Publication Abstract

Authors: Leung JW, Margolin FR, Dee KE, Jacobs RP, Denny SR, Schrumpf JD

Title: Performance parameters for screening and diagnostic mammography in a community practice: are there differences between specialists and general radiologists?

Journal: AJR Am J Roentgenol 188(1):236-41

Date: 2007 Jan

Abstract: OBJECTIVE: The purpose of this study was to compare the performance of general radiologists in interpretation of mammograms with that of breast imaging specialists in a high-volume community hospital-based private practice. MATERIALS AND METHODS: A retrospective observational study was conducted with data prospectively collected over a 5-year period in a community hospital-based practice in which 106,405 screening and 52,149 diagnostic mammograms were performed. The performance of three radiologists specializing in breast imaging was compared with that of six general radiologists. The following data were extracted and analyzed: recall rate, biopsy recommendation rate, and cancer detection rate. Statistical analysis was performed with a chi-square test and two-tailed calculation of p values. RESULTS: The recall rates of the specialists and generalists were nearly the same at 6.5% and 6.7%, respectively. The biopsy recommendation rate at recall from screening examinations was nearly the same for generalists and specialists (1.2% and 1.1%, respectively; p = 0.4504). This rate also was similar for diagnostic examinations (8.5% for generalists; 8.4% for specialists; p = 0.4086). The cancer detection rate in the screening setting was slightly higher for specialists than for generalists: 2.5 and 2.0 cancers per 1,000 cases, respectively (p = 0.0614). The cancer detection rate in the diagnostic setting was 24.2% higher among specialists (20.0 cancers per 1,000 cases) compared with generalists (16.1 cancers per 1,000 cases) (p = 0.0177). CONCLUSION: The only statistically significant difference between generalists and specialists was in cancer detection rate among patients undergoing diagnostic mammography. No statistically significant difference was identified between the two groups in terms of recall rate, biopsy recommendation rate, or percentage of favorable-prognosis cases of cancer detected. There was a trend toward greater cancer detection by specialists in the screening setting.