{"slug":"vitamin-d","name":"Vitamin D (25-OH)","aliases":["25-hydroxyvitamin D","25(OH)D","Calcidiol","Vitamin D level"],"kind":"biomarker","system":"evolutionary-medicine","unit":"ng/mL","unitNote":"Some laboratories report nmol/L; multiply ng/mL by 2.5 for nmol/L.","answer":"The evidence is firm that vitamin D below about 12 to 20 ng/mL harms bone and keeps company with higher mortality, and equally firm that randomized trials which raised already-replete adults from about 30 to 40 ng/mL or more prevented no cancers, heart attacks, fractures or deaths. Cohort risk sat lowest at about 20 to 36 and rose again above about 48 to 56, so no study defines an optimal vitamin D above the point where deficiency ends.","summary":"25-hydroxyvitamin D measures the body's vitamin D supply from sun, food and supplements. Laboratories print about 30 to 100 ng/mL as normal, the Institute of Medicine says about 20 covers nearly everyone's bone needs, and the large trials of the past decade found that adding vitamin D to replete adults changed no major outcome. The literature is clear at the low end and quiet above it.","whatItIs":[{"text":"Vitamin D is made in skin from ultraviolet light and taken in from a few foods and from supplements; the liver converts it to 25-hydroxyvitamin D, which is the measurement that defines vitamin D status. The active hormone, 1,25-dihydroxyvitamin D, is a different test: the 2011 Endocrine Society guideline recommends against using it to assess vitamin D status.","cites":["sempos2018","holick2011"],"grade":"A"},{"text":"Its established job is skeletal. The Institute of Medicine's 2011 review found the evidence for bone health consistent with cause and effect, and the evidence for cancer, cardiovascular disease, diabetes and autoimmune disease inconsistent, inconclusive as to causality and insufficient to set a requirement. Under the consensus of an international assay conference, values below 12 ng/mL carry a risk of rickets and osteomalacia.","cites":["ross2011","sempos2018"],"grade":"A"},{"text":"Low values are common. In a nationally representative US sample of 4,495 adults, 41.6 percent had a 25-hydroxyvitamin D at or below 20 ng/mL, and among 3.8 million laboratory results from US adults, 33 percent were below 20 and 60 percent below 30, with the share below 20 swinging from 21 percent to 48 percent across the year.","cites":["forrest2011","kroll2015"],"grade":"B"}],"labRange":{"typical":"About 30 to 100 ng/mL printed as normal by most US laboratories, with under 20 flagged as deficient and 20 to 29 as insufficient","claims":[{"text":"The printed range comes from the 2011 Endocrine Society guideline, which defined deficiency as below 20 ng/mL, insufficiency as 21 to 29 and sufficiency as 30 to 100. The same guideline found no sufficient evidence to screen people who are not at risk of deficiency, or to prescribe vitamin D for cardiovascular protection.","cites":["holick2011"],"grade":"A"},{"text":"The Institute of Medicine reached a lower threshold from the same literature: about 16 ng/mL meets the needs of half the population and about 20 meets the needs of at least 97.5 percent, based on bone health. It judged that values above 30 are not consistently associated with more benefit, that risks have been identified for some outcomes above 50, and that the prevalence of inadequacy in North America had been overestimated by cutoffs that exceed these levels.","cites":["ross2011"],"grade":"A"},{"text":"The 2024 Endocrine Society guideline, which replaces the 2011 one, found no clinical trial evidence to support routine 25-hydroxyvitamin D testing in the general population, including people with obesity or dark skin, and no clear evidence defining an optimal target level for disease prevention; it suggests against routine testing and against supplementing healthy adults under 75 above the dietary reference intake to lower disease risk.","cites":["demay2024"],"grade":"A"}]},"evidence":{"headline":"Lowest risk at about 20 to 36 ng/mL in the large cohorts, higher risk below about 20 and again above about 48 to 56; the randomized trials that raised replete adults from about 30 to 40 ng/mL or more found no fewer cancers, cardiovascular events, fractures