CHAPTER

Yoon Hang Kim, MD, MPH

Board-Certified in Preventive Medicine

Integrative & Functional Medicine

University of Arizona Osher Fellow | UCLA Medical Acupuncture | IFM Scholar

“The dose makes the poison.” — Paracelsus (1493–1541)

Abstract

Vitamins are among the most widely consumed supplements in the world, yet the cultural assumption that they are universally safe obscures a substantial body of evidence demonstrating harm at excessive doses. This chapter examines the full spectrum of vitamin-related pathology, from the classical deficiency syndromes that shaped twentieth-century nutrition policy, through the resurgence of deficiency in modern restrictive diets, to the well-documented toxicities of both water-soluble and fat-soluble vitamins at supraphysiologic doses. It gives particular attention to vitamin B6 neurotoxicity, the epidemiologic evidence linking high-dose B-vitamin supplementation to cancer risk, and the U-shaped dose–response curve that characterizes many vitamin–disease relationships.

Between frank deficiency and frank toxicity lies territory that conventional supplement-safety reviews tend to skip: functional insufficiency, altered requirement in disease, differences between molecular forms of the same vitamin, cofactor dependency, and legitimate therapeutic nutrient dosing under monitoring. This chapter maps that middle ground explicitly, and argues that recognizing it is a reason for more clinical rigor rather than less. The central message is unchanged and worth stating plainly: correcting deficiency saves lives, therapeutic nutrient dosing has a defensible place when it carries an indication and an endpoint, and chronic megadosing in the absence of documented need introduces measurable risk.

1. Introduction: The Paradox of Plenty

The history of vitamins is, at its core, a story of subtraction and restoration. For centuries, populations suffered from diseases we now know to be caused by the absence of specific micronutrients. Scurvy devastated navies, beriberi crippled rice-dependent populations, pellagra ravaged communities in the American South. The identification of vitamins in the early twentieth century, and the fortification and supplementation campaigns that followed, represent genuine triumphs of public health. These interventions have prevented incalculable suffering.

But the success of correcting deficiency created a logical error that persists to this day: the assumption that if some is good, more must be better. This chapter confronts that assumption directly. Drawing on clinical, epidemiologic, and mechanistic evidence, it makes the case that the dose–response relationship for most vitamins is not linear. It is U-shaped, with both deficiency and excess conferring harm. The optimal range is narrower than most patients, and many clinicians, realize, and the consequences of overshooting it are not trivial.

There is a second error, less discussed but equally consequential, that runs in the opposite direction. A clinician who concludes from the toxicity literature that all supplementation beyond the population reference intake is unjustified will undertreat the patient with malabsorption, the patient on chronic acid suppression, the post-bariatric patient, the patient with a genuinely increased requirement that no serum panel has yet flagged. Both errors are failures of the same kind: they substitute a rule for an assessment.

The working model used in this chapter is therefore not the familiar three-zone sequence of deficiency, adequacy, and excess. It is a five-zone model:

1. Overt deficiency, with a recognizable clinical syndrome
2. Functional insufficiency, in which biochemical or physiologic markers of inadequate nutrient function are present before a classical deficiency syndrome appears
3. Physiologic sufficiency, the broad plateau at which additional intake confers no further benefit
4. Targeted therapeutic dosing, in which a nutrient is used at pharmacologic doses for a defined indication, with monitoring and an endpoint
5. Excess and toxicity, in which cumulative exposure produces measurable harm

Zones two and four are where integrative and functional practice does most of its work, and where the evidence base is thinnest and the potential for self-deception greatest. They deserve rigor, not indulgence.

The relevance of this discussion has only grown. The dietary supplement industry generates over $50 billion annually in the United States alone, and surveys consistently show that more than half of American adults take at least one supplement regularly. Meanwhile, the resurgence of deficiency diseases in individuals following extreme dietary patterns, including carnivore and other very-low-produce diets, is a reminder that the fundamentals of nutrition science have not been superseded by ideology.

The chapter proceeds in seven parts. It reviews the classical deficiency diseases and their modern relevance; maps the zone of functional insufficiency and altered requirement; distinguishes the molecular forms in which each vitamin is actually consumed; examines the toxicity profiles of individual vitamins, with emphasis on vitamin B6 neurotoxicity and fat-soluble hypervitaminosis; evaluates the epidemiologic evidence linking high-dose B-vitamin supplementation to cancer risk; addresses nutrient interactions and cofactor dependency; and closes with a practical framework for therapeutic dosing and clinical counseling.

2. Classical Deficiency Syndromes: Lessons from History

Before examining the harms of excess, it is essential to recall why vitamins matter in the first place. The classical deficiency diseases remain among the most instructive case studies in the history of medicine, not only for what they taught us about micronutrients, but for what they reveal about the relationship between dietary patterns, social conditions, and disease.

2.1 Vitamin C and Scurvy

Scurvy, the prototypical deficiency disease, results from prolonged absence of dietary ascorbic acid. Clinical features include perifollicular hemorrhages, corkscrew hairs, bleeding and swollen gums, poor wound healing, joint effusions, and ultimately death from hemorrhage or infection. The disease was epidemic among sailors during the Age of Exploration; James Lind’s 1747 trial aboard HMS Salisbury is often cited as one of the earliest controlled clinical experiments, demonstrating that citrus fruits could prevent and treat the condition.[1,2]

What makes scurvy particularly instructive for the modern clinician is its potential re-emergence in populations that might seem unlikely candidates. Because muscle meat contains negligible vitamin C, and organ meats contain somewhat more, a diet that eliminates nearly all plant foods without supplementation or organ meat consumption is, on nutritional grounds, capable of depleting body stores over a period of months. Whether this translates into clinically apparent scurvy at meaningful frequency among adults following carnivore or similar very-low-produce diets has not, to the author’s knowledge, been established by systematic study. The concern is mechanistically plausible and should be treated as a hypothesis warranting clinical vigilance rather than a documented epidemiologic phenomenon.

Scurvy does continue to appear in other contemporary contexts: psychiatric illness, particularly avoidant/restrictive food intake disorder; chronic alcoholism; food insecurity; and elderly individuals with limited dietary variety. The practical point is that the diagnosis is easily missed, because bleeding and poor wound healing are commonly attributed to other causes before nutritional deficiency is considered.

