Patient-level estimates of the cost of complications in diabetes in a managed-care population

Objective: To develop incidence-based estimates of the cost of several diab etes-related complications. Design and setting: This was a retrospective cohort study in a large health maintenance organisation. A total of 8905 patients with type 1 (insulin-de pendent) and type 2 (non-insulin-dependent) diabetes mellitus and 36 520 ag e- and gender-matched controls without diabetes were observed from 1992 to 1995. Incidence rates of 6 major diabetes-related complications were comput ed for both populations. Annual health expenditures in the first and second year following diagnosis were computed for each complication. For comparis on, annual costs were derived for individuals without diabetes or the compl ication of interest. Main outcome measures and results: Over 3 years of observation, incidence r ates for the groups with and without diabetes were as follows: myocardial i nfarction 9.0 versus 3.2%; stroke 8.7 versus 3.8%; hypertension 26.2 versus 16.9%; end-stage renal disease 5.9 versus 1.4%; foot ulcer 7.9 versus 1.1% ; and eye disease 44.3 versus 2.8%. Expressed as a multiple of the average annual cost of care for those without diabetes [$US3400/year (1995 dollars) for those over 65 years of age] and the related complication of interest, excess expenditures for those with diabetes were as follows for the first y ear following diagnosis: no complications 1.59; myocardial infarction 4.1; stroke 3.5; hypertension 2.56; end-stage renal disease 4.32; foot ulcer 4.0 ; and eye disease 2.46. For younger cohorts (less prevalent in the sample), incremental costs for each complication were generally greater than in the older group. Conclusions: The high incidences and costs may support the value of aggress ive early intervention for patients with diabetes. These data will be usefu l for pharmacoeconomic modelling of the cost effectiveness of new and exist ing therapies for this condition.