Here we go again, around and around. Again, as previously stated in this thread the reason I found the French study flawed is because 1. It had a tiny sample size (IIRC 28 total people between both the control group and the experimental group. 2. 3 people from the tiny experimental group were sent to ICU because they got too sick to continue (one died) and these were not included in the final calculations. So yes, I made fun of that study.
As for this study, yes it was a review of outcomes of covid-19 patients, some that took the drug and some that did not. Was it a double-blind clinical trial? Nope, it was not, but it is based on a large number of patients which makes up for some inherent design flaws of the study. Is it proof that the drug is ineffective? No, it not definitive, and I never said it was. But since it's based on a large number it is another data point, another nail in the hydroxychloroqhine coffin. The reason I take the study seriously is that in a study that large, even with its acknowledged flaws, you would expect to see the efficacy of the therapy reflected in patient outcomes. And what was discovered is that the patient outcomes really weren't affected at all by the drug, or there was maybe a very small negative effect (I'm not sold on the negative effects of the drug, but I am becoming more and more convinced that it is useless in treating any viral disease as has been demonstrated in the past.) As for you aspirin example, if aspirin was a new drug and we wanted to see how it affected cardiac patients, it would be useful to look at outcomes of hundreds of these patients that took aspirin, and those that did not take it. To me it makes sense to add the study as a data point. YMMV.
success.