Setting: Undergraduate training on tuberculosis at Istanbul Medical School.
Objective: To assess whether tuberculosis-related questions asked in chest medicine examinations comply with the World Health Organization's (WHO) learning objectives for tuberculosis training, and to investigate students' skills in interpreting radiology and smears with respect to tuberculosis.
Design: Tuberculosis questions set for 4th year medical students (n = 838) were compared with WHO objectives (n = 23): 122 students were each asked to interpret 10 chest X-rays which had been read by three experts as 'typical tuberculosis lesions', 'normal' or 'non-tuberculosis lesions'; 114 students were also each given five smears which had been interpreted by an expert as 'negative' or 'positive' with respect to acid-fast-bacilli (AFB), and were asked to interpret them.
Results: Questions covered only nine of the WHO objectives. Among 117 types of questions, 91 (77.8%) were about clinical-radiological findings, treatment and drugs; 334 (65.1%) of 513 X-rays with typical tuberculosis lesions, 77 (21.1%) of 364 normal X-rays and 87 (25.4%) of 343 X-rays with non-tuberculosis lesions were identified as tuberculosis (kappa 0.45). Of 369 AFB-positive smears, 149 (40.4%) were evaluated as false negative, and of 185 AFB-negative smears, 48 (25.9%) were evaluated as false positive (kappa 0.49).
Conclusion: Examination questions set on tuberculosis at Istanbul Medical School do not adequately reflect WHO learning objectives. Students' skills in interpreting radiology and smears suggest that their practical training on tuberculosis is insufficient.