or deaths.","scale":{"min":0,"max":100,"ticks":[0,12,20,30,40,50,60,80,100]},"bands":[{"label":"Common laboratory reference range","low":30,"high":100,"tone":"lab","source":"Typical US laboratories; Endocrine Society 2011"},{"label":"Rickets and osteomalacia risk","low":0,"high":12,"tone":"caution","source":"Sempos 2018 consensus"},{"label":"Below the level that meets 97.5 percent of bone needs; higher mortality in cohorts","low":12,"high":20,"tone":"caution","source":"IOM 2011; Sempos 2013; Durup 2012"},{"label":"Lowest mortality in the Copenhagen cohort","low":20,"high":24,"tone":"evidence","source":"Durup 2012"},{"label":"Lowest mortality in NHANES, nadir about 32","low":24,"high":36,"tone":"evidence","source":"Sempos 2013"},{"label":"Safe and sufficient for bone by consensus; no outcome trial shows benefit above 30","low":30,"high":50,"tone":"neutral","source":"Sempos 2018; Bolland 2018; VITAL; D-Health"},{"label":"Higher mortality in cohorts above about 48 to 56; IOM flags risk above 50","low":50,"high":100,"tone":"caution","source":"Sempos 2013; Durup 2012; IOM 2011"}],"claims":[{"text":"VITAL randomized 25,871 US adults, men 50 and older and women 55 and older, to 2,000 IU of vitamin D3 a day or placebo for a median of 5.3 years. The mean baseline level was about 31 ng/mL and the vitamin D group rose to about 42 at one year. Invasive cancer (hazard ratio 0.96), major cardiovascular events (0.97) and death from any cause (0.99) did not differ, and the effect did not vary by baseline level, including the 12.7 percent who started below 20.","cites":["manson2019"],"grade":"A"},{"text":"D-Health randomized 21,315 Australians aged 60 and older to 60,000 IU of vitamin D3 a month or placebo for five years, lifting the mean level from about 31 to about 46 ng/mL. All-cause mortality did not fall (hazard ratio 1.04), and in exploratory analyses that excluded the first two years, cancer mortality was numerically higher in the vitamin D group (1.24). The authors suggested this regimen might not be appropriate for people who are already replete.","cites":["neale2022"],"grade":"A"},{"text":"Pooling 52 randomized trials with 75,454 participants, vitamin D supplementation did not change all-cause mortality (risk ratio 0.98) or cardiovascular mortality (0.98) but reduced cancer death by 16 percent (0.84). A separate meta-analysis of 10 trials found no change in cancer incidence (0.98), even where attained levels exceeded 40 ng/mL, and a 13 percent reduction in cancer mortality (0.87) across five trials, attributable largely to daily rather than bolus dosing. A modest, possibly real effect on dying of cancer is the one positive signal in the trial literature.","cites":["zhang2019","keum2019"],"grade":"A"},{"text":"For bone, 81 randomized trials with 53,537 adults showed no effect of vitamin D on total fracture (relative risk 1.00), hip fracture (1.11) or falls (0.97), no clinically relevant change in bone density, and no difference between higher and lower doses. In the VITAL fracture study of 25,871 adults not selected for deficiency or low bone mass, 2,000 IU a day did not reduce total fractures (hazard ratio 0.98), with no modification by baseline 25-hydroxyvitamin D.","cites":["bolland2018","leboff2022"],"grade":"A"},{"text":"The clearest benefit sits in older adults who start low. In eight double-blind trials of 2,426 people with a mean age of 65 or older, 700 to 1,000 IU a day reduced falls by 19 percent, and trials in which the achieved level reached about 24 ng/mL or more showed a 23 percent reduction, while lower doses and achieved levels below about 24 did not reduce falls.","cites":["bischoff2009"],"grade":"B"},{"text":"In prediabetes, 2,423 adults randomized to 4,000 IU a day or placebo in the D2d trial reached a mean of about 54 ng/mL from about 28 at baseline, and new diabetes was not significantly lower (hazard ratio 0.88). An individual-participant meta-analysis of three such trials found a 15 percent lower diabetes risk overall, a 3-year absolute reduction of 3.3 percent, and a much larger difference among those who happened to sustain a level of 50 ng/mL or more compared with 20 to 29; that last comparison is between people who responded differently