3 ms·
The paper says > but also for non-hospitalised cases who had biological confirmation of COVID-19 infection (N = 326) I think it applies to everyone?
by camgunz 5y ago
The paper says
> but also for non-hospitalised cases who had biological confirmation of COVID-19 infection (N = 326)
I think it applies to everyone?
- mlyle 5y agoSee figure 2. A standard deviation of IQ is about 15 IQ points. Only the ventilator patients have an effect anywhere near the claimed magnitude. Average across the COVID patients weighted by prevalence of outcomes (to match the overall population's COVID experiences) is more like 0.5-1.0 IQ points. And some or all of this effect may be generated by confounds.
- camgunz 5y agoFigure 2 shows a pretty clear correlation between severity of respiratory illness and cognitive deficits. If it were really confounded or a random effect, you wouldn't see 100% correlation like this. I do agree their N could be higher for all groups, but that's really hard, and again the correlation is 100%.
- mlyle 5y agoI'm not saying there's not a real effect here. Hell, if there wasn't, that would be unexpected: just about any sickness is correlated with measuring lower cognitively for a time afterwards, and being hospitalized and being on a vent are more so, independent of COVID-19. I'm just saying that: * The claim that there's a 7 IQ point drop from being a COVID-19 survivor is completely bogus, unless you're only a survivor after being on an ventilator. * The effect size and causative relationship are dubious, because the controls and matching on this study are imperfect. There's plenty of opportunity for significant confound. E.g. if dumb people were more likely to not take precautions against getting COVID, it would look like getting COVID was associated with lower cognitive performance in this study.