Cost effectiveness of microwave thermotherapy in patients with benign prostatic hyperplasia: part I-methods.

OBJECTIVES To present the method used to evaluate the cost effectiveness, from the societal perspective, of transurethral microwave thermotherapy relative to medical therapy (alpha-blocking agents) and transurethral resection of the prostate (TURP) for a hypothetical cohort of 65-year-old men with moderate-to-severe benign prostatic hyperplasia (BPH) symptoms. METHODS We constructed a decision-analytic Markov model with 25 health states describing the 3 treatments, 5 short-term clinical events, and 17 possible long-term outcomes. Each health state had an associated cost and utility. Utility weights, reflecting an individual's preference for a specific health outcome, range from 0, indicating death, to 100, indicating perfect health. Utility estimates were obtained by interviewing 13 men with moderate-to-severe BPH symptoms using the standard gamble preference measurement technique. On the basis of their risk attitudes, the patients were classified as risk averse or non-risk averse. The rates of remission, temporary and permanent adverse events, retreatment, and mortality were obtained from the Targis System (Urologix) randomized clinical trial, published reports, and a consensus panel. The costs during the 5 years after treatment initiation were estimated using national Medicare reimbursement schedules. The costs are reported in 1999 U.S. dollars. RESULTS Eliciting utility values from patients with BPH was feasible and generated internally consistent and externally valid measures. In the non-risk-averse group, the utility value for significant remission, moderate remission, no remission, and worsening BPH symptoms without an adverse event was 99.1, 97.1, 94.4, and 87.3, respectively. As expected, the risk-averse individuals (n = 6) exhibited higher utility values than those in the non-risk-averse group (n = 7). In the non-risk-averse group, thermotherapy was the preferred treatment, and in the risk-averse group, medical therapy was preferred. In both groups, TURP was the least preferred therapy. The initial thermotherapy procedure costs without complications were estimated at $2629, and the initial TURP procedure costs without complications were estimated at $4597. Time-dependent probabilities were developed to reflect treatment durability. CONCLUSIONS The resulting model parameters appear to be suitable for evaluating the cost effectiveness of thermotherapy relative to medical therapy and TURP in 65-year-old men with moderate-to-severe BPH symptoms.