Introduction: Time-restricted feeding has been associated with healthier weight, and as a weight-control intervention, can reduce caloric intake without calorie counting. Previous research showed that individuals with overweight who shortened their eating duration (time between first and last food intake of the day) experienced beneficial weight reduction. We explored this association in understudied US Hispanic/Latino individuals, as they experience a high prevalence of overweight and obesity. Hypothesis: We assessed the hypothesis that shorter eating duration is associated with healthier body mass index (BMI), percent body fat, or waist to hip ratio. Methods: We analyzed cross-sectional data from n=12079 Hispanic/Latino participants from the Hispanic Community Health Study/Study of Latinos (HCHS/SOL) who were not pregnant, not underweight, had reliable 24-hour dietary recall data, and had at least 3 accelerometer wear days. For each participant, we calculated eating duration, defined as the time between the first and last food intake of the day averaged over two 24-hour dietary recalls. We modeled eating duration in hours as the primary exposure in linear regression models with baseline measured BMI, percent body fat, or waist to hip ratio as outcomes of interest. We adjusted for age, sex, Hispanic background, field center, socioeconomic factors, sleep duration, glycemic status, diet quality, measured physical activity, and energy balance (ratio of energy intake to estimated energy needs). BMI was log transformed to help meet assumptions. Analyses accounted for complex survey design, missing accelerometer data, and accelerometer wear time. Results: Mean (SE) eating duration was 11.9 (0.04) hours. Those with longer eating duration tended to have higher energy intake (though less apparent for those with BMI at least 30), spend more time in moderate or vigorous physical activity, and were more likely to be employed full-time and perform shift work. After adjusting for energy balance and socioeconomic, medical and lifestyle factors, we found a significant positive association between eating duration and log-transformed BMI (regression coefficient 0.003 for eating duration in hours, SE 0.001, p=0.02). This significant association emerged only after adjusting for energy balance, demonstrating the importance of controlling for confounding between eating duration and intake. In sensitivity analysis, we excluded those with prevalent heart disease or diabetes, and obtained similar results. No statistically significant associations were noted between eating duration and the other obesity measures. Conclusions: Longer eating duration was associated with higher BMI in Hispanic/Latino persons after adjusting for covariates and energy balance from self-reported intake.
Read full abstract