Analytical approach:
A Markov model was developed and populated with efficacy and cost data from a range of sources. The model had a three-month cycle length and a three-year time horizon. The model included six health states, four based on serum creatinine levels of 5.0mg per dL (442μmol per L), 6.0mg per dL (530μmol per L), 7.0mg per dL (619μmol per L) and 8.0mg per dL (707μmol per L), plus haemodialysis, and death. All patients were assumed to start in the serum creatinine 5.0mg per dL health state. The authors reported that the analysis was conducted from a Japanese reimbursement perspective.
Effectiveness data:
The data from a single clinical trial conducted in Japan were used to estimate the transition probabilities for each treatment arm (Koide, et al. 1987, see 'Other Publications of Related Interest' below for bibliographic details). This multi-centre, double-blind, phase III clinical trial included 244 patients with severe chronic kidney disease (median serum creatinine level of 6.8mg per dL). The follow-up period was 24 weeks. The survival probabilities for patients undergoing haemodialysis were based on mortalities obtained from the United States Renal Data System for Asian patients with glomerulonephritis, who were undergoing dialysis.
Monetary benefit and utility valuations:
The utility values for each health state were from the published literature (Hogan, et al. 2002, and Smith, et al. 2004, see 'Other Publications of Related Interest' below for bibliographic details).
Measure of benefit:
The primary measure of benefit was quality-adjusted life-years (QALYs). A discount rate of 3% per annum was applied.
Cost data:
The cost categories included the costs of drug acquisition, out-patient treatment for chronic kidney disease (one out-patient visit per month), dialysis, hospitalisation for onset of dialysis, surgery for vascular access, hospitalisation for chronic kidney disease, and death. When patients moved from serum creatinine 7.0mg per dL to 8.0mg per dL it was assumed that they were hospitalised once for chronic kidney disease. The costs were from standard and published sources. They were reported in 2006 Japanese yen (JPY). A discount rate of 3% per annum was applied.
Analysis of uncertainty:
One-way sensitivity analysis was conducted on the transition probabilities, health utilities, and costs. Probabilistic sensitivity analysis was also conducted using beta distributions for the probabilities, and normal distributions for the costs and utilities.