2.2 Vitamin B1 (Thiamine) and Beriberi / Wernicke–Korsakoff Syndrome

Thiamine deficiency produces two major clinical syndromes. Beriberi, named from the Sinhalese for “I cannot, I cannot,” reflecting the profound weakness it causes, manifests as either wet beriberi (high-output cardiac failure, peripheral edema, tachycardia) or dry beriberi (peripheral neuropathy, muscle wasting, difficulty walking). Wernicke encephalopathy, the acute neurologic emergency of thiamine deficiency, presents with the classic triad of confusion, ataxia, and ophthalmoplegia; untreated, it can progress to Korsakoff syndrome, an irreversible state of anterograde amnesia and confabulation.[22]

Historically, beriberi was endemic in populations subsisting on polished white rice, which strips the thiamine-rich bran. Today, the primary at-risk population is individuals with alcohol use disorder, in whom poor dietary intake, impaired intestinal absorption, and increased metabolic demand converge. Thiamine deficiency is also reported after bariatric surgery, in hyperemesis gravidarum, in patients on prolonged parenteral nutrition without adequate supplementation, and in critically ill patients with refeeding syndrome.

Thiamine deserves note as the cleanest illustration of a principle developed later in this chapter. In Wernicke encephalopathy, the correct dose is far above any reference intake, the correct route is parenteral, and treatment precedes confirmatory testing. Nobody regards this as megadosing, because it has an indication, a rationale, an endpoint, and a monitoring plan. That is the distinction the chapter is built on.

2.3 Vitamin B3 (Niacin) and Pellagra

Pellagra, the disease of the four D’s (dermatitis, diarrhea, dementia, death), results from severe niacin deficiency. The characteristic photosensitive dermatitis, often in a Casal necklace distribution around the neck, was a diagnostic hallmark in the early twentieth-century American South, where populations dependent on unprocessed corn suffered enormously; maize contains niacin in a bound, biologically unavailable form unless treated with alkali. Joseph Goldberger’s epidemiologic investigations in the 1910s and 1920s demonstrated that pellagra was a nutritional disease rather than an infectious one, an insight initially met with fierce resistance.[21]

Today, pellagra is rare in developed nations due to niacin fortification of grain products, but it still occurs in alcoholism, anorexia nervosa, malabsorptive conditions, and in populations dependent on unfortified maize. Cases associated with isoniazid therapy, which interferes with niacin metabolism, and with carcinoid syndrome, which diverts tryptophan to serotonin synthesis, are also well described. Both are examples of disease-state and drug-induced alteration of requirement rather than inadequate intake.

2.4 Vitamin D and Rickets / Osteomalacia

Vitamin D deficiency impairs calcium and phosphate homeostasis, leading to rickets in children (growth plate abnormalities, skeletal deformities, bowing of long bones, craniotabes) and osteomalacia in adults (diffuse bone pain, proximal muscle weakness, increased fracture risk). Historically, rickets was rampant in industrialized cities where air pollution and indoor labor limited sunlight exposure, earning it the name “the English disease.”

Despite widespread awareness and the availability of supplementation, vitamin D deficiency remains prevalent globally. NHANES data consistently show that a substantial proportion of the U.S. population has serum 25-hydroxyvitamin D levels below 20 ng/mL, and that this proportion is higher among individuals with darker skin pigmentation, those who are obese, and the elderly.[19] The optimal serum level remains debated. The consequences of frank deficiency are not.

2.5 Vitamin A and Xerophthalmia

Vitamin A deficiency remains the leading cause of preventable childhood blindness worldwide, primarily in sub-Saharan Africa and South Asia. The clinical spectrum ranges from night blindness through conjunctival xerosis and Bitot spots to corneal ulceration and permanent blindness. Vitamin A deficiency also impairs immune function, increasing susceptibility to measles, diarrheal diseases, and respiratory infections; the WHO estimates that supplementation programs have prevented millions of childhood deaths.[20]

In developed nations, vitamin A deficiency is uncommon but does occur in malabsorptive conditions (cystic fibrosis, Crohn disease, celiac disease, chronic liver disease) and in individuals on extremely restricted diets.

2.6 The Relevance of Historical Deficiency to Modern Practice

These classical syndromes are not merely historical curiosities. They serve three functions in contemporary clinical thinking. First, they remind us that vitamins are essential, and that supplementation in the context of documented deficiency is unambiguously beneficial and sometimes lifesaving. Second, they provide the foundation for the cultural assumption, now deeply embedded in popular health consciousness, that vitamins are inherently good — an assumption that, when extrapolated to megadosing, leads directly to the harms described later in this chapter. Third, the resurgence of deficiency in individuals on restrictive diets underscores that even in an era of food abundance, deficiency is only one dietary ideology away.

There is a fourth function, less often noted. Almost every classical deficiency syndrome that a modern clinician will actually encounter arises not from a diet too poor in the nutrient, but from a physiologic state that alters its absorption, transport, or demand: alcohol use disorder, bariatric surgery, malabsorption, pregnancy, critical illness, drug interference. That observation is the bridge to the next section.

3. The Zone Between: Functional Insufficiency and Altered Requirement

The three-zone model of deficiency, adequacy, and excess has an obvious appeal: it is simple, it is teachable, and it maps neatly onto the reference intake tables. It is also incomplete. It treats “adequate” as a single state defined by a population statistic, and it treats the absence of a classical deficiency syndrome as evidence that nutrient function is normal. Neither assumption survives contact with individual patients.

3.1 What Functional Insufficiency Means, and What It Does Not

Functional insufficiency refers to a state in which a nutrient-dependent biochemical process is measurably impaired, or a physiologic demand is measurably unmet, in the absence of the classical deficiency syndrome. It is a defensible concept with a long history in nutrition science: the rise in methylmalonic acid that precedes megaloblastic anemia in B12 deficiency, the rise in homocysteine that precedes any clinical manifestation of folate depletion, the elevation of parathyroid hormone that precedes osteomalacia in vitamin D insufficiency. These are not fringe ideas. They are the reason serum B12 alone is recognized as an imperfect test.

Three cautions keep the concept honest.

