7. Measures of Discrimination of Diagnostic Tests

  1. Matthew Thompson1,2 and
  2. Ann Van den Bruel2

Published Online: 1 NOV 2011

DOI: 10.1002/9781119951827.ch7

Diagnostic Tests Toolkit

Diagnostic Tests Toolkit

How to Cite

Thompson, M. and Van den Bruel, A. (2011) Measures of Discrimination of Diagnostic Tests, in Diagnostic Tests Toolkit, Wiley-Blackwell, Oxford, UK. doi: 10.1002/9781119951827.ch7

Author Information

  1. 1

    Department of Family Medicine, Oregon Health & Science University, Portland, Oregon, USA

  2. 2

    Department of Primary Care Health Sciences, University of Oxford, Oxford, UK

Publication History

  1. Published Online: 1 NOV 2011
  2. Published Print: 7 OCT 2011

Book Series:

Book Series Editors:

  1. Carl Heneghan,
  2. Rafael Perera and
  3. Douglas Badenoch

ISBN Information

Print ISBN: 9780470657584

Online ISBN: 9781119951827



  • diagnostic tests, measures discriminated;
  • diagnostic accuracy results;
  • two by two tables;
  • outcome measures;
  • diagnostic odds ratio (DOR);
  • sensitivity and specificity;
  • SpPin–SnNout rule of thumb;
  • predictive values and prevalence;
  • effect of time;
  • CAGE questionnaire, alcohol abuse


This chapter contains sections titled:

  • Two by two tables

  • Outcome measures

  • Which measure should you use and why?

  • The effect of prevalence on predictive values

  • Effect of time on diagnostic accuracy

  • Effect of severity of the target condition on diagnostic accuracy

  • The referral filter

  • Rules of thumb for interpreting likelihood ratios

  • Further reading