Interventions to improve physical activity among socioeconomically disadvantaged groups: an umbrella review

Abstract Background People from socioeconomically disadvantaged population groups are less likely to be physically active and more likely to experience adverse health outcomes than those who are less disadvantaged. In this umbrella review we examined across all age groups, (1) the effectiveness of interventions to improve physical activity among socioeconomically disadvantaged groups, (2) the characteristics of effective interventions, and (3) directions for future research. Methods PubMed/MEDLINE and Scopus were searched up to May 2017 to identify systematic reviews reporting physical activity interventions in socioeconomically disadvantaged populations or sub-groups. Two authors independently conducted study screening and selection, data extraction (one author, with data checked by two others) and assessment of methodological quality using the ‘Assessment of Multiple Systematic Reviews’ scale. Results were synthesized narratively. Results Seventeen reviews met our inclusion criteria, with only 5 (30%) reviews being assessed as high quality. Seven (41%) reviews focused on obesity prevention and an additional four focused on multiple behavioural outcomes. For pre school children, parent-focused, group-based interventions were effective in improving physical activity. For children, school-based interventions and policies were effective; few studies focused on adolescents and those that did were generally not effective; for adults, there was mixed evidence of effectiveness but characteristics such as group-based interventions and those that focused on physical activity only were associated with effectiveness. Few studies focused on older adults. Across all ages, interventions that were more intensive tended to be more effective. Most studies reported short-term, rather than longer-term, outcomes and common methodological limitations included high probability of selection bias, low response rates, and high attrition. Conclusions Interventions can be successful at improving physical activity among children from socioeconomically disadvantaged groups, with evidence for other age groups weak or inconclusive. More high-quality studies in this population group are needed, which adopt strategies to increase recruitment rates and reduce attrition, report longer term outcomes, and provide adequate intervention details, to allow determination of the characteristics of effective interventions. We recommend that the benefits of physical activity be recognised more broadly than obesity prevention in future studies, as this may have implications for the design and appeal of interventions