kpfleger

The proper claim is that raising people from deficient (<20ng/ml) to a sufficient level (>30ng/ml ideally) should improve outcomes. Sometimes looking at a subset of those who started deficient is good enough to find such an effect, but sometimes it isn't, such as if the dose used was not high enough to raise people up enough, particularly common if the subjects weights were high which requires higher intakes to move the needle a lot, so other subset analysis look at norm-weight subjects. Proper study guidelines were published in the mid 2010s and meta-analyses like this one that aggregate many studies that don't conform to these guidelines are just as invalid as individual studies that don't.

For detailed treatment of this see this thread on X: https://x.com/KarlPfleger/status/1732514710715514883

Nonesense. It's well established that dietary D3 (or D2, but D3 is better) clearly incleases 25(OH)D serum levels. It is that serum biomarker whose deficiency (levels below 20ng/ml or 30ng/ml) that has the largest association with higher risk of dozens of diseases as well as all-cause mortality. You can defend your phrasing by saying that "not well" just means one has to take a lot of it. Yes, there is a wide variety of dose response, which is why it's best to test blood levels to titrate supplementation amount.

The academic paper this story is based on is here: https://nature.com/articles/s43587-024-00793-y Despite the linked story title only mentioning omega3, the paper was about 3 interventions, that, vitamin D, exercise and found reduced epigenetic age from the interventionts, with bigger anti-aging benefit as they were combined.

With respect to vitamin D specifically, this isn't by itself good evidence vitamin D reverses aging. But it is consistent with the totality of evidence that being vitamin D deficient probably (causally) speeds aging. And it mildly increases the overall weight of evidence of a connection.

As with much vitamin D research, it would have been better for the intervention to have been titrate vitamin D supplement amount to achieve an optimal target range (eg 30-60ng/ml) rather than using a fixed relative moderate dose (2000IU).

The problem with a fixed dose is that surely some subjects started severely deficient, some mildly, and some had sufficient levels. The moderate but not very high fixed dose will have helped many in the middle group climb out of deficiency, not been enough for many in the 1st group to bring them fully out of deficiency, and been unnecessary for those in the 3rd group, so the study ends up seeing a weaker overall effect averaged over all subjects. This is well known within the vitamin D world, eg see https://x.com/KarlPfleger/status/1732514710715514883, but somehow has not permeated to be widely enough understood, even within the scientific community (or the Hacker News community).

For an extensive list of papers supporting a link between vitamin D and speed of aging, see https://x.com/KarlPfleger/status/1390717755158974464

For more extensive discussion see my X thread here: https://x.com/KarlPfleger/status/1888649974621347950

No, ample evidence shows that vitamin D (or vitamin D deficiency at least) is causally related to multiple diseases. The evidence is pretty overwhelming. Eg here's a just published review on D deficiency causality for Covid based on applying Hill's Criteria for causality https://www.mdpi.com/2072-6643/17/3/599 and a prior 2022 paper on the same subject https://pubmed.ncbi.nlm.nih.gov/35308241/

Some vitamin D skeptics claim tat there's insufficient evidence from RCTs but few people question the causality of smoking's influence on cancer despite no RCTs for that---that relationship was established also via Hill's Criteria.

I put extended analysis of the results & caveats of what it means (correlation vs causation, etc.) on Twitter: https://twitter.com/KarlPfleger/status/1557814485070409728

I've collected the results from all 4 papers including the links to all 4 in a Google spreadsheet for easily lining up the results for each stage of Covid: https://docs.google.com/spreadsheets/d/1a5twV0x-TlHzd2AVr9e1... 1 tab for the links & underlying number of studies/patients & 1 tab for the results.

4 big systematic meta-analyses since Dec'21. Consistent results from all 4: Low D levels or not supplementing D has 1.5x-3x worse Covid risk at all stages (infection to mortality).

Some critical responses to this new study are starting to accumulate: Dr. Peter Attia's takedown: https://peterattiamd.com/vitamin-d-nejm-study/ Dr. Ronald Hoffman: https://drhoffman.com/article/whats-up-with-that-vitamin-d-s... Council for Responsible Nutrition: https://www.nutraingredients-usa.com/Article/2022/07/29/crn-... And a 3-part series from non-profit GrassrootsHealth (only 2 parts so far that I see): part 1: https://www.grassrootshealth.net/blog/go-part-1-3-headlines-... part 2: https://www.grassrootshealth.net/blog/go-part-2-3-vital-anal...

One reason getting these from the evolved route may be better is because feedback mechanisms stop making the beneficial substance when the body has enough, which may not happen via intake by other methods. This is the case for vitamin D from sun vs oral D supplements. Serum levels stop going up when they reach a natural adequate evolutionary range if it's coming from sun. I don't know if that's true of NO or not. Or of melatonin from the near-infra-red wavelengths, but I wouldn't be surprised.

I thought this piece spent too much time arguing about the wrong things (such as evolutionary stuff). I added 3 comments on what I think are the most important points with links to scientific research to the comments on Apr 22. Just search for Pfleger using find-in-page to read them.

