3 ms·
Certainly, but I think that still leaves a lot of room for a non-causal correlation. In a toy model where Covid-19 is uniformly present in N% of indoor spaces a
by SpicyLemonZest 6y ago
Certainly, but I think that still leaves a lot of room for a non-causal correlation. In a toy model where Covid-19 is uniformly present in N% of indoor spaces and 0% of outdoor spaces, we'd expect infection rates to correlate strongly with vitamin D levels even if vitamin D doesn't affect the virus at all.