4 ms·
Direct link to paper: https://www.sciencedirect.com/science/article/pii/S1094553920301462 https://www.sciencedirect.com/science/article/pii/S109455392...
by jmt_ 6y ago
Direct link to paper: https://www.sciencedirect.com/science/article/pii/S1094553920301462 https://www.sciencedirect.com/science/article/pii/S109455392...
- biagidp 6y agoThank you! That site's ad-to-content ratio was ramping up my clickbait detectors and made me bounce before I found this.
- beckler 6y agoI'll preface this by saying that I'm not even remotely a medical researcher, or a statistician. I would also love for someone to correct me if I get something wrong here... When I read this, I think the sample size (n=110) is too small for all the various claims they're correlating out of it. Yes, It's a large enough sample size for statistical significance, but I think there are too many variations involved to make any definitive conclusions. Specifically with the concomitant treatments. They also did not collect their own control data, but pulled SOC data from elsewhere. I don't think that's bad, but their control group was only 12 patients, and they even state in the paper that it's not large enough for sufficient results. I do think the results are encouraging from my armchair analysis, and I think it would be good to see the data from a larger study (and a far larger control group), but I think it's far too early to make any real conclusions from this.