A study testing how well a test correctly identifies who does or doesn't have a condition.
This design compares the results of an index test (e.g., a clinical exam, imaging finding, or screening tool) against a reference standard to determine measures such as sensitivity, specificity, and predictive values. Quality depends heavily on whether the reference standard is appropriate and applied blind to the index test result, and whether the sample reflects the population in which the test will actually be used. Poorly designed diagnostic accuracy studies (e.g., spectrum bias, verification bias) can substantially overstate a test's real-world usefulness.
This guide was auto-drafted and is pending editorial review.