Vitamin D test: what your number means and who actually needs one

Short answer

A vitamin D test measures 25-hydroxyvitamin D, written 25(OH)D, in your blood. The catch is that organizations disagree on the cutoff. US nutrition authorities call 20 ng/mL and above adequate (4, 6); the 2011 Endocrine Society guideline called anything under 30 insufficient (3); the 2024 one dropped thresholds entirely and suggests against routine testing in healthy adults (1).

If you have read three pages and come away with three different verdicts on the same number, you're not imagining it and it is not your fault. The cutoffs really do disagree, because they were written by different committees answering different questions, and nobody has gone back and merged them. This article lays the numbers side by side with their dates, so you can see which rule your result is being judged by.

Note: this article explains what guidelines and research say and is not medical advice; your clinician decides whether you need a vitamin D test and what to do with the result. Vitamin D toxicity comes almost always from supplements, not from food or sunlight: if you are taking high-dose vitamin D and develop nausea and vomiting, loss of appetite, constant thirst with heavy urination, confusion or muscle weakness, stop and get medical advice, and call 911 for a seizure, fainting or an irregular heartbeat (6). Deep aching bone pain, muscle weakness around the hips and thighs, or a second fracture from a minor fall is a reason to see a doctor promptly rather than to start a supplement and wait (6, 7).

Vitamin D test thresholds side by side: who says what, and when they said it

Every number below is real, current somewhere, and published by a serious body. They do not agree with each other. All values are serum 25(OH)D in ng/mL.

The disagreement is not sloppiness. Each group was answering a different question.

The Institute of Medicine panel was setting dietary intakes for a whole population and asked: what blood level covers bone health for nearly everyone? Its answer, in its own words, is that "the vitamin D–related bone health needs of approximately one-half of the population may be expected to be met at serum 25OHD concentrations between 30 and 40 nmol/L (12 and 16 ng/mL)," that most of the rest are covered between 16 and 20 ng/mL, and that "practically all persons are sufficient at levels of 50 nmol/L (20 ng/mL) and above" (4).

The Endocrine Society in 2011 was writing for clinicians treating individual patients who already had a reason to be tested, and it aimed higher, at 30 ng/mL, to leave a margin (3). AACE kept that bar in 2020, but only inside a narrow context: people with postmenopausal osteoporosis who are about to start bone drugs (7). A bar written for a 68-year-old with a spine fracture was never meant to judge a healthy 32-year-old's routine panel, but the number travels anyway.

What does a vitamin D level of 22 ng/mL actually mean?

Take one result: 25(OH)D of 22 ng/mL, drawn in November.

• By the Food and Nutrition Board table used by the NIH Office of Dietary Supplements, 22 is adequate — above the 20 ng/mL line (6).

• By the 2011 Endocrine Society guideline, 22 is insufficiency, inside the 21–29 band (3).

• By AACE's 2020 osteoporosis guideline, if you have osteoporosis, 22 is below the target of 30 and a reason to replete before starting treatment; if you don't have osteoporosis, that guideline isn't about you (7).

• By the 2024 Endocrine Society guideline, 22 has no verdict attached, because the panel concluded that "25(OH)D levels that provide outcome-specific benefits have not been established in clinical trials" (1).

Same blood, four readings. And this is not an exotic corner case: in NHANES data from 2001 to 2018, 40.9% of US adults had a 25(OH)D between 21 and 30 ng/mL — precisely the band where the verdict flips depending on which document your clinician reaches for. Another 22% were between 10 and 20, 2.6% were below 10, and 34.5% were above 30 (13).

So when one doctor glances at 22 and says "that's fine" and another says "you're deficient, start supplements," they may both be reading their own guideline correctly. What has changed is not your body. It is the ruler.

Why did the Endocrine Society stop recommending the vitamin D test?

This is the biggest turn in the field, and most pages describing "normal vitamin D levels" have not caught up with it.

