Prescription Drug Data and the National Diabetes Surveillance System Case Definition

Abstract

ABSTRACT OBJECTIVE To investigate the effect of supplementing the National Diabetes Surveillance System (NDSS) case definition with prescription drug claim information on the identification of diabetes cases. METHODS Saskatchewan Health's administrative databases (1991 to 2001) were used to identify individuals with diabetes using the current NDSS case definition and 2 randomly selected controls without diabetes. Among the controls, we identified those individuals who failed to satisfy the NDSS criteria but had at least 1 prescription of insulin or oral antidiabetic medication. Factors associated with not meeting the NDSS case definition were examined. RESULTS Only 0.55% of the control cohort had at least 1 recorded dispensation of insulin or oral antidiabetic medication. Compared to the additional drug claim data, the case capture of the standard NDSS case definition was estimated to be 94.4%. Females, younger individuals and those with fewer comorbidities were more likely to be missed. CONCLUSION The addition of prescription drug claim data did not substantially increase case capture.