Just how much lower it would get cannot be discerned from available data yet. It is very hard to believe from current data that it would go all the way to zero and also very hard to believe that fixing all vitamin D deficiency would not save any lives. The confidence interval range in between is still large, but it is absolutely (past) time to act on the data because it is so unlikely to not help (or even not help a lot). And the data is so far consistent with the possibility that it could help so much that it would have reduced the pandemic to no worse than prior flu seasons had everyone's vitamin D been brought up above 30ng/ml before SARS-CoV-2 arrived.

Did you read the open letter (original link of thread)? There is an entire paragraph and footnote with 2 refs talking about the problem with the current government guidelines.

The site itself was set up (by me) just to host this open letter, just as the site masks4all.co was set up to host the similar open letter that happened for masks in the spring. The credibility isn't of the site itself but of the signatories of the site, which are listed plainly (with any conflicts of interest). Look through the list yourself to judge their credibility.

This has been a big problem. Some people labeled science on vitamin D & covid19 as misinformation, lumping vitamin D in with things that have little to no evidence even when the vitamin D evidence was mounting rapidly. Part of the point of this open letter is to show that many serious people agree that vitamin D should be used in the pandemic. The inappropriate suppression of this message is now the mis-information. This should be clear from reading the letter and its very few references carefully.

See also the op-ed just posted to MedPage Today, and posted separately here on HackerNews: https://news.ycombinator.com/item?id=25680373

The FLCCC has been pushing Ivermectin, but I am not aware of an open letter analogous to this one or the similar one for masks that is signed by 100+ doctors & scientists.

I have not personally read the Ivermectin literature, but I think the FLCCC's methodology is good. (Paul Marik of FLCCC is a signatory on this vitamin D letter, BTW.) But it's notable that no one has Ivermectin deficiency, whereas more than half of people have insufficient vitamin D, especially in winter, so increased D intake is justified even in the absence of this COVID19 pandemic.

Subtitle: "We can't wait for perfect evidence" Op-ed by former surgeon general plus 2 other MDs.

185 experts. Over 100 medical doctors. Roughly 100 professors, including many of the most respected vitamin D experts in the world---check out the list of names and qualifications yourself.

Please read the link before posting about correlation or other opinions on the topic. The point is to note the aggregate credibility of the backers here. This is analogous to the open letter done for face masks last spring.

Terrible piece. First off, the majority of people with D levels of ~80ng/ml are fine, even quite healthy. Her problem was too much calcium, which is related but they (at least not in the article) don't address what her calcium intake was. Probably she was supplementing that too if the original reason for her supplementation was osteopenia. It's not clear if she would have had any problems if she had just dropped her calcium to low-normal intake. Quite a large fraction of Americans consume too much dairy.

Article that cherry pick a single example are terrible when they pick the example that is 100x less common than the other side of the story. 35-40% of the US has D < 20ng/ml. That's probably 100x greater than the number that have D >= this woman's levels (~80) and probably only 1% of folks who have that level have any kind of medical issue with it the way she did.

I completely disagree with the idea that D should be thought of as a medication in the sense of being careful with it. The problem currently is not enough care to make sure to get enough. 19 organizations (all the big important ones) consider 4000 IU/day of vitamin D to be a safe intake without the need to consult a doctor. The Endocrine Society and a few other orgs consider 10,000 IU/day safe. At the level of 5000 IU/day that she was taking, a very large % of people will remain clinically deficient. Many, many more than will have any kind of adverse issues.

The right dose for society to advocate as a default is not the dose that causes huge amount of harm on the low end only to avoid any kind of harm on the high end at all. That kind of asymmetric health optimization is poor public policy. But it's the way doctors think. Cover my ass and make sure to do no harm. The Hippocratic oath is a terrible way to optimize net health of society and it's unfortunate it has become entrenched in public health, not just the actions of individual doctors.

Some point the figure at big pharma. I've no evidence of any foul play, but the pharma industry would lose money if a large fraction of their revenue comes from drugs to treat conditions that are at least in part the consequence of inadequate vitamin D.

There seems to be more to it than that. Probably partially lack of belief that anything so simple, cheap, and easy could be such a help. I believe a similar skepticism surrounded mask use (and still does for a minority of folks).

And lack of the medical system (notably doctors) receiving much training in micronutrients.

And some amount of conservatism on the part of the bodies that do things like set RDAs. The IoM (now NAM) in the US was clearly being very conservative in its setting of the UL and its refusal to up the RDA when its statistical mistake in setting it was pointed out. And in the UK the recent NICE study was super-conservative (some would say narrow-minded) to only consider peer reviewed & published studies when it did a rapid review of D for C19 evidence.

But all this is just wild speculation to give you some idea since you asked. I would love to read the behind the scenes expose someday. Meantime I'm limiting myself to the evidence not looking into the conspiracy theories.

Correction (sorry, copy/paste error): The Mendelian randomization paper was by Kohlmeier. De Smet was one of the other studies.

