how to determine what test to use for a project?

#1
I'm trying to determine if there is a statistical significance in hospital admission rates (asthma and bronchitis) before and after tobacco regulations went in to affect. I got the crude and adjusted rates with the 95% Confidence Intervals. Can I just use the 95%CIs to show statistical difference? I've read other studies on the same subject and they've used a regression analysis to determine projected admission rates and compared it to the actual admission rates.

Is the regression model the only statistical analysis model I can use if I want to determine if the admission rates were significantly different before and after the tobacco laws were enacted?

Again, i have the admission rate of asthma and bronchitis diagnoses. I also have crude rates of asthma and bronchitis diagnoses WITH a history of tobacco use. I also have the data on sex and specific age groups.

What other data would I need to run a regression analysis? Or would you recommend another statistical analysis that could possible be better to detect a significant difference? Any help would be appreciated!
 

hlsmith

Less is more. Stay pure. Stay poor.
#2
Looks like multivariable logistic regressions. However for side note, establishing causality or a direction relationship would be difficult with your study design. I recommend this approach, because with it you can incorporate and control for the covariates that you mentioned.
 
#3
hlsmith, thank you for your response. I'll be just looking for a possible association instead of causality. Thanks for your recommendation!