CT screening for lung cancer: value of expert review of initial baseline screenings

• Published in: American journal of roentgenology (1976) Vol. 204; no. 2; pp. 281 - 286
• Main Authors: Xu, Dong Ming, Lee, In Jae, Zhao, Shijun, Rowena, Yip, Farooqi, Ali, Cheung, Edson H, Connery, Cliff P, Frumiento, Carmine, Glassberg, Robert M, Herzog, Gary, Peeke, Jeffrey, Scheinberg, Paul, Shah, Palmi, Taylor, Jana, Welch, Laura, Widmann, Mark, Yoder, Mark, Yankelevitz, David F, Henschke, Claudia I
• Format: Journal Article
• Language: English
• Published: United States 01.02.2015
• Subjects: Aged; Early Detection of Cancer - methods; Early Detection of Cancer - standards; Female; Humans; Lung Neoplasms - diagnostic imaging; Male; Middle Aged; Quality Control; Tomography, X-Ray Computed; lung cancer; specialized radiology training; CT screening; positive result; quality assurance
• ISSN: 0361-803X; 1546-3141
• DOI: 10.2214/AJR.14.12526

OBJECTIVE. Appropriate radiologic interpretation of screening CT can minimize unnecessary workup and intervention. This is particularly challenging in the baseline round. We report on the quality assurance process we developed for the International Early Lung Cancer Action Program. MATERIALS AND METHODS. After initial training at the coordinating center, radiologists at 10 participating institutions and at the center independently interpreted the first 100 baseline screenings. The radiologist at the institutions had access to the center interpretations before issuing the final reports. After the first 100 screenings, the interpretations were jointly discussed. This report summarizes the results of the initial 100 dual interpretations at the 10 institutions. RESULTS. The final institution interpretations agreed with the center in 895 of the 1000 interpretations. Compared with the center, the frequency of positive results was higher at eight of the 10 institutions. The most frequent reason of discrepant interpretations was not following the protocol (n = 55) and the least frequent was not identifying a nodule (n = 3). CONCLUSION. The quality assurance process helped focus educational programs and provided an excellent vehicle for review of the protocol with participating physicians. It also suggests that the rate of positive results can be reduced by such measures.