Does goal setting influence weight loss? A prospective analysis of a community weight loss programme (Preprint)

BACKGROUND Goal setting aids health-related behaviour change, but the influence of different types of goals on weight loss remains unclear. OBJECTIVE To investigate the association between three measures of goals setting and weight change during a 12-week digital behavioural weight loss programme delivered across the UK. METHODS Prospective longitudinal analysis of 36,794 UK adults with a body mass index of ≥25 kg/m2 that enrolled in the programme. Measures of goal setting were self-reported at enrolment including percentage weight loss goal (<5%, 5-10%, or >10%); overall goal preference (low, medium, or high); and weight loss motivation (appearance, health, fitness, or self-efficacy). Weight was measured at 4, 12, and 24 weeks. To measure sustained weight change, the primary outcome was weight at 24 weeks, assessed using mixed-model repeated-measures analyses to explore the effect of goal setting adjusted for known confounders. We explored drop-out rates by goal and whether engagement mediated the association between goals and weight loss. RESULTS Of the 36,794 participants included in the cohort, 13.1% (n=4,818) reported weight at 24 weeks. Most participants set goals of 5-10% weight loss (64.3%), but setting goals for >10% weight loss was associated with larger weight loss (mean difference between 5-10% vs >10%: 5.21kg; 95% CI: 5.01, 5.41; P<.001). There was no difference between goals of 5-10% and <5% (mean difference: 0.59kg; 95% CI: 0.00, 1.18; P=.05). Appearance was the most prevalent motivational factor (40.1%), but health or fitness were associated with greater weight losses than appearance (mean difference health vs appearance: 1.40kg; 95% CI: 1.15, 1.65; P<.001 and mean difference fitness vs appearance: 0.38kg; 95% CI: 0.05, 0.70; P=.03). Goal preference had no clear association with weight loss. Engagement with programme components was a significant independent predictor of weight loss but not a mediator of the effect of goal setting. At 24 weeks, those who set goals of >10% were less likely to drop-out compared to those setting 5-10% goals (OR: 0.40; 95% CI: 0.38, 0.42; P<.001); those who liked to set overall high goals were more likely to drop-out compared to those with medium goals (OR: 1.20; 95% CI: 1.11, 1.29; P<.001); and those motivated by fitness or health were less likely to drop-out compared with those motivated by appearance (OR: 0.92; 95% CI: 0.85, 0.995; P=.038 , and OR: 0.84; 95% CI: 0.78, 0.89; P<.001, respectively). CONCLUSIONS Setting higher and specific weight loss goals and being motivated by health or fitness were associated with greater weight loss and lower likelihood of drop-out. Randomised trials of setting these types of goals are needed to confirm causality.