Nationwide survey of comorbidity, use, and costs of all medications in Finnish diabetic individuals

OBJECTIVE - To investigate comorbidity and overall use and costs of medication for all Finnish individuals with diabetes treated with drugs compared with sex- and age-matched control subjects. RESEARCH DESIGN AND METHODS - According to a cross-sectional population study using national registries, 116,224 individuals purchased antidiabetic medications in Finland in 1995. The same number of nondiabetic individuals matched for sex, age, and area of residence were chosen as control subjects. Age at onset of diabetes was used as a criterion for distinguishing between type 1 and type 2 diabetes. The criterion could be applied in 74% of cases. A total of 16,955 individuals were defined as having type 1 diabetes, and 68,517 were defined as having type 2 diabetes. Comorbidity was determined by linkage with a national register including all individuals entitled to special reimbursement for drug treatment for a range of chronic diseases. Data on use and costs of all medications prescribed were obtained from drug purchase records. RESULTS - Cardiovascular diseases and uremia were, as expected, the chronic diseases most closely associated with diabetes. Use of almost all kinds of medication was significantly greater in individuals with type 1 and type 2 diabetes than in control subjects. The greatest differences were observed in relation to cardiovascular drugs and antibiotics. Unexpectedly low use of antiasthmatics was observed in individuals with both types of diabetes, low use of neuroleptics was observed in type 1 diabetic individuals, and low use of hormone replacement therapy was observed in women with type 2 diabetes. Total costs of medications for individuals with diabetes were 3.5 times greater than those for nondiabetic control subjects. The higher costs were mostly attributable to insulin therapy for individuals with type 1 diabetes. The higher costs for individuals with type 2 diabetes were related to the cost of medications other than antidiabetic medication. The possible selection bias in omitting diabetic individuals treated with diet only and individuals in whom diabetes type could not be determined must be considered in interpreting the results. CONCLUSIONS - Greater use by and costs of medications for individuals with diabetes than for nondiabetic individuals is related not only to antidiabetic treatment but also to all other kinds of medications. Although drug treatment and the prevalence of several chronic conditions were overall greater in individuals with diabetes versus other individuals, some exceptions merit further study.