to the same dose, not a randomized one.","cites":["pittas2019","pittas2023"],"grade":"A"},{"text":"The cohorts, unlike the trials, see risk at both ends. Among 247,574 Danish general-practice patients followed a median of 3.1 years, mortality was lowest at about 20 to 24 ng/mL; compared with 20, the hazard was 2.13 times at about 4 and 1.42 times at about 56. In 15,099 US adults followed 15 years, every level below about 24 carried significantly higher mortality than the reference of about 30 to 40, the nadir of risk sat at about 32 (95 percent confidence interval about 29 to 36), and levels of about 48 or more ran 1.2 to 1.5 times the reference. Both groups say the association may not be causal.","cites":["durup2012","sempos2013"],"grade":"B"},{"text":"Mendelian randomization, which uses inherited variants as a natural experiment, found no causal relationship. Across 386,406 Europeans, genetically higher 25-hydroxyvitamin D did not change coronary heart disease (odds ratio 0.98 per 4 ng/mL), stroke or all-cause mortality, in stratified analyses at every observed level of vitamin D, including people with low status. A first version of this analysis, published in 2021, reported benefit confined to deficient people and was retracted; the republished 2024 analysis is the one to quote, and it concludes that large reductions in mortality from long-term low-dose supplementation are unlikely even in people who start low.","cites":["erfc2024"],"grade":"B"},{"text":"Higher is not better for bone. In 311 healthy adults aged 55 to 70 randomized to 400, 4,000 or 10,000 IU of vitamin D3 a day for three years, levels reached about 31, 53 and 58 ng/mL, and radial bone density fell more with the higher doses (minus 1.2, 2.4 and 3.5 percent) with no difference in bone strength.","cites":["burt2019"],"grade":"B"},{"text":"The evolutionary reading rests on one small direct measurement: 35 Maasai pastoralists and 25 Hadzabe hunter-gatherers in Tanzania, living outdoors year-round with dark skin and moderate clothing, had a mean 25-hydroxyvitamin D of 115 nmol/L, about 46 ng/mL, with a range of about 23 to 68. No outcome was measured, and the authors themselves write that whether that level is optimal under a Western lifestyle is uncertain; it describes the level the skin produces under ancestral sun exposure, not a tested target.","cites":["luxwolda2012"],"grade":"C"}]},"levers":[{"name":"Sunlight","effect":"raises it, with a summer peak and a late-winter trough","claim":{"text":"Across 3.8 million US laboratory results, 25-hydroxyvitamin D peaked in September and bottomed in March at every latitude band, and parathyroid hormone moved in the opposite direction about four weeks behind. In 67 men in Omaha followed through a winter, summer skin synthesis stored in tissue supplied more than 80 percent of the winter requirement, which the authors estimated at 3,000 to 5,000 IU a day from all sources.","cites":["kroll2015","heaney2003"],"grade":"B"}},{"name":"Supplement dose","effect":"raises it, with diminishing returns as the dose climbs","claim":{"text":"In the Omaha winter study, steady daily vitamin D3 raised 25-hydroxyvitamin D by about 0.28 ng/mL per microgram, or about 7 ng/mL per 1,000 IU a day, from a baseline near 28. In 17,614 healthy volunteers tracked over 22,214 measurements, the response was exponential: about 4.8 ng/mL per 1,000 IU in the 0 to 1,000 IU range, falling to about 0.4 per 1,000 IU at 15,000 to 20,000 IU a day. The 2024 guideline suggests daily rather than intermittent high doses for adults over 50 who need vitamin D.","cites":["heaney2003","ekwaru2014","demay2024"],"grade":"B"}},{"name":"Body fat","effect":"lowers it for the same intake","claim":{"text":"In the same 17,614 volunteers, people with obesity ran about 7.9 ng/mL lower and people who were overweight about 3.2 lower than normal-weight people at a given dose, and body mass index predicted the level better than absolute weight. The 2011 guideline estimated that obese adults need two to three times the dose, while noting that no good study clearly justifies the multiplier.","cites":["ekwaru2014","holick2011"],"grade":"B"}},{"name":"Skin pigmentation","effect":"lowers it for the same