In their paper in the November issue of HEPATOLOGY, Younossi et al. 1 may have overestimated the cost effectiveness of interferon-ribavirin treatment. This mainly stems from their failure to take into account the fact that patients who achieve sustained response are not homogenous with respect to their risk of development of decompensated liver disease. Some of the predictive factors that are associated with higher response rate after interferon-ribavirin combination therapy as reported by Poynard et al., 2 including genotype, viral load, age, and fibrosis, have also been associated with a better prognosis in natural history studies. 3,4 Thus, patients who respond to therapy tend to have a lower risk of developing cirrhosis and its complications. Ignoring this overlap of predictive factors would lead to an overestimation of the costeffectiveness of antiviral treatment.
The relevance of this point is illustrated by the difference in the results between the study by Younossi et al. and our previous analysis. 5 In their Table 5, the authors reported that interferon monotherapy for 48 weeks was cost-saving. In other words, the cost of interferon treatment up front was more than compensated for by the reduction in future costs incurred by later complications of chronic liver disease. This is in contrast to our analysis in which interferon therapy increased the overall cost, although its cost-effectiveness remained within a reasonable range in comparison with other medical interventions. The methods used in both analyses were essentially identical including the response rate, discount rate, and the rate for discontinuing the treatment at 3 months in initial nonresponders. The only significant difference was that Younossi et al. did not incorporate the fact that responders to interferon therapy would be less likely to develop liver problems in the future. Instead, they assumed that patients with chronic hepatitis C would progress to cirrhosis at a uniform rate of 7.3% per year. This was based on a study by Yano et al., 6 in which the rate of histological progression was assessed in sequential biopsies in patients with suspected progressive disease. These assumptions predispose the model to an inherent bias in favor of benefits of antiviral therapy, including not only interferon monotherapy but also the combination of interferon and ribavirin.
Given the relatively high cost per quality-adjusted life year (QALY) figures reported by the authors (over $37,000), an unbiased estimate of the cost-effectiveness of combination therapy may in fact approach the arbitrary but widely accepted limit of $50,000 per QALY gained.