Course Level: Intermediate
Format Requirements: The workshop will consist of brief didactic presentations followed by small groups solving and discussing illustrative problems based on studies of diagnostic tests. Among other tests, the problems will discuss: a rapid antigen detection test for influenza; BNP for congestive heart failure; ultra-sensitive Troponin I for acute coronary syndrome; B-hCG for ectopic pregnancy, and the JAMA “Rational Clinical Examination” papers on diagnosing appendicitis and UTI in children. At the end of the session, we will hand out the answers to the problems. Participants should be comfortable with the basics of diagnostic testing such as the definitions of sensitivity, specificity and predictive value, and should also have some initial exposure to receiver operating characteristic (ROC) curves and likelihood ratios.
Background: This hands-on workshop is based on real examples from the medical literature. We will review multiple studies of diagnostic tests and show how the data can be (but often are not) presented to maximize the information to be gained from the test. We will also discuss various common but under-recognized biases and how they affect results. After this course, you will understand • the principles of diagnostic test assessment, including the calculation and use of interval likelihood ratios and the relationship between these likelihood ratios and the ROC curve • how to go beyond the area under the curve to get the most out of published ROC curves • when odds ratios are preferable to likelihood ratios and vice versa • underappreciated flaws, biases, and limitations in studies of diagnostic tests • how to minimize or resolve limitations and best present diagnostic test information to experienced clinicians.
Description and Objectives: The brief didactic sessions will cover the following topics (examples):
1) Review of dichotomous tests, sensitivity, specificity, LR(+), LR(-), and the false negative rate confusion (rapid antigen testing for H1N1 influenza)
2) Multilevel and continuous tests, interval likelihood ratios and the perils of making multi-level tests dichotomous (B-type natriuretic peptide for congestive heart failure)
3) Going beyond the area under the ROC curve and getting the most out of published ROC curves (peripheral WBC count for meningitis in infants; ultrasensitive troponin I for acute coronary syndrome)
4) Studies of diagnostic test accuracy -- beyond the check list: incorporation and spectrum biases (BNP revisited; using B-HCG levels in the diagnosis of ectopic pregnancy
5) Studies of diagnostic test accuracy: verification and double gold standard biases (JAMA Rational Clinical Examination: Does this Child Have Appendicitis?)
6) Odds ratios vs. likelihood ratios: Which to use? (JAMA Rational Clinical Examination: Does this Child Have a Urinary Tract Infection?)
We can adjust the schedule and material covered based on input from the participants at the beginning of the course.