Improving mental health through the regeneration of deprived neighbourhoods: a prospective controlled quasi-experimental study

Abstract

BackgroundPolicy makers often target deprived neighbourhoods for regeneration with the expectation that population health will improve, since housing and neighbourhoods of low quality, as well as the social and economic determinants of poor health, are concentrated in the most deprived areas. Our aim was to examine the effects of Communities First, a Welsh Assembly Government community-led programme of neighbourhood regeneration targeted at the 100 most deprived electoral wards in Wales on mental health. MethodsInformation on Communities First regeneration activities in 35 intervention lower super output areas (LSOAs) (n=4197) and 75 control LSOAs (6695) were linked to data from the eCATALyST study, a prospective cohort study, in 2001 (before regeneration) and 2008 (after regeneration) within the Secure Anonymised Information Linkage (SAIL) databank. Communities First was delivered through multiagency partnership boards in all 22 local authorities. Boards worked with residents to identify and secure funding for regeneration activities. The primary outcome was the change in Mental Health Inventory (MHI) score (a population-based measure of anxiety and depressive symptoms) between 2001 and 2008 recorded in eCATALyST. We examined the changes in mental health in intervention LSOAs compared with control LSOAs using propensity score matching (1:1 ratio) to balance the level of socioeconomic disadvantage across groups. Sensitivity analysis examined the impact of length of residence in an intervention area and six types of regeneration activity. Findings1500 regeneration projects were funded from 2001 to 2008. Before regeneration, mental health was worse in the intervention than in the control group (mean MHI score 66·6, SD 22·3 vs 71·0, SD 20·8). After propensity score matching, regeneration was associated with an improvement in the mental health of intervention compared with control group residents (β=1·54, 95% CI 0·50–2·59), suggesting that inequalities in mental health narrowed. We found evidence of a dose–response association between length of residence and improvements in mental health (ptrend=0·05). We could not attribute improvements to any one type of regeneration activity. InterpretationTargeted regeneration directed by the residents of deprived urban communities can help reduce inequalities in mental health. FundingThis study was funded by a grant from the National Institute for Social Care and Health Research (RFS-12-05) and undertaken with the support of the Centre for the Development and Evaluation of Complex Interventions for Public Health Improvement (MR/KO232331/1) and the Farr Institute of Health Informatics Research.