In June 2024 the Endocrine Society published Vitamin D for the Prevention of Disease, replacing its own 2011 guideline (1). The same organization that popularized the 30 ng/mL target now suggests against routine 25(OH)D testing in essentially every population it examined: adults under 50, adults 50 to 74, adults 75 and older, people who are pregnant, people with dark complexion, and people with obesity (1). Its press release puts it plainly: "We suggest against routine testing for 25-hydroxyvitamin D levels in any of the populations studied," and it specifies that this "includes 25-hydroxyvitamin D screening in people with dark complexion or obesity" (2).

Three things about that recommendation are worth holding onto, because they are easy to over-read in either direction.

It is a weak recommendation on very low certainty evidence. The guideline labels each of these testing statements that way (1). "Suggest against" is not "forbid." It means that in the panel's judgement most informed people would decline the test, and that a clinician with a specific reason can still order one.

It is not a statement that vitamin D doesn't matter. The same guideline suggests empiric supplementation — supplementation without testing first — for children and adolescents 1 to 18, for adults 75 and older, during pregnancy, and for adults with high-risk prediabetes (1). The shift is in what triggers action. The trigger used to be your number. Now, for these groups, the trigger is the group you are in.

The thresholds were not raised or lowered. They were withdrawn. The panel could not identify blood levels tied to specific outcomes, so it declined to publish any (1, 2). That is why you can no longer look up "the Endocrine Society normal range" for 2026 and find one. The absence is deliberate.

Two other bodies had already arrived in roughly the same place from different directions. The US Preventive Services Task Force gave screening for vitamin D deficiency in asymptomatic, community-dwelling, nonpregnant adults a Grade I in April 2021 — the evidence is insufficient to weigh benefits against harms — which is the same conclusion it reached in 2014 (5). As of October 2026 that 2021 statement is still the current one; the Task Force's active vitamin D work is a separate review of supplementation to prevent falls and fractures, not of screening. And back in February 2014, the American Society for Clinical Pathology put population-based 25(OH)D screening on its Choosing Wisely list, reasoning that for most otherwise healthy people the test does not change what happens next (8).

Meanwhile the test kept being ordered. A 2020 US primary-care study, gathering earlier estimates, reports more than a billion 25(OH)D tests a year worldwide, puts the share following a pattern of untargeted screening rather than targeted testing of high-risk patients somewhere between a quarter and three quarters depending on the study, and cites roughly 6 million tests classed as low-value among commercially insured and Medicare-insured Americans in a single year, at a cost over $800 million (14). That gap — a test ordered by the hundred million, and guidelines that mostly say don't — is the real reason the same number comes back with different verdicts.

Why your vitamin D result depends on which lab ran it

The second thing nobody tells you at the blood draw: 25(OH)D is a genuinely hard molecule to measure, and two labs testing the same tube can hand you numbers far enough apart to land on opposite sides of a cutoff.

The NIH Office of Dietary Supplements says it directly: measurement is complicated by "considerable variability of the available assays," so a result "can be falsely low or falsely high, depending on the assay used and the laboratory" (6).

How big is the spread in practice? A 2025 study run under the IFCC–IOF Committee on Bone Metabolism compared routine immunoassays against mass spectrometry across multiple laboratories (9). A few findings worth knowing before you compare two printouts:

• The share of individual samples a method got within ±5% of the reference value ranged from 0% for some immunoassay laboratories up to 87% for the best LC-MS/MS method.

• One laboratory showed a bias above 60% on a routine immunoassay.

• Two laboratories running the same manufacturer's assay differed ninefold in bias — so it is not only the method, it is the individual lab and even the reagent lot.

• Only slightly more than half of the methods met the desirable target for measurement uncertainty, of 10% or less.

• The authors' summary is that the work "highlights the persistent variability in 25-(OH)D assay performance despite ongoing standardization efforts" (9).

