Abstract
OBJECTIVES: Interventions for obesity that require individual behaviour change, such as behavioural weight management interventions, may contribute to health inequalities. We explored if there was evidence of inequalities in the attendance at and effectiveness of behavioural weight management interventions in adults. DESIGN: Two-stage individual participant data meta-analysis. DATA SOURCES: Eligible studies were extracted from a previous systematic review and an updated search was completed on PubMed. ELIGIBILITY CRITERIA: UK-based randomised controlled trials of behavioural weight management interventions suitable for use in primary care, published until 31 December 2021. DATA EXTRACTION AND SYNTHESIS: Multivariable regression analyses were conducted with weight at 12-month follow-up as the primary outcome and included an interaction between inequality characteristic and trial arm (control or intervention). Each model was adjusted for baseline weight, age and gender. Estimated interactions were combined across trials using a random-effects meta-analysis. Intervention attendance was defined as number of in-person sessions attended. Risk of bias was assessed using Cochrane's RoB 2 tool. RESULTS: Data from 13/16 eligible weight loss trials were analysed (complete case data n=5531 participants). The effect of the intervention on weight at 12 months was greater in male participants (-2.58 kg (95% CI -3.52 to 1.64)) than female participants (-1.71 kg (95% CI -2.79 to -0.63); p value for interaction=0.02, tau(2)=0) and greater for participants of white ethnicity (-2.74 kg (95% CI -4.30 to -1.19)), than those from an ethnic minority background (0.03 kg (95% CI -1.29 to 1.35); p interaction=0.04, tau(2)=0). Age, education, occupation, place of residence and household income did not significantly moderate effectiveness. We did not find evidence of inequalities in intervention attendance by ethnicity, occupation, gender/sex, area-level socioeconomic deprivation or age. CONCLUSIONS: Behavioural weight management interventions had smaller effects in people from ethnic minority backgrounds and larger effects in men. There was no evidence of other differences in intervention effectiveness or adherence. This is the first synthesis study to access individual participant data and quantitatively assess inequalities in these interventions. Future research should further explore reasons for differences in outcomes and consider how to prevent behavioural weight management interventions from potentially exacerbating health inequalities.