Ecological momentary assessment of between- and within-person sleep quality as a predictor of disordered eating behaviors among young women with disordered eating.

Although disordered eating and sleep problems often co-occur, there has been limited research on mechanisms underlying their association. Previous studies using ecological momentary assessment (EMA) to evaluate links between sleep quality and disordered eating behaviors (e.g., loss-of-control eating, restricting, purging, excessive exercise) have largely not found statistically significant temporal associations; however, studies with larger samples are warranted. We used EMA to evaluate sleep quality and next-day engagement in disordered eating behaviors. Participants were 137 women aged 18-25 (M = 20.22, SD = 1.78) with current disordered eating behaviors (DEBs). Participants completed seven days of EMA, in which they reported the past night's sleep quality and answered six surveys daily about engagement in DEBs. DEBs were aggregated at the day level and dichotomized as present or absent (n = 438 days). We conducted four multilevel logistic regressions, nesting days within subjects. We predicted loss-of-control eating, purging, restricting, and excessive exercise using past night's sleep quality and survey day as fixed effects with random intercepts by subjects. Results indicated that neither between- nor within-person sleep quality predicted engagement in any DEBs (all ps > .10). Findings replicated past studies that found no or few prospective associations between sleep quality and engagement in DEBs. It is possible that other sleep metrics (e.g., variability in sleep and wake timing) may predict DEBs or associations may vary by gender. Future studies should examine subjective and objective sleep indices across genders to better understand potential links between sleep and DEBs.