Specificity

The proportion of individuals who truly do not have a condition who are correctly identified as negative by a diagnostic test.

Source: Altman & Bland 1994

Specificity is the proportion of people who truly do not have a condition that a test correctly identifies as negative. It measures how well a test clears the healthy, so a highly specific test rarely labels a well person as diseased and produces few false positives. This makes specificity important for confirming disease, since a positive result from a very specific test strongly suggests the condition is genuinely present. How well a test spares the healthy is what it captures. Sackett and colleagues (1991) describe this measure.

Source: Sackett et al. 1991

Specificity is calculated as the number of true negatives divided by the total number of people who truly do not have the condition, that is true negatives plus false positives. It requires a reference standard to establish who genuinely lacks the condition. So specificity is calculated from the proportion of those without the condition who test negative, giving the true negative rate, which is a property of the test and does not depend on the prevalence of the condition, distinguishing it from predictive values, and it is estimated by comparing test results against a definitive reference standard in people known not to have the condition.

Source: Altman & Bland 1994

High specificity means a test correctly identifies most people who do not have the condition, producing few false positives, so a positive result from a highly specific test is fairly reliable for confirming the condition. This is why highly specific tests are valued when a false positive would be harmful or costly. So high specificity indicates that the test rarely flags the condition when it is absent, which makes a positive result useful for ruling disease in, though a highly specific test may still miss cases, so specificity is considered alongside sensitivity to judge overall performance.

Source: Sackett et al. 1991

Specificity is the proportion of those without the condition who test negative, the true negative rate, while sensitivity is the proportion of those with the condition who test positive, the true positive rate. Specificity concerns correctly excluding the condition when absent, and sensitivity detecting it when present. There is often a trade-off, since increasing specificity can lower sensitivity. So specificity and sensitivity are complementary measures of a test's performance, one capturing how well it rules the condition out and the other how well it finds it, and both are needed to characterise a diagnostic test fully.

Source: Altman & Bland 1994

Specificity is important because it shows how well a test correctly excludes a condition when it is absent, which matters where false positives would lead to harm, anxiety, or unnecessary further testing and treatment. A highly specific test gives confidence that a positive result is unlikely to be a false alarm. So specificity matters for judging a test's ability to avoid mislabelling healthy people as diseased, guiding its use where false positives are costly, and because it is a property of the test independent of prevalence, it characterises the test consistently across settings, complementing sensitivity and the prevalence-dependent predictive values.

Source: Altman & Bland 1994