Social inequalities in physical activity: do environmental and policy interventions help to reduce the gap? A pilot systematic review

Abstract

BackgroundPhysical inactivity is a mounting public health concern in many countries. A substantial proportion of the UK adult population do not achieve recommended levels of physical activity. Furthermore, growing evidence suggests that opportunities to achieve adequate levels of physical activity might be restricted in some disadvantaged social groups. Several systematic reviews have examined the effectiveness of interventions to promote physical activity, but little is known about the social distribution of intervention effects. Without greater emphasis on the distributional effect of broad environmental or policy interventions, some efforts to improve health might contribute to increasing inequalities. We report the findings of a pilot study that examines how distributional effects have been reported in systematic reviews and primary studies, what methods are available to synthesise distributional effects, and what conclusions can be drawn from these data. MethodsWe used the Active Living Research (ALR) reference lists (2002–10) to search for relevant studies. These lists are generated through systematic searches of PubMed, ISI Web of Science, and various journals. We obtained 2586 abstracts that were scanned for potentially relevant systematic reviews and primary studies; conference abstracts and unpublished studies were not included. Studies were coded by one reviewer using several published instruments that identify information about distributional effects, assess the suitability of study design, and assess the methodological execution of studies. With the PROGRESS-Plus instrument (Campbell Cochrane Equity Methods Group), relevant studies were assessed for intent to present social distributional effects in terms of baseline participant characteristics, intervention effects, adjusted associations, or interaction effects. We then synthesised relevant information from primary studies with a harvest plot to summarise evidence for the presence and direction of any social gradient in intervention effectiveness. FindingsWe identified 19 systematic reviews of environmental and policy interventions. Few (n=5, 26%) were prospectively designed to synthesise distributional effects. However, several studies presented post-hoc descriptive analyses of intervention effects by age and sex. Ten (52%) reviews presented baseline sociodemographic data, and ten reported adjusted associations. We identified no examples of reviews that reported distributional interaction effects. To further investigate the availability of information we analysed the availability of data in primary studies. Relevant information was reported more frequently in primary studies than in reviews: 37 (42%) of studies reported subgroup intervention effects; 16 (18%) reported sociodemographic interaction effects; 26 (30%) reported associations adjusted for sociodemographic characteristics; and, most notably, 80 (90%) reported baseline sociodemographic characteristics of their study samples. In synthesis of this information we noted that primary studies had most frequently examined differential effectiveness by comparing effectiveness by sex, age, ethnic origin, and socioeconomic status. For sex, effects seemed to be evenly distributed overall, although heterogeneity in gradients between studies suggested that some interventions might affect men and women differently. InterpretationPractitioners and policy makers need to understand how initiatives to improve health could affect inequalities. Findings from this pilot study suggest that existing systematic reviews may have insufficient information about the distributional effectiveness of physical activity interventions. However, information is available within existing primary studies of physical activity interventions that might support the investigation and synthesis of distributional effects. Therefore, generation of improved evidence about how public health interventions could affect health inequalities might be both necessary and feasible, with use of innovative methods of research synthesis. FundingThis work was undertaken by the Centre for Diet and Activity Research, a UKCRC Public Health Research Centre of Excellence, with funding from the British Heart Foundation, Economic and Social Research Council, Medical Research Council, NIHR, and the Wellcome Trust, under the auspices of the UK Clinical Research Collaboration.