sun","claim":{"text":"In a nationally representative US survey of 4,495 adults, 25-hydroxyvitamin D at or below 20 ng/mL was found in 82.1 percent of Black adults and 69.2 percent of Hispanic adults against 41.6 percent overall, with race, obesity and not drinking milk daily each independently associated. Whether the lower level carries the same meaning in darker skin is one of the open questions below.","cites":["forrest2011"],"grade":"B"}},{"name":"Prescription vitamin D or active vitamin D","effect":"raises it, or acts without raising it","claim":{"text":"High-dose prescription vitamin D and the active hormone analogs used in kidney and parathyroid disease are medications with their own monitoring; the active forms reduced falls in older adults by 22 percent in two trials of 624 people, about the same as ordinary vitamin D at 700 to 1,000 IU a day. Whether and when to use them belongs with a physician who knows the person's calcium, kidney function and parathyroid hormone, not a page.","cites":["bischoff2009","holick2011"],"grade":"A"}}],"caveats":[{"text":"Assays disagree. When 15 laboratories measured the same 50 serum samples against a reference method, nearly all mass-spectrometry results met the standardization targets of 10 percent or less imprecision and 5 percent or less bias, but only half of the immunoassays met the imprecision target and three of eight met the bias target. The conference that followed called assay variation a main source of the confusion in defining deficiency. A value near a cutoff can move across it by changing laboratories.","cites":["wise2017","sempos2018"],"grade":"B"},{"text":"Season matters as much as a supplement. Across 3.8 million US results, the share reading below 20 ng/mL swung from 21 percent to 48 percent over the year, with levels highest in September and lowest in March; a single draw should be read with the month in mind, and a repeat draw compared with the same season.","cites":["kroll2015"],"grade":"B"},{"text":"Body fat changes what a dose does. The same intake produces a lower level in heavier people, and a supplement adjusted for weight is a guess that a repeat measurement checks.","cites":["ekwaru2014"],"grade":"B"},{"text":"Toxicity is a matter of dose and calcium, and sits far above the laboratory range. In published cases of intoxication, 25-hydroxyvitamin D ran from 150 to 1,220 ng/mL with serum calcium from 11.1 to 23.1 mg/dL and symptoms of vomiting, dehydration, pain and loss of appetite, caused by manufacturing errors and overdosing by patients or prescribers. The 2011 guideline found most studies put the level of concern above 150 ng/mL; the Institute of Medicine set the tolerable upper intake at 4,000 IU a day for people 9 and older, corrected downward from 10,000 because of emerging concern about levels above 50.","cites":["galior2018","holick2011","ross2011"],"grade":"A"}],"seeAPhysician":["A very low value, under about 12 ng/mL, with bone or muscle pain, a fracture from a minor fall, or a condition that impairs absorption such as celiac disease, inflammatory bowel disease or prior weight-loss surgery.","A value above about 100 ng/mL, or any value while taking more than about 4,000 IU a day, which calls for a calcium check and a review of every supplement and fortified product in the house.","Thirst, frequent urination, nausea, vomiting, constipation, confusion or kidney stones while taking vitamin D, which can signal high calcium.","Sarcoidosis, lymphoma or another granulomatous disease, where immune cells make active vitamin D on their own and even ordinary doses can raise calcium.","Chronic kidney disease or a parathyroid disorder, where the active hormone and calcium, not 25-hydroxyvitamin D alone, guide treatment."],"openQuestions":["No trial has randomized healthy adults who start between 20 and 30 ng/mL to reach 40 to 60 and measured outcomes; the big trials enrolled mostly replete people and found nothing, and the cohort risk above about 50 has never been tested by randomization.","Whether the cancer-mortality reduction in the pooled trials is real, and whether it depends on daily dosing, body weight or baseline level, is unsettled; VITAL's own cancer-death estimate did not reach significance.","Whether