Over the past decade, no supplement has been as recommended as vitamin D. Unlike its antioxidant cousins, vitamin C and vitamin E, vitamin D is something we tend to be deficient in. In fact, many regulatory organizations have started to bump up their recommended daily intake from 400 IU/day to roughly 800 IU/day.
In the past few months, there has been some push back against vitamin D, most notably from The Lancet. They have recently published a plethora of meta-analyses, many of which conclude that vitamin D’s utility is overstated (you can see our analysis of one of their meta-analysis on vitamin D and bone health here). Most of these studies have been interpreted by the media in a very “black or white” manner - either vitamin D is useful, or it is useless.
It should be noted that no compound will be a panacea for everything. Everyone has their own diet, which leads to different deficient vitamins and minerals in the body.
While there are some outlandish associations with vitamin D with weak evidence (an easy one - that vitamin D leads to a testosterone boost), we thought it would be useful to take a step back and look at the big picture.
Vitamin D has been the “hot” supplement for the past decade, which means it inevitably got hyped beyond what it could actually do. We need to stop over-reacting to the latest research, and remember to keep an eye on the big picture.
We have addressed the recent meta-analysis to investigate vitamin D and its interactions with bone mineral density, and recently another meta-analysis which had a more encompassing goal was published investigating the role of vitamin D in vascular (heart and stroke related), skeletal (total fractures), and cancer outcomes. While the title seems interesting (and we’ll delve into it shortly), its conclusions are being circulating in some media sources as evidence that vitamin D “confers no health benefits” or that future trials are “unlikely” to support vitamin D supplementation.
The recent meta-analysis appears to be well conducted. They investigated double blind trials with either vitamin D in isolation or the combination with calcium, eliminated confounds such as dairy-enriched vitamin D products or interventions not given to both vitamin D and placebo groups, and eliminated studies with known comorbidities so they could more directly analyze the endpoints they wanted to (vascular, skeletal, and cancer outcomes). The study was funded by the Health Council of New Zealand which had no role in the design or conducting of the study. The study failed to find any interaction between vitamin D supplementation (usually in the range of 400-800 IU), with or without calcium, on any vascular endpoint (myocardial infarction, ischemic heart disease, cardiovascular disease, cerebrovascular disease, or stroke) or on cancer outcomes. No relationship was found even when controlling for baseline vitamin D status.
When investigating skeletal outcomes with regards to fractures (different from the previous meta-analysis covered), overall there was no significant association between supplementation and outcomes regardless of whether calcium was included or not. When investigating further, there was a benefit for reducing hip fracture in institutionalized patients. Community dwelling adults (those not in clinical settings and more likely to be purchasing vitamin D supplements on their own) did not, however, see a benefit to fractures that was deemed clinically significant. Clinical significance in this study was deemed a reduction in risk of 15% or more.
When looking at overall mortality (usually the ultimate end goal of interest), there was uncertainty in the results. When they assessed whether vitamin D could reduce the risk of death by 5% or more, the overall applicable sample size was only 60% of what was deemed statistically optimal; a proper statistical conclusion on mortality, due to this, could not be reached and it was deemed of uncertain status.
The latest The Lancet meta-analysis looked at vitamin D supplementation in the context of vascular, skeletal, and cancer situations.
This recent meta-analysis seemed well conducted and should not be disregarded. The relevant takeaways:
Cardiovascular disease is really not something vitamin D is catered to help treat or solve, and the failure of vitamin D having any benefits is no surprise. While aiming for a 15% reduction in risk may seem like the inefficacy of vitamin D was forced, it was chosen so that the magnitude was large enough to have practical benefits in the real world. Due to statistical reasons and subpar ‘ideal’ sample sizes, whether or not vitamin D reduces mortality by more than 5% cannot be answered with this study.
While a perfect study is pretty much impossible to conduct (hence why the body of evidence is greater than any one study), this one was pretty good. The discussion section does mention what evidence would be required to overturn the results which is always a good discussion to hear, and the reasoning that studies requiring over 50,000 people in sample size to yield a positive result have led to some articles stating that future articles are ‘unlikely to support vitamin D supplementation’ (shortsighted - it should be specifically mentioned for the studied parameters; vascular, skeletal, and cancer outcomes).
That being said, conclusions such as ‘no health effects’ are still much too grand to be associated to this study. This study could not assess everything, and claims that can still be made about vitamin D and its benefits include things such as colorectal cancer risk (this meta-analysis covered all cancer outcomes), multiple sclerosis, actual muscular control in the elderly (although the lack of reduction in fractures could argue against this), mood changes with regards to seasonal affective disorders, and diabetes risk. Some issues, such as the increase in testosterone and overall mortality risk, are still in a grey area.
The study was well-done, but to go from “had no effects on general vascular, skeletal, and cancer health” to “has no health effects” is a stretch (to say the least). It should also be emphasized that it looked at general cases, not specific cases.
As we previously cautioned, one must look at the entire body of evidence, not just at the latest research. With that in mind, if you look at our vitamin D page (more specifically, at our Human Effect Matrix), you will find a plethora of evidence that shows areas that vitamin D has a proven positive effect on. These include:
An improvement in muscular control in the elderly.
Vitamin D supplementation is strongly associated with a notable decrease in the risk of colorectal cancer, with some possible decreases in risk of other cancers such as breast. The Lancet meta-analysis assessed outcomes for those already diagnosed, and did not look at risk of cancer.
Appears to be a significant protective factor for the development of multiple sclerosis.
Usage for improving mood pertaining to seasonal affective disorders.
Potential (albeit thought to be minor) benefits to glucose control via aiding pancreatic function.
The original RDI of vitamin D, ~400 IU/day, was actually the minimum you need to not get rickets. The above benefits usually come from higher levels of vitamin D in the body, which can be usually obtained with roughly 2000 IU/day. If measuring the blood concentrations of vitamin D, usually you want to be either in the adequate range (50-75nM or 20-30ng/mL) or slightly higher. You do not want to be insufficient (25-50nM or 10-20ng/mL) or outright deficient (less than 25nM or 10ng/mL).
Vitamin D supplementation has a host of positive benefits. One must consider both the full body of research and also look past the quantitative (how long you live) and consider the qualitative (how well you live).
Overall, vitamin D is both cheap (roughly 5 cents for a daily dosage of 2000 IU) and safe (is not toxic except at absurdly high levels). While it may not fix everything, it has enough benefits to warrant taking.
Vitamin D is cheap, safe, effective, and most people are in a deficient state. If you do not get enough direct sunlight exposure, supplementation is the prudent choice.