I think there's an error on the page, but am not willing to delete it unless someone can confirm I am not mistaken (e.g. maybe the result is from some post-hoc analysis I'm not aware of??)
Major bleeding risk is increased in BOTH aspirin subgroups according to the authors, but there is a result shown on the page that suggests that it was only increased in the aspirin initiation sub-group. --Zoidberg
- Hi, I went back and double-checked the numbers. It is in fact correct. It is contained in the Supplementary appendix, pages 22-22. In regards to your concern, when taken as a whole, aspirin increases major bleeding for allcomers (the outcome that this trial was powered to detect). On subgroup analysis this was only significant for those initiating aspirin at the time of therapy. The P-value for interaction was non-significant.
- How should one take this result? Aspirin increases risk of surgically-related major bleeding. There is some evidence that surgical-related major bleeding risk is not increased for individuals on aspirin at baseline. However, this is a subgroup analysis and should be taken with a Morton's canister of salt. This can be used to inform complex medical decision making (eg, patient with known CAD and UA going in for a surgery -- continue on aspirin?) but not necessarily population-level guidelines (eg, every low-risk patient on aspirin for primary prevention needs to stay on aspirin no matter what). Would love to hear your thoughts. --Tim Plante (talk) 13:09, 1 February 2015 (UTC)