the same 25-hydroxyvitamin D cutoff applies to people with darker skin is unresolved; the 2024 guideline found no trial evidence to screen by complexion, and the NHANES mortality analysis was too small to evaluate Black and Mexican-American adults.","The prediabetes subgroup that sustained 50 ng/mL or more did strikingly better, but it was not randomized to that level; a trial that targets a level rather than a dose has not been done.","Free or bioavailable vitamin D, binding-protein variants and other metabolites might explain some of the disagreement between cohorts and trials, but their assays are not yet standardized."],"questions":[{"q":"Is a vitamin D of 31 enough?","a":"By every published standard, yes. The 2011 Endocrine Society guideline calls 30 to 100 ng/mL sufficient, the Institute of Medicine says about 20 meets nearly everyone's bone needs, and the large trials that raised people from about 31 to the low 40s found no fewer cancers, heart attacks, fractures or deaths. Risk in the cohorts sat lowest at about 20 to 36, which includes 31."},{"q":"What is an optimal vitamin D level?","a":"No study defines one above the point where deficiency ends. The literature is firm that below about 12 to 20 ng/mL harms bone and keeps company with higher mortality, and that pushing replete adults higher has not helped in randomized trials. Cohorts see the lowest mortality at about 20 to 36 and more risk again above about 48 to 56, and the Institute of Medicine flags possible risk above 50. A physician's zone inside 30 to 50 is a judgment, and this page says so."},{"q":"How much vitamin D should I take to reach 50?","a":"There is no dose that reliably lands a given person on a given number. In healthy volunteers each 1,000 IU a day raised the level by about 5 to 7 ng/mL at low doses and much less at high ones, heavier people rose less, and season moved the baseline. The more useful question is whether 50 is a target worth chasing: no trial shows that it is for a replete adult, and a dose above about 4,000 IU a day belongs under a physician's direction with a calcium check."}],"citations":[{"id":"sempos2018","authors":"Sempos CT, Heijboer AC, Bikle DD, et al.","title":"Vitamin D assays and the definition of hypovitaminosis D: results from the First International Conference on Controversies in Vitamin D","journal":"British Journal of Clinical Pharmacology","year":2018,"pmid":"29851137","doi":"10.1111/bcp.13652","type":"consensus","finding":"25-hydroxyvitamin D is the critical measurement of status; below 12 ng/mL carries rickets and osteomalacia risk; 20 to 50 ng/mL appears safe and sufficient for bone; assay variation has driven the confusion over deficiency.","url":"https://pubmed.ncbi.nlm.nih.gov/29851137/"},{"id":"holick2011","authors":"Holick MF, Binkley NC, Bischoff-Ferrari HA, et al.","title":"Evaluation, treatment, and prevention of vitamin D deficiency: an Endocrine Society clinical practice guideline","journal":"Journal of Clinical Endocrinology and Metabolism","year":2011,"pmid":"21646368","doi":"10.1210/jc.2011-0385","type":"guideline","finding":"Deficiency below 20 ng/mL, insufficiency 21 to 29, sufficiency 30 to 100; concern above 150; against using 1,25-dihydroxyvitamin D for status; obese adults need two to three times the dose.","url":"https://pubmed.ncbi.nlm.nih.gov/21646368/"},{"id":"ross2011","authors":"Ross AC, Manson JE, Abrams SA, et al.","title":"The 2011 report on dietary reference intakes for calcium and vitamin D from the Institute of Medicine: what clinicians need to know","journal":"Journal of Clinical Endocrinology and Metabolism","year":2011,"pmid":"21118827","doi":"10.1210/jc.2010-2704","type":"guideline","finding":"About 16 ng/mL meets half the population's bone needs and 20 meets 97.5 percent; no consistent benefit above 30; risks for some outcomes above 50; upper intake 4,000 IU a day.","url":"https://pubmed.ncbi.nlm.nih.gov/21118827/"},{"id":"demay2024","authors":"Demay MB, Pittas AG, Bikle DD, et al.","title":"Vitamin D for the Prevention of Disease: An Endocrine Society Clinical Practice Guideline","journal":"Journal of Clinical Endocrinology and Metabolism","year":2024,"pmid":"38828931","doi":"10.1210/clinem/dgae290","type":"guideline","finding":"Against