Having read this whole piece, I think my review is much more comprehensive in covering all relevant evidence:
http://agingbiotech.info/vitamindcovid19/
Note that the title changed and many improvements made since it was first posted to HN 3 weeks ago here:
https://news.ycombinator.com/item?id=23119949In the title of this piece, analogizing to chloroquine is irresponsible. A high % of readers will take the message to stay away from D and never read the rest.
I'm glad he has a similar top-level rec as me: Take all measures to avoid deficiency, probably by D3 of 2000-4000 IU/day (I'd prefer 4000). The piece unfortunately fails to make the other most important top-level comment: that all COVID-19 patients should have their D levels tested and more data on D & COVID-19 severity should be published urgently (possibly based on records that already exist). And all clinical trials for COVID-19 should be testing D levels as well.
Overall, this piece is not emphatic enough that vitamin D may significantly help during this COVID-19 crisis.
Several important responses to some of the top material in this piece:
The NYT article dismissing Holick is terrible and not worthy of being linked at the top of this piece. I won't go into the point-by-point criticisms of this article, but it's not worthy of a piece that is supposed to be weighing evidence.
The bullet point claiming the COVID-19 association could still be a fluke does not adequately acknowledge the causal aspects of the evidence: Many RCTs on D supplementation helping viral/respiratory infections (Martineau et al BMJ 2017), D causally affecting rate of aging in roundworms, causal inference model overwhelming supporting D's role as causal, studies on the correlation correcting for age, sex, comorbidities, and reverse causality, and causal mechanistic biological arguments. Some portion of the strength of the correlation seen in the 1000+ patient cases published in preprint studies so far may be due to other causal factors, but the causal evidence so far makes it very unlikely for the full extent of the large effect size seen to all be a fluke.
The piece wrote: "A healthy full body exposure to sunlight can generate 25000 IU of vitamin D in one sitting, without getting burned. 2000-4000 IU daily of D3 supplements will also work."
No. People wrongly believing that some sun gives them enough is probably one of the causes of widespread deficiency. From my review, in the "Practical considerations for avoiding D levels that are too low" section:
There are several studies showing that even a lot of sun exposure leaves many people with insufficient D levels. For one list of studies see Tsiaras & Weinstock, “Factors influencing vitamin D status” Acta dermato-venereologica, 2011: “Studies in Hawaii (97), South Florida (98), Southern Arizona (99), Brazil (100), rural India (101) and Queensland, Australia (102) found that significant proportions of the study populations had low vitamin D levels despite abundant sun exposure.” For example, in the Hawaiian study, 51% of the 93 young, non-obese adults had D4hr/day) in the sun at latitude 21 degrees. The variability in both seasonal sun intensity and daily hours of exposure for any given individual provide yet another reason that testing blood levels would be helpful. These studies show that even those who seem to get a lot of sun exposure should probably supplement if not testing to ensure adequacy.
Also, the bullet point summary in this piece fails to point out the history of the RDA being set too low by statistical mistake. (Which I've updated my review to include. The story is crazy---the original data used to set the RDA would have resulting in 8000 IU/day if it had been analyzed correctly.)