Long-term cost comparison between surgical and medical therapy for benign prostatic hyperplasia: a study using hospital billing data.

OBJECTIVES: To analyse actual long-term medical treatment of benign prostatic hyperplasia (BPH) and compare the incurred cost with that of patients with BPH who underwent early surgery. PATIENTS AND METHODS: Patients who were first diagnosed with BPH from 1 January 2008 to 31 December 2010 were identified using the Clinical Data Warehouse. Hospital billing data generated by the electronic hospital management system were collected until December 2015. For outpatient care, only procedures, materials and drugs directly related to the management of BPH were selected for the analysis. For inpatient care, all procedures, materials and drugs ordered on dates with continuity with BPH surgery date were included. The primary endpoint of the study was the total treatment-related direct costs of patients undergoing a long-term curative medical therapy for BPH (Group 1), which was arbitrarily defined as any medical therapy including a 5α-reductase inhibitor with a minimum medication possession ratio of 0.5 during ≥5 consecutive years, or ≥1 year until BPH surgery due to medical therapy failure. In all, 70 patients who underwent BPH surgery at <1 year of initial visit served as controls (Group 2). RESULTS: Amongst 137 patients in the Group 1, four patients underwent BPH surgery at a median of 57.8 months after the initial visit (2.9%). At a median follow-up of 76 months, the mean total treatment cost was significantly higher in Group 1 than in Group 2 ($3987 vs $3036 [USA dollars], P < 0.001). Similarly, the mean 'out-of-pocket' cost was significantly higher in Group 1 than in Group 2 ($1742 vs $1436, P = 0.005). When a linear increment of annual BPH treatment cost is assumed for Group 1 and all costs are assumed to be produced within the first year for Group 2, the total and out-of-pocket costs became equal at the end of the fifth year of medical treatment. For both total and out-of-pocket costs, medication-related costs occupied the largest proportion, exceeding half of the costs. CONCLUSIONS: We suggest patient counselling at the beginning of BPH treatment should include the likelihood that the cumulative out-of-pocket cost at 5 years of continuous medication will exceed that of early surgery. Our cost study using hospital billing data extractable from the electronic hospital management system may be a good model for cost studies that could provide valuable information to health providers and payers.