First, a marker is only as good as its validation. Methylmalonic acid, homocysteine, holotranscobalamin, erythrocyte transketolase activation, and erythrocyte glutathione reductase activation have literature behind them. Many panels sold as “functional nutrient testing” do not, and some report dozens of results with no established reference behavior, no demonstrated response to repletion, and no outcome data. Ordering a test that cannot change a decision is not assessment; it is expenditure.

Second, a marker is not an indication by itself. An abnormal functional marker identifies a hypothesis to be reconciled with dietary history, medications, disease state, and clinical phenotype. Homocysteine rises with renal impairment, hypothyroidism, and several drugs, none of which are corrected by folate.

Third, insufficiency is not a license for supraphysiologic dosing. If the shortfall is genuine, repletion generally requires a repletion dose, not a megadose, and it has a defined endpoint: the marker normalizes, or the hypothesis was wrong.

3.2 States That Alter Requirement

Requirement is not a fixed property of a person. It shifts with physiology, pathology, and pharmacology. The states below recur often enough in practice that they should be part of any nutritional assessment, and several of them appear elsewhere in this chapter as the settings in which classical deficiency still occurs.

• Malabsorption and altered anatomy: celiac disease, Crohn disease, pancreatic insufficiency, cholestasis, short bowel, and bariatric surgery, particularly procedures that bypass the duodenum and proximal jejunum
• Chronic inflammation and critical illness, which alter transport proteins and distribution and can make static serum levels difficult to interpret
• Medications: proton pump inhibitors and metformin (B12), methotrexate and certain anticonvulsants (folate), isoniazid, hydralazine, and penicillamine (B6), orlistat and bile acid sequestrants (fat-soluble vitamins), and long-term loop diuretics (thiamine)
• Life stage and reproductive demand: pregnancy, lactation, growth, and advanced age
• Dietary pattern: veganism (B12), carnivore and other very-low-produce patterns (vitamin C, folate), long-standing elimination diets, and food insecurity
• Organ dysfunction: renal disease altering vitamin D activation, hepatic disease altering storage and activation, and heart failure altering thiamine handling
• Substance use: alcohol use disorder, which impairs intake, absorption, and utilization simultaneously
• Genetic variation in absorption, transport, activation, and catabolism

The clinical value of this list is not that it justifies supplementation. It is that it identifies the patients in whom a normal-looking serum value should not end the inquiry, and equally, the patients in whom a supplement is likely to be doing something.

3.3 Static Versus Functional Markers

The distinction between a static marker, which measures the amount of a nutrient present in a compartment, and a functional marker, which measures whether a nutrient-dependent process is working, is central to assessment. Neither category is superior in the abstract. Static markers are better standardized; functional markers are often more sensitive to early depletion and more informative about tissue-level status. Table 1 summarizes the markers with the strongest support.

Table 1. Static and Functional Markers of Vitamin Status

Two practical rules follow. Interpret any nutrient marker in the light of concurrent inflammation, because several of them behave as negative acute-phase reactants. And when a functional marker is used to justify treatment, plan to repeat it, because a marker that never gets rechecked has functioned as a sales tool rather than a test.

3.4 Biochemical Individuality and Genetics

Biochemical individuality is real, and the clinical literature supports its relevance for specific, well-characterized variants. It is also the single most abused concept in integrative practice, and the abuse follows a predictable pattern: a common polymorphism is reported on a direct-to-consumer panel, is described to the patient as a defect, and becomes the justification for lifelong high-dose supplementation that no one plans to reassess.

Variants worth knowing, and what they actually support:

• MTHFR C677T and A1298C. The 677TT genotype is associated with reduced enzyme activity and modestly higher homocysteine, particularly at low folate intake. It is common, it is not a disease, and it does not establish that a patient “cannot methylate.” Where folate status is adequate, the phenotypic consequence is small.
• CYP24A1 loss-of-function variants. These impair catabolism of active vitamin D and can produce hypercalcemia at ordinary supplement doses. They may cluster in siblings, and they belong in the differential for unexplained hypercalcemia on modest vitamin D intake.[18]
• BCMO1 (BCO1) variants. These reduce conversion of beta-carotene to retinol, which is a relevant consideration for individuals relying entirely on provitamin A sources, and a reminder that “vitamin A intake” from a food frequency questionnaire conflates two different exposures.
• Transporter and receptor variation affecting B12, folate, and vitamin D binding protein. Well described in research settings; not yet actionable in routine practice.

The defensible position is straightforward. Genotype can raise or lower the prior probability of a nutritional problem. It does not establish that the problem is present, it does not substitute for a phenotypic or biochemical measurement, and it does not by itself justify a dose. Treat the patient, not the printout.

4. Vitamers: Why Form Matters, and Why Form Is Not Safety

Perhaps the most consistent oversimplification in both the popular and the academic literature on vitamins is the treatment of each vitamin as a single chemical entity. It is not. Each is a family of related molecules, or vitamers, that differ in absorption, transport, metabolism, tissue distribution, biological potency, and, in some cases, adverse effect profile. A study of one vitamer is not automatically a study of another.

Two errors follow from ignoring this. The conventional error is to generalize a finding from one form to the entire vitamin, as when the SELECT trial result for synthetic alpha-tocopherol is reported as evidence about “vitamin E.” The integrative error is the mirror image: to assume that because a form is more physiologic, better absorbed, or “natural,” it is therefore safe at any dose. Neither follows.

State the principle precisely: pharmacokinetics, metabolism, biological activity, and the available evidence may differ meaningfully by vitamer. Absence of toxicity data for an alternative form is not evidence of its absence of toxicity.

