Epidemiologic Methods to Estimate Insufficient Sleep in the US Population

Area of Science:

• Epidemiology
• Biostatistics

Background:

• Population-level sleep duration estimates can vary based on analytical methods.
• Race and ethnicity are significant factors influencing health disparities, including sleep patterns.

Purpose of the Study:

• To investigate the divergence in population-level estimates of insufficient sleep (<6 hours).
• To examine the role of race/ethnicity in insufficient sleep.
• To compare logistic and Poisson regression modeling techniques for estimating relative risk of insufficient sleep.

Main Methods:

• Utilized National Health and Nutrition Examination Survey (NHANES) data.
• Analyzed data from non-Hispanic Black and non-Hispanic White adults aged 20-85.
• Estimated odds ratios and relative risks using logistic regression and Poisson regression with robust variance.

Main Results:

• Non-Hispanic Black adults exhibited a higher proportion of insufficient sleep (19.2-26.1%) compared to non-Hispanic White adults (8.9-13.7%).
• All analyses indicated a significantly higher risk of insufficient sleep among non-Hispanic Black individuals.
• Logistic regression yielded considerably higher relative risk estimates than Poisson regression.

Conclusions:

• Non-Hispanic Black adults face a disproportionately higher risk of insufficient sleep.
• Poisson regression modeling provides more direct and potentially preferred estimates for population-level sleep risk compared to logistic regression.
• Methodological choices in statistical analysis can impact the understanding of health disparities in sleep duration.