A study of variability in the interpretation of chest x-rays in the detection of lung cancer.

As part of the American Cancer Society-VA Cooperative Pilot Study for the evaluation of radiologie and sputum cytologie screening in the early detection of lung cancer, it was possible to obtain some estimate of the variability in the interpretation of chest X-rays by radiologists. With respect to interindividual variability, the positive readings of one radiologist will be called positive by another radiologist in about 50-60% of instances and the negative readings in about 90-95% of instances. Similar degrees of agreement were found when a radiologist read a series of X-rays twice. This degree of variability is not unexpected and indicates the need for having 2 radiologists read X-rays independ ently in surveys in order to maximize the yield of suspect cases.