Diabetes is highly prevalent and is associated with dietary behaviors. Time-restricted eating, which consolidates caloric intake to a shortened eating duration, has demonstrated improvement in metabolic health. Timing of eating could also impact metabolism. Our objective was to examine whether the timing of eating was associated with metabolic health independently of eating duration. Data (n = 7619) are from four cycles (2005-2012) of the National Health and Nutrition Examination Survey (NHANES), a nationally representative U.S. survey that included surveys, physical examinations, and dietary recalls. The primary exposures are eating duration and eating start time estimated from two non-consecutive dietary recalls. Primary outcomes were fasting glucose and estimated insulin resistance using the homeostatic model assessment method (HOMA-IR). The mean (95% CI) eating duration was 12.0 h (11.9-12.0) and the mean (95% CI) start time was 8:21 (8:15-8:26). Earlier eating start time was significantly associated with lower fasting glucose and estimated insulin resistance but eating interval duration was not. Every hour later that eating commenced was associated with approximately 0.6% higher glucose level and 3% higher HOMA-IR (both p < 0.001). In this cross-sectional study, earlier eating start time was associated with more favorable metabolic measures, indicating that meal timing is another important characteristic of dietary patterns that may influence metabolism.
Keywords: circadian rhythm; diabetes mellitus; diet; epidemiology; insulin resistance.