Abstract MP001: The Number of Low and High Energy Dense Foods Consumed Affects Body Mass Index Reduction in a Randomized Controlled Weight Loss Trial

Abstract

Background: Decreasing dietary energy density is recommended to promote weight loss. However, it is unclear whether decreasing dietary energy density is best accomplished by increasing the number of low energy dense (LED) foods (≤1 kcal/g) or by decreasing the number of high energy dense (HED) foods (≥3 kcal/g) in the diet, and whether these actions have different effects on weight loss. We explored how the number of LED and HED foods consumed affected dietary energy density and change in BMI during the 6-month weight-loss phase of an 18-month weight loss trial. Methods: A secondary analysis of 202 adults 21-65y with a BMI between 27 and 45 kg/m 2 participating in a weight loss trial was conducted. Participants were randomized to 1 of 2 conditions – a standardized lifestyle intervention or a standardized lifestyle intervention plus a goal to limit consumption of non-nutrient-dense, energy-dense foods to 2 foods. Dietary intake, assessed via 3, 24-hour recalls, and BMI were measured at baseline and 6 months. Generalized linear models were used to examine the association between residualized changes in LED foods and HED foods and dietary energy density and change in BMI between baseline and 6 months. All analyses were adjusted for age, sex, race, education, treatment group, change in energy intake, and number of LED or HED foods. Results: At baseline, participants had a daily average energy density of 1.25 kcal/g, which decreased to 0.82 kcal/g at 6-months (p<0.0001). Similarly, daily number of LED foods consumed increased from 4.46 to 5.64 foods/day while daily HED foods consumed decreased from 4.02 to 2.98 foods/day (p<0.0001). In fully-adjusted models, a 1-unit increase in LED foods was associated with a 0.07 decrease in the overall energy density of the diet while a 1-unit decrease in HED foods was associated with a 0.08 decrease in energy density (p<0.05); no significant changes in BMI were detected. However, when we examined changes in BMI among those who decreased (>1 unit), maintained (< 1-unit), or increased (>1 unit) the number of LED or HED foods, significantly greater changes in BMI were observed among those with stable (-3.3 ± 0.4 kg/m 2 ) or increased LED food intake (-3.5 ± 0.4 kg/m 2 ) compared to those who decreased LED foods (-2.4 kg ± 0.5 kg/m 2 , p<0.05). No differences in BMI were observed among those who decreased, maintained, or increased the number of HED foods, though similar decreases in energy density were observed among those who decreased HED foods or increased LED foods (-0.6 kcal/g). Conclusions: Decreasing HED foods and increasing LED foods both reduced the overall ED of the diet, but increasing LED foods was associated with significant reductions in BMI.