OBJECTIVES: To determine the long-term incidence of tuberculosis (TB) and associated risk factors among individuals receiving HAART in South Africa. DESIGN: Prospective cohort study. METHODS: Microbiologically or histologically confirmed incident TB was identified in a hospital-based cohort of 346 patients receiving HAART between 1996 and 2005 in Cape Town. RESULTS: The TB incidence density rate was 3.5/100 person-years in the first year and significantly decreased during follow-up, reaching 1.01/100 person-years in the fifth year (P = 0.002 for trend). TB incidence during the study was highest among patients with baseline CD4 cell counts < 100 cells/microl and those with World Health Organization (WHO) clinical stage 3 or 4 disease (5.71 and 3.88/100 person-years, respectively). Risk of TB was independently associated with CD4 cell count < 100 cells/microl (adjusted risk ratio [ARR], 2.38; 95% confidence interval (CI), 1.01-5.60; P = 0.04), WHO stage 3 or 4 disease (ARR, 3.60; 95% CI, 1.32-9.80; P = 0.01) and age < 33 years (ARR, 2.86; 95% CI, 1.29-6.34; P = 0.01). Risk of TB was not independently associated with plasma viral load, previous history of TB, low socioeconomic status or sex. Despite similar virological responses to HAART, blood CD4 cell count increases were much smaller among patients who developed TB than among those who remained free of TB. CONCLUSIONS: Incidence of TB continues to decrease during the first 5 years of HAART and so HAART may contribute more to TB control in low-income countries than was previously estimated from short-term follow-up. Patients with advanced pretreatment immunodeficiency had persistently increased risk of TB during HAART; this may reflect limited capacity for immune restoration among such patients.