Military Special Forces trainees undergo intense psychological and physical stressors that often lead to respiratory infection. During 1998-2000, 477 Navy Special Forces trainees were enrolled in a double-blind trial of oral azithromycin (1 g given weekly) plus a placebo injection, compared with benzathine penicillin G (1.2 million U) plus azithromycin placebo tablets. Among the 464 subjects with complete data, 44 developed acute respiratory infection (20 with pneumonia) during the 2 weeks of most intense training; of these subjects, 12 (27.3%) had evidence of Chlamydia pneumoniae infection and 7 (15.9%) had evidence of Mycoplasma pneumoniae infection. Trainees who received azithromycin were less likely than were trainees who received benzathine penicillin G to develop acute respiratory infection (risk ratio, 0.50; 95% confidence interval [CI], 0.28-0.92) and less likely at the end of training to report episodes of breathing difficulty (odds ratio [OR], 0.59; 95% CI, 0.34-1.01) or sore throat (OR, 0.66; 95% CI, 0.41-1.05). Compared with benzathine penicillin G prophylaxis, weekly oral azithromycin was superior in preventing respiratory infection in this population at transient high risk.