Associations Between Late Evening Food Consumption and Exceeding Estimated Energy Needs in U.S. Adults: Results From What We Eat in America, NHANES 2013–2016
Abstract Objectives Eating later in the day has been associated with higher daily energy intake. However, it is unclear how this behavior relates to energy needs. The goal of this study was to determine whether late evening (LE) food and beverage consumption, overall and by previously identified prevalent LE food patterns, is associated with exceeding estimated daily energy needs. Methods One day of dietary intake collected from 9861 adults 20 + years of age (4829 men and 5032 women) in What We Eat in America, NHANES 2013–2016 was analyzed. LE consumption was defined as any food/beverage intake excluding plain water that began between 8:00 pm and 11:59 pm. Individuals’ energy needs were estimated based upon sex, age, and self-reported physical activity. T-tests identified differences in the percentages of adults exceeding their estimated energy needs by ≥25% between LE non-reporters and LE reporters, overall and by prevalent LE pattern, and logistic regression was applied to determine the relative risk (RR) of this outcome. Results On the intake day, 66.8% of men and 62.2% of women were LE reporters. The mean LE energy contribution among LE reporters was 579 kcal in men and 407 kcal in women, but ranged considerably among the prevalent patterns (268–924 kcal for men; 159–657 kcal for women; ANOVA for both P < 0.01). Among both men and women, LE consumption overall was associated with a higher likelihood of exceeding energy needs by ≥25% (men: 20.2% vs. 14.0%; RR = 1.44; 99% CI: 1.04, 2.00; women: 20.1% vs. 12.8%; RR = 1.57; 99% CI: 1.16, 2.11). However, when individual LE patterns were assessed, differences were observed among women only. Relative to non-reporter women, those in the following prevalent LE patterns were more likely to exceed estimated energy needs by ≥25%: the candy/nuts/seeds pattern (+13.6%; RR = 2.06; 99% CI: 1.20, 3.54), the savory snacks/alcohol pattern (+9.7%; RR = 1.76; 99% CI: 1.14, 2.71), and the milk/dairy drinks pattern (+16.8%; RR = 2.31; 99% CI: 1.43, 3.73). Conclusions Although LE intake overall was associated with a greater likelihood of exceeding estimated energy needs by ≥25% among both men and women, pattern-specific associations were only observed among women. Addressing both the frequency and content of LE consumption could attenuate its impact on potential excess energy intake in this population. Funding Sources ARS, USDA.