Implementation of mental health service recommendations in England and Wales and suicide rates, 1997-2006: a cross-sectional and before-and-after observational study.

BACKGROUND: Research investigating which aspects of mental health service provision are most effective in prevention of suicide is scarce. We aimed to examine the uptake of key mental health service recommendations over time and to investigate the association between their implementation and suicide rates. METHODS: We did a descriptive, cross-sectional, and before-and-after analysis of national suicide data in England and Wales. We collected data for individuals who died by suicide between 1997 and 2006 who were in contact with mental health services in the 12 months before death. Data were obtained as part of the National Confidential Inquiry into Suicide and Homicide by People with Mental Illness. When denominator data were missing, we used information from the Mental Health Minimum Data Set. We compared suicide rates for services implementing most of the recommendations with those implementing fewer recommendations and examined rates before and after implementation. We stratified results for level of socioeconomic deprivation and size of service provider. FINDINGS: The average number of recommendations implemented increased from 0·3 per service in 1998 to 7·2 in 2006. Implementation of recommendations was associated with lower suicide rates in both cross-sectional and before-and-after analyses. The provision of 24 h crisis care was associated with the biggest fall in suicide rates: from 11·44 per 10 000 patient contacts per year (95% CI 11·12-11·77) before to 9·32 (8·99-9·67) after (p<0·0001). Local policies on patients with dual diagnosis (10·55; 10·23-10·89 before vs 9·61; 9·18-10·05 after, p=0·0007) and multidisciplinary review after suicide (11·59; 11·31-11·88 before vs 10·48; 10·13-10·84 after, p<0·0001) were also associated with falling rates. Services that did not implement recommendations had little reduction in suicide. The biggest falls in suicide seemed to be in services with the most deprived catchment areas (incidence rate ratio 0·90; 95% CI 0·88-0·92) and the most patients (0·86; 0·84-0·88). INTERPRETATION: Our findings suggest that aspects of provision of mental health services can affect suicide rates in clinical populations. Investigation of the relation between new initiatives and suicide could help to inform future suicide prevention efforts and improve safety for patients receiving mental health care. FUNDING: National Patient Safety Agency, UK.