.S. cities and racial/ethnic diversity with mostly non-Hispanic White samples. The purpose of our study was to investigate the association between NSES and short sleep duration and determine if this relationship varies by race among a diverse sample of adolescents. Methods: Participants (n=17,746) are from Wave I of the National Longitudinal Study of Adolescent to Adult Health, a nationally representative multi-ethnic study of health behaviors during adolescence and outcomes during adulthood. Sleep duration was self-reported as total hours of sleep at night, and categorized as short (6-13 years: <9 hours, 14-17 years: <8 hours, 18-25 years: <7 hours) vs. normal (6-13 years: 9-11 hours, 14-17 years: 8-10 hours, 18-25 years: 7-9 hours ). NSES was a summary z-score consisting of three census-block measures: unemployment rate, proportion of households with income below poverty level, and public assistance. Participants were categorized as living in low (>median z-score) or high (≤ median z-score) NSES. Log-binomial regression models examined the association between NSES (low vs. high) and short sleep duration and tested for racial differences. Models adjusted for race, sex, age, and highest parental education. Results: The study population was 51% female, 66% White, 15% African-American, 12% Hispanic, 4% Asian, 3% other racial group, and mean age was 15.4 years (Standard Error=0.11). The prevalence of short sleep duration was 41% and 48% for low NSES. There was no significant association between NSES and short sleep duration before (Prevalence Ratio (PR):1.02; 95% Confidence Interval (CI): 0.96-1.09) and after adjusting for individual sociodemographic characteristics (PR: 1.00; 95% CI: 0.93-1.06). There was also no significant race-by-NSES interaction (p=0.21). Conclusion: Findings suggests NSES is not associated with short sleep duration regardless of race and that other factors may play a larger role in sleep among this sample of adolescents. The inconsistency with findings from prior studies could be due to our inclusion of a larger, racially diverse, and nationally representative sample. Future studies should explore whether other neighborhood and contextual factors influence sleep and if racial differences exist in these relationships.