Biomarkers

25-OH Vitamin D Optimal Levels: Why the Thresholds Were Withdrawn

On this page
  1. What the 2024 guideline actually changed
  2. Where 20 and 30 ng/mL came from
  3. Your result is partly a property of the laboratory
  4. What happened when large trials supplemented people anyway
  5. So what do you do with a number you already have?

There is no current expert-endorsed optimal 25-OH vitamin D level, because in 2024 the Endocrine Society withdrew the thresholds it had published in 2011 and stopped recommending routine testing in generally healthy adults. The guideline's stated reason is blunt: there is no clear 25(OH)D level below which supplementation is required, and a single threshold may not be appropriate for an entire population. So if you are holding a lab report that says your level is "insufficient", the number that word is measured against is a number its own authors no longer stand behind.

Number you have probably seenWhere it came fromStatus now
20 ng/mL (50 nmol/L)Institute of Medicine, 2011: the level meeting the bone-health needs of at least 97.5% of the populationStill the basis of the US and Canadian reference intakes
30 ng/mL (75 nmol/L)Endocrine Society, 2011: the target for "sufficiency"Withdrawn in 2024. No longer endorsed
"Insufficient" below 30Lab reference ranges built on the 2011 targetMany labs still print it

Unit note, because it causes more confusion than anything else on the report: 1 ng/mL = 2.5 nmol/L (2.496, to be exact). A result of 32 ng/mL and a result of 80 nmol/L are the same result. If you have ever compared two tests and found a fourfold jump, check the units before anything else.

You could be asking this about yourself

has my vitamin D held up since the summer?

Connect BioTrakk to Claude, ChatGPT or Gemini. Log a reading in one sentence, and your own history becomes something you can ask about — in the chat you already have open.

Free. No BioTrakk account to make first. Claude takes one click.

What the 2024 guideline actually changed

The Endocrine Society's 2024 guideline is easy to misread in both directions, so here is what it does and does not say.

What it says: for generally healthy adults, the panel suggests against routine 25(OH)D testing, and against routine follow-up testing to guide dosing. It does not endorse a target level, and it does not endorse the 2011 definitions of deficiency, insufficiency and sufficiency. It suggests that healthy adults under 75 are unlikely to benefit from taking more vitamin D than the Institute of Medicine reference intake, which for adults aged 19 to 70 is 600 IU per day.

What it does not say: that vitamin D does not matter, or that nobody should be tested. The guideline covers generally healthy people. It explicitly excludes anyone with a condition that changes vitamin D absorption or metabolism, and it still points to groups where it suggests supplementing without testing first, among them adults aged 75 and over, pregnant people, children and adolescents, and adults with high-risk prediabetes.

The distinction worth holding on to is between treating a number and treating a person. The guideline did not decide that low vitamin D is fine. It decided that the number, measured in a healthy adult, was not reliably telling anyone what to do.

Where 20 and 30 ng/mL came from

Both numbers were produced by committees, three years apart, reading overlapping evidence, and they disagreed by 50%.

The IOM's 20 ng/mL was never meant to be a personal target. It is a population figure: the serum level at which the committee judged the skeletal needs of at least 97.5% of people to be met. Read it the way it was written and it says that almost everybody is fine at 20 ng/mL. Read it as a cut-off for individuals, which is how it is usually used, and it inverts into a claim that everybody below 20 has a problem. That is a different statement, and the report does not make it.

The Endocrine Society's 2011 figure of 30 ng/mL went further and became the default on lab reports, which is why so many people have been told they are "insufficient". Withdrawing it in 2024 does not mean 30 ng/mL was harmful advice. It means the evidence that getting an otherwise healthy adult from 24 to 32 changes anything they would notice did not hold up.

A practical consequence: the word on your report is doing less work than it appears to. "Insufficient" is a comparison against a reference range your lab chose, not a diagnosis.

Your result is partly a property of the laboratory

This is the part that gets left out, and it is the most useful thing on this page if you intend to track the marker over time.

25(OH)D is hard to measure, and different assays give different answers on the same blood. In a study of 494 patients measured by two methods, one assay read on average about 5 ng/mL lower than the other. That sounds small until you apply a threshold to it. At a 20 ng/mL cut-off, 52% of the same patients needed treatment by one assay and 36% by the other. At 30 ng/mL it was 79% against 64%. The two assays agreed on an individual's category in only 41% to 68% of cases, depending on how agreement was defined.

Nobody's vitamin D changed. The machine did. The authors' conclusion was that a single universal threshold may not be adequate to define vitamin D status, and that a laboratory result may not reflect a patient's actual level.

Standardisation efforts have narrowed this, and have not closed it: a 2020 review of assay standardisation describes meaningful variability persisting among the most widely used assays even after recalibration to reference methods, with the bias in several immunoassays depending on the concentration itself.

So the one rule that survives all of this: if you are comparing two of your own results, they have to come from the same lab and the same assay, or you are not comparing them at all.

What happened when large trials supplemented people anyway

Two trials are worth knowing about, because together they explain why the guidance moved.

VITAL randomised 25,871 US adults to 2000 IU of vitamin D3 per day or placebo, with a median of 5.3 years of treatment, and did not select people for low levels. It found no significant reduction in total invasive cancer (hazard ratio 0.96, 95% CI 0.88–1.06) and none in major cardiovascular events (0.97, 0.85–1.12). Both intervals comfortably include 1, which is the statistical way of saying the trial could not distinguish the result from no effect.

