BackgroundConsuming different food groups and nutrients can have differential effects on body weight, body composition, and insulin sensitivity. ObjectiveThe aim was to identify how food group, nutrient intake, and diet quality change relative to usual-diet controls after 16 weeks on a low-fat vegan diet and what associations those changes have with changes in body weight, body composition, and measures of metabolic health. DesignSecondary analysis of a randomized clinical trial conducted between October 2016 and December 2018 in four replications. Participants/settingParticipants included in this analysis were 219 healthy, community-based adults in the Washington, DC, area, with a body mass index (BMI) between 28 and 40, who were randomly assigned to either follow a low-fat vegan diet or make no diet changes. InterventionA low-fat, vegan diet deriving approximately 10% of energy from fat, with weekly classes including dietary instruction, group discussion, and education on the health effects of plant-based nutrition. Control group participants continued their usual diets. Main outcome measuresChanges in food group intake, macronutrient and micronutrient intake, and dietary quality as measured by Alternate Healthy Eating Index-2010 (AHEI-2010), analyzed from 3-day diet records, and associations with changes in body weight, body composition, and insulin sensitivity were assessed. Statistical analyses performedA repeated-measure analysis of variance model that included the factors group, subject, and time was used to test the between-group differences throughout the 16-week study. Interaction between group and time was calculated for each variable. Within each diet group, paired comparison t tests were calculated to identify significant changes from baseline to 16 weeks. Spearman correlations were calculated for the relationship between changes in food group intake, nutrient intake, AHEI-2010 score, and changes in body weight, body composition, and insulin sensitivity. The relative contribution of food groups and nutrients to weight loss was evaluated using linear regression. ResultsFruit, vegetable, legume, meat alternative, and whole grain intake significantly increased in the vegan group. Intake of meat, fish, and poultry; dairy products; eggs; nuts and seeds; and added fats decreased. Decreased weight was most associated with increased intake of legumes (r = −0.38; P < 0.0001) and decreased intake of total meat, fish, and poultry (r = +0.43; P < 0.0001). Those consuming a low-fat vegan diet also increased their intake of carbohydrates, fiber, and several micronutrients and decreased fat intake. Reduced fat intake was associated with reduced body weight (r = +0.15; P = 0.02) and, after adjustment for changes in BMI and energy intake, with reduced fat mass (r = +0.14; P = 0.04). The intervention group’s AHEI-2010 increased by 6.0 points on average, in contrast to no significant change in the control group (treatment effect, +7.2 [95% CI +3.7 to +10.7]; P < 0.001). Increase in AHEI-2010 correlated with reduction in body weight (r = 0.14; P = 0.04), fat mass (r = −0.14; P = 0.03), and insulin resistance as measured by the Homeostasis Model Assessment (HOMA-IR; r = −0.17; P = 0.02), after adjustment for changes in energy intake. ConclusionsWhen compared with participants’ usual diets, intake of plant foods increased, and consumption of animal foods, nuts and seeds, and added fats decreased on a low-fat vegan diet. Increased legume intake was the best single food group predictor of weight loss. Diet quality as measured by AHEI-2010 improved on the low-fat vegan diet, which was associated with improvements in weight and metabolic outcomes. These data suggest that increasing low-fat plant foods and minimizing high-fat and animal foods is associated with decreased body weight and fat loss, and that a low-fat vegan diet can improve measures of diet quality and metabolic health.