Podcast 297: Forget about all that vitamin D testing!!

Summary

Clinical Conversations, Episode 297: Forget about all that vitamin D testing!!

Podcast Host: NEJM Group
Date: July 28, 2022
Participants:

• Alan Brett (B): Professor of Medicine, University of Colorado; Editor-in-Chief, NEJM Journal Watch
• Stephen Cummings (C): Senior Scientist, Sutter Health; Emeritus Professor of Epidemiology, UCSF
• Joe Elia (A): Host

---

Episode Overview

This episode focuses on the clinical ramifications of the VITAL trial—the largest randomized study to date examining whether vitamin D supplementation prevents fractures in otherwise healthy adults. The central message: Vitamin D supplementation does not prevent fractures, even in people with low measured vitamin D levels. The discussion further explores vitamin D's physiological role, the pitfalls of testing, and clinical scenarios where supplementation or testing might still be justified.

---

Key Discussion Points and Insights

The VITAL Trial: Results and Meaning

• Background

  • VITAL trial involved 26,000+ generally healthy, community-dwelling adults aged 50+.
  • Participants were randomized to receive vitamin D3 (2,000 IU/day) or placebo for 5 years.
  • Previously published results showed no benefit for cancer or cardiovascular outcomes.
  • Omega-3 randomization did not influence fracture outcomes.

• Main Findings

  • "Simply, it was negative. There was no effect of that vitamin D supplementation on risk of any kind of fracture." (C, [01:57])
  • No subgroup—including those with low vitamin D levels (<20 ng/mL), previous fragility fractures, those on osteoporosis medications, or various racial/weight groups—derived fracture-prevention benefit from vitamin D supplementation. (C, [01:57]–[02:44])

• Reliability of Results

  • Prior observational findings showing benefit are likely due to confounding (people with low Vitamin D tend to be sicker).
  • “All of the large randomized placebo controlled trials have been negative for fracture falls and other conditions. It's the randomization that's key and the size of it allows you to look at subgroups. So it is the gold standard.” (C, [03:17])

Vitamin D: Hormone, Not a Vitamin

• Vitamin D operates as a tightly regulated hormone in the body, not just a dietary supplement.
• Measured levels often reflect the protein-bound complex, complicating interpretation.
• The body maintains active vitamin D (1,25-dihydroxy) levels through robust physiological regulation—lack of dietary vitamin D doesn’t usually indicate a true deficiency unless regulation fails (e.g., severe malnutrition, certain illnesses). (C, [04:13])

Clinical Scenarios

1. Healthy Older Adults with Osteoporosis

• Case: Healthy 65-year-old woman, osteoporotic T-score, good diet/sunlight, starting bisphosphonates.
• Recommendation:
  "I'll choose answer number one, don't test and don't treat...there is no potential benefit of giving her vitamin D." (C, [06:43])
  No testing or supplementation needed unless clear risk factors or signs of severe deficiency exist.

2. Frail Elderly with Poor Nutrition/Sunlight

• Case: Nursing home resident, poor diet, little sunlight—possibly severely deficient.
• Recommendation:
  Give vitamin D (e.g., 2,000 IU/day) with calcium without necessarily testing, as evidence (old French trial) suggests rare, truly deficient patients benefit. (C, [06:43])

---

Vitamin D Harms and Misconceptions

• Increased Risk of Falls and Weakness
  “Some of the trials, several of the trials, ... showed an increased risk of falls, not a decrease, but an increased risk of falls. ... vitamin D may actually exacerbate weakness and increase the risk of falls.” (C, [07:57])
• No Mortality Benefit
  “It did not reduce mortality in the vital trial or other large trials, even in those who were low. ... it does not support longevity or improve your chances of getting to 100.” (C, [08:48])

