HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Beam, C. A. Articles by Weinstein, S. P. Search for Related Content PubMed PubMed Citation Articles by Beam, C. A. Articles by Weinstein, S. P.

Evaluation of Proscriptive Health Care Policy Implementation in Screening Mammography¹

¹ From the H. Lee Moffitt Cancer Center & Research Institute, 12902 Magnolia Dr, Tampa, FL 33612-9497. From the 2002 RSNA scientific assembly. Received December 4, 2002; revision requested February 6, 2003; final revision received May 13; accepted May 19. Supported by National Cancer Institute grant CA-74110. Address correspondence to C.A.B. (e-mail: beamca@moffitt.usf.edu).

View larger version (14K): [in a new window]   Figure 1a. (a) Graph shows quantiles of accuracy in U.S. radiologists (as extrapolated from data in the study sample). Median is 0.50 quantile. By following the horizontal line to the right, where it intersects the graph, and then down to where it intersects the horizontal axis, we estimate median accuracy (measured as partial ROC area of the reader) in U.S. radiologists to be approximately 0.66. (b) Graph shows implications of health care policy goal of increasing median accuracy by 10%. To move the population median from 0.66 to 0.76 (an amount indicated by the right-pointing arrow under x axis), we follow the arrows upward to the distribution and then to the left to determine that the target value of accuracy of 0.76 is approximately the 0.75 quantile in U.S. radiologists. That is, about 75% of U.S. radiologists have an accuracy (partial ROC area) less than or equal to 0.76. Thus, to achieve an increase in median accuracy of 10%, we must shift median upward from 0.50 to 0.75 (ie, across about 25% of the population). The fact that to accomplish this shift requires restricting 50% of the currently practicing U.S. radiologists from interpreting screening mammograms is outlined in the Appendix.

View larger version (16K): [in a new window]   Figure 1b. (a) Graph shows quantiles of accuracy in U.S. radiologists (as extrapolated from data in the study sample). Median is 0.50 quantile. By following the horizontal line to the right, where it intersects the graph, and then down to where it intersects the horizontal axis, we estimate median accuracy (measured as partial ROC area of the reader) in U.S. radiologists to be approximately 0.66. (b) Graph shows implications of health care policy goal of increasing median accuracy by 10%. To move the population median from 0.66 to 0.76 (an amount indicated by the right-pointing arrow under x axis), we follow the arrows upward to the distribution and then to the left to determine that the target value of accuracy of 0.76 is approximately the 0.75 quantile in U.S. radiologists. That is, about 75% of U.S. radiologists have an accuracy (partial ROC area) less than or equal to 0.76. Thus, to achieve an increase in median accuracy of 10%, we must shift median upward from 0.50 to 0.75 (ie, across about 25% of the population). The fact that to accomplish this shift requires restricting 50% of the currently practicing U.S. radiologists from interpreting screening mammograms is outlined in the Appendix.

View larger version (19K): [in a new window]   Figure 2. Implications of proscriptive health care policy.

View larger version (17K): [in a new window]   Figure A1. Graph depicts concepts and defines terminology involved in determining percentage of population that must be restricted to achieve health care policy goals expressed in terms of median accuracy.