6 ms·
Yet another observational study with no real comparison. The REACT dataset used by the study relies mostly on self reported surveys and data from people sick e
by nostrebored 3y ago
Yet another observational study with no real comparison. The REACT dataset used by the study relies mostly on self reported surveys and data from people sick enough to go to the doctor. This has obvious problems.
The idea that there is a control arm of people who haven’t gotten Covid large enough to power a real control here is farcical. Again, a problem of self selection for people committed enough to maintain a large level of isolation for multiple years.
Any causal study here should be looking at a differential in pre and post covid cognitive scores with a control of known uninfected participants in the same time window.
- Animats 3y ago> The idea that there is a control arm of people who haven’t gotten Covid large enough to power a real control here is farcical. Why? Whether someone has ever been infected is testable. About 20% of the US population had not had COVID as of some point in 2023.
- nostrebored 3y agowhich tests are you talking about? Antibody titers only work for a few months, and as far as I understand they’re the standard test. Why do you believe that 20% of the U.S. population hasn’t had covid?