[api]

One difficulty in determining if vitamin D actually helps with COVID is determining if vitamin D levels are directly affecting outcomes or whether they are a proxy for something else. For example: are low levels a proxy for people who don't go outside much and therefore don't exercise?

That's a huge problem in medical studies generally.

[pjkundert]

There is strong evidence of this, summarized: https://www.outsideonline.com/2380751/sunscreen-sun-exposure...

Huge inverse correlation between Vitamin D and all-cause mortality -- but almost complete failure of vitamin D supplementation vs. cancer, heart disease and stroke. What?

Turns out that sunlight produces vitamin D -- and a host of other difficult to detect artifacts, which seem to be at the root of the reduction in all-cause mortality.

Seems interesting. Or, carry on -- I'm sure there's nothing to see here, and your pasty-white overweight doctor's advice to avoid sun exposure is right...

[Dylan16807]

Doctors advising people to avoid sun exposure? Is that strawman based on anything?

I hope you don't mean the advice to use sunscreen when you're going to be outside so long you'd get sunburn without it.

[jerf]

How old are you?

I can assure you, whether it was the intent or not, that the message got out in the 1980s and 1990s that all sunlight exposure is uniformly bad and its only effect is to increase your likelihood of skin cancer, so everyone should avoid it and unconditionally use strong sunscreen for any exposure.

[radford-neal]

See https://www.canada.ca/en/health-canada/services/sun-safety/s... for one example of an official health authority advising people to zealously avoid sun exposure, with no mention whatever of vitamin D deficiency or other downsides of not getting any sun.

[pdonis]

The studies referenced in the Outside Online article failed to collect a crucial piece of data: what were the actual blood levels of Vitamin D in the participants?

Without that piece of data you have no idea what, if any, difference the supplements were actually making vs. placebo.

[perl4ever]

Maybe sun = uv exposure, which changes the microbiome of your skin and insides to more uv resistant strains which are healthier.

[chl]

There are now 24 trials (of some kind) listed on ClinicalTrials.gov:

https://clinicaltrials.gov/ct2/results?term=vitamin+d&cond=C...

At some point, high-quality RCT results will be in.

[OJFord]

In other words: establishing causation (or lack of it) is more useful than merely correlation. Yes. But hardly breaking news, or even specific to medicine.

[antishatter]

I mean yes, but OP's comment is especially notable with Vit D because it's deficiency is very common and often a marker for other common age related diseases.

It is highly likely that Vitamin D isn't saving these people's lives, it's that the folks without vitamin D are older and/or have disease.