D-Health randomised 21,315 Australians aged 60 and over to 60,000 IU of vitamin D3 monthly or placebo for five years, again without screening for low levels, and found no reduction in all-cause mortality.

Read these carefully, because they are narrower than the headlines they generated. They tested supplementing unscreened populations, most of whom were not deficient to begin with. A trial that gives extra vitamin D to people who already have enough is, in most cases, testing whether surplus helps, and the answer was no. Neither trial shows that correcting genuine deficiency does nothing, and neither was designed to.

That is the honest shape of the evidence: strong against supplementing by default in healthy adults, much thinner on where any individual's floor actually is.

So what do you do with a number you already have?

If a test has already been run, or your clinician has a reason to run one, a few things make the result more useful than it would otherwise be:

  • Write down the lab, the assay and the units. Without them a future comparison is guesswork. This is the single highest-value thing on this list, and the easiest to skip.
  • Record the season and the date. 25(OH)D has a large seasonal swing in anyone living far from the equator. A February result and an August result from the same person are not the same measurement, and comparing across seasons without noting it is the most common self-tracking error with this marker.
  • Treat a value near a threshold as near a threshold. Given assay differences of around 5 ng/mL, a result of 29 and a result of 31 are not meaningfully different, whatever word the report prints next to them.
  • Keep the trend, not the point. Three results across two years, same lab, same season, tell you something a single result never can.
  • Take any intake decision to a clinician. This page deliberately contains no dose. What to take, if anything, depends on your history, your other medications and your calcium, and it is not a question a reference range answers.

In practice the failure is logistical rather than medical: the results arrive as PDFs months apart, the lab changes when the clinic changes, the units change with it, and nobody records which assay produced which number. Logging each panel in BioTrakk (photograph the lab report or type the values in) keeps one trend line per marker with the conditions attached, so a real move is distinguishable from a change of laboratory. The blood test tracker page describes how that storage works, and if you want to test something against your own baseline rather than a population average, that is what an n-of-1 experiment is for: BioTrakk reports "not distinguishable from chance" when that is the answer, which on this marker it often will be. Where a supplement sits in a daily routine, separately from whether to take one, is covered in the supplement timing chart.

BioTrakk is not a medical device and this is not medical advice. 25-OH vitamin D is a clinical laboratory test: interpreting yours, and deciding anything on the basis of it, is a conversation with a clinician. Nothing here diagnoses, treats or prevents any condition, and this page recommends no dose.

Frequently asked questions

What is the optimal 25-OH vitamin D level?

As of 2024 there is no expert-endorsed optimal level for generally healthy adults. The Endocrine Society withdrew its 2011 targets and no longer endorses the levels that defined sufficiency, insufficiency or deficiency, on the grounds that there is no clear level below which supplementation is required and that one threshold may not suit an entire population. The Institute of Medicine figure of 20 ng/mL (50 nmol/L) remains in use, but it describes the bone-health needs of at least 97.5% of a population rather than a personal target.

Why does my lab still say I am insufficient below 30 ng/mL?

Because many laboratory reference ranges were built on the Endocrine Society 2011 target of 30 ng/mL, which was withdrawn in 2024. Reference ranges on reports are not updated in step with guidelines. The word on the report is a comparison against the range your lab chose, not a diagnosis.

Can two labs give different vitamin D results from the same blood?

Yes, and the difference can change your category. In a study of 494 patients measured by two assays, one read about 5 ng/mL lower on average than the other; at a 20 ng/mL cut-off, 52% of the same patients fell below it by one assay and 36% by the other. The two assays agreed on an individual category in only 41% to 68% of cases. Compare your own results only within the same lab and assay.

How do I convert ng/mL to nmol/L?

Multiply by 2.5 (precisely 2.496): 30 ng/mL is 75 nmol/L, and 20 ng/mL is 50 nmol/L. To go the other way, divide by 2.5. The United States typically reports ng/mL while most of Europe reports nmol/L, which is why the same person can appear to have two very different results.

Did the large trials show vitamin D supplements do not work?

They showed that supplementing unscreened adults who were mostly not deficient did not reduce the outcomes measured. VITAL gave 2000 IU daily to 25,871 adults for a median 5.3 years and found no significant reduction in invasive cancer or major cardiovascular events; D-Health gave 60,000 IU monthly to 21,315 adults aged 60 and over for five years and found no reduction in all-cause mortality. Neither trial was designed to test whether correcting genuine deficiency helps, so neither answers that question.

Should I get my vitamin D tested?

The 2024 Endocrine Society guideline suggests against routine testing in generally healthy adults, including those with obesity or darker skin. It does not apply to people with conditions affecting vitamin D absorption or metabolism, and it identifies groups where supplementation is suggested without testing, including adults aged 75 and over, pregnant people, children and adolescents, and adults with high-risk prediabetes. Whether you are in one of those groups is a question for your clinician, not for a reference range.

Track it. Test it. Know for sure.

Log food, supplements, sleep and biomarkers by chatting with BioTrakk on Telegram, WhatsApp or the web — then let it find what actually moves your numbers.

Free on Telegram & Web. No app to install.