Hi,
I'm a student at university doing my dissertation project.
I'm looking at what could independently predict readmission to hospital.
We have done a cox regression analysis but the results seem to be the wrong way.
It shows that a factor increases the likelihood of readmission but when looking at the data directly it should be the opposite way round to show a decreased risk.
We have coded the factor so it is binary, and readmission the same, with readmission being 1 and not readmitted being 0.The time course is in months and has been censored.
We have tried swapping the readmitted to 0 and not readmitted to 1 however, this just makes all our variables insignificant.
after exhausting you tube and our lectures knowledge i would really appreciate it if someone could help.
Many Thanks
Ceri
I'm a student at university doing my dissertation project.
I'm looking at what could independently predict readmission to hospital.
We have done a cox regression analysis but the results seem to be the wrong way.
It shows that a factor increases the likelihood of readmission but when looking at the data directly it should be the opposite way round to show a decreased risk.
We have coded the factor so it is binary, and readmission the same, with readmission being 1 and not readmitted being 0.The time course is in months and has been censored.
We have tried swapping the readmitted to 0 and not readmitted to 1 however, this just makes all our variables insignificant.
after exhausting you tube and our lectures knowledge i would really appreciate it if someone could help.
Many Thanks
Ceri