---

When Testing Might Be Appropriate

• True severe deficiency is rare, but in patients with classic symptoms or high-risk features (frailty + no sun exposure, malabsorption):
  • D testing may be useful as a diagnostic test when there is a specific clinical suspicion (e.g., osteomalacia, unexplained bone pain), not as routine screening. (C, [09:20], [10:09])
  • If clinical suspicion is high, may not need to test—reasonable to empirically treat. (C, [10:09])

---

The Case Against Routine Vitamin D Testing

• Tests Can Be Misleading and Variable

  • “From assay to assay,... you can get extremely divergent results whether you measure it in the winter or the summer. So to say, oh, your level was 18 and we gave you vitamin D and now it's 25, you cross the magic threshold is misleading.” (B, [11:27])
  • “It's a bad test for several reasons. ... the country spends nearly $300 million a year on vitamin D testing—number seven on the hit list of Medicare expenditures. We have... better things to do with our health care money than measuring vitamin D.” (C, [11:59])

• Potential for Overtreatment/Harm

  • “Telling a patient that they have deficiency on the lab test drives so many to buy and stay on vitamin D and even take higher doses than safe.” (C, [11:08])

---

Notable Quotes & Memorable Moments

• "Forget about all that vitamin D testing!!" (Episode title)
• “All of the large randomized placebo controlled trials have been negative for fracture falls and other conditions. It's the randomization that's key and the size of it allows you to look at subgroups. So it is the gold standard.” (C, [03:17])
• "Don't test and don't treat." (C, [06:43])
• “Vitamin D may actually exacerbate weakness and increase the risk of falls.” (C, [07:57])
• “It does not support longevity or improve your chances of getting to 100.” (C, [08:48])
• “It's a bad test for several reasons. ... We have better things to do with our health care money than measuring vitamin D.” (C, [11:59])

---

Key Timestamps

• 00:00 — Introduction & overview of the VITAL trial
• 01:57 — Dr. Cummings on main findings: vitamin D supplementation does not prevent fractures
• 03:17 — Discussion of why randomized trials are more reliable than observational data
• 04:13 — Vitamin D as a hormone and what 'deficiency' means physiologically
• 06:43 — Clinical scenarios: recommendations for healthy older adults vs. frail elderly
• 07:57 — Harms: increased risk of falls and weakness with vitamin D
• 08:48 — Vitamin D and lack of mortality benefit
• 09:20–10:09 — When (if ever) to test for vitamin D: true deficiency and osteomalacia
• 11:08 — Dangers of routine testing and overtreatment
• 11:27 — Test variability and cost implications
• 11:59 — Commentary on cost-effectiveness, closing remarks

---

Conclusion

The panel delivers a clear, evidence-based message: For most adults, routine vitamin D testing and supplementation do not prevent fractures or convey other health benefits, and may even cause harm by increasing fall risk. Only rarely, in cases of true deficiency or severe at-risk states (e.g., frailty, malnutrition, malabsorption), does vitamin D supplementation remain clinically justified—with or without confirmatory testing. Routine population-level testing is discouraged: it's costly, unreliable, and needlessly alarms patients.

Essential clinical takeaway:
“Forget about all that vitamin D testing!!”

Transcript

[A]

Do Vitamin D Supplements Prevent Fractures or anything else? This week the New England Journal of Medicine published results of a major randomized trial, the VITAL study, and the answer is overwhelmingly no. But there are some exceptions. To discuss the trial, we have one of the co authors of a fascinating editorial that appeared with this week's results. Stephen Cummings is a senior scientist who directs the Coordinating center for Sutter Health and is emeritus professor of Epidemiology at the University of California, San Francisco. Alan Brett will lead the discussion. He is professor of Medicine at the University of Colorado and editor in chief of the NEJM Journal Watch series. Welcome.

