Differences in weight-loss outcomes among race-gender subgroups by behavioural intervention delivery mode: An analysis of the POWER trial

Abstract

Prior in-person behavioural intervention studies have documented differential weight loss between men and women and by race, with Black women receiving the least benefit. Remotely delivered interventions are now commonplace, but few studies have compared outcomes by race-gender groups and delivery modality. We conducted a secondary analysis of POWER, a randomized trial (NCT00783315) designed to determine the effectiveness of 2 active, lifestyle-based, weight loss interventions (remote vs. in-person) compared to a control group. Participants with obesity and at least one cardiovascular disease risk factor (N = 415) were recruited in the Baltimore, MD area. Data from 233 white and 170 Black individuals were used for this analysis. Following an intention-to-treat approach, we compared the mean percent weight loss at 24 months by race-gender subgroups using repeated-measures, mixed-effects models. Everyone lost weight in the active interventions however, weight loss differed by race and gender. white and Black men had similar results for both interventions (white: in-person (−7.6%) remote (−7.4%); Black: in-person (−4.7%) remote (−4.4%)). In contrast, white women lost more weight with the in-person intervention (in-person (−7.2%) compared to the remote (−4.4%)), whereas Black women lost less weight in the in-person group compared to the remote intervention at 24 months (−2.0% vs. −3.0%, respectively; p for interaction <.001). We found differences between the effectiveness of the 2 weight loss interventions—in-person or remote—in white and Black women at 24 months. Future studies should consider intervention modality when designing weight loss interventions for women.