routine 25-hydroxyvitamin D testing and against empiric supplementation above the dietary reference intake in healthy adults under 75; no clear evidence of an optimal target; daily rather than intermittent dosing.","url":"https://pubmed.ncbi.nlm.nih.gov/38828931/"},{"id":"forrest2011","authors":"Forrest KY, Stuhldreher WL","title":"Prevalence and correlates of vitamin D deficiency in US adults","journal":"Nutrition Research","year":2011,"pmid":"21310306","doi":"10.1016/j.nutres.2010.12.001","type":"cross-sectional","finding":"41.6 percent of 4,495 US adults at or below 20 ng/mL; 82.1 percent of Black and 69.2 percent of Hispanic adults.","url":"https://pubmed.ncbi.nlm.nih.gov/21310306/"},{"id":"kroll2015","authors":"Kroll MH, Bi C, Garber CC, et al.","title":"Temporal relationship between vitamin D status and parathyroid hormone in the United States","journal":"PLoS One","year":2015,"pmid":"25738588","doi":"10.1371/journal.pone.0118108","type":"cross-sectional","finding":"In 3.8 million US results, 25-hydroxyvitamin D peaked in September and troughed in March; 33 percent below 20 and 60 percent below 30; deficiency varied from 21 to 48 percent by season.","url":"https://pubmed.ncbi.nlm.nih.gov/25738588/"},{"id":"manson2019","authors":"Manson JE, Cook NR, Lee IM, et al.","title":"Vitamin D Supplements and Prevention of Cancer and Cardiovascular Disease","journal":"New England Journal of Medicine","year":2019,"pmid":"30415629","doi":"10.1056/NEJMoa1809944","type":"rct","finding":"In 25,871 US adults, 2,000 IU a day for a median 5.3 years did not reduce invasive cancer (HR 0.96), major cardiovascular events (0.97) or death (0.99); baseline mean 30.8 ng/mL.","url":"https://pubmed.ncbi.nlm.nih.gov/30415629/"},{"id":"neale2022","authors":"Neale RE, Baxter C, Romero BD, et al.","title":"The D-Health Trial: a randomised controlled trial of the effect of vitamin D on mortality","journal":"Lancet Diabetes and Endocrinology","year":2022,"pmid":"35026158","doi":"10.1016/S2213-8587(21)00345-4","type":"rct","finding":"In 21,315 Australians 60 and older, 60,000 IU a month for five years did not reduce all-cause mortality (HR 1.04); exploratory cancer mortality HR 1.24 after excluding the first two years.","url":"https://pubmed.ncbi.nlm.nih.gov/35026158/"},{"id":"zhang2019","authors":"Zhang Y, Fang F, Tang J, et al.","title":"Association between vitamin D supplementation and mortality: systematic review and meta-analysis","journal":"BMJ","year":2019,"pmid":"31405892","doi":"10.1136/bmj.l4673","type":"meta-analysis","finding":"52 trials, 75,454 participants: no change in all-cause (RR 0.98) or cardiovascular mortality; cancer death reduced 16 percent (0.84).","url":"https://pubmed.ncbi.nlm.nih.gov/31405892/"},{"id":"keum2019","authors":"Keum N, Lee DH, Greenwood DC, Manson JE, Giovannucci E","title":"Vitamin D supplementation and total cancer incidence and mortality: a meta-analysis of randomized controlled trials","journal":"Annals of Oncology","year":2019,"pmid":"30796437","doi":"10.1093/annonc/mdz059","type":"meta-analysis","finding":"No reduction in cancer incidence (RR 0.98), even where attained levels exceeded 100 nmol/L; cancer mortality RR 0.87, mostly from daily dosing.","url":"https://pubmed.ncbi.nlm.nih.gov/30796437/"},{"id":"bolland2018","authors":"Bolland MJ, Grey A, Avenell A","title":"Effects of vitamin D supplementation on musculoskeletal health: a systematic review, meta-analysis, and trial sequential analysis","journal":"Lancet Diabetes and Endocrinology","year":2018,"pmid":"30293909","doi":"10.1016/S2213-8587(18)30265-1","type":"meta-analysis","finding":"81 trials, 53,537 adults: no effect on total fracture (RR 1.00), hip fracture (1.11) or falls (0.97), no clinically relevant bone density change, no difference between higher and lower doses.","url":"https://pubmed.ncbi.nlm.nih.gov/30293909/"},{"id":"leboff2022","authors":"LeBoff MS, Chou SH, Ratliff KA, et al.","title":"Supplemental Vitamin D and Incident Fractures in Midlife and Older Adults","journal":"New England Journal of Medicine","year":2022,"pmid":"35939577","doi":"10.1056/NEJMoa2202106","type":"rct","finding":"In 25,871 VITAL participants, 2,000 IU a day did