The official French threshold of 30 ng/mL for vitamin D defines the absence of skeletal deficiency, not the biological optimum for longevity. Recent meta-analyses converge on a protective range of 20 to 50 ng/mL (50-125 nmol/L) of serum 25-hydroxyvitamin D for all-cause mortality, with risk rising both below and above this window. The distinction is not semantic. It conditions the relevance of supplementation and the dose to target.
This divergence between clinical and biological thresholds has long been sidestepped. The VITAL trial published in 2025 in the American Journal of Clinical Nutrition provided the first evidence, from a large-scale randomized controlled trial, that daily vitamin D3 supplementation slows telomere attrition (PubMed). The debate between deficiency and optimum is no longer theoretical.
Meta-analyses converge on a serum 25(OH)D concentration between 50 and 125 nmol/L (20-50 ng/mL) as the zone of minimum mortality, with a significant rise in risk both below and above.
The official threshold: a minimum target, not a Healthspan goal
French health authorities define three vitamin D status zones: deficiency (< 10 ng/mL), insufficiency (10-30 ng/mL), and sufficiency (> 30 ng/mL). This classification was designed to prevent rickets in children and osteomalacia in adults, not to optimize extra-skeletal functions.
The reference review published by Roger Bouillon in Endocrine Reviews in 2019 clarified this point. Skeletal and extra-skeletal thresholds do not overlap (PubMed). Fracture prevention requires about 800 IU/day with adequate calcium. Extra-skeletal functions (regulation of cell proliferation, immune modulation, muscle function, vascular and metabolic properties) respond to higher thresholds.
This nuance escapes most clinical reading frames. A patient with 32 ng/mL is labeled "not deficient" and the conversation ends. Yet 32 ng/mL is precisely the lower bound of the Healthspan zone, not its optimum.
Where the protective range lies: what meta-analyses show
The dose-response meta-analysis published by Garland and colleagues in the American Journal of Public Health in 2014 synthesized 32 prospective studies, totaling more than 566,000 participants (PubMed). All-cause mortality risk was nearly twice as high for serum concentrations between 0 and 9 ng/mL compared to concentrations above 50 ng/mL.
The most instructive study on the shape of the curve remains the Danish CopD cohort, published in the Journal of Clinical Endocrinology & Metabolism in 2012 by Durup and colleagues (PubMed). Across 247,574 general practice subjects in Copenhagen, the authors described a reverse J-shaped curve rather than a simple linear relationship. Minimum mortality was observed at 70 nmol/L (28 ng/mL). Below 10 nmol/L, risk was multiplied by 2.31. Above 140 nmol/L (56 ng/mL), it rose again by 42%.
| Serum 25(OH)D status | Concentration | Longevity implication |
|---|---|---|
| Severe deficiency | < 10 nmol/L (< 4 ng/mL) | Mortality risk ×2.31 |
| Insufficiency | 10-50 nmol/L (4-20 ng/mL) | Functional reserve degraded |
| Skeletal sufficiency | 50-75 nmol/L (20-30 ng/mL) | Plateau being established |
| Healthspan zone | 75-125 nmol/L (30-50 ng/mL) | Minimum mortality observed |
| Risk rebound zone | > 140 nmol/L (> 56 ng/mL) | CV risk rising by 42% |
| Toxicity | > 375 nmol/L (> 150 ng/mL) | Possible hypercalcemia |
This reverse J-shape distinguishes vitamin D from nutrients with a linear effect. More is not better. The biological target is not a threshold to exceed but a range to reach.
VITAL 2025: the telomeric evidence of a geroprotective effect
The VITAL trial (VITamin D and OmegA-3 TriaL) randomized 25,871 American adults (women ≥ 55 years, men ≥ 50 years) to either 2,000 IU/day of vitamin D3 or placebo, with a median follow-up of 5.3 years. The initial publication in the New England Journal of Medicine in 2019 reported the absence of effect on the primary endpoints (invasive cancer, major cardiovascular events) (PubMed). The verdict seemed harsh for systematic supplementation.
The telomere substudy published in May 2025 in the American Journal of Clinical Nutrition reversed the reading (PubMed). Among participants with leukocyte telomere length measured at baseline and 4 years, vitamin D3 supplementation reduced telomere attrition by 140 base pairs. This corresponds to about three years of cellular aging avoided compared to placebo.