There are two common methods. Immunoassays are the automated workhorses in most hospital and commercial labs: fast, cheap, and the most variable. LC-MS/MS — liquid chromatography with tandem mass spectrometry — separates the molecule physically before measuring it and is the reference approach; in that 2025 comparison both mass spectrometry methods met every performance criterion (9). An international Vitamin D Standardization Program exists to pull assays into line, and the ODS notes that most of the older studies linking vitamin D to health outcomes used unstandardized measurements, which muddies their conclusions too (6).

Units add a second layer of confusion. US labs usually report ng/mL; much of the rest of the world reports nmol/L. One ng/mL equals 2.5 nmol/L (6). So "my level is 60" is alarming-sounding in ng/mL and unremarkable in nmol/L, where it works out to 24 ng/mL. Before you panic or relax, check the unit printed on the report.

What this means for you, practically: a difference of a few ng/mL between two results is not reliable evidence that anything changed. If you are tracking your level over time, use the same lab and the same method, keep the printouts, and treat small moves as noise. This is the same discipline that makes any repeated measurement readable, whether it is a cortisol test or a number from a wearable.

Total vs free 25(OH)D, and the unsettled argument about skin tone

Most 25(OH)D in your blood is not floating free. It travels bound to a carrier called vitamin D–binding protein (DBP), and a standard test measures the total — bound plus free. The question that has been argued for over a decade is whether total is the right thing to measure for everyone.

In 2013, a study in the New England Journal of Medicine reported that Black American participants had markedly lower total 25(OH)D than White participants (15.6 vs 25.8 ng/mL) but also much lower DBP (168 vs 337 µg/mL), so their estimated bioavailable 25(OH)D came out essentially the same (2.9 vs 3.1 ng/mL). Their parathyroid hormone was only slightly higher and their femoral-neck bone density was higher, not lower. The authors calculated that between 77% and 96% of the Black participants would be labelled deficient by a 20 or 30 ng/mL cutoff, and argued that this label did not match their bones (10).

That finding was widely repeated — and then substantially challenged. A 2016 study in the Journal of Clinical Endocrinology & Metabolism measured DBP with several different assays in the same samples. The racial difference appeared only with the monoclonal assay used in the earlier work, and the authors attributed it to that assay reading one common genetic variant of the protein poorly, a variant frequent in people of African ancestry. With polyclonal and proteomic methods there was no meaningful difference. Their conclusion was that "Previously reported racial differences in DBP concentration are likely from monoclonal assay bias," that free 25(OH)D was in fact lower in African American participants too, and that total 25(OH)D remains a valid marker of status regardless of race (11).

Where that leaves things in 2026, honestly: the question is open, free 25(OH)D is not a routine clinical test, and the practical guidance has moved past it. The 2024 Endocrine Society guideline addresses people with dark complexion directly and suggests against routine 25(OH)D screening for them — not because their levels are reassuring, but because the panel found no evidence that screening changes outcomes (1). What is not in dispute is the mechanism behind lower readings: more melanin reduces how much vitamin D the skin makes from sunlight, and Black Americans do typically have lower serum 25(OH)D, while the health consequences of that remain unclear (6).

If you have darker skin and a low number, the useful conversation with a clinician is not about the label. It is about whether anything in your history — bone pain, fractures, malabsorption, a bone-density result — makes the number worth acting on.

Is it normal for vitamin D to drop in autumn and winter?

Yes, and the size of the drop is bigger than most people expect. This is ordinary physiology, not a developing illness.

In the Cardiovascular Health Study, researchers fitted seasonal curves to 25(OH)D measurements in older US adults. The average level across the year was 25.1 ng/mL, and the average peak-to-trough difference was 9.6 ng/mL — large relative to the mean. The swing was bigger for men, bigger at study sites farther north, and bigger in people who were more physically active, presumably because they were outdoors more in summer. Parathyroid hormone and bone-specific alkaline phosphatase cycled too, in the opposite direction, which is what you would expect if the seasonal change in 25(OH)D is biologically real rather than a measurement artefact (12).