not reduce total (HR 0.98), nonvertebral (0.97) or hip fractures (1.01), with no modification by baseline 25-hydroxyvitamin D.","url":"https://pubmed.ncbi.nlm.nih.gov/35939577/"},{"id":"bischoff2009","authors":"Bischoff-Ferrari HA, Dawson-Hughes B, Staehelin HB, et al.","title":"Fall prevention with supplemental and active forms of vitamin D: a meta-analysis of randomised controlled trials","journal":"BMJ","year":2009,"pmid":"19797342","doi":"10.1136/bmj.b3692","type":"meta-analysis","finding":"Eight trials, 2,426 older adults: 700 to 1,000 IU a day cut falls 19 percent; achieved levels of 60 nmol/L or more cut falls 23 percent; lower doses and levels did not; active forms cut falls 22 percent.","url":"https://pubmed.ncbi.nlm.nih.gov/19797342/"},{"id":"pittas2019","authors":"Pittas AG, Dawson-Hughes B, Sheehan P, et al.","title":"Vitamin D Supplementation and Prevention of Type 2 Diabetes","journal":"New England Journal of Medicine","year":2019,"pmid":"31173679","doi":"10.1056/NEJMoa1900906","type":"rct","finding":"In 2,423 adults with prediabetes, 4,000 IU a day reached a mean 54.3 ng/mL against 28.2 at baseline; new diabetes HR 0.88, not significant.","url":"https://pubmed.ncbi.nlm.nih.gov/31173679/"},{"id":"pittas2023","authors":"Pittas AG, Kawahara T, Jorde R, et al.","title":"Vitamin D and Risk for Type 2 Diabetes in People With Prediabetes: A Systematic Review and Meta-analysis of Individual Participant Data From 3 Randomized Clinical Trials","journal":"Annals of Internal Medicine","year":2023,"pmid":"36745886","doi":"10.7326/M22-3018","type":"meta-analysis","finding":"Vitamin D cut diabetes risk 15 percent (HR 0.85), absolute 3.3 percent over three years; those sustaining 50 ng/mL or more versus 20 to 29 had HR 0.24, a non-randomized comparison.","url":"https://pubmed.ncbi.nlm.nih.gov/36745886/"},{"id":"durup2012","authors":"Durup D, Jørgensen HL, Christensen J, Schwarz P, Heegaard AM, Lind B","title":"A reverse J-shaped association of all-cause mortality with serum 25-hydroxyvitamin D in general practice: the CopD study","journal":"Journal of Clinical Endocrinology and Metabolism","year":2012,"pmid":"22573406","doi":"10.1210/jc.2012-1176","type":"cohort","finding":"In 247,574 Copenhagen patients, mortality was lowest at 50 to 60 nmol/L; HR 2.13 at 10 nmol/L and 1.42 at 140 nmol/L versus 50; causality not inferred.","url":"https://pubmed.ncbi.nlm.nih.gov/22573406/"},{"id":"sempos2013","authors":"Sempos CT, Durazo-Arvizu RA, Dawson-Hughes B, et al.","title":"Is there a reverse J-shaped association between 25-hydroxyvitamin D and all-cause mortality? Results from the U.S. nationally representative NHANES","journal":"Journal of Clinical Endocrinology and Metabolism","year":2013,"pmid":"23666975","doi":"10.1210/jc.2013-1333","type":"cohort","finding":"In 15,099 US adults over 15 years, risk was higher at every level below 60 nmol/L, lowest at 81 nmol/L (73 to 90), and 1.2 to 1.5 times the reference at 120 nmol/L or more.","url":"https://pubmed.ncbi.nlm.nih.gov/23666975/"},{"id":"erfc2024","authors":"Emerging Risk Factors Collaboration/EPIC-CVD/Vitamin D Studies Collaboration","title":"Estimating dose-response relationships for vitamin D with coronary heart disease, stroke, and all-cause mortality: observational and Mendelian randomisation analyses","journal":"Lancet Diabetes and Endocrinology","year":2024,"pmid":"38048800","doi":"10.1016/S2213-8587(23)00287-5","type":"mendelian-randomization","finding":"Republished after the 2021 version was retracted: in 386,406 Europeans, genetically predicted 25-hydroxyvitamin D showed no causal relationship with coronary disease (OR 0.98 per 10 nmol/L), stroke or mortality at any level.","url":"https://pubmed.ncbi.nlm.nih.gov/38048800/"},{"id":"burt2019","authors":"Burt LA, Billington EO, Rose MS, Raymond DA, Hanley DA, Boyd SK","title":"Effect of High-Dose Vitamin D Supplementation on Volumetric Bone Density and Bone Strength: A Randomized Clinical Trial","journal":"JAMA","year":2019,"pmid":"31454046","doi":"10.1001/jama.2019.11889","type":"rct","finding":"In 311 healthy adults aged 55 to 70, 4,000 and 10,000 IU a day for three years lowered radial bone density (minus 2.4 and 3.5 percent versus minus 1.2 at 400 