Monitoring drug concentrations in humans to optimize efficacy and reduce toxicity is not a new concept in clinical pharmacology. It has been successfully applied to many different classes of drugs. As a result of considerable concentration and response data, the concept of therapeutic drug monitoring (TDM) has been expanded to certain antiretroviral compounds. In particular, protease inhibitors and nonnucleoside reverse transcriptase inhibitors may be viable candidates for TDM, and limited clinical trial data suggest monitoring plasma concentrations of these agents may indeed clinically benefit patients with HIV infection. A primary distinction between TDM of antiretroviral drugs compared with other drugs is that multiple agents are concomitantly used to treat HIV infection. As with all illnesses that require self-administered drug therapy, poor adherence is a major impediment to success. However, in the treatment of HIV infection, inadequate drug concentrations will result in the appearance or evolution of drug resistance mutations that can endanger present and future drug treatment options. Procedures for sample collection, cross-validation of analytical procedures, and interpretation of assay results should be standardized. More clinical data are needed to confirm this approach and methods of implementing TDM should be further explored. This position paper offers guidelines to aid clinicians who choose to incorporate TDM into the routine care of their patients.