2 ms·
Not necessarily. If they were able to sufficiently test for confounding variables, it'd be a reasonable conclusion. The news article specifically states they we
by elcritch 11y ago
Not necessarily. If they were able to sufficiently test for confounding variables, it'd be a reasonable conclusion. The news article specifically states they were able to control and remove overlap with existing dementia cases. Question is how well it's done, and I don't have time to examine the OP. Post if you notice anything?
- joveian 11y agoHere is a comment from the authors about residual confounding: https://archinte.jamanetwork.com/article.aspx?articleID=2428897 https://archinte.jamanetwork.com/article.aspx?articleID=2428...