IU) with no bone-strength difference.","url":"https://pubmed.ncbi.nlm.nih.gov/31454046/"},{"id":"heaney2003","authors":"Heaney RP, Davies KM, Chen TC, Holick MF, Barger-Lux MJ","title":"Human serum 25-hydroxycholecalciferol response to extended oral dosing with cholecalciferol","journal":"American Journal of Clinical Nutrition","year":2003,"pmid":"12499343","doi":"10.1093/ajcn/77.1.204","type":"rct","finding":"In 67 men through an Omaha winter, 25-hydroxyvitamin D rose about 0.70 nmol/L per microgram of daily vitamin D3; summer tissue stores supplied more than 80 percent of winter need, estimated at 3,000 to 5,000 IU a day.","url":"https://pubmed.ncbi.nlm.nih.gov/12499343/"},{"id":"ekwaru2014","authors":"Ekwaru JP, Zwicker JD, Holick MF, Giovannucci E, Veugelers PJ","title":"The importance of body weight for the dose response relationship of oral vitamin D supplementation and serum 25-hydroxyvitamin D in healthy volunteers","journal":"PLoS One","year":2014,"pmid":"25372709","doi":"10.1371/journal.pone.0111265","type":"cross-sectional","finding":"In 17,614 volunteers, the dose response was exponential (12.0 nmol/L per 1,000 IU at 0 to 1,000 IU, 1.1 at 15,000 to 20,000); obese and overweight people ran 19.8 and 8.0 nmol/L lower.","url":"https://pubmed.ncbi.nlm.nih.gov/25372709/"},{"id":"wise2017","authors":"Wise SA, Phinney KW, Tai SS, et al.","title":"Baseline Assessment of 25-Hydroxyvitamin D Assay Performance: A Vitamin D Standardization Program (VDSP) Interlaboratory Comparison Study","journal":"Journal of AOAC International","year":2017,"pmid":"28822355","doi":"10.5740/jaoacint.17-0258","type":"laboratory","finding":"Across 15 laboratories and 50 samples, nearly all mass-spectrometry assays met the 10 percent CV and 5 percent bias criteria; only half of immunoassays met the CV criterion and three of eight the bias criterion.","url":"https://pubmed.ncbi.nlm.nih.gov/28822355/"},{"id":"galior2018","authors":"Galior K, Grebe S, Singh R","title":"Development of Vitamin D Toxicity from Overcorrection of Vitamin D Deficiency: A Review of Case Reports","journal":"Nutrients","year":2018,"pmid":"30042334","doi":"10.3390/nu10080953","type":"review","finding":"In 13 published case reports, intoxication presented with 25-hydroxyvitamin D of 150 to 1,220 ng/mL and calcium of 11.1 to 23.1 mg/dL, from manufacturing errors and overdosing.","url":"https://pubmed.ncbi.nlm.nih.gov/30042334/"},{"id":"luxwolda2012","authors":"Luxwolda MF, Kuipers RS, Kema IP, Dijck-Brouwer DA, Muskiet FA","title":"Traditionally living populations in East Africa have a mean serum 25-hydroxyvitamin D concentration of 115 nmol/l","journal":"British Journal of Nutrition","year":2012,"pmid":"22264449","doi":"10.1017/S0007114511007161","type":"cross-sectional","finding":"Maasai and Hadzabe adults with lifelong tropical sun exposure averaged 115 nmol/L (about 46 ng/mL), range 58 to 171 nmol/L; optimality under a Western lifestyle called uncertain.","url":"https://pubmed.ncbi.nlm.nih.gov/22264449/"}],"written":"2026-10-07","updated":"2026-10-07","reviewed":true,"reviewedOn":"2026-10-10","related":{"essays":["morning-sunlight-the-cheapest-health-tool-you-re-ignoring","do-you-need-a-concierge-doctor-if-you-feel-fine"],"glossary":["Reference range","Optimal range"],"symptoms":["fatigue","joint-pain","mood-stress"],"entries":["ferritin","fasting-glucose","hba1c"]},"url":"https://taggehealth.com/labs/vitamin-d","author":"Daniel Tagge, MD","publisher":"Tagge Health","license":"Free to read and cite with attribution to the URL above.","inPractice":"I like to see 25-hydroxyvitamin D between about 40 and 60 ng/mL, drawn in the same season each year and read next to calcium. The window comes from an evolutionary reading: people who still live outdoors year-round in the sun where our species began average about 46 ng/mL, and I treat that as the level the body was built to run at. I want to be plain about what that is: a judgment from physiology, not a trial result. The trials have not shown that lifting a replete adult from 30 to 50 prevents anything, so I do not chase the number with large doses; I prefer sun and daily dosing, make sure nobody sits below 20, and check calcium in anyone on more than a modest dose."}