[B]

24 hours after the fracture outcome of the vital study has been published and there's already been a lot of media attention, some balanced and some not so balanced, and and I hope we can bring some solid clinical perspective to this topic. So I'm just going to start with a very brief description of the background of this study. It goes by the acronym VITAL involved 26,000 community dwelling, relatively healthy adults in their 50s or older who are randomized to receive 2,000 international units of vitamin D3 or placebo for five years. Now, listeners might recall that over the past few years several other outcomes have been already published. For example, vitamin D did not prevent cardiovascular events or cancer, and there have been several other outcomes that have been looked at as well. You might also recall that the study additionally randomized people to omega 3 supplements or placebo, but omega 3 supplementation did not influence the vitamin D fracture outcomes that were that we're going to discuss now. So let's get started and I'd like to ask Steve to tell us what the study showed both overall and in some of the interesting subgroups.

[C]

Simply, it was negative. There was no effect of that vitamin D supplementation on risk of any kind of fracture. And the subgroups that were looked at were a variety of them, including weight and race. But most importantly, they were able to look at the subgroup of people, people who had low vitamin D levels at the beginning. That's deficiency defined as less than 20. And there was no benefit in those who had low vitamin D levels to begin with. There was no reduction in fractures in that group.

[B]

And so other subgroups that they looked at were people with previous fragility fractures and people who are already taking osteoporosis medications. And what did those subgroups show?

[C]

There's no effect in any subgroup. And the study was large enough and being randomized, you can be sure that there were no benefits of vitamin D for almost any other characteristic that you

[B]

can think of now, this is one study, albeit a very large one, that comes up against the background of a lot of previous research, both randomized, observational and so forth. Why is this one more trustworthy, say, than previous research?

[C]

Previous research that found a benefit of vitamin D was observational. People with low levels tended to have poorer outcomes. Those were observational. All of the large randomized placebo controlled trials have been negative for fracture falls and other conditions. It's the randomization that's key and the size of it allows you to look at subgroups. So it is the gold standard.

[B]

As we were setting up this interview, you mentioned how important it is to think of vitamin D as a hormone and not a vitamin. And by virtue of being a hormone, it's involved with a lot of other homeostatic processes in the body that target various organs, not just bone. And I was wondering if you could just talk a little bit more about vitamin D as a hormone and also what the meaning of deficiency is in this context.

[C]

Yes, it's part of a highly regulated system of vitamin D as a pro hormone and125 vitamin D, which is what the body focuses on to maintain adequate concentrations of 125 which is the active hormone. And so that system is well regulated. Vitamin D itself, when you take it, is bound to the protein, vitamin D binding protein. When you measure. People measured vitamin D, they were actually measuring the complex of 25D and its binding protein. So you can see that it's not really a vitamin that's essential in your diet. It's part of a system that your body maintains, regulates very carefully.

[B]

Kind of circling back to the subgroup issue, you've given a pretty definitive position on that, but it's going to be on the minds of a lot of our clinician listeners and readers. So I just wanted to address that one more time. You've said that the sample size was sufficient that we can have confidence in some of those subgroups I mentioned. But let me circle back to that by giving you two common clinical scenarios. One would be, for example, a primary care clinician who's seeing a healthy 65 year old, and according to guidelines, let's say a 65 year old woman. According to guidelines, bone density testing is done and let's say the results come back with a T score of -3.0. She qualifies for osteoporosis, you talk to her and she and you decide to go ahead and start a bisphosphonate. Let's say the patient gets plenty of sunlight, has a good Diet, so forth and so on. It seems to me that there are three possible ways to go there. One is that you don't check a level and you don't give her vitamin D. You don't check a level and you just pick a dose and give her vitamin D, or you check a level and sort of let that guide you in some way. So that's the first one. And then the second one would be the frail elderly person who gets no sunlight, has a poor diet, in or out of a nursing home. Those are patients that many people screen who are not really well represented in this trial. So Both the healthy 65 year old on osteoporosis meds and the frail elderly person.

