The aim of this study was to evaluate the annual incidence of tuberculosis infection among newly employed doctors and nurses in Korea. The annual incidence of tuberculosis infection ranged from 3.3% to 5.7%, based on the definition of conversion of an interferon-γ release assay, which suggests that stricter preventive strategies against nosocomial TB infection should be employed. Follow-up interferon-γ levels measured after 3 months of isoniazid and rifampicin treatment showed considerable variation. Therefore, serial testing with interferon-γ release assays after treatment of latent TB infection may be insufficient for evaluating the effects of treatment due to the variable responses.