A hospital's quality improvement committee is reviewing the performance characteristics of various diagnostic assays under different clinical and epidemiological scenarios. Match each described scenario (left) with its expected impact on the assay's predictive values or error rates (right). Assume all other variables remain constant unless otherwise specified.
- The screening assay is moved from a general population setting to a high-risk infectious disease ward, significantly increasing the pretest probability.Increases the positive predictive value and decreases the negative predictive value.
- The diagnostic threshold is decreased so that a lower concentration of the biomarker is sufficient for a positive result.Increases test sensitivity and the false positive rate, while decreasing the false negative rate.
- The diagnostic threshold is increased to strictly minimize the number of healthy individuals being unnecessarily subjected to invasive follow-up testing.Increases test specificity and the false negative rate, while decreasing the false positive rate.
- The assay is deployed in a region where a recent mass prophylactic campaign has drastically reduced the incidence and prevalence of the infection.Decreases the positive predictive value and increases the negative predictive value.
Answer
Scenario 1 matches with increased PPV and decreased NPV. Scenario 2 matches with increased sensitivity and false positive rate. Scenario 3 matches with increased specificity and false negative rate. Scenario 4 matches with decreased PPV and increased NPV.
Prevalence changes directly impact predictive values: higher prevalence leads to a higher PPV and lower NPV, while lower prevalence leads to a lower PPV and higher NPV. Cutoff threshold changes directly impact intrinsic test properties: a lower cutoff increases sensitivity and the false positive rate, whereas a higher cutoff increases specificity and the false negative rate.
Step-by-Step Solution
Key Concept
The relationship between disease prevalence, predictive values (PPV/NPV), and how diagnostic cutoff threshold adjustments alter sensitivity, specificity, and error rates.