• Vitamin B6. Pyridoxine hydrochloride is the form in nearly all supplements and in essentially all the neuropathy case literature. Pyridoxal 5’-phosphate (P5P) is the active coenzyme form and bypasses hepatic pyridoxal kinase activation. There is mechanistic work suggesting that high concentrations of pyridoxine itself may inhibit PLP-dependent enzymes, which would make the neuropathy an effect of the supplemental form rather than of vitamin B6 activity as such.[5,6] This is a plausible and interesting hypothesis. It is not a demonstration that P5P is safe at doses that make pyridoxine neurotoxic, and P5P should not be dosed as though it were exempt.
• Folate. Food folates, folic acid, folinic acid (5-formyl-THF), and L-5-methyltetrahydrofolate are not interchangeable exposures. Folic acid is synthetic, requires reduction by dihydrofolate reductase, and at higher intakes appears in circulation unmetabolized; 5-MTHF is the circulating form and bypasses that step. Nearly all of the fortification and cancer-signal literature concerns folic acid. The neural tube defect evidence, which is the strongest evidence in the entire field, also concerns folic acid. Substituting 5-MTHF on theoretical grounds in a patient for whom the folic acid evidence is the relevant evidence is a defensible choice only if it is made knowingly.
• Vitamin B12. Cyanocobalamin is stable, inexpensive, and the form used in most trials. Methylcobalamin, adenosylcobalamin, and hydroxocobalamin differ in retention and in specific clinical niches; hydroxocobalamin is preferred in tobacco amblyopia and in cyanide toxicity, and cyanocobalamin is reasonably avoided in advanced renal disease. For routine repletion, the evidence does not establish clinical superiority of any form, and dose and route matter far more than the cobalamin moiety.
• Vitamin A. Preformed retinol and retinyl esters are absorbed largely unregulated and carry the fracture and teratogenicity signals. Provitamin A carotenoids undergo regulated conversion and do not carry the same fracture risk, though beta-carotene supplementation carries its own well-established signal for lung cancer in smokers, which is not a reason to consider it benign.
• Vitamin E. Synthetic all-rac-alpha-tocopherol at 400 IU/day is the exposure in the mortality and prostate cancer literature. Mixed tocopherols and tocotrienols differ biologically, and high-dose alpha-tocopherol suppresses gamma-tocopherol concentrations, a plausible mechanism by which isolated supplementation may not reproduce the effects of dietary vitamin E. Whether mixed forms are safer at equivalent doses has not been tested at the scale required to know.
• Vitamin D. Cholecalciferol (D3) raises and maintains serum 25-hydroxyvitamin D more effectively than ergocalciferol (D2) at equivalent doses, particularly with intermittent dosing. Both are toxic at sufficient dose. Calcitriol and other activated analogues are drugs with a much narrower margin and require calcium monitoring as a matter of course.
• Niacin. Nicotinic acid, nicotinamide, inositol hexanicotinate, and extended-release formulations behave differently. Flushing is a nicotinic acid effect; nicotinamide does not lower LDL cholesterol; sustained-release preparations carry disproportionate hepatotoxicity. Reporting all of these as “niacin” is a genuine source of clinical error.

The practical implication for the clinician is modest but real: record the form, not just the nutrient, in the medication list. A patient taking 100 mg of pyridoxine and a patient taking 100 mg of P5P have not necessarily taken the same risk, and neither has a patient taking 1 mg of folic acid and a patient taking 1 mg of 5-MTHF. Neither, however, has grounds for reassurance in the absence of an indication.

5. When Vitamins Become Toxins: Evidence of Harm at High Doses

The assumption that water-soluble vitamins are categorically safe because they are excreted in the urine is one of the most persistent and dangerous myths in popular nutrition. It is true that water-soluble vitamins have wider therapeutic windows than their fat-soluble counterparts. Wider does not mean unlimited, and chronic high-dose supplementation of several water-soluble vitamins produces well-characterized toxicity syndromes.

5.1 Vitamin B6 (Pyridoxine): The Paradigm of Water-Soluble Toxicity

Vitamin B6 toxicity is the most clinically important example of water-soluble vitamin harm, because it is common, dose-dependent, and frequently unrecognized until neurologic damage has occurred.

5.1.1 Clinical Presentation

Chronic high-dose pyridoxine causes a predominantly sensory peripheral neuropathy. Patients present with progressive numbness, tingling, and burning pain in the hands and feet in a stocking-glove distribution. As the neuropathy advances, proprioceptive loss leads to sensory ataxia: difficulty walking in the dark, a positive Romberg sign, and impaired fine motor coordination. Deep tendon reflexes are diminished or absent. In severe cases the neuropathy extends proximally and involves the trunk and face.[3,4]

The neuropathy is axonal and primarily affects large-diameter sensory fibers. Nerve conduction studies typically show reduced sensory nerve action potential amplitudes with preserved conduction velocities, consistent with axonal degeneration rather than demyelination. The pathology is a distal dying-back process affecting the dorsal root ganglion neurons.

5.1.2 Dose–Response and Threshold

Toxicity is well established at doses of 100–250 mg/day taken chronically, but the threshold may be lower than historically appreciated. The Tolerable Upper Intake Level set by the Institute of Medicine is 100 mg/day, yet pharmacovigilance data and systematic reviews have identified cases of neuropathy at doses within the range of commonly marketed supplements, 25–50 mg/day, particularly with prolonged use over months to years.[6] Individual susceptibility evidently varies, and the UL may not protect all individuals.

The RDA for adults is 1.3–1.7 mg/day, a figure that highlights the extraordinary gap between physiologic need and the doses contained in many commercial supplements and energy formulations, which often provide 5,000 to 10,000 percent of the RDA per serving. Note also that B6 is among the most common inadvertent stacking exposures in practice: a B-complex, a separate B6, an energy drink, and a magnesium-with-B6 preparation can silently sum to a neurotoxic daily dose.

5.1.3 Reversibility and Prognosis

The neuropathy is generally reversible on discontinuation, but recovery can take months to years, and after prolonged high-dose exposure some residual deficit may persist. The key clinical point is early recognition: once the diagnosis is made, the intervention is simply stopping the supplement.

5.2 Niacin (Vitamin B3): Hepatotoxicity and Metabolic Harm

Niacin occupies a unique position among vitamins because it has been used at pharmacologic doses of 1.5–3 g/day as a lipid-modifying agent. At these doses the side-effect profile is substantial: flushing (prostaglandin-mediated cutaneous vasodilation), hepatotoxicity, hyperglycemia and worsening insulin resistance, hyperuricemia with precipitation of gout, gastrointestinal irritation, and activation of peptic ulcer disease.

The hepatotoxicity is particularly notable. Sustained-release formulations carry a higher risk of liver injury than immediate-release forms, and cases of fulminant hepatic failure have been reported. The mechanism appears to involve dose-dependent hepatocellular injury, and liver function monitoring is mandatory when niacin is used therapeutically.

