The Functional Assessment of Human Immuno-deficiency Virus (HIV) Infection (FAHI) quality of life instrument was developed using a combination of conceptual and empirical strategies. The core, general health-related quality of life instrument is the Functional Assessment of Cancer Therapy-General (FACT-G) questionnaire. The FACT-G was selected to enable comparison of data across two similar, life-threatening conditions and because of its desirable psychometric properties. Initial data on both the relevance (applicability) of the FACT-G to the HIV population and the generation and testing of questions for an HIV-specific subscale were encouraging. Consequently, the FACT-G and a 9-item HIV-specific subscale were combined and tested in 196 patients in three categories: an English-speaking stress management sample from Chicago, illinois (n = 110); an English-speaking urban, mixed race sample from Chicago (n = 71); and a Spanish-speaking urban sample from Chicago and San Juan, Puerto Rico (n = 64). With the exception of the Social Well-being subscale, the subscales of the FACT-G demonstrated good internal consistency reliability across all three samples (alpha range = 0.72-0.88). Total FAHI scores produced consistently high alpha coefficients (0.89-0.91). Concurrent validity data included moderately strong associations with other measures of similar concepts and an ability to distinguish groups of patients by activity level and disease severity. Sensitivity to change in mood disturbance and responsiveness to a stress management intervention were also evident. The 9-item HIV-specific subscale demonstrated relatively low alpha coefficients (range = 0.53-0.71) and marginal sensitivity to change, leading to supplementation of content with an additional 11 items, creating a 20-item HIV-specific subscale that is currently being tested. Clinical trial and clinical practice investigators are encouraged to use the FACT-G in its current (version 3) form when evaluating group differences and within-group change over time. It should prove particularly useful when comparing clinical trial and clinical practice data for cancer vs. HIV-infected patients and in the evaluation of treatments for HIV disease and HIV-related malignancy. The supplemental 20 questions comprising the revised HIV-specific subscale are undergoing further testing, and may ultimately enhance the value of this measurement system.