Most of the observational studies are weak and I’ve seen pretty over-exaggerated claims on higher end of Vitamin D from scientists like Dr. Rhonda Patrick (she is sensational and overhypes some things, but not to the point of a quack) to the random typical quacks claiming ultra mega-dosing vitamins are good for you when there is no “gold standard” trial proving this and a few trials against this.
The current “official” recommendation is 600-800 IU depending on adult vs older adult. Generally, Vitamin D levels are not tested unless there is a good reason to due to cost effectiveness and issues with assays commonly used in commercial labs. (For reference, I use LC-MS/MS based on NIST standard to be as confident as possible on the accuracy since I’m trying to go for a fairly tight window to minimize any potential harms)
I’m going a bit against the official rec by taking Vitamin D 2,000 IU USP grade with cold water fish/roe & quality fish oil (which adds a labeled tiny amount of mixed tocopherols - with frequent independent third party testing and my own tests to avoid rancidity) based on the VITAL trial on NEJM.
Keep in mind - I’ve been really careful with genetics before and tested baseline as well as ongoing levels which is not generally cost-effective. I also avoid sunlight exposure from all angles as much as possible for skin cancer prevention (only have a tiny amount in the early morning indirectly). So individual supplementation will vary with a lot of caveats.
https://www.nejm.org/doi/full/10.1056/NEJMoa2202106
In this vitamin D and omega 3 (VITAL) trial, 25,871 participants (mean age 67 years and mean baseline of approximately 30 ng/mL) were randomly assigned to vitamin D 2000 IU or placebo. After a median follow-up of 5.7 years, the cumulative incidence of autoimmune disease was lower in the treatment group (0.95 versus 1.2 percent, HR 0.78, 95% 0.61-0.99). I have family that has autoimmune disorders so I’ve been following many options for things like “pre-RA” and other proposed pre-disease states based on known genetics associated with HLA (which can vary by reference population and ethnicity) and other issues. I’m quite familiar with the immune system in general. If I’m right, then I might make a big gain on life expectancy, since autoimmune disorders like RA add 3x CV risk. Other people might not find that potential gain compelling enough in their circumstance.
There are no prospective studies to define optimal Vitamin D levels for extraskeletal health with widespread definitive conclusions though but I suspect >=80 ng/mL is very likely “bad” because there isn’t enough of a feedback mechanism for this hormone and <80 is what one would get if from all-day constant direct sun exposure. We also saw decreases in bone mineral density at >=40 ng/mL or around 4,000 IU, so may be potential harmful dose if it gets you to >40. >50 is a common agreed upon upper limit, where >60 is where more associations of potential negative effects occur.
We also have an official recommended upper limit of 4,000 IU (assuming no malabsorption etc). So I aim for 30-40 personally with 2,000 IU per day. Preferably closer to 35-40 based on some small association with tiny additional cancer prevention. That’s my personal interpretation of the literature where benefit and harm is balanced towards possibly insignificant likelihood of harm for me personally. Some experts previously claimed around 25-30 is best for maximum PTH suppression - a fair position to take.
Vitamin D levels have a genetic basis for variations with a high degree of heritability in twin studies. Some may have malabsorption but it’s much less likely if you are “healthy”. So when you talk to your doc, must take that and a lot of other things into account. Not as simple as it’s made out as there are actually still knowledge gaps on this hormone. Don’t let the “vitamin” name make it seem like it must be innocuous - it’s always about getting close to the “right” amount at the “right” time often with limited imperfect data anytime you take anything. The deepest experts in the field for new therapies often get it wrong statistically and the likelihood of a non-expert getting it wrong is probably even more likely on average, especially if they are an influencer trying to sell you something with an affiliate link or they are tied to some ideology.
The thing is contrarianism is usually wrong and if one is trading one potential small benefit for a small likely known harm, one is probably bad at math, risk management, and recognizing cognitive biases. So if you are going contrarian against official guidelines like myself, you better be pretty darn familiar with the potential tradeoff and evidence.