The UL for niacin from supplements and fortified foods is 35 mg/day, set for the flushing endpoint, well below the pharmacologic doses used for dyslipidemia. This gap is not an inconsistency; it is the clearest illustration in the reference intake tables of the difference between a nutritional intake and a drug dose.

5.3 Vitamin C: Kidney Stones and Oxalate Metabolism

Vitamin C is widely perceived as harmless at any dose, and this perception is not fully supported by the evidence. Ascorbic acid is metabolized to oxalate, and high-dose supplementation, typically 1,000 mg/day or more, significantly increases urinary oxalate excretion. Meta-analyses of metabolic studies demonstrate a consistent and statistically significant rise in 24-hour urinary oxalate at high ascorbate intakes,[17,31] and prospective cohort data, including the Health Professionals Follow-Up Study, have linked vitamin C supplement use with modestly increased kidney stone risk in men.[16]

The absolute risk in average-risk individuals is likely small, and some authors have argued that the concern is overstated. For patients with a personal or family history of calcium oxalate stones, hyperoxaluria, or chronic kidney disease, the recommendation to avoid chronic high-dose vitamin C is well supported.

Two qualifications belong here. Intravenous ascorbate produces plasma concentrations unattainable orally and is a distinct pharmacologic intervention with its own literature, its own contraindications (G6PD deficiency, renal impairment, oxalate nephropathy risk), and its own monitoring requirements; it should not be reasoned about using oral supplement data. And gastrointestinal tolerance, often cited as a natural ceiling on oral dosing, limits absorption but does not limit the oxalate exposure of what has already been absorbed.

5.4 Fat-Soluble Vitamins: Narrower Margins, Greater Consequences

Fat-soluble vitamins accumulate in body stores and are not rapidly excreted, giving them inherently narrower safety margins. Vitamin K toxicity is extremely rare. Vitamins A, D, and E have well-characterized toxicity profiles that are clinically relevant.

5.4.1 Vitamin A (Retinol): Bone Loss and Fracture Risk

Chronic excess of preformed vitamin A has been associated with bone loss and increased hip fracture risk. In the Nurses’ Health Study, women in the highest category of retinol intake (≥2,000 µg/day) had substantially elevated hip fracture risk relative to those in the lowest category.[13] Women currently taking a specific vitamin A supplement showed a 40 percent increased risk that did not reach statistical significance (RR 1.40, 95% CI 0.99–1.99), a distinction frequently lost in secondary citation of this study and worth preserving. Beta-carotene did not contribute significantly to fracture risk. Pooled analyses of prospective studies support an association between high retinol intake and hip fracture (RR approximately 1.40), while also finding that low blood retinol is associated with increased hip fracture risk (RR approximately 1.27), a bidirectional pattern consistent with the U-shaped framework advanced in this chapter.[12] Mechanistically, retinoids stimulate osteoclastogenesis and bone resorption, and animal models show cortical bone thinning with excess retinol.

Acute vitamin A toxicity, classically described after consumption of polar bear or seal liver, presents with headache, vomiting, elevated intracranial pressure, and skin desquamation. This is largely a historical and Arctic-medicine phenomenon, but it illustrates the principle that even natural dietary sources can deliver toxic doses.

As noted in section 4, beta-carotene does not carry the same fracture risk, because conversion to retinol is regulated. The concern is specifically preformed retinol from supplements and animal sources, and the distinction should be preserved in both directions: regulated conversion is not the same as harmlessness, as the beta-carotene lung cancer trials in smokers demonstrated.

5.4.2 Vitamin D: Hypercalcemia and Acute Kidney Injury

Vitamin D toxicity manifests through its downstream effect on calcium homeostasis. Excess increases intestinal calcium absorption, leading to hypercalcemia, which produces polyuria, polydipsia, dehydration, nausea, confusion, cardiac arrhythmias, and in severe cases acute kidney injury from nephrocalcinosis and calcium deposition in renal tubules.

The published case literature includes children and adults developing severe hypercalcemia and acute kidney injury from sustained high-dose supplementation. A systematic review of case reports of vitamin D toxicity arising from overcorrection of deficiency documents this pattern across a range of doses and durations, with resolution typically following cessation and supportive management.[18] Clinicians evaluating unexplained hypercalcemia should recall that inherited disorders of vitamin D catabolism, such as CYP24A1 loss-of-function variants, can produce a similar picture at ordinary supplement doses and may cluster in siblings, a distinction with direct management consequences.

The UL for vitamin D is 4,000 IU/day for adults, though some clinical contexts warrant higher doses under monitoring. The essential point is that high-dose vitamin D requires periodic monitoring of serum 25-hydroxyvitamin D and calcium. At those doses it is a medication, not a supplement.

5.4.3 Vitamin E: Mortality and Prostate Cancer

High-dose vitamin E supplementation has been subjected to extensive evaluation in large randomized trials, and the results have been consistently unfavorable. A meta-analysis of 19 clinical trials found that supplementation at 400 IU/day or more was associated with a small but statistically significant increase in all-cause mortality.[14] The SELECT trial found that vitamin E at 400 IU/day increased the risk of prostate cancer by 17 percent over a median follow-up of 5.5 years, a result both statistically significant and clinically relevant in a study of more than 35,000 men.[15]

High-dose vitamin E has also been associated with increased risk of hemorrhagic stroke, plausibly through interference with vitamin K-dependent coagulation and platelet aggregation. This is particularly relevant for patients on anticoagulant therapy. As noted above, these data concern synthetic alpha-tocopherol; the trials do not settle the question for mixed tocopherols, and no comparably powered trial of mixed forms exists to settle it in either direction.

6. B Vitamins and Cancer Risk: The Epidemiologic Evidence

The relationship between B-vitamin supplementation and cancer risk is among the most important and underappreciated topics in nutritional epidemiology. B vitamins are essential for one-carbon metabolism, DNA synthesis, and methylation, processes central to normal cell division and repair. The same metabolic machinery can, in theory, fuel the growth of pre-existing neoplastic cells when substrate is provided in excess.

The evidence is complex, site-specific, and not uniform in direction. The sections that follow present observational signals of harm alongside randomized data that do not corroborate them. Readers should treat the resulting picture as an unresolved hypothesis with clinically relevant implications for specific subgroups, not as a demonstrated general carcinogenicity of high-dose B vitamins.

