September 26, 2023

Normalizing Predictive Values for Appropriate Interpretation of Diagnostic Test Results

As clinicians, we often evaluate diagnostic test results like imaging studies to guide patient care. An important consideration is how disease prevalence in the tested population impacts the meaning of positive and negative predictive values. For example, a recent study using cardiac CT angiography (CTA) to diagnose coronary artery disease reported an excellent 99% negative predictive value. However, disease prevalence was only 5% in the study population. When predictive values were normalized to a more average 50% prevalence, the negative predictive value dropped to 80% – still useful, but not as definitively reliable to rule out disease as initially thought. Carefully attending to the prevalence and normalizing predictive values is crucial for appropriately interpreting test results.  In this 2009 study, a positive CTA was highly meaningful for true disease when prevalence is average, but clinicians should apply caution relying solely on negative results to confidently exclude important diagnoses, regardless of initially reported predictive values. Thoughtful normalization allows diagnostic test results to be meaningfully applied across diverse patient populations.

Heston TF (2009). Detection of myocardial infarction by CT angiography. Heart 2009;95:1108. http://dx.doi.org.offcampus.lib.washington.edu/10.1136/hrt.2009.169508