Abstract
Abstract Introduction Circadian disruptions are associated with increased risk for morbidity and mortality. However, it is unclear whether these associations vary by race/ethnicity. We aim to explore, 1) Examine whether rest-activity rhythm (RAR) patterns vary across race/ethnicity among a representative sample of the US adult population, and 2) Examine the interaction of race/ethnicity and sex in RAR patterns. Methods The study was based on the National Health and Nutrition Examination Survey (2013–2014) with data from participants who wore a physical activity monitor (PAM ActiGraph accelerometer model GT3X+) continuously for seven consecutive days. After excluding pregnant participants and those with sleep minutes < 120 and >1200, the analytic sample included was 4,722 adults. The cosinor method was used based on PAM minutes time-series data to RAR outcome variables: mesor, amplitude, acrophase. and robustness. The main predictor of interest was race/ethnicity (Asian, Black, Hispanic, multi/other, and White). A sex-race/ethnicity interaction was tested to assess if racial/ethnic differences in RAR patterns differ between males and females. Other covariates included were age, marital status education federal poverty level, and employment status. Eight adjusted Generalized linear models (GLM) : four with race/ethnicity as the main predictors and four multiplicative interaction models with the product term of race/ethnicity and sex. An adjusted Wald test was used to test for interaction. Results Compared with White adults, Hispanic adults had increased mesor levels (ß=0.10; 95% CI:0.07;0.13), increased RAR amplitude (ß=0.10; 95% CI:0.06;0.13), and increased robustness (ß=0.07; 95% CI:0.03;0.11) whereas Black adults had decreased amplitude (ß=-0.11; 95% CI:-0.16;-0.07) and decreased robustness ( (ß= -0.14; 95% CI:-0.20; -0.09) compared to White adults. Similarly, Asians had decreased amplitude (ß=-0.06; 95% CI:-0.10;-0.01) and decreased robustness (ß=-0.11; 95% CI:-0.16; -0.06). A significant sex-race/ethnicity interaction was found for amplitude F(4,12)=3.94;p=0.029 and robustness (4,12)=6.02;p< 0.001. Conclusion RAR is associated with race/ethnicity, and this association varies by sex. Notably, Hispanic adults had increased mesor, amplitude, and robustness compared to Whites. Conversely, Black and Asian populations shared decreased amplitude and robustness compared to Whites. Future studies may consider further investigation of circadian health by race/ethnicity and sex for community intervention. Support (if any) K01HL135452, K07AG052685, R01AG072644, R01HL152453, R01MD007716, R01HL142066, R01AG067523, R01AG056031, and R01AG075007