0823 Social Determinants of Sleep Disorders Among Multiethnic Americans in the NIH All of Us Research Program

Abstract Introduction Studies show that social characteristics associated with good health, such as a high level of education, being married, and being employed, are associated with healthier sleep health. Accordingly, we leveraged All of Us (AOU), an NIH research, to analyze sleep disorders prevalence and sociodemographic covariates. Methods Phecodes for diagnoses were derived from the ICD-9/10 billing codes in participants’ electronic health records (EHRs). We used 0 phecodes of the parent category of sleep disorders (insomnia and sleep apnea) (phecode 327) in a person’s EHR to classify individuals as not having a diagnosis and 2+ phecodes as having a diagnosis (those with only one instance in their EHR were dropped). Logistic regression analysis was conducted to assess associations between sleep disorders and race/ethnicity, sex, age, and education level using the R statistical framework. Results A total of 214, 206 participants were available for analysis; of whom 61.3% were female (mean [SD] age, 51.7 [16.6] years). Women were 1.27xmore likely to have a diagnosed sleeping problem (95% CI: 1.31,1.23; p=1.31×10-61). We observed higher rates of sleep disorders in older individuals; compared to persons between 18-30 years old, those aged 65 years or older were 4.74 times more likely to have a diagnosable sleep disorder (95% CI: 3.83, 4.33; p< 10-300). We also observed a trend of greater education attainment corresponding to an increased likelihood of sleep problems; this was most extreme when comparing individuals with some college versus those who did not complete high school (OR=1.38; 95% CI: 1.30, 1.46; p=3.65×10-29). Individuals with reported income < 25K/year were 1.44x more likely to be diagnosed than those earning >100K (95% CI:1.51, 1.37; p=1.05×10-51). Individuals identified as Black were least likely to have a diagnosis (OR=0.52; 95% CI: 0.50, 0.54; p=2.2that 2×10-208). Conclusion Among All Of Us populations, self-identified Blacks, men, younger adults, higher-income, and non-college-educated patients are less likely to have recorded insomnia or sleep apnea diagnoses. Some of these unexpected within-group differences could be due to self-reported versus clinically recorded diagnoses and access to healthcare. Planned analyses aim to explore further social determinants of health by incorporating geospatial and available self-report data on exposures Support (if any) 1HL152453, R01MD007716, R01HL142066