6.1 Vitamin B6 and B12: The Lung Cancer Signal

The most discussed epidemiologic data linking individual B vitamins to cancer come from the VITAL (Vitamins and Lifestyle) cohort, a prospective study of over 77,000 adults aged 50 to 76. In analyses published by Brasky and colleagues, long-term individual supplementation with vitamin B6 above 20 mg/day was associated with a 30 to 40 percent increased risk of lung cancer in men, with the highest dose category (more than 20 mg/day for 10 or more years) showing approximately 82 percent increased risk. For vitamin B12 above 55 µg/day, the corresponding increase was approximately 98 percent in men. These associations were strongest in current smokers.[7]

Critically, the associations were observed with individual B6 and B12 supplements, not with B vitamins obtained from food or from multivitamins, where doses are typically lower. They were also not seen in women, suggesting a sex-specific interaction, possibly related to hormonal influences on one-carbon metabolism or to the higher baseline lung cancer risk in male smokers.

Other observational studies have reported associations between elevated circulating B12 and incident cancer, including lung, liver, and hematologic malignancies, though the direction of causation is debated. Reverse causation, in which subclinical or undiagnosed malignancy elevates B12 through increased haptocorrin production or hepatic release, is a well-recognized confounder.

The VITAL findings have attracted substantive methodological criticism. Because lung cancer risk is dominated by smoking intensity, residual confounding by pack-years and by chronic obstructive pulmonary disease, both of which differed markedly between cases and non-cases, offers a plausible alternative explanation. Several studies and at least one systematic review examining circulating vitamin B6 have reported inverse relationships with lung cancer risk, the opposite direction to the VITAL supplementation signal.[8] The Norwegian trials of folic acid plus B12 in patients with ischemic heart disease reported increased cancer incidence and mortality, which is often cited alongside VITAL, though that population and that intervention differ from long-term individual B6 supplementation.[9]

The honest summary is that VITAL is a well-conducted observational finding that has not been replicated in randomized data and remains contested. It warrants caution in the specific population studied, above all the male smoker taking individual high-dose B6 or B12 without an indication, rather than a general conclusion about B-vitamin carcinogenicity.

6.2 Folate and Folic Acid: The Dual-Agent Hypothesis

Folate occupies perhaps the most nuanced position of any vitamin with respect to cancer risk, because it plays genuinely opposing roles depending on timing and context, a phenomenon termed the dual-agent or dual-modulator hypothesis.

Before summarizing that evidence, it is worth stating precisely what the exposure under discussion is. Most of the literature concerns synthetic folic acid, at supplemental or fortification doses, in populations whose baseline folate status is generally adequate and whose B12 status is variable. It is not, for the most part, a literature about food folate, about folinic acid, about 5-MTHF, or about repleting a deficient patient. Folate status, folate dose, folate form, and concurrent B12 sufficiency are four separate variables, and conflating them is the main reason clinicians come away from this literature with the wrong impression.

Adequate folate status is protective against several cancers. Low folate is associated with increased risk of colorectal cancer, through impaired DNA repair and aberrant methylation, cervical dysplasia, and possibly breast and pancreatic cancers. Folate’s role in preventing neural tube defects is unambiguous, and folic acid fortification of grain products in the United States and Canada, mandated since 1998, is among the most successful public health interventions of the modern era.

On the other hand, high-dose folic acid supplementation, in the synthetic form that bypasses normal intestinal regulation of folate absorption and can appear unmetabolized in circulation, has raised concerns about promoting the growth of existing pre-neoplastic and neoplastic lesions. Several lines of evidence bear on this:

• Colorectal adenoma recurrence. The Aspirin/Folate Polyp Prevention Study, a randomized trial, found that folic acid at 1 mg/day for up to 6 years was associated with a non-significant increase in advanced adenoma recurrence and a statistically significant increase in the number of adenomas per patient, particularly in those with higher baseline folate.[10]
• Breast cancer. The literature here is genuinely mixed and should not be summarized in one direction. Some meta-analyses of prospective studies report that higher dietary folate intake is associated with reduced breast cancer risk,[32] while other analyses suggest a U-shaped pattern in which risk rises again at high supplemental intakes, particularly with concurrent alcohol use. A specific inflection point near the RDA has been proposed but is not firmly established across cohorts.
• Prostate cancer. Pooled and trial analyses have suggested a modest positive association between high folic acid intake and prostate cancer risk, though findings are heterogeneous.[11]

Against these signals, the randomized evidence is reassuring and must be given equal weight. A meta-analysis of individual participant data from randomized trials involving approximately 50,000 individuals found no significant increase in overall or site-specific cancer incidence during roughly five years of folic acid supplementation.[23] The Women’s Antioxidant and Folic Acid Cardiovascular Study likewise found no significant increase in cancer risk with combined folic acid, B6, and B12.[24] Trial durations may be too short to capture late promotion effects, but on present evidence the promotional hypothesis is unproven rather than confirmed.

The mechanistic rationale is biologically plausible: folate provides the one-carbon units needed for de novo nucleotide synthesis. In a cell that has already undergone malignant transformation, excess folate may accelerate proliferation. In normal cells, adequate folate supports faithful DNA replication and repair. The clinical implication is that folate supplementation should be targeted to those who need it — women of childbearing potential, individuals with documented deficiency or malabsorption, those on methotrexate or other antifolates — rather than prescribed broadly at high doses to people with no reason to take it.

One further point deserves emphasis because it is a genuine patient safety issue rather than a contested epidemiologic one: high-dose folate can correct the megaloblastic anemia of B12 deficiency while the neurologic damage progresses unchecked. B12 status should be established before committing a patient to sustained high-dose folate, in any form.

6.3 The Broader Pattern: U-Shaped Risk Curves

The B-vitamin data fit within a broader pattern well described across nutritional epidemiology: the U-shaped or J-shaped dose–response curve, in which both the lowest and the highest levels of intake or circulating concentration are associated with increased risk, while moderate levels are associated with the lowest risk.

This pattern has been documented for vitamin D, where both very low and very high 25-hydroxyvitamin D levels have been associated with increased mortality in large cohorts; for vitamin A, where deficiency increases infection risk and excess increases fracture risk; for vitamin E, where deficiency causes hemolytic anemia and excess is associated with hemorrhagic stroke and prostate cancer; and for folate, where deficiency increases colorectal cancer and neural tube defect risk while excess may promote existing neoplasia.

