Population Risk Factors for Hospitalization for Stroke in Scotland

Background. There is considerable heterogeneity of cerebrovascular disease (CVD) incidence throughout Scotland. We sought to relate this geographical variation to differences in population risk factors. Method. A computer search of linked national hospital discharge data was performed. Individuals were selected if they had been coded with a diagnosis of CVD (ICD-9 430-438), 1984-1986. Age-sex specific rates were correlated to deprivation score and estimated population risk factors for the 22 local government districts of the Scottish Heart Health study (1984-1986). Results. Men were more commonly admitted in all districts except one, with an overall male :female ratio of 1.43 (P < 0.0001). Rates for men were significantly higher in populations with elevated mean diastolic blood pressure, greater mean alcohol consumption, and where fewer men ate fruit. The most significant associations of male CVD hospital admissions were with population density and deprivation. A similar pattern was found for women, where CVD was significantly associated with elevated diastolic blood pressure, more smokers and fewer non-smokers, and fewer eating fruit. Again, rates were higher in densely populated, deprived districts. Multiple linear regression identified the proportion of current smokers as most explanatory of interdistrict rate differences for women, and population density and deprivation for men. Sex difference was explained by vitamin C intake. Conclusion. Geographical variation in CVD rates in Scotland is related to identifiable population risk factors for atherosclerosis and urban environment.