Thursday, February 8, 2018
Prevalence contributes to Positive Predictive Value
Positive predictive Value, sensitivity and prevalence are related. This can be seen in the app:
Prevalence vs PPV app
From Bayes Rule:
PPV = P(D|+) = P(+|D) P(D) = P(+|D) P(D)
P(+) P(+|D) P(D) + P(+|~D) P(~D)
P(D|+) = Positive Predictive Value (PPV)
P(D) = Prevalence
P(+|D) = Sensitivity
P(+|~D) = 1- Specificity
P(~D) = 1-Prevalence
This is the relationship among PPV, sensitivity and prevalence illustrated in the app.
Subscribe to:
Posts (Atom)