3 ms·
The fact that they were able to show a dose-dependent response is pretty damning. I think they controlled pretty well for confounding variables, so I'd consider
by jmcgough 11y ago
The fact that they were able to show a dose-dependent response is pretty damning. I think they controlled pretty well for confounding variables, so I'd consider this more than "suggestive".
- KingMob 11y agoNot really, a dose-dependent response can also be explained by hidden variables: people taking stronger doses of medications are likely to be sicker than ones taking lighter doses and therefore, more likely to develop dementia. (E.g., who's probably hurt worse? The person on aspirin or the one on morphine?)