To investigate the reason of high incidence of annual patients with tuberculosis (TB) in a military school previously known by screening tuberculin skin test (TST) and finding out the proportion of annual infection risk (PAIR), the prevalance of TB infection and the distribution for each grades. Our study is a cross-sectional epidemiologic study made about TB infection. TST were screened for all students in the school. 5 TU PPD was injected to every student and after 72 hours, the results were evaluated by measuring the diameter of enduration. Test was repeated after 10 days for negative reactions. Age, sex, the number of BCG wound, smoking and dwelling for last 5 years were asked from the students and their answers were recorded. More than 10 mm enduration for cases who had no BCG and 15 mm enduration for cases who had BCG were accepted positive. Chest roentgenogram was taken for each student enrolled into the study. Infection prevalance and PAIR were calculated after tests and measurements. The total number of students was 948. Of 917 (96.7%) were male and 31 (3.3%) were female. The mean age was 19.72 +/- 1.25. The mean of TST was 12.79 +/- 5.96 mm for all students. According to the number of BCG scar, the numbers of students, percentage and the mean of TST were like that 70 (7.3%) cases no BCG scar 8.41 +/- 7.87 mm, 393 (41.4%) students one BCG scar, 11.94 +/- 6.26 mm, 343 (36.1%) cases two BCG scars, 13.74 +/- 5.12 mm, 142 (14.9%) students three or more then three scars, 14.97 +/- 4.11 mm. In the students who had no BCG, TST positivity was 50%. TB infection prevalance of entire school and PAIR were 46% and 3.44% (respectively). In this study, we found that increased number of BCG wound associated with the increased diameter of TST enduration. The proportion of unvaccinated students was similar to the same age population in our country but it showed differences in the distribution of regions. The students who started first grade had serious TB infection risk in their first school year. We think that PAIR values derived from TST conversions done in high risky community by screening annual TST could show all aspects of TB infection risk in those community.