[C]

For the case number one, I'll choose answer number one, don't test and don't treat. In addition to this vital trial, there are good trials showing that vitamin D does not improve bone density, it does not improve bone structure, and therefore really there is no potential benefit of giving her vitamin D. And there really, since we know that the level doesn't matter in vital, there's no reason to measure vitamin D in that patient. Let's take on the second one about the patient who's in a nursing home. Poor diet, no sunlight. There was a trial long ago in France that showed that those patients benefited from the combination of about 1,200 units of vitamin D along with calcium supplementation. These people were not in the trial. It's possible that they have severe deficiency. You don't need to check a level. In that case, I think the consensus would be you give her vitamin D. 2000 units a day is reasonable, but along with calcium. So I think that that's reasonable practice in those really frail patients.

[B]

Would you like to talk about what we know about potential harms of vitamin D?

[A]

Sure.

[C]

I don't think it's well recognized that some of the trials, several of the trials, really well done trials, generally using somewhat higher doses showed an increased risk of falls, not a decrease, but an increased risk of falls. In addition, there's been a recent meta analysis of a number of studies showing that vitamin D tends to cause weakness. So that less able to do exercise, getting up out of a chair or other kinds of maneuvers that involved strength. So contrary to our impressions and old beliefs, vitamin D may actually exacerbate weakness and increase the risk of falls.

[B]

Some people talk about vitamin D and mortality, improving mortality. Steve, want to talk about that for a moment?

[C]

Sure. One of the most common ways it's marketed online as a pro longevity drug, enhancing longevity and many people take it because they believe it will support longevity. It did not reduce mortality in the vital trial or other large trials, even in those who were low. You can tell your patients it does not support longevity or improve your chances of getting to 100.

[B]

As we're getting to the end here, I just wanted to mention one thing to make sure people don't confuse what we're talking about here with true deficiency of vitamin D, which is probably pretty rare in the United States, but it happens. And in my practice over the years, I've seen several people, typically frail people with no sun exposure, somebody with malabsorption who had diffuse or focal bone pain, who ended up having imaging evidence of bone lesions which represented osteomalacia. And there are. That should be on our differential diagnosis of sick people with bone pain who might be at risk for true vitamin deficiency physiologically and not just according to some arbitrary cutoff. Is that a fair way?

[C]

Yes. Yeah. If it's uncertain what the cause of all those symptoms are, we suggested that it could be used as a diagnostic test, is not a screening test. But if you've got a clinical situation where osteomalacia or severe deficiency is one of the things to consider, by all means you can use D levels as a diagnostic test and look for levels less than 12 or 10. Absolutely. And however, in some cases where you believe that the patient is going to have is prone to have severe deficiency malabsorption syndromes, malabsorptions on top of frailty, then you probably don't need a level, just treat. So, absolutely, this trial did not address those patients at risk of severe deficiency who may have severe deficiency on testing and warrant treatment if found.

[B]

Thank you. I think that's a reasonable note to end on. Any other thoughts, Steve?

[C]

Well, I just really encourage doctors not to do vitamin D testing because telling a patient that they have deficiency on the lab test drives so many to buy and stay on vitamin D and even take higher doses than safe.

[B]

Yeah. Another thing I think that's not widely recognized is the fact that from assay to assay, studies have been done splitting samples and being and having 25 hydroxy D levels done in 5 or 10 different labs. You can get extremely divergent results whether you measure it in the winter or the summer. So to say, oh, your level was 18 and we gave you vitamin D and now it's 25, you cross the magic threshold is misleading.

[C]

It's a bad test for several reasons. And the country spends nearly $300 million a year on vitamin D testing number seven on the hit list of Medicare expenditures. We have other things, better things to do with our health care money than measuring vitamin D. Right.

[B]

So we'll end with this commentary on cost effectiveness in the health care system. Thank you for joining us. You've been a busy guy. Other media outlets have been contacting you these last 24 hours and. And we'll see how the country reacts to this. Thank you.

[C]

Thanks very much for having me.

[A]

That was the 297th edition of Clinical Conversations, all of which are available [e-mail] we come to you from the writers and editors of the NEJM Group. Our executive producer is Kristen Kelly. And I'm Joe Elia. Thank you for listening.