Key points are not available for this paper at this time.
BACKGROUND: Infection with human immunodeficiency virus (HIV) increases the risk of tuberculosis (TB), but no study has assessed how this risk changes with time since HIV seroconversion. METHODS: The incidence of pulmonary TB was estimated in miners with and those without HIV infection in a retrospective cohort study. HIV test results were linked to routinely collected TB, demographic, and occupational data. The rate ratio (RR) for the association between HIV status and TB was estimated by time since HIV seroconversion, calendar period, and age. RESULTS: Of the 23, 874 miners in the cohort, 17, 766 were HIV negative on entry, 3371 were HIV positive on entry, and 2737 seroconverted during follow-up (1962 had a seroconversion interval of < or =2 years). A total of 740 cases of TB were analyzed. The incidence of TB increased with time since seroconversion, calendar period, and age. TB incidence was 2. 90 cases/100 person-years at risk (pyar) in HIV-positive miners and was 0. 80 cases/100 pyar in HIV-negative miners (adjusted RR, 2. 9 95% confidence interval CI, 2. 5-3. 4). TB incidence doubled within the first year of HIV infection (adjusted RR, 2. 1 95% CI, 1. 4-3. 1), with a further slight increase in HIV-positive miners for longer periods, up to 7 years. CONCLUSION: The increase in the risk of TB so soon after infection with HIV was unexpected. Current predictive models of TB incidence underestimate the effect of HIV infection in areas where TB is endemic.
Sonnenberg et al. (Tue,) studied this question.