OBJECTIVES: This retrospective study aimed to assess the risk of tuberculosis infection for the employees of a Japanese hospital using baseline interferon-gamma release assay (IGRA). The risk was defined as exposure to the hospital environment. METHODS: In total, 870 hospital employees including 161 new employees, 582 for baseline assay, and 127 for contact examination (709 subjects in the post-employment group) were examined from December 2010 to April 2012. The new employees were considered as the "non-exposure" group, whereas the post-employment group was considered as the "exposure" group. Multiple logistic regression analyses were used to calculate the odds ratio (OR) for IGRA positivity, adjusted for gender, smoking history, and alcohol intake (model 1), and for years of employment (model 2). RESULTS: The exposure group was significantly associated with an increased risk of positive IGRA results, even when adjusted for years of employment (OR: 4.1; 95% confidence interval: 1.4-17.6; P = 0.007). Subgroup analyses stratified by profession indicated a significantly increased OR for laboratory technicians, doctors, and nurses in both models. No correlation was observed between the length of employment and IGRA positivity. CONCLUSION: Exposure to the hospital environment increased the risk of tuberculosis infection for employees irrespective of the length of employment. Laboratory technicians, doctors, and nurses were at the highest risk of infection.