Cost-effectiveness of a chronic disease management model for cirrhosis: Analysis of a randomized controlled trial.

Abstract

In this follow-up study to a randomized controlled trial of a chronic disease management (CDM) model in cirrhosis, our aim was to assess the relative cost-effectiveness of this model compared with usual care during the 12-month study period, using incremental costs per death avoided as the primary outcome. Mean differences in hospitalization costs, deaths avoided, and change in Chronic Liver Disease Questionnaire (CLDQ) total scores were presented with 95% non-parametric bootstrapped confidence intervals. Results were also presented using a cost-effectiveness plane (CEP) and cost-effectiveness acceptability curve. The CDM intervention was more expensive, by 18 521 AUD per participant, but more effective (% of deaths at 12months: 10% vs 15% and 0.67units increase per patient in CLDQ total scores). The resultant incremental cost-effectiveness ratios were 370425 AUD per death avoided (95% confidence interval: -14564 AUD to 2059373 AUD) and 27547 AUD per unit improvement in the CLDQ total score (95% CI: 7455 AUD to 143874 AUD). The CEPs demonstrated some uncertainty around cost-effectiveness. The cost-effectiveness acceptability curves demonstrated that at willingness to pay values of 400000 AUD per additional death avoided and 40000 AUD per unit improvement in the CLDQ, there was at least a 70% probability of CDM being more cost-effective than usual care. At 24months, CDM was much more effective (12% less deaths but now also cheaper by 985 AUD per patient). The analysis of data from a randomized controlled trial suggests that the CDM intervention used is likely to be cost-effective, relative to usual care, due to fewer patient deaths.