Calculate the Youden Index (J) to evaluate diagnostic test performance. Combines sensitivity and specificity into a single metric independent of disease prevalence.
Last updated: March 2026
The Youden Index (J), named after statistician William Youden, is a single statistic that combines sensitivity and specificity to evaluate the overall performance of a diagnostic test. It's calculated as: J = Sensitivity + Specificity − 1.
The index ranges from −1 (completely inaccurate) to 1 (perfectly accurate). A value near 1 indicates excellent test performance, while values near 0 suggest no diagnostic value. Negative values indicate the test performs worse than chance.
Unlike overall accuracy, the Youden Index doesn't depend on disease prevalence, making it particularly useful when comparing tests across different populations or conditions.
| TP (True Positive) | Disease present, test positive (correct) |
| FN (False Negative) | Disease present, test negative (missed) |
| FP (False Positive) | Disease absent, test positive (false alarm) |
| TN (True Negative) | Disease absent, test negative (correct) |
Scenario: A COVID-19 rapid test was evaluated on 200 patients (100 confirmed positive, 100 negative by PCR).
Results
Interpretation: With a Youden Index of 0.85, this rapid test has good diagnostic value and would be suitable for screening purposes.
Accuracy doesn't distinguish between sensitivity and specificity and is affected by disease prevalence. The Youden Index provides a more balanced evaluation by combining both metrics independently of prevalence.
Sensitivity is the test's ability to correctly identify those with the disease (true positive rate). Specificity is its ability to correctly identify those without the disease (true negative rate).
Yes. A negative J indicates the test performs worse than chance—worse than simply guessing. This means the test is actively misleading.
It's most useful when you want a single number to compare diagnostic tests or when disease prevalence varies across populations, as J is prevalence-independent.
The Youden Index represents the maximum vertical distance from the diagonal (chance line) to the ROC curve. It identifies the optimal operating point on the curve.
It's moderate. Generally, J > 0.5 is considered good. A test with J = 0.5 has some diagnostic value but isn't as reliable as one with J > 0.75.
Related Tools
Evaluate classification models.
Classification metrics.
Diagnostic test accuracy.
Updated diagnostic probability.
Risk ratio comparison.
Event probability.