Game of Stones NIHR Synopsis

Abstract

Background Men with obesity infrequently engage with weight management services. Objectives To determine: i) percentage weight loss at 12 and 24 months for text messages with or without financial incentives compared to control; ii) secondary outcomes; iii) costeffectiveness; iv) moderators of effectiveness; iv) participant and stakeholder perspectives. Design and methods Assessor blinded randomised controlled trial. UK NHS perspective cost-effectiveness over 24 months and modelled lifetime horizon. Mixed methods process evaluation. Setting and participants 585 men with body mass index ≥30kg/m2 enrolled (July 2021-May 2022) in Belfast, Bristol and Glasgow; final follow-up June 2024. Interventions Random allocation to 12 months of behavioural text messages plus financial incentives (N=196), same texts alone (N=194), or 12-month waiting list control group offered 3 months of texts between 12-15 months (N=195). A £400 financial incentive was lost if weight losstargets were not met. Main outcome measures Weight change as a percentage of baseline weight at 12- and 24-months comparing control with i) texts with financial incentives and ii) texts alone. Results Of 585 men (mean age 51 years; mean weight: 119 kg), 227 (39%) lived in lower socioeconomic areas, 146 (25%) reported a mental health condition, and 253 (40%) had multiple long-term conditions. Follow-up was completed by 426 (73%) at 12 months and 377 (64%) at 24 months. At 12 months, mean percentage weight changes (SD) were −4.8% (6.1) [−5.7 kg], −2.7% (6.3) [−3.0 kg], and −1.3% (5.5) [−1.5 kg] for the incentives, text-only, and control groups, respectively. Compared to control, weight loss was significantly greater with incentives [−3.2% (97.5% CI, −4.6 to −1.9; p<0.001)] but not texts alone [−1.4%; CI −2.9 to 0.0; p=0.053]. At 24 months, changes were −3.9% [−4.6 kg], −2.6% [−3.1 kg], and −2.2% [−2.6 kg], no significant between-group differences. Intervention costs were £243 for texts with incentives, £110 for texts alone. There were no significant differences between 24-month costs or QALYs. Long-term modelling found texts with incentives vs. control were: QALY difference (95% CI): 0·02 (0·007, 0·029); cost difference: £176 (£43; £311); Incremental cost-effectiveness ratio (ICER): £9,748 (£7,705, £11,791). For texts alone vs. control: QALY difference: 0·03 (0·015, 0·037); cost difference: £16·5 (-£117; £152); ICER: £628 (£-5,914, £5,384) There were no moderator effects for socio-economic, health or wellbeing status for either comparison versus control. The texts with incentives group had higher engagement in weight goal setting, food changes, self-weighing, confidence, satisfaction and quality of life compared to the control. Limitations Generalisability to women, diverse ethnic groups, and people with low literacy is uncertain. Not generalisable to people with no mobile phone access. Retention was lowest in the text messages alone group. Conclusions Texts with financial incentives have a modest but important effect to12 months with clinically relevant weight loss maintenance to 24 months, are cost-effective and equally effective regardless of socioeconomic or health characteristics. Future work Implementation, adapt for women, other cultures and longer-term follow-up. Study registration: ISRCTN91974895 https://doi.org/10.1186/ISRCTN91974895 Funding: This project was funded by the National Institute for Health and Care Research Public Health Research programme (PHR 129703)