Previously, I had a question about CI for Patient-Years (py) and there was a great answer from @TamasFerenci. Now I need to find out the $p$ value for the py of two different groups.

I have a table which contains events per 1,000 py:

                           Group A    Group B
Hospitalization/1,000 py   8.8        38.0
Death/1,000 py             18.0       36.6
...                        ...        ...

There is already a question about significance test for two poisson distributions but I am not sure how this applies on py, because in case of py we have rates and not the original data (observed number of events and observed follow-up time). Can I we estimate whether the two rates are significantly different using poisson.test(x= 38, T= 1000, r= 8.8)$p.value for hospitalization? Or do I need the original data with observed number of events and observed follow-up time?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.