Gordon H. Guyatt
@GuyattGH
Promoting use of the best evidence and patient values and preferences to inform optimal clinical and health policy decisions. https://t.co/JxbX7cZMgy
ID:2559531679
https://www.linkedin.com/in/guyattgh/ 10-06-2014 18:19:48
4,3K Tweets
13,6K Followers
565 Following
Landmark study: 29 analyst teams used same data set to see if dark-skin soccer players get more red cards. Result: extreme different estimates from 0.89 to 2.93, 29 significant, 9 not. Until replication credibility of results of complex data limited. journals.sagepub.com/doi/10.1177/25…
Two articles highlight big problems when platform trials include non-contemporaneous controls. In addressing A vs no A, analysis should only use randomizations that included possibility of A. Should never use controls of B vs non-B.
nejm.org/doi/full/10.10…
sciencedirect.com/science/articl…
Latest evidence of utter failure of non-inferiority trials to set appropriate non-inferiority margins. Clinicians should ignore these margins & statistical claims of non-inferiority & instead think how their patients would balance benefits/harms/burdens. pubmed.ncbi.nlm.nih.gov/37421996/
After revisions for regulators, we're just about to start recruitment for large, pragmatic #ARTStrial that recruits uro, gyne & abdominal surgery patients in which net benefit of pharmacologic prophylaxis uncertain. If interested, please comment or message #EBM #EAU24 European Association of Urology (EAU)