The surprising failure of vitamin D in deficient kids


Welcome to Impact Factor, your weekly dose of commentary on a new medical study. I’m Dr F. Perry Wilson of the Yale School of Medicine.

If you’ve watched this space over the years, you’ll know that I’m not the biggest proponent of vitamin D supplementation. My basic gripe is that you’ve got all these observational studies linking lower levels of vitamin D to everything from dementia to falls to cancer to COVID infection, and then you do a big randomized trial of supplementation and don’t see an effect.

And the explanation is that vitamin D is not necessarily the thing causing these bad outcomes; it’s a bystander – a canary in the coal mine. Your vitamin D level is a marker of your lifestyle; it’s higher in people who eat healthier foods, who exercise, and who spend more time out in the sun.

And yet ... if you were to ask me whether supplementing vitamin D in children with vitamin D deficiency would help them grow better and be healthier, I probably would have been on board for the idea.

And, it looks like, I would have been wrong.

Yes, it’s another negative randomized trial of vitamin D supplementation to add to the seemingly ever-growing body of literature suggesting that your money is better spent on a day at the park rather than buying D3 from your local GNC.

We are talking about this study, appearing in JAMA Pediatrics.

Briefly, 8,851 children from around Ulaanbaatar, Mongolia, were randomized to receive 14,000 international units of vitamin D3 or placebo every week for 3 years.

Before we get into the results of the study, I need to point out that this part of Mongolia has a high rate of vitamin D deficiency. Beyond that, a prior observational study by these authors had shown that lower vitamin D levels were linked to the risk of acquiring latent tuberculosis infection in this area. Other studies have linked vitamin D deficiency with poorer growth metrics in children. Given the global scourge that is TB (around 2 million deaths a year) and childhood malnutrition (around 10% of children around the world), vitamin D supplementation is incredibly attractive as a public health intervention. It is relatively low on side effects and, importantly, it is cheap – and thus scalable.

Back to the study. These kids had pretty poor vitamin D levels at baseline; 95% of them were deficient, based on the accepted standard of levels less than 20 ng/mL. Over 30% were severely deficient, with levels less than 10 ng/mL.

The initial purpose of this study was to see if supplementation would prevent TB, but that analysis, which was published a few months ago, was negative. Vitamin D levels went up dramatically in the intervention group – they were taking their pills – but there was no difference in the rate of latent TB infection, active TB, other respiratory infections, or even serum interferon gamma levels.


But to be fair, the TB seroconversion rate was lower than expected, potentially leading to an underpowered study.

Which brings us to the just-published analysis which moves away from infectious disease to something where vitamin D should have some stronger footing: growth.

Would the kids who were randomized to vitamin D, those same kids who got their vitamin D levels into the normal range over 3 years of supplementation, grow more or grow better than the kids who didn’t?

And, unfortunately, the answer is still no.

At the end of follow-up, height z scores were not different between the groups. BMI z scores were not different between the groups. Pubertal development was not different between the groups. This was true not only overall, but across various subgroups, including analyses of those kids who had vitamin D levels less than 10 ng/mL to start with.

So, what’s going on? There are two very broad possibilities we can endorse. First, there’s the idea that vitamin D supplementation simply doesn’t do much for health. This is supported, now, by a long string of large clinical trials that show no effect across a variety of disease states and predisease states. In other words, the observational data linking low vitamin D to bad outcomes is correlation, not causation.

Or we can take the tack of some vitamin D apologists and decide that this trial just got it wrong. Perhaps the dose wasn’t given correctly, or 3 years isn’t long enough to see a real difference, or the growth metrics were wrong, or vitamin D needs to be given alongside something else to really work and so on. This is fine; no study is perfect and there is always something to criticize, believe me. But we need to be careful not to fall into the baby-and-bathwater fallacy. Just because we think a study could have done something better, or differently, doesn’t mean we can ignore all the results. And as each new randomized trial of vitamin D supplementation comes out, it’s getting harder and harder to believe that these trialists keep getting their methods wrong. Maybe they are just testing something that doesn’t work.

What to do? Well, it should be obvious. If low vitamin D levels are linked to TB rates and poor growth but supplementation doesn’t fix the problem, then we have to fix what is upstream of the problem. We need to boost vitamin D levels not through supplements, but through nutrition, exercise, activity, and getting outside. That’s a randomized trial you can sign me up for any day.

Dr. Wilson is associate professor, department of medicine, Yale University, New Haven, Conn. He reported no relevant conflicts of interest.

A version of this video transcript first appeared on

Recommended Reading

Opt-out HIV testing in EDs can help identify undiagnosed cases
MDedge Pediatrics
Study supports banning probiotics from the ICU
MDedge Pediatrics
More vaccinated people dying of COVID as fewer get booster shots
MDedge Pediatrics
More work needed to optimize STI screening in primary care settings
MDedge Pediatrics
U.S. flu activity already at mid-season levels
MDedge Pediatrics