Neighborhood disparities in stroke and socioeconomic, urban-rural factors using stroke registry
Abstract
IntroductionDespite major improvements in management over recent decades, stroke remains a devastating disease in Europe and geographic disparities persist. Determining the spatial distribution of stroke may be useful for both epidemiological research and health services planning. The aim of this study was to estimate stroke incidence rates in Pays de Brest (Western France) and to explore the presence of geographic clusters associated with socioeconomic and urban-rural characteristics.MethodsThe study included incidence stroke cases aged 60 years and more from the Stroke Brest registry for the period 2008 to 2013 aggregated at the census block level. Poisson and negative binomial regression models were used to explore determinants of disease risk. Geographically weighted regression (GWR) was used to allow estimation of local regression coefficients.ResultsBetween 2008 and 2013, 3088 cases aged 60 years were identified in Pays de Brest. The incidence of stroke was 6.67 per 1 000 inhabitants. Living in a rural neighborhood with a higher level of blue collar and of farmer was associated with higher age adjusted stroke incidence rate ratio (IRR) 1.23, [95%CI 1.09-1.39]. Higher risk was also found in a city center neighborhood with a higher level of blue [IRR 1.22, 95%CI 1.04-1.49]. Local Poisson GWR models achieved the best fit and suggested evidence of spatially varying regression coefficients and clusters of higher incidence.ConclusionsOur study contributes to a better understanding of the relationship between stroke risks and contextual characteristics such as socioeconomic circumstances and urban rural setting. This will help targeting areas for specific public health prevention.