Does mental health and wellbeing influence the outcomes and experiences of men during a text message delivered weight management programme with or without financial incentive (Game of Stones)?

Problem

Behavioural weight management interventions (BWMI) can improve physical and mental health. However, an individual's mental state may affect engagement and some interventions, including those with financial incentives, may negatively affect mental health. Pre-specified aims were to understand whether mental health conditions moderate weight change outcomes and integrate qualitative data to better understand findings.

Approach

Participants (n=585) were randomly allocated to daily automated texts (SMS); texts with endowment financial incentives (SMS+I); or waiting list control (WLC). Secondary exploratory mental health and wellbeing outcomes were: Patient Health Questionnaire (PHQ-4), Warwick and Edinburgh Mental Wellbeing Scale (WEMWBS), EQ-5D-5L Anxiety and Depression (AD) sub-scale and Weight Self-Stigma Questionnaire (WSSQ). Sub-groups included: the secondary outcome measures (high and low scores); doctor diagnosed mental health condition (MHC) and possible latent mental health condition (LMHC- defined as men with no self-reported MHC but either a WEMWBS low score or high score on other secondary outcomes). Moderating effects were explored using linear regression models that included treatment-by-subgroup interactions, using 99.5% confidence intervals (CI). Interviews were conducted at 12 months with 54 men (SMS+I (n=30); SMS (n=24)), including men with: self-reported MHC (n=16); LMHC (based on PHQ-4 >3) (n=13), no reported MH difficulties (n=12). Data were analysed using Framework method.

Findings

At baseline, 146 (25%) men reported a MHC; 142 (24%) a possible LMHC. Mean percentage weight change from baseline (SD) was -4.8% (6.1) for SMS+I, -2.7% (6.3) for SMS, and -1.3% (5.5) for WLC. Results showed no significant differences in secondary outcomes for men in the SMS+I or SMS compared to WLC. For SMS+I compared to WLC, the mean difference in percentage weight at 12 months from baseline for men with MHC was 1.0 (99.5% CI 0.77 (-3.93, 5.47); p=0.64). The mean difference in percentage weight for SMS compared to WLC was -3.08 (99.5% CI -3.22 (-8.06, 1.63); p=0.06).Men’s views of the texts were similar across trial groups and mental health status. There was limited information to suggest that texts, with or without incentives, had an adverse effect on mental health. Some men in SMS+I with MHC or LMHC did suggest having poorer mental health e.g. depression or seasonal affective disorder, was a barrier to behaviour change. Men from both trial groups, with MHC or LMHC discussed being unmotivated by money or the incentive was not motivating enough to drive change, compared to those with no MHC.

Consequences

Texts with or without incentives did not adversely affect mental health and wellbeing outcomes for men living with obesity. Neither did mental health or wellbeing status at baseline moderate weight loss. Qualitative data provides insights into varying experiences among men. There was some limited evidence that mental health status might affect an individual’s ability or willingness to make weight loss changes.