Systematic evidence reviews conducted for the U.S. Preventive Services Task Force, and subsequent global overviews of supplement use, have consistently concluded that routine vitamin and mineral supplementation in non-deficient adults is not supported for the primary prevention of cardiovascular disease or cancer.[25,26] Taken together with the toxicity data reviewed above, this supports a working position that the therapeutic window is narrower than commonly assumed. It should be noted that a formally nonlinear dose–response has been demonstrated for some vitamins and specific endpoints; it is not established as a general law across all micronutrients. The U-shaped curve is a useful heuristic and a well-supported observation in several instances, not a proven universal.

7. Nutrient Interactions and Cofactor Dependency

Vitamins are studied one at a time and consumed all at once. Almost every framework in this chapter so far has evaluated nutrients in isolation, which is how the evidence was generated but not how physiology works. Several interactions are consequential enough to change clinical decisions.

• Folate, B12, and B6 in one-carbon metabolism. These three nutrients operate in a single interlocking pathway. Supplementing one in isolation shifts flux through the others; correcting homocysteine may require attention to all three plus riboflavin, which is the cofactor for MTHFR; and, as noted above, folate can mask the hematologic manifestation of B12 deficiency. One-carbon metabolism should be assessed as a system.
• Vitamin D, calcium, magnesium, and PTH. Vitamin D increases intestinal calcium absorption, which is both the mechanism of its benefit and the mechanism of its toxicity. Magnesium is required for the enzymes that hydroxylate vitamin D at both the hepatic and renal steps, and magnesium depletion can blunt the response to supplementation. PTH is the integrating signal and is often the most informative single test in an ambiguous vitamin D picture. Any decision to combine high-dose vitamin D with calcium supplementation should be made deliberately, because the hypercalcemia risk is additive.
• Vitamin A and vitamin D signaling. Retinoid X receptor heterodimerization means these two vitamins share downstream signaling machinery, and their effects are not independent. This is the most-cited and least-quantified interaction in the integrative literature; it is mechanistically real and its clinical dosing implications are, at present, undefined. It should be described as such rather than used to justify specific ratios.
• Vitamin E and vitamin K. High-dose alpha-tocopherol interferes with vitamin K-dependent coagulation, which is the plausible mechanism of the hemorrhagic stroke signal and the reason high-dose vitamin E warrants explicit review in any patient on warfarin or a direct oral anticoagulant.
• Vitamin C and iron. Ascorbate substantially enhances non-heme iron absorption. This is a therapeutic asset in iron deficiency, taken together with the iron dose, and a genuine hazard in hereditary hemochromatosis or transfusional iron overload.
• Zinc and copper. Sustained zinc supplementation, above roughly 40 mg/day, induces intestinal metallothionein and can produce copper deficiency, with sideroblastic anemia, neutropenia, and a myelopathy that is often mistaken for B12 deficiency. This is one of the most commonly missed supplement-induced syndromes in practice, and it arises almost entirely from unmonitored long-term self-supplementation.
• Fat-soluble vitamin competition and absorption. All four fat-soluble vitamins depend on bile acids and dietary fat for absorption, and they compete at shared transport steps. High-dose supplementation of one may modestly affect the others; more importantly, any condition or medication that impairs fat absorption impairs all four simultaneously.

The general lesson is that isolated single-nutrient megadosing is not merely unnecessary in most cases; it is the dosing strategy most likely to create a second problem while addressing the first.

8. Therapeutic Nutrient Dosing Versus Megadosing

Much of the confusion in this field comes from using one word, “supplementation,” for two entirely different activities. Thiamine at 500 mg intravenously for suspected Wernicke encephalopathy, niacin at 2 g/day for refractory dyslipidemia, vitamin D at 50,000 IU weekly for documented severe deficiency, and pyridoxine at pharmacologic doses for isoniazid toxicity or specific inborn errors are all supraphysiologic nutrient dosing. So is a patient taking 100 mg of B6 daily for eleven years because a website suggested it. The difference is not the dose. It is everything around the dose.

Therapeutic nutrient dosing, properly conducted, has eight components. A regimen missing several of them is not therapy; it is megadosing with a rationale attached after the fact.

1. Indication. A specific clinical problem or documented deficit that the nutrient is being used to address, stated before treatment begins.
2. Biochemical or physiologic rationale. A mechanism connecting the nutrient to the indication that would survive being written down and read by a skeptical colleague.
3. Baseline assessment. The relevant measurement obtained before dosing, including the tests that would detect harm as well as those that would detect benefit.
4. Defined dose, form, and duration. Specified in advance, with the vitamer named, and with an expected time to response.
5. Monitoring plan. Scheduled, matched to the known toxicity of the agent: liver enzymes for niacin, calcium and 25-hydroxyvitamin D for vitamin D, neurologic examination for B6, INR review for vitamin E in anticoagulated patients.
6. Therapeutic endpoint. What outcome or marker will define success, and by when.
7. Toxicity surveillance. Explicit attention to the adverse effects catalogued in section 5, including a total daily dose calculated across every product the patient takes, not just the one being prescribed.
8. De-escalation strategy. What happens when the endpoint is met, and what happens when it is not. Every high-dose regimen should have a stopping condition. In practice, the absence of one is the single most reliable marker distinguishing therapy from habit.

The asymmetry is worth naming directly, because it is where the integrative and conventional critiques of each other usually collide. Recognizing biochemical individuality, functional insufficiency, and vitamer differences does not lower the evidentiary bar for high-dose treatment. It raises it. If the argument for a given regimen is that this patient’s requirement differs from the population norm, then the burden falls on the clinician to demonstrate that it does, to measure the response, and to stop when the demonstration fails. Individualization is a commitment to more measurement, not a license for less. Used in the other direction, “everyone is different” becomes an unfalsifiable justification for anything, and the field earns the skepticism it receives.

9. Practical Clinical Framework

Translating this evidence into practice requires a framework that balances the genuine benefits of correcting deficiency and the legitimate role of monitored therapeutic dosing against the measurable risks of chronic, unmonitored megadosing. Table 2 summarizes the key thresholds, red flags, and monitoring considerations for the vitamins discussed.