Put that next to the cutoffs and the consequence is uncomfortable. A person whose summer peak is 28 ng/mL and whose winter trough is 18 crosses the 20 ng/mL line every year without anything going wrong. Tested in August, they are "adequate" by every table in this article. Tested in February, the same person is "deficient" by the 2011 Endocrine Society definition and "inadequate" by the Food and Nutrition Board's. Nothing about them changed except the date of the blood draw.

Three things follow:

1. Write the date on the result. A number without a month is close to uninterpretable.

2. Don't compare a February result with an August one and call the difference a trend.

3. A seasonal dip is not, by itself, a diagnosis. Which is exactly why guideline panels have become cool on routine testing: a test that labels a large share of a northern population deficient every winter, without evidence that acting on the label improves anything, is a test doing little work (1, 5).

Autumn is also the season when people start feeling flatter and start looking for a blood-test explanation. If that is what brought you here, it is worth reading about what actually causes persistent fatigue and about how thyroid problems get separated from everything else that looks like them before settling on one vitamin.

Who actually needs a vitamin D test?

The guidelines converge here, even where they disagree about cutoffs: test when a result would change what happens next.

Groups where testing is commonly considered appropriate, drawn from the 2011 Endocrine Society guideline's at-risk framing, the AACE osteoporosis guideline, the ASCP Choosing Wisely rationale, and the NIH ODS list of groups at risk of inadequacy (3, 6, 7, 8):

• Osteoporosis, osteomalacia or a fragility fracture — AACE recommends measuring in people at risk, especially with osteoporosis, and confirming status before starting bone medication (7).

• Conditions that limit fat absorption — celiac disease, Crohn's disease, ulcerative colitis, cystic fibrosis, some liver disease. Vitamin D is fat-soluble, so absorption fails with the fat (6).

• After gastric bypass surgery, because the part of the small intestine that absorbs vitamin D is bypassed (6).

• Chronic kidney disease and other conditions that change vitamin D metabolism (8).

• Medicines that affect vitamin D metabolism, such as anticonvulsants and glucocorticoids (8).

• Very limited sun exposure — homebound or institutionalized people, people who cover for religious reasons, people working entirely indoors (6).

• Unexplained high calcium, abnormal parathyroid hormone, or unexplained bone pain and proximal muscle weakness — here the test is part of working out a specific problem, not a screen.

And where the current advice is "probably not": a healthy adult with no symptoms, no bone diagnosis, no malabsorption and no medication complication, who simply wants to know. The 2024 Endocrine Society guideline suggests against it; the USPSTF says the evidence doesn't support a verdict either way; ASCP says don't do it as population screening (1, 5, 8).

That is advice about the test, not about vitamin D intake. The Food and Nutrition Board's Recommended Dietary Allowance is 600 IU (15 mcg) a day for adults 19 to 70 and 800 IU (20 mcg) over 70, and the Tolerable Upper Intake Level for adults is 4,000 IU (100 mcg) a day (6). What dose, if any, suits you is a conversation with your own clinician — as it is with any supplement, including the ones marketed for sleep, where the evidence is thinner than the packaging suggests.

Living with a borderline vitamin D result: the questions people ask next

My result is 22 ng/mL and my doctor said it's fine. Was that wrong?

Probably not wrong — just read against a different ruler than the one you found online. By the Food and Nutrition Board categories the NIH uses, 20 ng/mL and above is adequate for bone and overall health, so 22 is inside the normal band (6). By the Endocrine Society's 2011 definition, 21–29 was labelled insufficiency (3), and that definition is what most consumer pages still quote. The 2024 Endocrine Society guideline withdrew thresholds altogether (1). If you want a concrete answer, the productive question for your doctor is not "is 22 normal" but "given my history, would a different number change anything you'd do?"

Should I retest, and how soon?

That depends entirely on why you were tested. If you have a bone diagnosis or a malabsorption condition and you are being repleted, your clinician sets the interval, often after two to three months of treatment (7). If you were tested out of curiosity and nothing is being treated, repeating the test mostly buys you a second number from the same uncertain measurement, possibly on a different assay, possibly in a different season (9, 12). If you do retest, use the same laboratory and note the month.

