« Rooting for Results | Main | CANOE and Diabetes as a Surrogate Endpoint »

May 30, 2010

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a0120a692721d970b0133ef5860e9970b

Listed below are links to weblogs that reference Composite Endpoints and the CREST Trial:

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Please correct me if I am wrong.. but what I read is stenting is associated with a 50% decrease in MI rate but a doubling of stroke rate with likely equal mortality. .. Now.. I think I'd choose an MI over stroke if I were to survive. As its likely the MI won't have nearly the devastating impact on my livelihood as a neurologic disaster. Did I get this right?
JFS

I agree with the comment from JFS, but that's not how I saw the information presented in the media.

Also, it's worth keeping in mind that a halving of MI rates and a doubling of stroke rates obscures a bit the absolute differences in MI and stroke rates. Mortality was also "doubled" with stenting (0.7 versus 0.3 percent); this was not statistically significant (p=0.18), but is hardly reassuring that stenting is noninferior with regard to mortality. It's hard to imagine choosing stenting over endarterectomy given these results.

@JFS,
you contort to use relative differences. Most patients patients and doctors think better in absolute numbers.

I agree with @JFS; realtive numbers are an abstraction in thinking that will be lost on the layman. Absolute values everytime - mis communication is worse than no communication.

The comments to this entry are closed.