Epidemiologic Methods to Estimate Insufficient Sleep in the US Population

Bibliographic Data

ID15515901
AuthorsGirardin Jean‐Louis (0000-0001-6777-2724, Columbia University Irving Medical Center, corresponding author), Arlener D Turner (0000-0002-0493-7736, Columbia University Irving Medical Center), Azizi Seixas (0000-0003-0843-2679, Columbia University Irving Medical Center), Peng Jin (0000-0001-5095-5743, University Medical Center), Diana Margot Rosenthal (0000-0002-1238-5003, University of London), Mengling Liu (0000-0001-9758-8522, University Medical Center), George Avirappattu (0000-0002-6572-6406, Kean University)
Year2020
Volume17
Issue24
Pages9337-9337
Publication date2020-12-14
Open AccessYes
TypeARTICLE
VenueInternational Journal of Environmental Research and Public Health (JOURNAL)
Journal identifiersISSN: 1661-7827 • E-ISSN: 1660-4601
PublisherMultidisciplinary Digital Publishing Institute (PUBLISHER • CH)
DOI10.3390/ijerph17249337
PMID33327388
OpenAlexW3111739205
LanguageEN
Citations received4
References cited14

This study explored the divergence in population-level estimates of insufficient sleep (<6 h) by examining the explanatory role of race/ethnicity and contrasting values derived from logistic and Poisson regression modeling techniques. We utilized National Health and Nutrition Examination Survey data to test our hypotheses among 20-85 year-old non-Hispanic Black and non-Hispanic White adults. We estimated the odds ratios using the transformed logistic regression and Poisson regression with robust variance relative risk and 95% confidence intervals (CI) of insufficient sleep. Comparing non-Hispanic White (10176) with non-Hispanic Black (4888) adults (mean age: 50.61 ± 18.03 years, female: 50.8%), we observed that the proportion of insufficient sleepers among non-Hispanic Blacks (19.2-26.1%) was higher than among non-Hispanic Whites (8.9-13.7%) across all age groupings. The converted estimated relative risk ranged from 2.12 (95% CI: 1.59, 2.84) to 2.59 (95% CI: 1.92, 3.50), while the estimated relative risks derived directly from Poisson regression analysis ranged from 1.84 (95% CI: 1.49, 2.26) to 2.12 (95% CI: 1.64, 2.73). All analyses indicated a higher risk of insufficient sleep among non-Hispanic Blacks. However, the estimates derived from logistic regression modeling were considerably higher, suggesting the direct estimates of relative risk ascertained from Poisson regression modeling may be a preferred method for estimating population-level risk of insufficient sleep

Environmental health · Population · Sleep (system call · Computer Science · Environmental Science · Medicine · Obesity, Physical Activity, Diet · Sleep and related disorders · Sleep and Work-Related Fatigue

Tools

Open DOIOpen Access