Pubmed is sort of like arxiv.org and the fact this has an nih.gov base url should not be taken as an endorsement of it by the NIH.
That said I do actually believe that more then the recommended amount of vitamin D can be beneficial and the recommendations are in need of reevaluation.
Just to back this up: the journal it's published in has a very very low impact factor of 0.235 (i.e. it publishes 4x more papers that it gets citations, probably most of it's papers are never cited by anyone) [1]. If this paper was considered a big a deal by the wider research community you would expect it to be published in a high profile journal.
This sounds extremely low to me, but looking at the list of top journals sorted by Impact Factor on https://journal.komci.org/SearchJournal.php (you have to choose "Impact factor" in the "Sort results by" dropdown, as it does not generate a useful URL), the top journal only has an IF of 0.791.
I don't really know anything about the medical publishing world, but I would have expected a top journal to be cited much more?
I mean, it does. The "fairy tale" that one hears in high school doesn't, but we put aside fairy tales when we became adults.
If you are done with the "undergrad level" of Popper and Kuhn it is worth reading Imre Lakatos's work on philosophy of science. It contains a moment where one realizes that research programs live or die by this "impact factor" and that this living or dying is a key part of the overall methodology of science. The gist is that science is actually participating in a survival-of-the-fittest evolution with certain foundational ideas as the "genes" which "reproduce". So scientific ideas are actually good or bad in no small part due to their ability to create further scientific research along similar lines. A low impact-factor therefore directly says "along this particularly important-to-science axis, this journal sucks."
You are right that "the impact factor of a journal is meaningful and provides a simple/preliminary heuristic for measuring up _some_ aspects of a paper published in it" - but this is not what you wrote.
1. Imre Lakatos and maturity are great, both implying that you should not apply the aforementioned rule of thumb to an individual paper - an individual in the population - whether it was published in Nature or an insignificant contender.
2. Your memetic approach is also good, but incomplete: the objective function in case of these journals is maximizing the impact factor - so we can conclude that "PrevMed is less successful in maximizing the impact factor than some competitors, or it is a younger journal, or ..." Yes, imact factor and quality correlate in the long run, but we are not at undergrad level.
3. "A low impact-factor ... directly says" - Not directly. Also, most of the journals - not to mention conferences - do not even have an impact factor.
4. "...this journal sucks" - Most of the people writing in these kind of journals have given up a lot to contribute something modest. The editor of this journal is probably emailing with reviewers at 1am or so. Just saying...
But to stretch the analogy a bit further would it be fair to say that impact factor is very like sexual selection for extreme display traits that otherwise are detrimental to the wellbeing of the species?
Yes impact factor matters to current science as practiced but there is plenty of good criticism to show (at least as it is currently calculated) that it is a lousy measure of what is likely to end up being true, reproducible and useful.
> would it be fair to say that impact factor is very like sexual selection for extreme display traits that otherwise are detrimental to the wellbeing of the species?
If I read every PoS article vaguely related to my research, I'd never get anything done. In practice, I don't pay attention to impact factor. But I do pay attention to who's publishing. And that's basically the same as impact factor, in practice.
> that it is a lousy measure of what is likely to end up being true, reproducible and useful.
I don't think so.
High impact factor publications are MUCH more likely to be quality science than low impact factor publications (at least in my area).
The major venues would have to get at least two orders of magnitude worse before they became bad indicators of quality.
Of course, and obviously, that does not entail that all work published in high impact factor journals is high-quality.
I think the fundamental problem is just that you vastly under-estimate the enormous volume of utter crap there is out there.
What? It's nothing of the sort. Pubmed is not a preprint server. The paper in question was accepted by a journal. The quality may not have been great - I agree - but it is not like the archive at all.
He means it’s a resource for finding any/all articles, irrespective of source journal or quality - specifically pointing out that people shouldn’t infer anything from the NIH url.
Seems that way, I know, to some of my friends who don't work in the same field as me (health research), but pubmed doesn't include all biomedical journals, and it's not a place where anyone can archive papers independently. Which isn't to say that everything in pubmed is high quality or that it's a great gatekeeper, but yeah, it definitely excludes material.
If taking EPIC series morbidity and mortality endpoint analysis at face value, best outcomes on average are around 2500 IU/d intake. (In generally healthy individuals.)
So this recommending 4000 IU/d is not too far fetched.
Yeah to be clear I think there is lots of good science supporting increased vitamin D intake possibly including some of the work cited/described here. This article just doesn't seem notable, novel or particularly well written beyond the grabby title.
Fair enough. I was just trying to make the point that something being on pubmed doesn't mean its anything close to an NIH recommendation and chose arxiv as it would be familiar to a CS audience.
Lots of people in health research do treat pubmed as a gatekeeper though (sometimes foolishly imo). It is more exclusive than both Google Scholar and arxiv -- but also like you say, not a guarantee of quality + not the kind of resource where you can read one article and be like "here's what NIH says."
So semantics, I guess. Just felt compelled to mention it, probably more because it frustrates me to see health researchers use pubmed as their only gateway to scholarly lit.
vitamin D testing was a huge money maker to labs. the EU cut this test from its health plans and a few equipment makers (in labs, equipment works like big iron, you get it leased for free while your reagent subscription is active) are bankrupting.
Pubmed is sort of like arxiv.org and the fact this has an nih.gov base url should not be taken as an endorsement of it by the NIH.
That said I do actually believe that more then the recommended amount of vitamin D can be beneficial and the recommendations are in need of reevaluation.