Does feeling exhausted mean my vitamin D is low?

Fatigue is the most common reason people ask for this test and one of the least specific symptoms in medicine. Trials of vitamin D supplementation in the general population — including people with low baseline levels — have not shown consistent benefit for fatigue, and the same is true for falls, fractures, cancer, dementia, depression and respiratory infections (13). That doesn't mean your tiredness isn't real. It means a 25(OH)D result is unlikely to explain it on its own, and that things like short or broken sleep, thyroid disease and iron status deserve a place in the same conversation — see why eight hours can still leave you tired.

Is a direct-to-consumer vitamin D test worth paying for?

Two things make it a weak purchase for a healthy adult. First, the result arrives without the clinical context that decides what it means, and the thresholds printed on the report are the vendor's choice among several competing ones. Second, the assay underneath is usually an immunoassay, which is where most of the measured variability lives (9). If you have a reason to be tested — a bone diagnosis, malabsorption, a medication — the same money is better spent on a test ordered by the clinician who will act on it.

I already take a vitamin D supplement. Does that change how I read the result?

Yes, in the obvious direction: a result taken while supplementing reflects the supplement. Tell your clinician the dose and how long you have been on it, because a "normal" level on 2,000 IU a day says something different from the same number with no supplement at all. It also matters for the upper end — the adult Tolerable Upper Intake Level is 4,000 IU a day, and routine intakes above it are where risk starts to accumulate (6).

Can you take too much vitamin D?

You can, and it comes from supplements rather than sun or food. Toxicity involves marked hypercalcemia — too much calcium in the blood — and 25(OH)D levels typically above 150 ng/mL (375 nmol/L). Hypercalcemia can cause nausea, vomiting, muscle weakness, neuropsychiatric disturbances, pain, loss of appetite, dehydration, excessive thirst and heavy urination, and kidney stones; in extreme cases, kidney failure, soft-tissue calcification, cardiac arrhythmias and death (6). The Food and Nutrition Board also flags levels above 50 ng/mL as a possible concern, with adverse effects particularly above 60 ng/mL (4, 6). If you are on a high dose and feel unwell in these specific ways, that is a call-your-doctor situation, not a wait-and-see one.

Does my vitamin D number tell me anything about my bones?

Less than the marketing implies, and more in some contexts than others. Severe deficiency, below 12 ng/mL, is linked to rickets in children and osteomalacia in adults (6). Above that, a single 25(OH)D value is a poor stand-in for bone health: the direct measure is a bone-density scan, and that is where the AACE pathway starts for people at risk (7). If bone loss is the actual worry — around menopause, for example — the risk factors and the tests that matter are a better place to start than a vitamin level.

Does a sunny holiday or a winter indoors really move the number?

Enough to cross a cutoff. The seasonal peak-to-trough difference averaged 9.6 ng/mL in one large US cohort, and it was larger at higher latitudes and in people who spent more time active outdoors (12). Season, time of day, day length, cloud cover and skin melanin all affect how much vitamin D the skin makes (6). This is one more reason to date every result and to distrust comparisons made across seasons.

How to bring this up with your doctor

What to ask for, by name. If you think you have a reason to be tested, ask for a serum 25-hydroxyvitamin D (25(OH)D) test, and say why: a fracture from a minor fall, a malabsorption diagnosis, bariatric surgery, a long course of glucocorticoids or anticonvulsants, almost no sun exposure. Naming the reason is what separates a targeted test from the untargeted screening every major body advises against (1, 5, 8).

What to bring. Previous 25(OH)D results with their dates and the lab that ran them; your current supplement dose and how long you have taken it; the months of the year your results were drawn. If you track your days, bring a dated written record of how you have actually been — sleep, energy, the weeks it got worse — rather than a general impression.

