Variation of diabetes mellitus prevalence in general practice and its relation to deprivation.

This is an observational study to compare age standardized diabetes prevalences and relate these to socio-economic measures of deprivation. It includes data from eight general (family) practices in the Bristol, UK, area with no ethnic minorities affecting diabetes prevalence. A total population of 71 599 was covered, including 181 Type 1 and 901 Type 2 diabetic patients, 91 of whom were controlled with insulin, 499 with oral hypoglycaemics, and 311 with diet alone. Actual Type 1 and Type 2 diabetes prevalences were standardized to what they would be if each practice had the UK national age profile. Total standardized diabetes prevalence varied from 1.31% to 2.51% (p < 0.001) and Type 2 diabetes prevalence from 0.97% to 2.29% (p < 0.001). There was no significant variation in the prevalence of Type 1 diabetes. The Spearman rank correlation coefficient indicated a significant association between standardized diabetes prevalence and two measures, the Jarman and Townsend indices, of deprivation in the electoral ward where each practice was situated. Total standardized diabetes prevalence was significantly correlated with each of the Jarman and Townsend indices (r = 0.76, p < 0.05). Standardized Type 2 diabetes prevalence was similarly significantly correlated to each deprivation index (rs = 0.74, p < 0.05). Type 2 diabetes prevalence is affected by socio-economic factors with implications for health targets and capitation based budgets.