The result is doubly important. First, it comes from a large-scale randomized controlled trial, the most demanding standard of evidence. Second, the effect was observed independently of participants' baseline status, suggesting a distributable extra-skeletal benefit. The omega-3 arm of the same trial showed no significant telomeric effect, which argues against a global methodological bias.
The plateau and the J-curve: why "more" is not "better"
Acute vitamin D toxicity is rare and only occurs at massive doses, on the order of 50,000 to 1,000,000 IU/day for several months. A retrospective study by McCullough and colleagues on nearly 4,700 hospitalized patients receiving 5,000 to 50,000 IU/day for up to seven years reported no cases of hypercalcemia attributable to vitamin D (PubMed).
But the absence of acute toxicity does not mean the absence of biological cost. The CopD cohort documented the rise in mortality starting at 56 ng/mL, and the European meta-analysis covering 26,916 individuals found no additional benefit for concentrations above 125 nmol/L (50 ng/mL) (PubMed). The plateau is real.
The mode of administration also matters. The French practice of administering 100,000 IU or 200,000 IU as a quarterly or semi-annual bolus corrects serum status but does not reproduce the extra-skeletal benefits of daily protocols. A narrative review by Mazess, Bischoff-Ferrari and Dawson-Hughes, published in JBMR Plus in 2021 under the eloquent title Bolus Is Bogus, synthesized the trials comparing these two modalities (PubMed). Bolus dosing produces a transient rise in 25(OH)D followed by a rapid decline. This pulsatility does not reproduce continuous physiological regulation, and several trials have reported an increased risk of falls and fractures after high-dose bolus in elderly subjects.
For longevity, the rule is simple. Daily supplementation at moderate dose (1,000 to 2,000 IU/day depending on profile) outperforms high-dose bolus, regardless of the serum level achieved.
Three sources of vitamin D: ultraviolet, diet, supplementation
Vitamin D stands apart from all other micronutrients through its predominantly endogenous origin. In an adult living at temperate latitudes, cutaneous synthesis triggered by UVB radiation (290-315 nm) theoretically covers 80 to 90% of daily needs. The photoconversion of 7-dehydrocholesterol into pre-vitamin D3 in the deep epidermal layers is rapid: 10 to 15 minutes of midday sun exposure on 50% of body surface can generate between 10,000 and 25,000 IU, the equivalent of a full week of supplementation.
This efficient physiology runs into a major geographical constraint. Above the 42nd parallel north (Marseille marks the southern French boundary), the angle of solar incidence drops too low in winter to allow cutaneous synthesis. Continental France, located between 42° and 51° N, receives no effective UVB from October to March. Five to six months of seasonal insufficiency thus structurally affect the population, regardless of time spent outdoors.
Dietary intake compensates poorly for this insufficiency. Naturally rich sources are scarce: 100 g of wild salmon delivers about 600 to 800 IU, herring reaches 1,600 IU, and cod liver oil exceeds 1,300 IU per teaspoon. Egg yolk, beef liver, and UV-exposed mushrooms provide only 30 to 60 IU per serving. The INCA3 survey conducted by ANSES estimates the average dietary intake of the French adult population at 120-150 IU per day, or 15 to 20% of the official recommendation.
Unlike the United States, Finland, or Canada, France does not systematically fortify milk or cereals with vitamin D. This regulatory difference partly explains why the prevalence of vitamin D insufficiency reaches 70 to 80% of the French adult population at the end of winter, compared with 25 to 35% in the United States according to NHANES data.
| Source | Vitamin D intake | Practical limit |
|---|---|---|
| Cutaneous synthesis (summer, midday) | 10,000-25,000 IU / 15 min | None from October to March in France |
| Cod liver oil | 1,360 IU / teaspoon | Low organoleptic tolerance |
| Wild salmon (100 g) | 600-800 IU | Cost and consumption frequency |
| Herring (100 g) | 1,600 IU | Same |
| Egg yolk | 30-50 IU | Marginal contribution |
| Fortified milk (United States) | 100 IU / 250 mL | Not commercialized in France |
| Supplementation (oily D3) | 400-5,000 IU per dose | Only year-round controllable lever |
Special cases: why the same dose does not produce the same level
Standardized vitamin D supplementation masks considerable interindividual variability in achieved serum levels. Understanding the factors that modulate this response avoids the pitfall of the one-size-fits-all dose.
