Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) —collectively known as “test charac-teristics” —are important ways to express the usefulness of diagnostic tests. The 2 x 2 tables from which these terms are derived are familiar to some physicians ( Table ).
Sensitivity is the “true positive rate,” equivalent to a/a+c. Specificity is the “true negative rate,” equivalent to d/b+d. PPV is the proportion of people with a positive test result who actually have the disease (a/a+b); NPV is the proportion of those with a negative result who do not have the disease (d/c+d).
Sensitivity and specificity are fixed for a particular type of test. PPV and NPV for a particular type of test depend upon the prevalence of a disease in a population. For example, though current screening tests for HIV have high sensitivity and specificity, the low prevalence of HIV in the general population cannot justify universal screening since the majority of positive tests would be falsely positive (ie, low PPV).
TABLE
2 x 2 table for diagnostic test results
| Disease present (+) | Disease absent (−) | Totals | |
|---|---|---|---|
| Test result positive (+) | a | b | a + b |
| Test result negative (−) | c | d | c + d |
| Totals | a + c | b + d | — |
How to Remember these Terms
Begin by assuming that you have 4 patients. For the first 2 you know only their disease status; for the last 2 patients you know only their test result.
You know your patient’s disease status:
- Sensitivity: “I know my patient has the dis-ease. What is the chance that the test will show that my patient has it?”
- Specificity: “I know my patient doesn’t have the disease. What is the chance that the test will show that my patient doesn’t have it?”
- PPV: “I just got a positive test result back on my patient. What is the chance that my patient actually has the disease?”
- For NPV: “I just got a negative test result back on my patient. What is the chance that my patient actually doesn’t have the disease?”
Keeping these 4 questions in mind as you run across these frequently used terms will help you interpret diagnostic tests accurately and efficiently, without having to think about more awkward 2 x 2 tables.
Correspondence
Goutham Rao, MD, 3518 Fifth Avenue, Pittsburgh, PA 15261. E-mail: raog@msx.upmc.edu.