Complication Rate comparison

#1
I'm trying to compared the rate of a complication occurring in two different types of treatment.

I have full set of demographic data for the patients and wondering what tests I should be running to be able to form a complete discussion following my results.

Edit: I think I'll add more detail about what I'm doing

I'm retrospectively looking at patients who've had two types of intervention. I want to look at the rate of a complication occurring between the two to assess which operation is better.

In addition, patients were given two questionnaires (using LIKERT scale) before the operation and after the operation, I also want to use this in my evaluation of which treatment is better.
 
Last edited:

hlsmith

Omega Contributor
#2
A basic comparison of the interventions can be done using chi-square or Fisher's exact test.

If you want to examine the other variables' influence on intervention complication, I would use multiple logistic regression.


You can generate odds ratios with both approaches, and suggested recommendations are based on complication being a binary variable (e.g., complication: yes/no).
 
#3
Thanks for your reply hslmith


Are those all the statistical tests you would recommend for the intervention? I just want to make sure I exhaust everything in my write up.

Also, with regards to the questionnaire. Would you recommend an analysis of each section of the questionnaire or the questionnaire overall score?