0733 Elucidating Sleep-Related Health Disparities across The Socioeconomic Spectrum Leveraging a Large-Scale Sleep Registry

Abstract

Abstract Introduction Sleep disparities have been implicated as a contributor to overall health disparities in socially disadvantaged groups. Despite epidemiological studies reporting sleep deprivation and poorer sleep quality among those with low socioeconomic status and group minorities, little is known about the extent to which sleep disorders such as sleep-disordered breathing (SDB), varies across the socioeconomic spectrum. Methods A retrospective cohort study was conducted utilizing data from the Cleveland Clinic Sleep Laboratory Registry. All adults who underwent diagnostic or split (baseline diagnostic) polysomnogram (PSG) or home sleep apnea test (HSAT) were included in the study. Area Deprivation Index (ADI), a biomarker of neighborhood socioeconomic disadvantage, was calculated by national rank, i.e. 25th, 50th and 75th percentiles; higher quartiles reflect greater deprivation. Generalized linear models adjusted for age, race, sex, body mass index(kg/m2) and primary payer were used to investigate association of ADI with SDB breathing measures (apnea hypopnea index, (AHI) and sleep-related hypoxemia (percentage of total sleep time <90%SaO2, [TST<90]). Results The analytic sample included 81,212 sleep studies; 60,013(74%) were PSG and 21,199(26%) HSAT with age: 52.0[41.0, 62.3], 49% females, 19% black race, with BMI=34.5±8.5 kg/m2, 44% with Medicaid and Medicare. Median ADI National Rank 59.0[39.0, 81.0] with higher 4th quartiles in PSG versus HSAT:29.1% vs 16.3%,p<0.001. In the PSG group, ADI was associated with hypoxia measures: TST<90(coefficient p<0.0001), model R2=0.171; mean SaO2(p<0.0001), R2=0.189; minimum SaO2(p<0.0001), R2=0.169; all measures were higher with higher ADI quartiles. In the HSAT group, ADI was associated with mean SaO2(p<0.0001), R2=0.188 and minimum SaO2(p<0.0001), R2=0.181 with all measures being higher with higher ADI quartiles. AHI was associated with ADI(p=0.0032), R2=0.239; but least squares mean AHI did not differ among ADI quartiles in PSG and HSAT groups. Interactions were observed between ADI and age, BMI and male sex (p<0.05), but not race. Conclusion Sleep-related hypoxia was greater among patients living in areas of higher deprivation when considering rankings of neighborhoods by socioeconomic disadvantage. Further understanding of the reason for this sleep disorder-related disparity is needed, i.e. further characterizing theoretical domains and social and geographic determinants of income, education, employment, and housing quality with the overarching goal to improve disparities in health. Support (If Any)