Demographic characteristics associated with circadian rest-activity rhythm patterns: a cross-sectional study

Jingen Li,Virend K Somers,Francisco Lopez-Jimenez,Junrui Di,Naima Covassin

DOI: https://doi.org/10.1186/s12966-021-01174-z

2021-08-18

Abstract:Background: Rest-activity rhythm (RAR), a manifestation of circadian rhythms, has been associated with morbidity and mortality risk. However, RAR patterns in the general population and specifically the role of demographic characteristics in RAR pattern have not been comprehensively assessed. Therefore, we aimed to describe RAR patterns among non-institutionalized US adults and age, sex, and race/ethnicity variation using accelerometry data from a nationally representative population. Methods: This cross-sectional study was conducted using the US National Health and Nutrition Examination Survey (NHANES) 2011-2014. Participants aged ≥20 years who were enrolled in the physical activity monitoring examination and had at least four 24-h periods of valid wrist accelerometer data were included in the present analysis. 24-h RAR metrics were generated using both extended cosinor model (amplitude, mesor, acrophase and pseudo-F statistic) and nonparametric methods (interdaily stability [IS] and intradaily variability [IV]). Multivariable linear regression was used to assess the association between RAR and age, sex, and race/ethnicity. Results: Eight thousand two hundred participants (mean [SE] age, 49.1 [0.5] years) were included, of whom 52.2% were women and 67.3% Whites. Women had higher RAR amplitude and mesor, and also more robust (pseudo-F statistic), more stable (higher IS) and less fragmented (lower IV) RAR (all P trend < 0.001) than men. Compared with younger adults (20-39 years), older adults (≥ 60 years) exhibited reduced RAR amplitude and mesor, but more stable and less fragmented RAR, and also reached their peak activity earlier (advanced acrophase) (all P trend < 0.001). Relative to other racial/ethnic groups, Hispanics had the highest amplitude and mesor level, and most stable (highest IS) and least fragmented (lowest IV) RAR pattern (P trend < 0.001). Conversely, non-Hispanic blacks had the lowest peak activity level (lowest amplitude) and least stable (lowest IS) RAR pattern (all P trend < 0.001). Conclusions: In the general adult population, RAR patterns vary significantly according to sex, age and race/ethnicity. These results may reflect demographic-dependent differences in intrinsic circadian rhythms and may have important implications for understanding racial, ethnic, sex and other disparities in morbidity and mortality risk.