RESUMO
SETTING: University hospital and university campus in Lima, Peru. OBJECTIVE: To demonstrate the risk of latent tuberculosis infection (LTBI) in the hospital relative to the community. DESIGN: Prospective cohort study of university students measuring prevalence, boosting, and conversion of tuberculin skin tests (TSTs) among health care students (HCS) and non-health care students (NHCS). RESULTS: Among the HCS relative to NHCS, prevalence of initial positive TST was 20.9% vs. 12.2% (P < 0.001), and conversion rate was 1.1% vs. 0% (P = 0.423) at the 10 mm cut-off and 11.8% vs. 0% at the 6 mm cut-off (P = 0.00005). Multivariate analysis showed that the HCS group had a higher risk of baseline positive TST compared with the NHCS group after controlling for confounding factors (OR 1.7, 95% CI 1.1-2.6). CONCLUSION: HCS are at greater risk than NHCS for having positive baseline TSTs and for TST conversion at the 6 mm cut-off. We conclude that the hospital we studied in Lima, Peru, poses a greater risk than the surrounding community for tuberculosis infection, and greater attention to hospital infection control measures is warranted. A higher rate of skin test boosting among the HCS cohort suggests the possibility of transient, non-progressive LTBI, which merits further study.