topic Re: case mix adjusted mortality rate calculation in Discussions

case mix adjusted mortality rate calculation

Sandeep123 — Fri, 09 Jun 2023 00:30:05 GMT

I have a large data set , where I need to compare the mortality rate among two groups. The groups differ in age , diagnostic categories , gender etc.

I can do logistic regression and it gives me odds ratio of death in one group vs other after including the confounding variables..but it does not give me a 'adjusted rate'

How do I calculate and compare the case mix adjusted mortality rates between two groups ?

Re: case mix adjusted mortality rate calculation

dale_lehman — Tue, 09 Mar 2021 13:57:20 GMT

I suspect you will have to provide more information, a concrete example if possible, to get an answer you can use. But I would suggest the following from your brief description: Your Y variable appears to be death/no death. More usually, the Y variable would be the time to death (a continuous variable). If you use a continuous variable, then you can create a column for the column mean time to death - by age, diagnostic category, gender, etc. - and use that as your response variable. It will then be an adjusted time. I suppose if you have your Y variable (death/no death) coded as 0,1, you could do the same thing, and then make it a nominal variable and use the logistic regression.

Re: case mix adjusted mortality rate calculation

Sandeep123 — Wed, 10 Mar 2021 21:40:58 GMT

I am attaching the model dialog box.

I do not have the time to death. Only interested in mortality rate ( yes/no) ..which is coded as "survived' or "died". I converted it to 0, 1 but that does not seem to change the output when I run the model.