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.         



No comments:

Post a Comment