Date Published: August 15, 2017
Publisher: Oxford University Press
Author(s): James White, Giles Greene, Daniel Farewell, Frank Dunstan, Sarah Rodgers, Ronan A. Lyons, Ioan Humphreys, Ann John, Chris Webster, Ceri J. Phillips, David Fone.
Neighborhood-level interventions provide an opportunity to better understand the impact of neighborhoods on health. In 2001, the Welsh Government, United Kingdom, funded Communities First, a program of neighborhood regeneration delivered to the 100 most deprived of the 881 electoral wards in Wales. In this study, we examined the association between neighborhood regeneration and mental health. Information on regeneration activities in 35 intervention areas (n = 4,197 subjects) and 75 control areas (n = 6,695 subjects) was linked to data on mental health from a cohort study with assessments made in 2001 (before regeneration) and 2008 (after regeneration). Propensity score matching was used to estimate the change in mental health in intervention neighborhoods versus control neighborhoods. Baseline differences between intervention and control areas were of similar magnitude as produced by paired randomization of neighborhoods. Regeneration was associated with an improvement in the mental health of residents in intervention areas compared with control neighborhoods (β = 1.54, 95% confidence interval: 0.50, 2.59), suggesting a reduction in socioeconomic inequalities in mental health. There was a dose-response relationship between length of residence in regeneration neighborhoods and improvements in mental health (P-trend = 0.05). These results show that targeted regeneration of deprived neighborhoods can improve mental health.
The protocol for this prospective controlled quasi-experimental study or natural experiment has been published previously (21). We linked data from 3 sources for this study: regeneration data from the Caerphilly County Borough Council (Caerphilly, Wales), the eCATALyST study (22), and the Secure Anonymised Information Linkage (SAIL) Databank (23). The eCATALyST study received ethical approval for the baseline population survey (2001) from the former Gwent Local Research Ethics Committee and for the follow-up survey (2008) from the South East Wales Research Ethics Committee, Panel C. The linking of data sets received approval from the SAIL Information Governance Review Panel at Swansea University (Swansea, Wales).
There were 1,500 funded regeneration projects in Caerphilly County Borough during the 7-year follow-up period (2001–2008) at a cost of £82,857,180 ($161,571,501 at midyear 2008 currency exchange rate). Of these projects, over one-half (59.1%; £16,489,716 ($32,154,946)) were classified as community-based projects, with the remaining classified as improvements in housing and the physical environment (22.3%; £55,670,516 ($108,557,506)), projects in educational settings (8.0%; £5,534,839 ($10,792,936)), health improvement (4.7%; £1,321,639 ($2,577,196)), interventions to reduce crime (4.2%; £1,742,129 ($3,397,152)), and the provision of vocational training or business support (1.7%; £2,098,341 ($4,091,765)).
We estimated the impact of socioeconomic regeneration on mental health in a natural experiment by linking data from a regeneration program delivered to deprived communities to data from a prospective cohort study that collected data both before and after intervention. We found that the targeted regeneration of deprived neighborhoods narrowed inequalities in mental health, and we found some evidence for a dose-response association between length of residence in a regeneration area during the study period and improvements in mental health.