The extent of use of alternative therapies, psychosocial and disease-specific variables predictive of alternative therapy use, and factors motivating the use of alternative therapies in HIV-infected patients receiving highly active antiretroviral therapy (HAART) have not been well defined. Types of alternative therapies used, demographic and medical data, coping (Billing and Moos inventory of coping with illness styles), social support (Irwing and Sarason questionnaire), sense of personal control (Pearlin's Mastery scale), quality of life (Medical Outcome Study scale), health beliefs, and adherence rate were prospectively assessed in 118 HIV-infected patients receiving HAART. Of 38% (45/118) of the patients who used alternative therapies, 56% (25/45) began using alternative therapies since the initiation of HAART. While Caucasian patients were more likely to use alternative therapies than all other patients (P = 0.015), new users of alternative therapies were more likely to be African-American (P = 0.022). Alternative therapy users reported less satisfaction with their emotional support (P = 0.027), and had greater psychological distress (P = 0.048), but were more likely to utilize problem-focused coping (P = 0.015). Patients who used alternative therapies were less likely to believe that HAART was beneficial (P = 0.06). Physicians were unaware of patients' alternative therapy use in 40% (18/45) of all patients who used alternative therapies, in 67% of herbal therapy users, and in 100% of dietary supplement users. Adherence to antiretroviral therapy, CD4 count, and HIV-RNA level were neither predictive nor affected by alternative therapy use. Despite scepticism about the benefits of HAART, resort to alternative therapies did not undermine adherence with antiretroviral therapy. Although able actively to cope with their illness, users of alternative therapies had greater psychological distress and were less satisfied with their emotional support. Interventions aimed at promoting their psychological well-being and enhancing the emotional support should be considered in these patients.