Sensitivity and Specificity of Lung Cancer Screening in Osaka, Japan

Sensitivity and specificity were evaluated for lung cancer screening conducted at 8 municipalities in Osaka Prefecture during 1981–1985. As a screening policy, all attendants were examined by miniature chest X‐ray, and the high‐risk group, defined as those who smoked cigarettes or had bloody sputum, were also examined by 3‐day pooled sputum cytology. A total of 33,599 screening tests for 19,028 people who were 40 years old or more at the time of screening were conducted, resulting in 33,490 miniature chest X‐ray examinations for 18,992 people and 11,420 sputum cytologies for 7,070 people. As a result, 43 lung cancer cases were detected. All test‐negatives were followed by means of record linkage with the files of the Osaka Cancer Registry up to the end of 1986. There were 24 cases who were diagnosed as having lung cancer without having given a positive screening result in 1981‐1986. Assuming the preclinical detectable phase of lung cancer to be one year uniformly, the sensitivity and specificity for the lung cancer screening were estimated to be 71.6% and 95.3%, respectively. The feasibility of increasing the sensitivity is discussed.