turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- Health Care and Pharma
- /
- SAS in Health Care Related Fields
- /
- non-inferiority analysis

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

09-28-2006 01:58 PM

My question is how to perform a non-inferiority analysis with SAS. Which PROC statements should be use? Is this completed in one step or are there several steps in this type of analysis?

The study is an observational cohort and the methods will be something like this: Our hypothesis is that Drug X 5mg is non-inferior to the same Drug X 10mg, evaluating percent lowering of cholesterol from baseline at 3 months. The cost is almost twice for 10mg than 5mg.

Literature states that Drug X 10mg lower cholesterol 15% from baseline. We feel that if Drug X 5mg lowers cholesterol by 12% this would clinically be non-inferior to the 10mg strength and thus not worth the additional extra cost. There will also be 4-5 covariates to control for potential bias.

Any help would be appreciated. I have search the SAS database and internet without success.

The study is an observational cohort and the methods will be something like this: Our hypothesis is that Drug X 5mg is non-inferior to the same Drug X 10mg, evaluating percent lowering of cholesterol from baseline at 3 months. The cost is almost twice for 10mg than 5mg.

Literature states that Drug X 10mg lower cholesterol 15% from baseline. We feel that if Drug X 5mg lowers cholesterol by 12% this would clinically be non-inferior to the 10mg strength and thus not worth the additional extra cost. There will also be 4-5 covariates to control for potential bias.

Any help would be appreciated. I have search the SAS database and internet without success.

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to deleted_user

09-29-2006 08:31 AM

I have read that one way to perform non-inferiority (= one-sided) tests is to build two-sided tests with the alpha risk twice the wanted value. Eg : if you want to test non-inferiority with 5% alpha risk, test equality with 10% alpha risk.

One way to do this with SAS using covariates to stratify analysis is through GLM (fixed effects only) or MIXED (fixed and random effects) procedures.

I am not sure if this way of performing one-sided tests is really exact, or just a workaround with close results to the true method (or something totally insane, by the way, but I'm sure I've read some people doing so).

Olivier

One way to do this with SAS using covariates to stratify analysis is through GLM (fixed effects only) or MIXED (fixed and random effects) procedures.

I am not sure if this way of performing one-sided tests is really exact, or just a workaround with close results to the true method (or something totally insane, by the way, but I'm sure I've read some people doing so).

Olivier