Current issues of ACP Journal Club are published in Annals of Internal Medicine


Interferon-α2b might prolong life and save costs in HBeAg-positive chronic hepatitis B

ACP J Club. 1995 Nov-Dec;123:85. doi:10.7326/ACPJC-1995-123-3-085

Source Citation

Wong JB, Koff RS, Tinè F, Pauker SG. Cost-effectiveness of interferon-α2b treatment for hepatitis B e antigen-positive chronic hepatitis B. Ann Intern Med. 1995 May 1;122:664-75.



To estimate the cost-effectiveness of interferon-α2b in patients with chronic hepatitis B infection who were positive for hepatitis B e antigen (HBeAg).


Decision analytic model to estimate the cost-effectiveness of interferon-α2b. The effectiveness of interferon in causing the loss of HBeAg or hepatitis B surface antigen (HBsAg) was derived from a meta-analysis of 9 randomized controlled trials comparing interferon with standard care. The natural histories of patients with different serologic states were estimated from published studies and opinions of a panel of experts. Because no trials assessing the effect of interferon on hard clinical end points (mortality or morbidity of end-stage liver disease) had been done, the natural histories of the different serologic states were used to predict the subsequent development of decompensated cirrhosis, hepatocellular carcinoma, and death. Cost data were obtained from a correction of the fee structure at the New England Medical Center.


Interferon trials in the United States, Europe, Egypt, Hong Kong, and Argentina. Natural history data from world literature.


552 patients with HBeAg-positive chronic hepatitis B (meta-analysis patients).


Interferon-α2b with a total dose ranging from 112 MU to 560 MU.

Main Cost and Outcome Measures

Life expectancy, quality-adjusted life expectancy, costs, and marginal cost-effectiveness ratios.

Main Results

In the first year, interferon increased the likelihood of becoming negative for HBeAg from 9.1% to 45.6% (95% CI for the 36.5% absolute risk improvement [ARI], 23.7% to 49.2%) and of becoming negative for HBsAg from 1.7% to 7.7% (CI for the 6.0% ARI, 2.8% to 9.3%). For a 35-year-old man with HBsAg/HBeAg-positive chronic hepatitis, interferon therapy, compared with standard care, increased life expectancy by 3.1 years (from 24.8 y to 27.9 y) and quality-adjusted life expectancy by 3.4 years (from 22.4 y to 25.8 y). Interferon decreased lifetime costs by $6600 (from $60 200 to $53 600) and discounted lifetime costs by $2100 (from $32 700 to $30 600). Even with the model biased strongly in favor of standard care, the marginal cost-effectiveness ratio of interferon did not exceed $12 000 per life-year gained.


Interferon-α2b should prolong life and lower costs in patients with chronic hepatitis B who are positive for hepatitis B e antigen.

Sources of funding: In part, National Library of Medicine and Schering-Plough Laboratories.

For article reprint: Dr. S.G. Pauker, New England Medical Center, 750 Washington Street, Box 302, Boston, MA 02111, USA. FAX 617-636-4838. E-mail


The time between initial infection and the appearance of symptomatic end-stage liver disease in chronic viral hepatitis is usually measured in decades. For this reason, trials evaluating interferon therapy use surrogate or intermediate end points, aminotransferase normalizations or serologic phenomena. Wong and colleagues assumed that patients with chronic viral hepatitis who responded to interferon would have the same natural histories that they would have had if they had arrived at that serologic state spontaneously; they also assumed that the residual nonresponders would behave like any other group of HBeAg-positive patients with chronic hepatitis B.

We already know that interferon responders tend to have the same clinical characteristics as patients who have spontaneous improvements (1). It could be postulated that patients are immunologically programmed from the beginning to take a particular clinical course and that interferon, acting as an immunostimulant, merely accelerates recovery in those who were destined to recover anyway. In this case, interferon may not prevent patients from developing end-stage liver disease. Thus, nonresponders would have a much poorer clinical course than calculated, compensating for the responders with the better outcomes. In the Appendix of their article, Wong and colleagues considered this possibility and concluded that if nonresponders were very unlikely to lose HBeAg, standard treatment was the preferred course.

These calculations cannot establish any efficacy or cost-efficacy for interferon therapy. As difficult as they might be to do, only long-term trials can prove that altering the surrogate end point reduces the morbidity or mortality of chronic viral hepatitis. If efficacy is established, then the cost-efficacy of interferon-α2b can be more readily calculated.

Ronald L. Koretz, MD
University of California at Los Angeles Medical School Sylmar, California, USA