Epidemiology of High-Risk Sleep Phenotypes in a Diverse Cohort of Older Adults

TL;DR

This study examines how common short and long sleep patterns are in older adults and how they relate to health issues like depression and physical impairment.

Contribution

The paper provides new insights into the epidemiology of high-risk sleep phenotypes in a racially and ethnically diverse older adult cohort.

Findings

ISSD was more common in Black participants, while LS was more common in non-Hispanic White participants.

Depression and anxiety were higher in both ISSD and LS groups compared to normal sleepers.

Physical impairment was significantly higher in the LS group compared to normal sleepers.

Abstract

Insomnia with objective short sleep duration (ISSD) and objective long sleep duration (LS) are high-risk sleep phenotypes that have not been examined in a diverse group of older adults.

We studied 1001 adults in the DORMIR study (average age 64.4[8.6] years, 65.7% female; 28.4% Black, 31.6% Hispanic, 40.1% non-Hispanic White). Insomnia was defined as self-reporting any of the following > =3 times/week: trouble getting to sleep within 30min, waking in the middle of the night/early morning or taking sleep medication. Objective overnight sleep duration was assessed by actigraphy (averaged over 5.9[0.9] days). We examined ISSD (insomnia and sleep duration< 6h), LS (sleep duration>8h), and normal sleep (NS; no insomnia and sleep duration=6-8h). Using ANOVA F-tests or chi-square, we assessed phenotype prevalence overall and according to race/ethnicity, and also examined, within each race/ethnicity group, whether comorbidities (e.g., depression, anxiety, dementia status, physical impairment [Short Physical Performance Battery< 10]) differed among phenotypes.

ISSD prevalence was 8.1% overall and was higher among Blacks (13.4%; p < 0.001). LS prevalence was 22.1% overall and was higher among Whites (27.2%; p < 0.001). Compared to NS, depression was higher in ISSD and LS across all groups. Anxiety was higher in Hispanics and Whites with ISSD and LS compared with NS. Across all groups, physical impairment was significantly higher in LS compared with NS. Dementia status was not different across phenotypes.

The prevalence of high-risk sleep phenotypes differs across race/ethnicity groups, while comorbidities associated with phenotypes were similar. Future work should examine the association of phenotypes with adverse outcomes.