BACKGROUND: Finding effective referral policies for weight management services would have important public health implications. AIM: Here we compare percentage weight change by referral methods, BMI categories and participants who have had attended weight loss programmes multiple times. DESIGN AND SETTINGS: A prospective cohort study of 15,621 participants referred to 12-week behavioural weight loss programmes funded by the public health service in Birmingham, UK. METHODS: Comparisons were made between GP versus self-referrals, BMI ≥40kg/m2-<40kg/m2 and multiple referrals compared to only one referral. Linear mixed modelling was used to assess percentage weight change after adjusting for covariates. RESULTS: Participant's mean age was 48.5 years, 78.7% were of white ethnicity, 90.3% female and mean baseline BMI was 36.3kg/m2. There were no significant differences in percentage weight loss, between participants that self-referred and those that were referred by their general practitioner (GP) and no significant differences between baseline BMI categories. Referral to a weight loss programme more than once was associated with less weight loss at subsequent attendances (0.92%, 95% CI 0.70-1.14, p<0.001). CONCLUSION: Allowing self-referral to a weight loss programme widens access without compromising amount of weight lost. These programmes are beneficial for all categories of obesity, including those with a BMI ≥40kg/m2. Attending weight management programmes more than once results in less weight loss and that swapping to a different program may be advisable.

Original publication




Journal article


Obes Res Clin Pract

Publication Date





709 - 717


Behaviour, Obesity, Primary healthcare, Treatment