Programme characteristics associated with weight change following behavioural weight loss programmes: a systematic review and meta-analysis

Objective: To determine if characteristics of behavioural weight loss programmes (BWMPs) influence weight change trajectories following programme end. Design: Systematic review, meta-analysis and meta-regression. Data sources: Trial registries, 11 electronic databases, forward citation searching (from database inception; latest search Dec-19). Eligibility criteria for selecting studies: Randomised trials of BWMPs in adults with overweight/obesity reporting outcomes at >=12 months, including at and after programme end, were identified by two independent reviewers. Two reviewers independently screened with discrepancies resolved through discussion. 5% of the studies identified in the searches met the inclusion criteria. Data extraction and synthesis: One reviewer extracted data; these were checked by a second. Risk of bias was assessed using Cochrane’s Risk of Bias tool (v1). Main outcome measure: We calculated weight trajectories (kg/month; translated to kg/year for interpretability) after programme end in intervention versus control groups using a mixed model with a random intercept. We tested for associations between trajectories and pre-specified variables. Results: We analysed data from 249 trials (n=59,081) with mean length of follow-up 2 years, longest 30 years. Fifty-six percent of studies were at unclear risk of bias, 21% at low risk, and 23% at high risk. Weight regain was faster in intervention versus no intervention control groups at 0.12 to 0.32 kg/year, but between-group difference was maintained for at least five years. Each kilogram of weight lost at programme end was associated with 0.13-0.19 kg/year faster weight regain. Financial incentives for weight loss were associated with 1-1.5kg/year faster regain. Compared to those with no meal replacements, interventions involving partial meal replacement were associated with faster regain, but not after adjustment for weight loss during the programme. Participant access to the BWMP outside the study was associated with slower regain. Programmes that diminished in intensity were also associated with slower weight regain in the multivariable analysis, although the point estimate suggested that the association was small. Other characteristics did not explain the heterogeneity in weight regain. Conclusion: Faster weight regain after weight loss was associated with greater initial weight loss, but greater initial weight loss was still associated with added benefits for at least five years after programme end, after which there was a paucity of data. Continued availability of the programme predicted slower weight gain and provision of financial incentives predicted faster regain; no other clear associations were found. Though to the best of our knowledge this is the most comprehensive review in this field, limitations persist in regards to reporting, risk of bias, and a paucity of data at greater than 5 years from programme end.