PURPOSE: The purpose of this study was to evaluate the incidence and causes of chronic diarrhea in patients with AIDS over a period of time that included the pre-HAART (highly active antiretroviral therapy) era and the introduction of HAART. METHODS: The study cohort was comprised of patients receiving primary care at a university-associated outpatient HIV clinic from January 1, 1995 to December 31, 1997. Patients were identified retrospectively through a clinical database and were included in the study if their diarrhea had persisted for longer than two weeks and their CD4 cell count at time of symptoms was <200 cells/mm3. Further data were obtained by chart review. RESULTS: Over the 36-month period, the occurrence of chronic diarrhea did not change significantly, ranging from 8 to 10.5% per year in patients with CD4 cell counts <200 cells/mm3. The percentage of patients diagnosed with opportunistic infectious etiologies decreased over the three-year period from 53% (1995) to 13% (1997). The percentage of patients diagnosed with noninfectious causes increased from 32% to 70% over this same time period. CONCLUSIONS: Over the three years of the study, the incidence of chronic diarrhea in AIDS patients in our clinic did not change. The etiologies of diarrhea did change significantly, with an increased incidence of noninfectious causes and a decreased incidence of opportunistic infectious causes. This shift in etiologies coincides with the introduction and increased use of HAART in our clinic population (1996).