From: thomas on
Hello,

I wonder if somebody could help me.

I want to estimate OR with CI for a specific cut-off, whereby cut-off
should not input as a probability (then I could use ctable and pprob-
options in model as shown below) but as a value xi from independent
variable x.

1.Is there a related option etc which I can use within proc logistic?

proc logistic data=data;
model disease=x /ctable pprob=0.4;
run;

2. I know how to calculate OR and CI from classification table, but
does exist an automatic version?

An additional question:

3. Does exits an option to use standardized values of x within
logistic regression (without using proc stdize in advance). Again I
want to estimate OR.

Thank you,
Hans