Abstract Introduction Alignment between circadian and sleep timing (“phase angle of entrainment”, PAE) is defined as the difference in time between dim light melatonin onset (DLMO) and sleep onset. We and others have identified significant sex differences in PAE. It is important to determine whether modifiable behavioral factors, like caffeine and alcohol use, contribute to these sex differences. Caffeine and alcohol use differ by sex and influence circadian rhythms and sleep timing, making them plausible mechanisms through which sex differences in PAE may arise. In this study, we examined the extent to which self-reported caffeine and alcohol use in young adults contribute to variations in PAE and whether these associations differ by sex. Methods 118 college students (54% female) from a single university participated in this hybrid, in-laboratory/at-school ~30-day study. Sleep timing and Sleep Regularity Index (SRI) were calculated using actigraphy and sleep diaries. During the 30 days, circadian timing (salivary DLMO) was assessed once, and caffeine and alcohol use were determined using ~7-day photographic food diaries. Multiple linear regression models were performed with PAE as the outcome variable, and sex (M or F), caffeine use (yes/no), alcohol use (yes/no), and SRI as predictor variables. Results There were no sex differences in caffeine or alcohol use in this dataset. Caffeine use was associated with larger PAE (β=0.96, p=0.045; Yes Caffeine=3.4 h±1.4 h; No Caffeine=2.5 h±1.3 h), ii) Lower SRI was associated with larger PAE (β=-0.03, p=0.03), iii) Sex (p=0.11) and alcohol use (p=0.96) were not significant predictors in this multivariate model. Conclusion College students in our sample who consumed caffeine or had less regular sleep had (on average) a larger PAE. A larger PAE may have meaningful consequences for students: misalignment between internal circadian rhythms and scheduled activities can impair learning, cognitive performance, and overall health. Sex differences in PAE were reported previously in this data set; this effect was not a significant predictor in the current models that included caffeine, alcohol use, and sleep regularity. Therefore, behavioral factors such as caffeine use and sleep regularity may contribute to some of the sex-related variability in PAE. Support (if any) NIH R01GM105018, UL1 RR025758, Samsung Electronics, and NEC
McCullar et al. (Fri,) studied this question.