Jane V Carter1, Jianmin Pan2, Shesh N Rai2, Susan Galandiuk3. 1. Price Institute of Surgical Research, Hiram C. Polk Jr MD Department of Surgery, University of Louisville, Louisville, KY. 2. Department of Bioinformatics and Biostatistics, School of Public Health and Information Sciences, University of Louisville, Louisville, KY. 3. Price Institute of Surgical Research, Hiram C. Polk Jr MD Department of Surgery, University of Louisville, Louisville, KY. Electronic address: s0gala01@louisville.edu.
Abstract
BACKGROUND: It is vital for clinicians to understand and interpret correctly medical statistics as used in clinical studies. In this review, we address current issues and focus on delivering a simple, yet comprehensive, explanation of common research methodology involving receiver operating characteristic (ROC) curves. ROC curves are used most commonly in medicine as a means of evaluating diagnostic tests. METHODS: Sample data from a plasma test for the diagnosis of colorectal cancer were used to generate a prediction model. These are actual, unpublished data that have been used to describe the calculation of sensitivity, specificity, positive predictive and negative predictive values, and accuracy. The ROC curves were generated to determine the accuracy of this plasma test. These curves are generated by plotting the sensitivity (true-positive rate) on the y axis and 1 - specificity (false-positive rate) on the x axis. RESULTS: Curves that approach closest to the coordinate (x = 0, y = 1) are more highly predictive, whereas ROC curves that lie close to the line of equality indicate that the result is no better than that obtained by chance. The optimum sensitivity and specificity can be determined from the graph as the point where the minimum distance line crosses the ROC curve. This point corresponds to the Youden index (J), a function of sensitivity and specificity used commonly to rate diagnostic tests. The area under the curve is used to quantify the overall ability of a test to discriminate between 2 outcomes. CONCLUSION: By following these simple guidelines, interpretation of ROC curves will be less difficult and they can then be interpreted more reliably when writing, reviewing, or analyzing scientific papers.
BACKGROUND: It is vital for clinicians to understand and interpret correctly medical statistics as used in clinical studies. In this review, we address current issues and focus on delivering a simple, yet comprehensive, explanation of common research methodology involving receiver operating characteristic (ROC) curves. ROC curves are used most commonly in medicine as a means of evaluating diagnostic tests. METHODS: Sample data from a plasma test for the diagnosis of colorectal cancer were used to generate a prediction model. These are actual, unpublished data that have been used to describe the calculation of sensitivity, specificity, positive predictive and negative predictive values, and accuracy. The ROC curves were generated to determine the accuracy of this plasma test. These curves are generated by plotting the sensitivity (true-positive rate) on the y axis and 1 - specificity (false-positive rate) on the x axis. RESULTS: Curves that approach closest to the coordinate (x = 0, y = 1) are more highly predictive, whereas ROC curves that lie close to the line of equality indicate that the result is no better than that obtained by chance. The optimum sensitivity and specificity can be determined from the graph as the point where the minimum distance line crosses the ROC curve. This point corresponds to the Youden index (J), a function of sensitivity and specificity used commonly to rate diagnostic tests. The area under the curve is used to quantify the overall ability of a test to discriminate between 2 outcomes. CONCLUSION: By following these simple guidelines, interpretation of ROC curves will be less difficult and they can then be interpreted more reliably when writing, reviewing, or analyzing scientific papers.
Authors: Tyler J Loftus; Elisha M Collins; Camille G Dessaigne; Amber N Himmler; Alicia M Mohr; Ryan M Thomas; Charles E Hobson; George A Sarosi; William J Zingarelli Journal: Surg Endosc Date: 2017-04-13 Impact factor: 4.584
Authors: Joshua I Rosenbloom; Candice L Woolfolk; Leping Wan; Molly J Stout; Methodius G Tuuli; George A Macones; Alison G Cahill Journal: Am J Obstet Gynecol Date: 2019-05-30 Impact factor: 8.661
Authors: Jane V Carter; Henry L Roberts; Jianmin Pan; Jonathan D Rice; James F Burton; Norman J Galbraith; Maurice R Eichenberger; Jeffery Jorden; Peter Deveaux; Russell Farmer; Anna Williford; Ziad Kanaan; Shesh N Rai; Susan Galandiuk Journal: Ann Surg Date: 2016-10 Impact factor: 12.969
Authors: David A Stevens; Yonglong Zhang; Malcolm A Finkelman; Demosthenes Pappagianis; Karl V Clemons; Marife Martinez Journal: J Clin Microbiol Date: 2016-08-24 Impact factor: 5.948
Authors: Juan C Rojas; Patrick G Lyons; Teresa Jiang; Megha Kilaru; Leslie McCauley; Jamila Picart; Kyle A Carey; Dana P Edelson; Vineet M Arora; Matthew M Churpek Journal: Ann Am Thorac Soc Date: 2020-07
Authors: Sladjana Lukic; Maria Luisa Mandelli; Ariane Welch; Kesshi Jordan; Wendy Shwe; John Neuhaus; Zachary Miller; H Isabel Hubbard; Maya Henry; Bruce L Miller; Nina F Dronkers; Maria Luisa Gorno-Tempini Journal: Brain Lang Date: 2019-05-02 Impact factor: 2.381