Introduction: Meal timing may affect type 2 diabetes risk due to the circadian regulation of glucose metabolism. However, the optimal meal timing for lowering diabetes risk remains unknown. Objective and Hypothesis: We aimed to identify the optimal meal timing for diabetes prevention. Since glucose tolerance peaks in the morning, we hypothesized that eating more in the morning is associated with a lower risk of diabetes when daily diet quantity and quality are held constant. Methods: We included 7716 participants from the Hispanic Community Health Study/Study of Latinos (HCHS/SOL) who were free of diabetes at baseline (2008-2011) and attended clinic visit 2 (2014-2017). Incident diabetes was identified during the annual phone follow-up (through self-reported diabetes) or at visit 2 (through self-reported diabetes or antihyperglycemic medication usage or laboratory tests using the ADA criteria). Dietary intakes at baseline were assessed using two 24-hour dietary recalls, and meal timing was categorized into early morning (6:00-8:59 AM), late morning (9:00-11:59 AM), afternoon (12:00-5:59 PM), evening (6:00-11:59 PM), and night (0:00-5:59 AM). We calculated energy intake (EI) in kilocalories and energy-adjusted glycemic load (GL) in each meal timing. We estimated their associations with incident diabetes using the Cox proportional hazards models, accounted for the HCHS/SOL complex survey design and adjusted for age, sex, field center, Hispanic/Latino background, income, education, smoking, alcohol intake, sleep duration, physical activity, BMI, family history of diabetes, hypertension, dyslipidemia, total energy intake, and diet quality (Health Eating Index for energy intake in each meal timing as the exposure; total GL and fiber intake for GL in each meal timing as the exposure). Results: At baseline, the analytic population had a mean (SE) age of 39.0 (0.3) years, and 50.9% were female. Over a median (range) follow-up of 5.8 (0.8-9.6) years, 1,262 incident diabetes cases were documented. EI and GL exhibited strong positive correlations throughout the day. EI and GL in late morning were inversely associated with diabetes risk, whereas those in other meal timings were not associated with diabetes risk. In fully adjusted models, the hazard ratios (HRs) of diabetes were 0.94 (95% CI: 0.91 to 0.97) for a 100-kcal increase of EI and 0.93 (0.89 to 0.97) for a 10-unit increase of GL in late morning. The associations were stronger and showed a linear dose-response relationship in males and adults with overweight/obesity. Replacing energy intake or GL, from early morning, afternoon, or evening with late morning equivalents was associated with comparable reductions in diabetes risk. Conclusions: This study identifies late morning as a favorable meal timing for lowering diabetes risk in Hispanic/Latino adults, providing a novel perspective on type 2 diabetes prevention that warrants confirmation.
Read full abstract