I like the idea of the Vitamin D link, and it seems well established. But the idea of this paper I find too crude. There can easily big cofactor if you look at

I like the idea of the Vitamin D link, and it seems well established. But the idea of this paper I find too crude. There can easily big cofactor if you look at the average vitamin D levels per state. Like when the is less Vitamin D might be more people have office jobs leading to higher obesity (for the mortality rate). For infection rate could also be meeting people in offices/indors. Also areas with less sunlight have people indoors more often.