We would like to thank Cusinato et al. for their thoughtful comments. The relative prevalence of hepatitis C virus (HCV) genotypes was 87% (genotype 1), 7% (genotype 2), 5% (genotype 3), and 1% (genotype 4). Clearly, HCV genotype is strongly associated with sustained virological response (SVR) rates and with the likelihood of receiving antiviral treatment. HCV genotype was also associated with development of cirrhosis or hepatocellular carcinoma (HCC) in our data set, but the results were not statistically significant. Specifically, for cirrhosis, the adjusted odds ratio (OR) was 0.74 (95% confidence interval [CI]: 0.52-1.06) for genotype 2 patients and 1.06 (95% CI: 0.71-1.56) for genotype 3 patients, relative to genotype 1 patients as the reference group. For HCC, the adjusted OR was 0.53 (95% CI: 0.16-1.73) for genotype 2 patients and 0.82 (95% CI: 0.25-2.7) for genotype 3 patients, relative to genotype 1 patients. We did not find the proportion of human immunodeficiency virus (HIV)/HCV coinfected patients who ever received antiviral treatment (18%) surprisingly low. Only 22% (36,898 of 170,119) of all HCV mono-infected patients who were in Veterans Affairs (VA) care in 2011 ever received antiviral treatment. Lack of antiviral treatment may be partly explained by the presence of absolute and relative contraindications to antiviral treatment, which may be more common in VA than non-VA patients, by the relatively low success rate of antiviral treatment and its significant side effects. It is beyond the scope of this work to examine the patient, provider, and facility predictors of receipt of antiviral treatment. However, it is crucial to point out that these represent treatment rates in an unselected population of HIV/HCV coinfected patients (i.e., all patients who were seen at least once in any VA facility throughout the country in year 2009) and cannot be compared with treatment rates from tertiary referral centers or prospectively followed cohorts of HIV-infected patients. A more-appropriate comparison would be the proportion of all patients with HIV/HCV coinfection in the entire United States who ever received antiviral treatment. Shire et al. reported that SVR rates to HCV antiviral treatment with pegylated interferon (Peg-IFN) and ribavirin (RBV) in HIV/HCV coinfected patients in five randomized controlled trials and two prospective cohort studies ranged from 27% to 44%. We reported that among all patients in VA care in 2009 who ever received antiviral treatment with either Peg-IFN or regular IFN with or without RBV (defined as at least two prescriptions of IFN), 17% achieved SVR. The lower SVR rate that we reported is the result of differences in treatment regimens and the fact that we did not report on results of clinical trials; rather, we reported SVR rates in unselected patients in a real-world setting. There are multiple viral, patient, provider, and facility factors that might account for these differences in SVR rates observed in clinical trials versus SVR rates observed in real clinical practice. The fact that we defined antiviral treatment as receipt of two or more prescriptions of IFN (“intention to treat”) also accounts for lower SVR rates, as Cusinato et al. rightly point out. An analysis of predictors of SVR is fairly complicated and not directly relevant to the presented research. We merely wanted to determine whether achieving SVR is associated with clinical outcomes (cirrhosis and HCC). George N. Ioannou M.D., M.S.*, Christopher L. Bryson M.D., M.S. , Noel S. Weiss M.D., Dr.P.H. , John D. Scott M.D., M.Sc.§, Edward J. Boyko M.D., M.P.H. , * Division of Gastroenterology, Veterans Affairs Puget Sound Health Care System and University of Washington, Seattle, WA, Division of Internal Medicine, Veterans Affairs Puget Sound Health Care System and University of Washington, Seattle, WA, Department of Epidemiology, University of Washington, Seattle, WA, § Division of Infectious Diseases, University of Washington, Seattle, WA.
Read full abstract