Background
Behavioural weight management programmes can prevent type 2 diabetes and reduce risk factors for complications and medication use in those with type 2 diabetes. However, few are sufficiently scalable, and weight is typically regained. We evaluated whether scalable behavioural weight management programmes can achieve and maintain weight loss and reduce type 2 diabetes incidence and complications in two populations: people at high risk, and those recently diagnosed.
Objectives
To quantify the impact of scalable behavioural weight management programmes on type 2 diabetes incidence, related complications and health resource use (work package 1).
To develop and test the feasibility and acceptability of a scalable weight loss maintenance programme (work package 2).
Design and methods
Work package 1
Work package 1.1: to evaluate the effectiveness and cost-effectiveness of a scalable behavioural weight management programme for individuals at risk of type 2 diabetes, we conducted a 5-year follow-up of the Weight loss Referrals for Adults in Primary care randomised controlled trial (N = 1267).
Work package 1.2: to evaluate the effectiveness and cost-effectiveness of a scalable weight loss programme for individuals recently diagnosed with type 2 diabetes, we conducted a new randomised controlled trial (N = 577).
For both trials, we evaluated effectiveness (change in weight and glycated haemoglobin) and within-trial cost-effectiveness from a National Health Service perspective. Semistructured interviews explored the experiences of participants and practitioners.
Work package 1.3: using the National Institute for Health and Care Research School for Public Health Research Diabetes Prevention microsimulation model, we evaluated the lifetime cost-effectiveness of the interventions evaluated in work package 1.1 and work package 1.2. To reduce uncertainty and increase confidence in our findings, we compared findings to a separate microsimulation model (National Health Service Health Checks model).
Work package 2
Work package 2.1: we used an evidence-, theory- and person-based approach to develop a programme for weight loss maintenance (Supporting Weight Management).
Work package 2.2: we evaluated the feasibility and acceptability of Supporting Weight Management in a 6-month randomised controlled trial (N = 61) with embedded qualitative study of participants (N = 28) and coaches (N = 2).
Work package 2.3: we used the School for Public Health Research Diabetes model and expected value of sample information methods to estimate the value of alternative trial designs in reducing uncertainty in the cost-effectiveness estimates.
Setting and participants
Adults living with overweight and obesity who are at risk of type 2 diabetes or who have a recent diagnosis. Primary care and community settings in England.
Interventions
Scalable behavioural interventions for weight management.
Work package 1.1: brief advice, 12-week weight watchers or 52-week weight watchers.
Work package 1.2: six-month weight watchers with tailored diabetes education and dietitian support via telephone [diabetes education and behavioural weight management programme (intervention group in the GLoW trial)] or diabetes education only.
Work package 2: Supporting Weight Management (a digital, acceptance-based, guided self-help programme for weight loss maintenance) or standard advice (leaflet on weight loss maintenance).
Main outcome measures
Primary outcomes were weight or glycated haemoglobin. Secondary outcomes included healthcare resource use, quality of life and psychosocial measures.
Results
Work package 1.1: mean 5-year weight change was: −0.46 (standard deviation 8.31) kg in the brief intervention group, −1.95 (9.55) kg in the 12-week programme group and − 2.67 (9.81) kg in the 52-week programme. The adjusted difference in weight change between groups was not statistically significant. During the trial, the 12-week programme incurred the lowest cost and produced the highest quality-adjusted life-years. Qualitative interviews identified key differences in the cognitive and behavioural strategies used by ‘maintainers’ and ‘regainers’.
Work package 1.2: we found no evidence for an intervention effect on change in glycated haemoglobin but diabetes education and behavioural weight management programme participants lost more weight than diabetes education at 6 [−1.77 (−2.86 to −0.67) kg] and 12 months [−1.38 (−2.56 to −0.19) kg]. During the trial, diabetes education and behavioural weight management was cost-effective compared with diabetes education. Qualitative interviews identified how the intervention was effective and key barriers to effectiveness.
Work package 1.3: in the lifetime cost-effective analysis, 52-week weight watcher was cost-effective compared with brief intervention, 12-week weight watcher and a simulated do-nothing scenario, producing more health benefit and cost savings. A lifetime cost-effectiveness of diabetes education and behavioural weight management compared with diabetes education was cost-effective with an incremental cost-effectiveness ratio of £2290 per quality-adjusted life-year gained.
Work package 2.1: we codeveloped the Supporting Weight Management intervention for weight loss maintenance. It involves 14 psychoeducational sessions delivered via a web-based platform over 18 weeks, and telephone support from trained, non-specialist coaches.
Work package 2.2: pre-trial average weight loss was 8 kg; during the trial Supporting Weight Management participants lost a further 2.2 (± 6.4 standard deviation) kg and control participants gained 2.2 (± 6.6) kg. Supporting Weight Management participants described that Supporting Weight Management reinforced existing knowledge and supported skill development for weight loss maintenance.
Work package 2.3: the simulated justifiable cost of a weight maintenance intervention was £104.64 (body mass index ≥ 28) and £88.14 (type 2 diabetes), based on a cost-effectiveness threshold of £20,000. A large prospective randomised controlled trial to evaluate Supporting Weight Management was found to be cost-effective in expected value of sample information analysis.
Limitations
The coronavirus disease discovered in 2019 pandemic resulted in changes to study and intervention procedures, delayed recruitment, and higher-than-anticipated attrition. The ethnic profile of participants limits the generalisability of findings to minority ethnic groups.
Conclusions
This research supports scalable behavioural weight management programmes as cost-effective for preventing and treating type 2 diabetes. Findings highlight the need for targeted support for maintaining weight loss. Acceptance-based guided self-help is an acceptable and feasible approach to support weight maintenance.
Future work
We have identified a range of research needs going forward, including developing and testing interventions to prevent weight regain and improve long-term health economic outcomes, improving the reach into and engagement with under-represented groups, as well as enhancing and validating long-term health economic models using real-world weight trajectories to better predict lifetime cost-effectiveness and equity impacts.
Study registration
The study WRAP is registered as ISRCTN64986150; GLoW is registered as ISRCTN18399564 and the SWiM feasibility study is registered as ISRCTN12685964.
Funding
This award was funded by the National Institute for Health and Care Research (NIHR) Programme Grants for Applied Research Programme (NIHR award ref: RP-PG-0216-20010) and is published in full in Programme Grants for Applied Research; Vol. 14, No. 22. See the NIHR Funding and Awards website for further award information.
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)