The use of GP databases in comparative research: the case of diabetes care in Sweden and the Netherlands.

Background: The prevalence of diabetes mellitus (DM) is increasing globally. In many European countries, management of DM has become a responsibility of primary care. Objectives: To compare pharmaceutical treatment of DM in Sweden and the Netherlands for patients with and without cardiovascular co-morbidity using routine primary care registration databases. Methods: Data (2005) were used from the Skaraborg Primary Care Reasearch Database (SPCRD) and the Netherlands Information Network of General Practice (LINH) Setting: general practices/health care centers, subjects: 19,000 patients with diabetes mellitus Main outcome measure: pharmaceutical treatment of patients with diabetes (yes/no) Analysis: Multilevel analysis from which the percentage of patients in each country twith certain treatment was calculated. Results: Overall, pharmaceutical treatment of diabetes patients with antihyperglycmic drugs and statins is more likely in the Netherlands compared to Sweden, but insulin is a more preferred treatment in Sweden (25% in Sweden versus 16% in the Netherlands) Higher treatment rates in the Netherlands are fully attributable to patients without cardiovascular co-morbidity. Dutch DM patients without cardiovascular co-morbidity receive the same pharmaceutical treatment as their counterparts with cardiovascular co-morbidity. Swedish GPs, however, make a clear distinction between DM patients with and without cardiovascular comorbidity. A striking difference between the two countries is that while in the Netherlands 40% of all DM patients without cardiovascular co-morbidity were prescribed statins, none of their Swedish peers was. Conclusions: Our study (re-) emphasizes that there are considerable between-country differences in the management of DM and that cardiovascular co-moborbidity cannot be ignored in the explanation of cross-country variation in pharmaceutical treatment of DM.(aut. ref.