Overweight and obesity. Adipose tissue sequesters vitamin D, soluble in fats, in a slow-release storage compartment. The reference review by Bouillon published in Endocrine Reviews documents that obese subjects receiving an equivalent dose reach serum concentrations 30 to 50% lower than those of normal-weight subjects (PubMed). To reach the protective 30-50 ng/mL range, current recommendations call for doses 2 to 3 times higher in obese individuals (BMI > 30).
Skin pigmentation. Melanin acts as an endogenous solar filter. For the same exposure time, a person with phototype VI (Black skin) synthesizes about six times less vitamin D than a person with phototype II. Populations with dark skin living at northern latitudes thus accumulate a structural chronic deficit, particularly pronounced during months without effective cutaneous synthesis.
Aging. Cutaneous synthesis capacity drops with age. At 70, photoconverted production is about four times lower than that of a 20-year-old adult, due to epidermal thinning and decreasing 7-dehydrocholesterol precursor. This decline compounds with often reduced sun exposure and less efficient intestinal absorption. Geriatric guidelines (HAS, GRIO) raise the target intake to a minimum of 800-1,000 IU/day in older adults without additional risk factors.
Pregnancy and breastfeeding. Needs increase significantly. The Haute Autorité de Santé recommends systematic supplementation of 800-1,000 IU/day from the second trimester onward, with reinforced evidence on supporting fetal bone mineralization and regulating maternal calcemia.
Chronic kidney disease. Renal 1-alpha-hydroxylase, which converts 25(OH)D into the active form 1,25(OH)₂D, sees its activity decline with kidney function. At advanced stages of insufficiency, D3 supplementation no longer suffices: prescribing calcitriol (the already-hydroxylated active form) becomes necessary, under nephrology supervision. This renal dependence is partially reflected by serum creatinine, but as recalled in our article on creatinine as an obsolete metric, this marker often masks early kidney dysfunction.
Enzyme-inducing medications. Glucocorticoids, antiepileptics (phenytoin, phenobarbital), azole antifungals, and certain antiretrovirals accelerate vitamin D catabolism through induction of cytochrome CYP3A4. People on long-term chronic treatment require closer monitoring and increased intake.
This heterogeneity explains why the one-size-fits-all dose fails. The gap between theoretical intake and the actually achieved serum concentration can reach a factor of five, regardless of protocol quality.
Measuring your level: seasonal variability and dosage choice
The only reliable biomarker of vitamin D status is the serum concentration of 25-hydroxyvitamin D, or 25(OH)D. This circulating form integrates cutaneous synthesis, dietary intake, and supplementation, with a half-life of about three weeks that makes it a stable indicator. The active form 1,25(OH)₂D, despite its central biological role, should never be measured routinely: its short half-life (4 to 6 hours) and rapid hormonal regulation make it unsuitable for nutritional status monitoring.
Two families of methods coexist in laboratories. Automated immunoassays (Architect, LIAISON, Elecsys) dominate routine practice through their high throughput and moderate cost. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) is the reference method: it precisely distinguishes 25(OH)D2 from 25(OH)D3 and offers superior accuracy. Inter-method variability can reach 25%, complicating the comparison of results from different laboratories. The international Vitamin D Standardization Program (VDSP), launched in 2010, aims to harmonize these gaps but its diffusion in France remains partial.
Seasonality must be integrated into any interpretation. The average amplitude between the winter nadir (late March) and the summer peak (late August) reaches 50 to 70% in continental France. A level measured in September structurally overestimates the average status; a level measured in March reveals the most unfavorable threshold. For a realistic mapping, the reference measurement is taken at the end of winter, when reserves accumulated in summer are depleted.
The recommended frequency follows from this dynamic. An initial measurement before supplementation, then a follow-up 3 to 6 months after starting the protocol to validate the target reached. Once equilibrium is established, an annual check in March suffices. In France, reimbursement by the national health insurance has been restricted since 2014 to explicit medical indications (osteoporosis, malabsorption, immunosuppressive treatment). Out-of-indication testing costs 15 to 25 € at retail pharmacies.
