Specificity is one of the 4 basic diagnostic test metrics in addition to sensitivity, positive predictive value and negative predictive value. Specificity is a measure of how good a diagnostic test is at identifying people who are healthy and is calculated by dividing the number of true negatives (TN) by the number of people without disease, i.e. true negatives and false positives (FP):
TN/(TN + FP)
The formula shows that a high specificity is achieved by maximizing true negatives and minimizing false positives. Because highly specific tests have few false positives, a specific test when positive may be used to “rule in” disease since a positive result is likely to be a true positive.
Specificity is also called the true negative rate and can be expressed as a conditional probability:
P(Test negative|Disease negative)
In an ROC curve, 1-specificity is plotted along the x-axis and can be renamed the false positive rate, the opposite of the true negative rate.