This piece is not as thorough as my vitamin D & COVID-19 review: http://agingbiotech.info/vitamindcovid19/ or its 1-page bullet-point summary: http://agingbiotech.info/vitamindcovid19facts/

A lot of words on correlation != causation without noting any of the causal evidence. Even my 1-pager highlights the obvious responses to this over-used inequality:

1. RCTs show D supplements effective against respiratory infection (Martineau BMJ'17: 25 RCTs, now updated as preprint expanded to ~40RCTs covering ~30,000 people.

2. Causal evidence D is protective against lung injury (in rats) related to ACE2.

3. D extends lifespan in worms (which don't have bones), and we all know how much of a risk factor age is for C19. [This one not in my 1pager, only the full review.]

4. The number of plausible biological mechanism arguments is very large and expanding. See Linda Benskin's excellent review for the most comprehensive review of that evidence up through mid-June. More recently, the active form has been shown to have direct action against SARS-CoV-2.

5. Causal inference model shows that D's effect on C19 is causal [Davies et al].

6. Mendelian randomization shows that the correlations that would need to explain its data are far fetched (eg, systematic racism is worse in the US the farther north you go, by more than 5x) [De Smet et al]

7. Controlled intervention trial shows benefit from D+mag+B12 [Chuen Wen Tan et al]

I don't talk about it in my reviews, but there is also a set of guidelines for when you can infer causation from observational data called Hill's criteria and one paper did apply that to D related data and the evidence so far met all the criteria.

Long pieces that try to create uncertainty around vitamin D in the context of COVID-19 by repeatedly questioning the correlational data without noting any of the relevant causal evidence are far too common these days, and a bit irresponsible at this point.

Karl

Every time a new D & C19 item comes up, 75+% of the comments are already addressed in my review: http://agingbiotech.info/vitamindcovid19/ This new study is already incorporated there with a couple paragraphs of good discussion in strengths, weaknesses, importance, etc. Broadly, it's consistent with prior evidence but significant for being the 1st controlled trial (even if not random, which would be unethical at this point and thus should be impossible now). It's too bad the control group had so much more comorbidity. Hopefully their multivariate model is trustable for taking baseline differences in the 2 arms into account. The effect size is huge (OR=0.15).

Other common comment topics here (correlation vs causation, sun exposure, burden of proof for correcting deficiency, etc.) are all covered in the review already. If something important isn't covered, email me.

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=23119949

In 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 D<30ng/ml despite 29hr/week (>4hr/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.)

Author of the review here. Happy to have addition suggestions for papers to include in the review. I looked at the Reddit group you linked and saw only citations of Alipio et al and Grant et al (from several posts), both of which I already cite. The review says of the Grant et al paper that it "broadly reviews evidence for D to help with infectious diseases, pneumonia, and inflammatory cytokines that damage the lungs and discusses some of the aspects of COVID-19 infection that are known to relate to things known to be also relevant to vitamin D." So not quite sure what it is that you are suggesting be covered that is not.

The review is not meant to be a thorough description of the biology at the intersection of vitamin D & COVID-19. The point of bring up the refs I have in section 8 is to point out that there are good causal mechanistic biological arguments for D's usefulness against C19, to help justify a call for additional data gathering.

Your reddit group looks fine and all, but it isn't the same as a single document that concisely collects all the best evidence for the potential of vitamin D in a well organized way.

As with many comments in this thread, carefully reading the originally linked review: http://agingbiotech.info/vitamindcovid19/ would help. One of the 3 recent preprints specifically noted repeated testing of C19 patients and they did an ANOVA statistical test and determined that D blood levels were not declining in the patients over the course of the infection.

This doesn't prove causality, but it rules out the reverse causality you hypothesize in this post. There is a good, concise (3-paragraph) sub-section on the causality question in the discussion section called "Correlational data supported by many pieces of causal evidence". Mandatory reading for anyone who reacts with "correlation doesn't imply causation" as a criticism. Karl (author of the review)

It is hard to reconcile the fact that it's so easy to get enough D from sun in a short time with the fact that deficiency levels are roughly 40% in the US, with insufficiency (20-30ng/ml) much, much higher than 40% (and also high is much of the rest of the world). Look at the world chart in the paper (copied from Palacios et al review): 73% insufficiency in Austrelia, 61% in Britain even in summer, etc.).

Author of originally linked post here. I welcome your constructive suggestions to make the review better. I intend it to be the best review on the internet of the best-available data on vitamin D & COVID-19 and think that it probably is currently.

The best method of increasing D levels to avoid deficiency is not the main point of the article. I have added a paragraph to note the debate on whether sun is better than supplements, but covering that debate adequately is beyond the intended scope. The urgent point is to get more data on D levels by patient vs. case severity/outcome. Such records should already exist in sufficient quantity if we can just get the right people from the right orgs (Kaiser? VA? big hospitals?) to simply aggregate the records and analyze or partner with researchers who will.

I have specifically tried to collect the best papers on this into a detailed report: http://agingbiotech.info/vitamindcovid19/ which has been specifically linked to in this other HN thread: https://news.ycombinator.com/item?id=23119949

See also: AgingBiotech.info Specifically AgingBiotech.info/companies (esp. the clinical stage & clinical trials columns).