An evaluation of economic evidence included in published randomised controlled trials of interventions to prevent obesity in children.

Abstract

Background: Childhood obesity is a public health policy priority. Policy-makers need an understanding of the costs of interventions to prevent childhood obesity alongside their effectiveness when tested in randomised controlled trials. It is not known what cost data have been included in published randomised controlled trials of childhood obesity prevention interventions. This study aimed to summarise these costs and identify associated economic evaluations published separately.

Methods: This review summarises data extracted from studies included in two Cochrane systematic reviews of interventions to prevent obesity in children aged 5-11 and 12-18 years old. Eligible interventions could be delivered in any setting and studies were randomised controlled trials reporting (standardised or unstandardised) body mass index outcome data at a minimum follow-up of 12 weeks post baseline. Databases searched included Cochrane Central Register of Controlled Trials, MEDLINE, EMBASE and PsycINFO. Searches were limited to between 1990 and 2023. Any cost data reported in the publications were extracted, in addition to citations of linked economic analyses. Data were tabulated and summarised using a narrative approach.

Results: Of the 244 randomised controlled trials included in the Cochrane systematic reviews, 85 (35%) included costs related to the trial, intervention, school, health sector, out-of-pocket or productivity costs, or were linked to a full economic evaluation published separately. Of the studies reporting costs, five (6%) studies reported results of a full economic evaluation within the trial paper. The majority of research costs were incentives for participation, identified in 41 (48%) studies. Where these are not part of the intervention itself, these would not usually be included in economic evaluations. Thirty (35%) studies included intervention costs, although reporting was inconsistent. The payer of the intervention was unclear in most publications, making the attribution of costs to sectors difficult. Only one study reported healthcare resource use data, but some estimated the cost of obesity-related chronic conditions in linked decision models.

Limitations and future work: The findings of this review are limited to randomised controlled trials only and interventions for children aged 5 years and over. Other study designs may provide important cost data. Future work could explore the cost data needs of public health policy-makers and the impact of including incentives on the effectiveness and cost-effectiveness in public health randomised controlled trials.

Conclusions: This is a novel synthesis of costs reported in randomised controlled trials of interventions to prevent childhood obesity. Overall, the reporting of any type of costs was low (35% of studies). The most common type of reported costs were intervention costs (e.g. staff costs, materials and training) and participant incentives for data collection, although five studies included full economic evaluations. This study complements published systematic reviews of economic evaluations in this area by offering new insights into the prevalence, type and quality of cost information included in the effectiveness literature.

Funding: This article presents independent research funded by the National Institute for Health and Care Research (NIHR) Public Health Research programme as award number NIHR131572.