Why most prediction models cannot support treatment decisions

[@JCResDoc94]

36:00 the depressing truth. (we are only good at the tails. & even then.) -jc

[@SowaliGonsya]

Add to get the prediction

[@JCResDoc94]

30:00 it is all so close to witchcraft. i know it is the best we can do. & we rarely have to go into a trance or kill a 2nd chicken. but damn. sometimes. all i see are variables, confounds & no way to know where your pt is on the curve anyhow. -jc

[@JCResDoc94]

all models are wrong. thats what we were taught. gr8 entry. -jc