Cost-effectiveness of repeat medical procedures: kidney transplantation as an example

The constraints on medical-care resources can give rise to the question of the cost- effectiveness of permitting repeat medical procedures when some patients may die without undergoing even a first procedure. Using kidney transplantation as an example, this study estimates the cost-effectiveness of patients' having available the option of a repeat medical procedure in the event the first procedure fails. Specifically, the anal ysis examines the effect on transplant candidates of having the option of kidney re transplantation, if and when retransplantation might be needed. Data sources include the U.S. Renal Data System (USRDS) Case-Mix Severity Study, Health Care Financ ing Administration (HCFA) data, and a MEDLINE search. Outcome measures include life expectancy, quality-adjusted life expectancy, lifetime costs of medical care, and mar ginal cost-effectiveness from a societal perspective. By avoiding lifelong dialysis after graft failure, first-transplant candidates gain an average of 47 quality-adjusted days with a retransplantation policy, despite the prolongation of time to first transplant by an average of 30 quality-adjusted days. The lifetime cost of medical care per first- transplant candidate is $1,210 higher with a retransplantation policy compared with the no-retransplantation policy; its societal cost-effectiveness is estimated to be $9,656 per quality-adjusted life-year saved. The retransplantation policy provides the greatest improvement in quality-adjusted life expectancy for younger candidates. In the case of kidney transplantation, the cost-effectiveness of a repeat transplant, on average, com pares favorably with those of other medical strategies in common practice. As re sources become increasingly constrained, this study demonstrates a framework for considering the cost-effectiveness of repeat medical procedures. Key words: kidney transplantation; decision analysis; cost-effectiveness analysis; ethics; rationing. (Med Decis Making 1997;17:363-372))