Surrogate markers are an important component in the process of investigating management and prevention strategies, and for increasing understanding of viral diseases. The importance of surrogate markers and applied statistical models is particularly true for HIV. For HIV infection, the development of such methods provides new approaches for evaluation of HIV therapies and vaccines, and for the study of HIV transmission and its pathogenesis. The complex natural history of hepatitis B infection demonstrates that viral load is not the only predictor of transmission of this virus; for hepatitis C infection, viral load per se is not a prognostic factor for disease progression, but cumulative viral load may affect the outcome, and therapy is aimed at eliminating active viral replication. © 2004 Elsevier B.V. All rights reserved.