Maximizing the benefits of investments in obesity research requires effective interventions to be adopted and disseminated broadly across populations (scaled-up).However, interventions often need considerable adaptation to enable implementation at scale, a process that can reduce the effects of interventions. A systematic review was undertaken for trials that sought to deliver an obesity intervention to populations on a larger scale than a preceding randomized controlled trial (RCT) that established its efficacy. Ten scaled-up obesity interventions (six prevention and four treatment) were included. All trials made adaptations to interventions as part of the scale-up process, with mode of delivery adaptations being most common. A metaanalysis of body mass index (BMI)/BMI z score (zBMI) from three prevention RCTs found no significant benefit of scaled-up interventions relative to control (standardized mean difference [SMD] = 0.03; 95% CI, −0.09 to 0.15, P = 0.639 − I 2 = 0.0%). All four treatment interventions reported significant improvement on all measures of weight status. Pooled BMI/zBMI data from prevention trials found significantly lower effects among scaled-up intervention trials than those reported in prescale-up efficacy trials (SMD = −0.11; 95% CI, −0.20 to −0.02, P = 0.018 − I 2 = 0.0%). Across measures of weight status, physical activity/sedentary behaviour, and nutrition, the effects reported in scaled-up interventions were typically 75% or less of the effects reported in pre-scale-up efficacy trials. The findings underscore the challenge of scaling-up obesity interventions.

Background Sustainment has been defined as the sustained use or delivery of an intervention in practice following cessation of external implementation support. This review aimed to identify and synthesise factors (barriers and facilitators) that influence the sustainment of interventions (policies, practices, or programmes) in schools and childcare services that address the leading risk factors of chronic disease. Methods Seven electronic databases and relevant reference lists were searched for articles, of any design, published in English, from inception to March 2020. Articles were included if they qualitatively and/or quantitatively reported on school or childcare stakeholders’ (including teachers, principals, administrators, or managers) perceived barriers or facilitators to the sustainment of interventions addressing poor diet/nutrition, physical inactivity, obesity, tobacco smoking, or harmful alcohol use. Two independent reviewers screened texts, and extracted and coded data guided by the Integrated Sustainability Framework, an existing multi-level sustainability-specific framework that assesses factors of sustainment. Results Of the 13,158 articles identified, 31 articles met the inclusion criteria (8 quantitative, 12 qualitative, 10 mixed-methods, and 1 summary article). Overall, 29 articles were undertaken in schools (elementary n=17, middle n=3, secondary n=4, or a combination n=5) and two in childcare settings. The main health behaviours targeted included physical activity (n=9), diet (n=3), both diet and physical activity (n=15), and smoking (n=4), either independently (n=1) or combined with other health behaviours (n=3). Findings suggest that the majority of the 59 barriers and 74 facilitators identified to impact on intervention sustainment were similar across school and childcare settings. Factors predominantly relating to the ‘inner contextual factors’ of the organisation including: availability of facilities or equipment, continued executive or leadership support present, and team cohesion, support, or teamwork were perceived by stakeholders as influential to intervention sustainment. Conclusions Identifying strategies to improve the sustainment of health behaviour interventions in these settings requires a comprehensive understanding of factors that may impede or promote their ongoing delivery. This review identified multi-level factors that can be addressed by strategies to improve the sustainment of such interventions, and suggests how future research might address gaps in the evidence base. Trial registration This review was prospectively registered on PROSPERO: CRD42020127869, Jan. 2020.

Background The ‘scale-up’ of effective physical activity interventions is required if they are to yield improvements in population health. The purpose of this study was to systematically review the effectiveness of community-based physical activity interventions that have been scaled-up. We also sought to explore differences in the effect size of these interventions compared with prior evaluations of their efficacy in more controlled contexts, and describe adaptations that were made to interventions as part of the scale-up process. Methods We performed a search of empirical research using six electronic databases, hand searched reference lists and contacted field experts. An intervention was considered ‘scaled-up’ if it had been intentionally delivered on a larger scale (to a greater number of participants, new populations, and/or by means of different delivery systems) than a preceding randomised control trial (‘pre-scale’) in which a significant intervention effect (p < 0.05) was reported on any measure of physical activity. Effect size differences between pre-scale and scaled up interventions were quantified ([the effect size reported in the scaled-up study / the effect size reported in the pre-scale-up efficacy trial] × 100) to explore any scale-up ‘penalties’ in intervention effects. Results We identified 10 eligible studies. Six scaled-up interventions appeared to achieve significant improvement on at least one measure of physical activity. Six studies included measures of physical activity that were common between pre-scale and scaled-up trials enabling the calculation of an effect size difference (and potential scale-up penalty). Differences in effect size ranged from 132 to 25% (median = 58.8%), suggesting that most scaled-up interventions typically achieve less than 60% of their pre-scale effect size. A variety of adaptations were made for scale-up – the most common being mode of delivery. Conclusion The majority of interventions remained effective when delivered at-scale however their effects were markedly lower than reported in pre-scale trials. Adaptations of interventions were common and may have impacted on the effectiveness of interventions delivered at scale. These outcomes provide valuable insight for researchers and public health practitioners interested in the design and scale-up of physical activity interventions, and contribute to the growing evidence base for delivering health promotion interventions at-scale. Trial registration PROSPERO CRD42020144842.