Read on Twitter
1/9 Started our yearly Critical Appraisal of the Landmark Literature course with the 4th years today (on Zoom, 
). Seems like a good time for a quick tweetorial on reading RCTs. These are the studies we rely on most in medicine but they can be misleading. What would you add?
). Seems like a good time for a quick tweetorial on reading RCTs. These are the studies we rely on most in medicine but they can be misleading. What would you add?
2/9
How RCTS can be misleading #1
For studies with subjective endpoints, was the control really adequate. https://www.amjmed.com/article/S0002-9343(18)30754-X/fulltext
How RCTS can be misleading #1
For studies with subjective endpoints, was the control really adequate. https://www.amjmed.com/article/S0002-9343(18)30754-X/fulltext
3/9
How RCTS can be misleading #2
Always be cautious when surrogate endpoints are used. https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(04)17355-8/fulltext
How RCTS can be misleading #2
Always be cautious when surrogate endpoints are used. https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(04)17355-8/fulltext
4/9
How RCTS can be misleading #3
Consider the âpre-studyâ probability that the treatment is effective. https://jamanetwork.com/journals/jama/fullarticle/2658299
How RCTS can be misleading #3
Consider the âpre-studyâ probability that the treatment is effective. https://jamanetwork.com/journals/jama/fullarticle/2658299
5/9
How RCTS can be misleading #4
When the results of a trial only barely reach statistical significance, consider the fragility index. https://jamanetwork.com/journals/jamasurgery/fullarticle/2712857
How RCTS can be misleading #4
When the results of a trial only barely reach statistical significance, consider the fragility index. https://jamanetwork.com/journals/jamasurgery/fullarticle/2712857
6/9
How RCTS can be misleading #5
Remember that trials terminated early systematically overestimate the benefit of the intervention. https://jamanetwork.com/journals/jama/fullarticle/185591
How RCTS can be misleading #5
Remember that trials terminated early systematically overestimate the benefit of the intervention. https://jamanetwork.com/journals/jama/fullarticle/185591
7/9
How RCTS can be misleading #6
Never forget about publication bias and p-hacking. https://www.ncbi.nlm.nih.gov/pubmed/22853650
How RCTS can be misleading #6
Never forget about publication bias and p-hacking. https://www.ncbi.nlm.nih.gov/pubmed/22853650
8/9
How RCTS can be misleading #7
Always ponder generalizability. Your patient almost certainly would have been excluded from the trial you are reading. Is it OK to still apply the conclusions to your patient?
How RCTS can be misleading #7
Always ponder generalizability. Your patient almost certainly would have been excluded from the trial you are reading. Is it OK to still apply the conclusions to your patient?
9/9
How RCTS can be misleading #8
And, for #8, just seems like this needs to be cited⊠https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.0020124
How RCTS can be misleading #8
And, for #8, just seems like this needs to be cited⊠https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.0020124
