Effect of future costs on cost-effectiveness of medical interventions among young adults - The example of intensive therapy for type 1 diabetes mellitus

OBJECTIVES, Recent research based on a lifetime utility maximization model has suggested that cost-effectiveness analyses should account for all futur e costs, including medical costs for related and unrelated illnesses and no nmedical costs. This work has also shown that analyses that omit future cos ts are biased to favor interventions among the elderly that extend life ove r interventions that improve quality of life. However, the effect of includ ing future costs on the cost-effectiveness of interventions among the young has not been studied. This article examines the effect of including future costs on the cost-effectiveness of intensive therapy for type 1 diabetes m ellitus among young adults, METHODS. By modifying a cost-effectiveness model based on the Diabetes Cont rol and Complications Trial to include future costs, the effect of includin g future costs on the cost-effectiveness of intensive therapy for type 1 di abetes mellitus among young adults was examined. Future costs added to the model included future costs for medical expenditures for illnesses unrelate d to diabetes and future nonmedical expenditures net of earnings. RESULTS. Intensive therapy among young adults led to approximately equal in creases in the expected number of years lived before age 65, when people ge nerally produce more than they consume, and after age 65, when the opposite tends to hold. Because the discounted value of savings due to lower mortal ity before age 65 exceeded the discounted value of later increases in costs due to lower mortality after age 65, accounting for future costs decreased the cost-effectiveness ratio from $22,576 to $9,626 per quality-adjusted l ife-year. CONCLUSIONS. The inclusion of future costs can significantly improve the co st-effectiveness of interventions that decrease mortality among young adult s. The common practice of excluding future costs may bias cost-effectivenes s analyses against such interventions.