Predictive values (PPV, NPV)

The positive predictive value (PPV) is the share of those testing positive who are truly positive; the negative predictive value (NPV) is the share of those testing negative who are truly negative. Unlike sensitivity and specificity they depend on how common the condition is in the group. Likelihood ratios (LR+ = sensitivity / (1 − specificity), LR− = (1 − sensitivity) / specificity) do not.

Example: PPV = 79%, NPV = 95% at a prevalence of 33%: of ten flagged children about eight are truly affected.

Caution: The values only hold for groups with a prevalence similar to the sample. If affected and unaffected people were sought separately (case-control study), PPV and NPV cannot be interpreted.