An investigation into the spatial epidemiology of suicide in England and Wales

Abstract

Background: The incidence of suicide exhibits marked geographical variability. Explanations for the observed patterns are not well understood. Previous area-based studies of suicide in Britain have focused on relatively large areas and/or have not investigated whether associations varied between different age groups. Aims: Two interrelated aspects of the geography of suicide were investigated: (a) the magnitude and spatial patterning of its geographical variation and (b) the extent to which socio-economic characteristics of areas explain the observed patterns. Analyses were based on parliamentary constituencies (N=569) and electoral wards (N=9265) in England and Wales, 1988-1994.Methods: Random effects regression models (negative Binomial models in Stata and Bayesian hierarchical models in winBUGS) were used to derive and map age-and sex-specific "smoothed" area estimates of rate ratios by incorporating evidence of heterogeneity and spatial autocorrelation into their estimation. Associations between levels of suicide and a range of 1991 census-derived socio-economic characteristics of the areas were investigated. Furthermore, the patterning of any unexplained variation was assessed in a series of maps of residual rate ratios after controlling for the effect of area characteristics.Results: There was strong evidence of heterogeneity and spatial patterning at both levels of geographic aggregation. Although the geography of suicide differed across age/sex groups, two main common patterns emerged: clusters of high rates in (a) remote and coastal parts of the country and (b) central parts of cities (with low rates in their periphery). Possible indicators of levels of social fragmentation in an area, such as single-person households and unmarried population, were most consistently associated with suicide in all age/sex groups. These associations appeared independent of levels of socio-economic deprivation and stronger in the younger age-groups. While area characteristics accounted for 50%-75% of the observed geographic variation, up to 2-fold differences in rates remained unexplained. Residual variability also clustered spatially.Conclusions: Suicide is an important contributor to area health inequalities. This thesis described its geography and identified a number of area characteristics associated with area differences. Spatial patterning in unexplained variability may be a product of additional local risk factors not accounted for in the models. Further research should seek to identify such factors as well as clarify the extent to which the observed area differences reflect the high-risk demographic composition of such areas or true area influences on the mental health and suicide risk of their populations.