Slide 1: What do you know about diagnostic odds ratios and how they are used?
Slide 2: Most measures of diagnostic tests are dichotomous (sensitivity and specificity, positive predictive value and negative predictive value, positive likelihood ratio and negative likelihood ratio). These measures provide a lot of information and can help you understand the utility of a particular test result in a particular patient, but each measure by itself does not allow different tests to be compared to each other.
Slide 3: A diagnostic odds ratio (DOR) provides a single measure of test performance and can be calculated in several ways: [A][LR+/LR-], [B][(true positive/false positive)/(false negative/true negative)], [C][((PPV/(1-PPV))/((1-NPV)/NPV)], [D][(sens/(1-sens))/((1-spec)/spec)]. Yes, these are all mathematically equivalent!
Slide 4: How good is your diagnostic test– interpreting a DOR: Ranges from 0 to infinity. The higher the number, the better the test is at discriminating patients who have the disease from patients who don’t. DOR of 1 = the test is unable to discriminate between patients who do and do not have the disease. DOR < 1 = improper test interpretation (a negative result is more commonly attributed to diseased patients, whereas a positive result is more commonly attributed to nondiseased persons). Slide 5: What's the point? A DOR is useful to compare tests to each other, for example, for the purposes of conducting a meta-analysis. However, when interpreting the results of a test with your patient, you still need to consider the actual test result, the test's properties, and the local disease prevalence.
- Glas AS, Lijmer JG, Prins MH, Bonsel GJ, Bossuyt PM. The diagnostic odds ratio: a single indicator of test performance. J Clin Epidemiol. 2003 Nov;56(11):1129-35.
Tags: diagnostic odds ratio