What to say, out loud. "My result was [X] ng/mL, drawn in [month], at [lab]. I've read that one guideline calls that insufficient and another calls it adequate. Which one are you using, and would a different number change what you'd do for me?"

What to ask next. "Do I have a reason to be tested at all, or is this a number without a decision attached?" · "Should we be looking at calcium or parathyroid hormone instead?" · "If we supplement, when would we recheck, and at which lab?" · "Could something else explain how I feel — thyroid, iron, sleep?"

If you're brushed off and you still feel unwell. Ask for the reasoning to be written in your notes, and ask the more specific question: not "can we check my vitamin D," but "what would explain this symptom, and what would rule it in or out?" A borderline vitamin level is a weak explanation for a strong symptom, and a clinician who declines to chase the vitamin may still agree that the symptom needs chasing. Bone pain, muscle weakness in the hips and thighs, or a second low-impact fracture deserves a prompt appointment regardless of what any vitamin number says.

Start understanding your body

Learn what affects your energy, stress, sleep, and daily state. Get the app.

Explore your data

How Welltory helps — and what it cannot do

The limits first, in plain terms. Welltory is a general wellness product, not a medical device. It does not diagnose, detect, predict, monitor, prevent, treat or mitigate any condition. It does not measure vitamin D and it cannot find a vitamin D deficiency — no wearable or phone can. Nothing in the app substitutes for a blood test, and the app's measurements may be less accurate for people under 18 and during pregnancy. If a number in Welltory and a symptom disagree, the symptom goes to a clinician.

What it can do is give the vague complaint a dated shape. Fatigue is the single most common reason people go and get a vitamin D test, and it is also the least specific symptom there is. One blood value from one morning cannot tell you whether anything actually changed in the weeks you felt worse. A dated record of your own sleep, resting heart rate, heart rate variability (HRV) and stress minutes can at least show whether something moved in those same weeks — and that is usually a more useful thing to bring to an appointment than a single number.

You note what happened; the app records the rest. Add a tag from the plus sign at the top of the feed — "wiped out," "slept badly," "started vitamin D," "flat since the clocks changed." When the Today screen marks a stress or rest stretch, it asks "What happened?"; tap a suggested tag, type, or talk, and fix the tags before saving.

Then look at the weeks, not the day. Sleep analysis shows how your heart rate behaved overnight; resting heart rate, HRV, stress minutes and Battery are recorded the same way every day, which is what makes them comparable to your own history (Battery and stress minutes need iOS with an Apple Watch or Oura; detailed sleep analysis needs an Apple Watch). Autumn is a season when several things shift at once — less daylight, later bedtimes, more indoor time — and a dated series separates "this started in October" from "this has been going on since spring."

Over a couple of months, My Patterns shows what your tags have in common. My Patterns collects tags from stress and rest stretches (iOS with an Apple Watch or Oura) and needs roughly seven tagged events in a month before patterns appear, with history from the month before. Two bad weeks won't produce one.

A record you made, for the appointment. The Journal (Premium) keeps your measurements, tags and notes together; from the web app you can export a CSV (Dashboard → choose a chart → Export). A dated line like "tired every afternoon since early October, sleeping 7h, resting heart rate unchanged" is concrete in a way that "I've been exhausted" is not. Whatever shows up there is an association to discuss with your doctor, not a cause — and certainly not a vitamin D level.

The red flags stay the same size. Nausea and vomiting, loss of appetite, constant thirst with heavy urination, confusion or muscle weakness while taking high-dose vitamin D needs medical advice now, and a seizure, fainting or an irregular heartbeat means 911 (6). Bone pain or a second fracture from a minor fall needs a doctor, whatever your app shows.

How we made it

Made with AI tools, then edited and fact-checked by the Welltory team. See our Editorial & AI policy.

Data analysis by Jane Smorodnikova, co-founder of Welltory and the person who built the methodology behind how we read physiological data.

Written by Tatsiana Yashyna.

Discounts for blog readers: up to 36% off

See what affects your energy, stress, sleep, and daily state with Welltory

Explore your data