Background Approximately one-third of children in England leave primary school overweight or obese. There is little evidence of effective obesity prevention programmes for children in this age group.

Objective To determine the effectiveness and cost-effectiveness of a school-based healthy lifestyles programme in preventing obesity in children aged 9–10 years.

Design A cluster randomised controlled trial with an economic and process evaluation.

Setting Thirty-two primary schools in south-west England.

Participants Children in Year 5 (aged 9–10 years) at recruitment and in Year 7 (aged 11–12 years) at 24 months’ post-baseline follow-up.

Intervention The Healthy Lifestyles Programme (HeLP) ran during the spring and summer terms of Year 5 into the autumn term of Year 6 and included four phases: (1) building a receptive environment, (2) a drama-based healthy lifestyles week, (3) one-to-one goal setting and (4) reinforcement activities.

Main outcome measures The primary outcome measure was body mass index (BMI) standard deviation score (SDS) at 24 months post baseline measures (12 months post intervention). The secondary outcomes comprised waist circumference SDS, percentage body fat SDS, proportion of children overweight and obese at 18 and 24 months, accelerometer-assessed physical activity and food intake at 18 months, and cost-effectiveness.

Results We recruited 32 schools and 1324 children. We had a rate of 94% follow-up for the primary outcome. No difference in BMI SDS was found at 24 months [mean difference –0.02, 95% confidence interval (CI) –0.09 to 0.05] or at 18 months (mean difference –0.02, 95% CI –0.08 to 0.05) between children in the intervention schools and children in the control schools. No difference was found between the intervention and control groups in waist circumference SDS, percentage body fat SDS or physical activity levels. Self-reported dietary behaviours showed that, at 18 months, children in the intervention schools consumed fewer energy-dense snacks and had fewer negative food markers than children in the control schools. The intervention effect on negative food markers was fully mediated by ‘knowledge’ and three composite variables: ‘confidence and motivation’, ‘family approval/behaviours and child attitudes’ and ‘behaviours and strategies’. The intervention effect on energy-dense snacks was partially mediated by ‘knowledge’ and the same composite variables apart from ‘behaviours and strategies’. The cost of implementing the intervention was approximately £210 per child. The intervention was not cost-effective compared with control. The programme was delivered with high fidelity, and it engaged children, schools and families across the socioeconomic spectrum.

Limitations The rate of response to the parent questionnaire in the process evaluation was low. Although the schools in the HeLP study included a range of levels of socioeconomic deprivation, class sizes and rural and urban settings, the number of children for whom English was an additional language was considerably lower than the national average.

Conclusions HeLP is not effective or cost-effective in preventing overweight or obesity in children aged 9–10 years.

Future work Our very high levels of follow-up and fidelity of intervention delivery lead us to conclude that it is unlikely that school-based programmes targeting a single age group can ever be sufficiently intense to affect weight status. New approaches are needed that affect the school, the family and the wider environment to prevent childhood obesity.