Estimating the impact of changes in weight and BMI on EQ-5D-3L: a longitudinal analysis of a behavioural group-based weight loss intervention

Abstract

Abstract Purpose To estimate the association between changes in BMI and changes in Health-Related Quality of Life (EQ-5D-3L). Methods The WRAP trial was a multicentre, randomised controlled trial with parallel design and recruited 1267 adults (BMI ≥ 28 kg/m2). Participants were allocated to Brief Intervention, a Commercial weight management Programme (WW, formerly Weight Watchers) for 12 weeks, or the same Programme for 52 weeks. Participants were assessed at 0, 3, 12, 24, and 60 months. We analysed the relationship between BMI and EQ-5D-3L, adjusting for age and comorbidities, using a fixed effects model. Test for attrition, model specification and missing data were conducted. Secondary analyses investigated a non-symmetric gradient for weight loss vs. regain. Results A unit increase in BMI was associated with a − 0.011 (95% CI − 0.01546, − 0.00877) change in EQ-5D-3L. A unit change in BMI between periods of observation was associated with − 0.016 017 (95% CI − 0.0077009, − 0.025086) change in EQ-5D-3L. The negative association was reduced during weight loss, as opposed to weight gain, but the difference was not statistically significant. Conclusions We have identified a strong and statistically significant negative relationship between BMI changes and HRQoL. These estimates could be used in economic evaluations of weight loss interventions to inform policymaking. Clinical trial registration This trial was registered with Current Controlled Trials, number ISRCTN82857232.