In a previous observational study with six years of data from Saskatchewan Health, we observed that management costs for people in the years prior to their diagnosis of diabetes were approximately 50% that of management costs for those diagnosed with diabetes, that management costs for Registered Indians were substantially higher that that for the general population, and that the presence of concurrent diseases was associated with significant increases in management costs. These observations require confirmation utilizing an extended observation period and comparison with a control group of subjects who do not develop diabetes.

A cohort of subjects who have or develop diabetes during the years 1991 to 2001 will be identified using standard case definitions from the Saskatchewan Health databases. In essence, the observation period for subjects utilized in the 1991-1996 cost study will be extended, whilst subjects developing diabetes from 1997 through 2001 will be added into the cohort. We will also obtain health resource utilization information for a cohort of Saskatchewan residents who do not meet the case definition for diabetes during the study period.

Initial data analysis will involve descriptive statistics of resource utilization and associated costs with an estimation of the mean cost per subject for each calendar year. This database will allow us to estimate the incremental management costs for major diabetes-related complications as well as comparisons of management costs between subjects with and without diabetes. The length of this observation period will facilitate a number of epidemiologic analyses such as incidence of diabetes and important concurrent illnesses.

Research Team

Jeff Johnson, Scot Simpson, Sumit Majumdar, Phil Jacobs, Arto Ohinmaa, William Osei, Winanne Downey


  • Institute of Health Economics
  • Aventis Canada
  • Pfizer Canada