The therapeutic response among patients infected with hepatitis C virus (HCV) of genotype 1 remains suboptimal. We examined a large database of patients treated with combination therapy (ribavirin plus an interferon [IFN]). We hypothesized that the longer the duration that the virus load was rendered undetectable in serum, the better would be the probability of a sustained viral response (SVR). A model predicting SVR was generated; it included the following covariates: duration of continuous nondetectability of an HCV load in serum, estimated creatinine clearance, and whether the isolate was of genotype 1. The validation model demonstrated positive and negative predictive values as well as sensitivity and specificity exceeding 90%. The model predicted that patients infected with HCV genotype 1 require continuous nondetectability of virus load in serum for 36 and 32 weeks, to attain 90% and 80% probabilities, respectively, of a SVR. The average time to clear serum of genotype-1 virus was 30.4 weeks, which indicates that the 48-week duration of therapy provided a suboptimal probability of a SVR. For some patients, suboptimal therapy with pegylated IFN plus ribavirin may need to be of longer duration than the currently recommended 48 weeks. This hypothesis requires prospective validation.