Thread: Weighting adjusted rates

1. Weighting adjusted rates

Hi All,

I have a patient database with columns for observed and expected values for a number of outcomes. The observed values are binary, expected values came from predicting probabilities after carrying out logistic regression. Now, to calculate a risk-adjusted rate for an outcome I can do :

sum(Observed)/sum(Expected)

However, I know that each patient has spent a proportion of their time (total hospital stay) in one or several of 10 wards. I have columns for prop_wardA...prop_wardJ to tell me the proportions of time spent in each ward.

Is there a way I can use this information, along with the above equation, to get a weighted adjusted rate for each ward, reflecting the time patients spent on each ward?

Thanks for your help

2. Re: Weighting adjusted rates

I was thinking something like:

sum(observed)/sum(expected) x sum(prop_wardA)/sum(LOS)

where LOS is total length of stay. Does this work when you sum over all patients though?

 Tweet

Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts