Are sleep duration and eating behaviors associated with glucose metabolism in working-age adults at the US-Mexico border?
Uncontrolled eating and short sleep duration are independently associated with higher A1c levels in adults at the US-Mexico border, suggesting distinct pathways affecting metabolic health.
Abstract Introduction Prior research studying border populations has shown that disrupted and insufficient sleep is linked to unhealthy eating patterns and poorer overall sleep quality, shaped in part by the sociocultural factors unique to this region. The present secondary analysis sought to examine the overlapping roles of sleep and eating behaviors on glucose metabolism in border-dwelling adults. Methods Data were collected from 992 working-age adults (age 25-60) living at the US-Mexico Border. Glucose metabolism was assessed using a point-of-case blood spot assay of hemoglobin A1c (A1c). Sleep variables included actigraphic (Mini-Motionlogger) and self-reported (Pittsburgh Sleep Quality Index) sleep duration, and insomnia symptoms (Insomnia Severity Index). Eating behavior was assessed with three-factor eating questionnaire, with 3 subscales: Cognitive Restraint (CR), Uncontrolled Eating (UE), and Emotional Eating (EE). Linear regression analyses adjusted for age, sex, and acculturation examined associations between A1c and both sleep (actigraphic and self-reported sleep duration and insomnia severity) and eating (CR, UE, EE). Forward stepwise analyses examined unique contributions of these 6 variables on A1c. Results Adjusted for covariates, shorter actigraphic sleep duration was associated with higher A1c (B=-0.08 / hour, p 0.05), but self-reported sleep duration and insomnia were not associated with A1c. Greater UE was associated with higher A1c (B=0.02, p 0.05), as was greater EE (B= 0.05, p 0.01), but CR was not. Greater UE was also associated with insomnia (B=0.22, p 0.0005), actigraphic (B=-1.05, p 0.05) and self-reported (B=-2.28, p 0.05) sleep duration; greater EE was associated with insomnia (B=0.50, p 0.0005) and both shorter actigraphic (B=-2.28, p 0.05) and self-reported (B=-0.05, p 0.05) sleep duration. CR was not associated with sleep variables. A forward stepwise model adjusted for covariates found that significant variance was explained by UE (B=0.02, p 0.05) and actigraphic sleep duration (B=-0.08, p 0.05). Conclusion Although unhealthy eating patterns are also separately associated with sleep duration and quality, both Uncontrolled Eating and actigraphic short sleep duration are related to A1c likely via distinct pathways. Integrated, culturally informed interventions targeting both sleep and eating behaviors may hold promise for improving the metabolic health of adults at the U.S.-Mexico Border. Support (if any) R01MD011600, R01MH135978
Arce et al. (Fri,) studied this question.