Cofactors and mode of administration: what science has settled
Vitamin D does not act in isolation. Its conversion into the active 1,25-dihydroxyvitamin D form requires several hepatic and renal enzymes, including 25-hydroxylase and 1-alpha-hydroxylase. All of these enzymes use magnesium as an obligatory cofactor. A review published in the Journal of the American Osteopathic Association in 2018 established that magnesium deficiency can compromise vitamin D activation, independently of the serum 25(OH)D level (PubMed).
This interdependence explains a frequent phenomenon: patients supplemented with vitamin D whose serum level does not progress as expected, because their limiting magnesium status blocks the activation step. Supplementing vitamin D without considering magnesium is like adding fuel to an engine with faulty spark plugs. This point is examined in more detail in our article on magnesium and longevity.
Vitamin K2 (as MK-7, derived from natto fermentation) plays a complementary role. It activates the matrix Gla-protein, which prevents calcium precipitation in the vascular wall. Vitamin D mobilizes intestinal calcium; K2 directs its tissue destination. Without K2, the rise in calcemia induced by D3 may theoretically contribute to arterial calcification, although long-term clinical trials on this interaction remain limited.
The choice of galenic form is also established. Vitamin D3 (cholecalciferol) is approximately 1.7 times more effective than D2 (ergocalciferol) at raising serum 25(OH)D. Current guidelines systematically favor D3, ideally in oily form to facilitate intestinal absorption, which depends on dietary fats. As recalled in our article on the illusion of absolute dosing, the raw figure of the dose ingested only makes sense in light of the serum concentration actually reached.
Vitamin D occupies a singular educational position. No other micronutrient reveals as clearly the difference between a population approach ("avoid deficiency") and a precision approach ("reach the biologically optimal range").
The gap between these two logics is measured in years of preserved telomere length, in cardiovascular risk avoided, in muscle capacity maintained. It is also measured in avoidable clinical errors: pointless boluses, doses standardized across a heterogeneous population, apparently "normal" statuses that mask a suboptimal biological zone.
Serum 25(OH)D testing remains the only objective way to position an individual within this map. Without measurement, blind supplementation can only produce an average result, and the average is never the biological target.
Frequently asked questions
References
- Zhu H, Manson JE, Cook NR, et al. Vitamin D3 and marine ω-3 fatty acids supplementation and leukocyte telomere length: 4-year findings from the VITamin D and OmegA-3 TriaL (VITAL) randomized controlled trial. Am J Clin Nutr. 2025;122(1):39-47 (PubMed).
- Bouillon R, Marcocci C, Carmeliet G, et al. Skeletal and Extraskeletal Actions of Vitamin D: Current Evidence and Outstanding Questions. Endocr Rev. 2019;40(4):1109-1151 (PubMed).
- Garland CF, Kim JJ, Mohr SB, et al. Meta-analysis of all-cause mortality according to serum 25-hydroxyvitamin D. Am J Public Health. 2014;104(8):e43-e50 (PubMed).
- Durup D, Jørgensen HL, Christensen J, et al. A reverse J-shaped association of all-cause mortality with serum 25-hydroxyvitamin D in general practice: the CopD study. J Clin Endocrinol Metab. 2012;97(8):2644-2652 (PubMed).
- Manson JE, Cook NR, Lee IM, et al. Vitamin D Supplements and Prevention of Cancer and Cardiovascular Disease. N Engl J Med. 2019;380(1):33-44 (PubMed).
- McCullough PJ, Lehrer DS, Amend J. Daily oral dosing of vitamin D3 using 5000 TO 50,000 international units a day in long-term hospitalized patients: Insights from a seven year experience. J Steroid Biochem Mol Biol. 2019;189:228-239 (PubMed).
- Gaksch M, Jorde R, Grimnes G, et al. Vitamin D and mortality: Individual participant data meta-analysis of standardized 25-hydroxyvitamin D in 26916 individuals from a European consortium. PLoS One. 2017;12(2):e0170791 (PubMed).
- Mazess RB, Bischoff-Ferrari HA, Dawson-Hughes B. Vitamin D: Bolus Is Bogus—A Narrative Review. JBMR Plus. 2021;5(12):e10567 (PubMed).
- Uwitonze AM, Razzaque MS. Role of Magnesium in Vitamin D Activation and Function. J Am Osteopath Assoc. 2018;118(3):181-189 (PubMed).