Table 2. Doses Associated With Concern and Clinical Red Flags

Note: The “Dose Associated With Concern” column deliberately mixes categories. It includes formal Tolerable Upper Intake Levels (niacin 35 mg/day; vitamin D 4,000 IU/day), observational cut-points at which risk signals appear (retinol above 2,000 µg/day), and trial doses associated with harm (vitamin E 400 IU/day). These are not equivalent forms of evidence, and the column should not be read as a list of established toxicity thresholds. Doses refer to the forms studied, which in most cases are the common supplemental vitamers described in section 4. Clinicians should consult the Institute of Medicine / National Academies Dietary Reference Intake tables for authoritative UL values.

9.1 Six Principles for Clinical Counseling

Principle 1: Assess before you treat. Supplementation should follow an assessment, not a preference. That assessment includes dietary history, clinical phenotype, medications, disease state, relevant biomarkers, and, where validated, functional or metabolic markers. Testing is a core part of that assessment and, for any sustained or high-dose regimen, it is not optional: the clinician who cannot state what was measured before starting has no way to know whether the regimen is working, and no way to recognize when it has begun to cause harm. What this principle does not endorse is the equation of a normal serum value with optimal cellular physiology, or the reflexive ordering of unvalidated panels whose results cannot change a decision. Measure what informs the decision, and let the result actually decide.

Principle 2: Respect the U-shaped curve, and use reference intakes as a floor rather than a target. Population dietary reference values were designed to meet the needs of nearly all healthy individuals. They are a useful nutritional baseline and a poor individualized therapeutic target, since requirements shift with disease state, absorption, medications, life stage, and dietary pattern. That said, departure from the population reference should be a reasoned clinical decision with a justification and a monitoring plan, not a default. For the great majority of patients with no such justification, intake near the reference range remains the right answer, and the marketed dose is usually far above it.

Principle 3: Distinguish food from supplements. The toxicity and cancer data reviewed in this chapter are driven almost entirely by supplemental vitamins, not by vitamins obtained from food. Dietary sources are largely self-limiting — it is difficult to eat enough liver to develop chronic vitamin A toxicity — and food-matrix effects modify bioavailability and metabolism in ways that isolated supplements do not replicate.

Principle 4: Record the form, and total the dose across every product. “B6 100 mg” is an incomplete entry: pyridoxine and P5P are different exposures, as are folic acid and 5-MTHF, retinol and beta-carotene, alpha-tocopherol and mixed tocopherols. Equally important, the clinically relevant number is the sum across the multivitamin, the B-complex, the individual supplement, the energy drink, and the fortified food, which is frequently several times what the patient believes they are taking.

Principle 5: Know the high-risk intersections. Male smokers on individual high-dose B6 or B12; patients with a history of adenomatous polyps on high-dose folic acid; anyone on sustained folate with unverified B12 status; postmenopausal women on retinol supplements; patients with kidney stone history on high-dose vitamin C; anyone on anticoagulants taking high-dose vitamin E; patients on long-term zinc without copper monitoring. These intersections of supplement, dose, and patient context are where the signal is strongest and where a five-minute medication review has the highest yield.

Principle 6: When high-dose therapy is indicated, run it as a drug protocol. There are genuine situations for pharmacologic nutrient therapy: niacin for refractory dyslipidemia, high-dose vitamin D for documented severe deficiency, parenteral thiamine for Wernicke encephalopathy, pyridoxine for isoniazid toxicity. In these cases the vitamin is functioning as a drug and should be treated accordingly, with the eight components set out in section 8 — indication, rationale, baseline, defined dose and form, monitoring, endpoint, toxicity surveillance, and a de-escalation plan.

10. Modern Deficiency in an Era of Excess: The Paradox Continues

It would be a mistake to conclude this chapter with the impression that supplementation is categorically harmful. The opposite error, undervaluing vitamins entirely, is just as dangerous, and in some clinical populations more immediately so.

The resurgence of scurvy in individuals following strict carnivore diets, the persistence of thiamine deficiency in alcohol use disorder and post-bariatric populations, the ongoing global burden of vitamin A deficiency in children, and the prevalence of vitamin D insufficiency in northern-latitude populations all remind us that deficiency is not a solved problem. Nor is functional insufficiency in the patient on chronic acid suppression, the patient with untreated celiac disease, or the patient whose medication list quietly antagonizes three nutrients at once.

The solution is not to abandon supplementation but to apply it with the rigor we would apply to any other medical intervention: the right molecule, in the right form, at the right dose, for the right patient, with the right monitoring, and with a clear plan for when to stop.

The prevailing cultural narrative — that vitamins are universally safe, that natural supplements cannot harm, that more is always better — is not supported by the evidence reviewed here. The counter-narrative, that nutrients are irrelevant beyond the prevention of frank deficiency, is not supported either. The truth is more nuanced, more interesting, and ultimately more useful: vitamins are powerful, biologically active compounds with genuine therapeutic potential and genuine toxicity. Treating them as harmless is as much a failure of clinical thinking as ignoring their benefits.

11. Conclusions

The arc of vitamin science bends from deficiency to sufficiency and, increasingly, from sufficiency to excess. The twentieth century taught us that vitamins cure specific deficiency diseases. The twenty-first is teaching us two lessons at once: that the territory between overt deficiency and megadose toxicity is real clinical terrain, containing functional insufficiency, altered requirement, and legitimate monitored therapeutic dosing; and that chronic megadosing in the absence of a defined indication introduces measurable risk, including neurotoxicity, hepatotoxicity, bone loss, kidney injury, and, for certain B vitamins and populations, epidemiologic signals of increased cancer risk that remain incompletely resolved.

Holding both lessons at once is harder than holding either alone, and it is the actual work of integrative and functional practice. The clinician’s task is to guide patients through the narrow therapeutic corridor between two forms of harm, which requires familiarity with the evidence, willingness to challenge popular assumptions in both directions, and the clinical humility to acknowledge that the optimal dose is usually far lower than the marketed dose — and that when it is not, one has to be able to say why, and to check. As Paracelsus understood five centuries ago, the dose makes the poison, and vitamins are no exception.

References

1. Lind J. A Treatise of the Scurvy. Edinburgh: Sands, Murray and Cochran, for A. Millar; 1753.

2. Carpenter KJ. The History of Scurvy and Vitamin C. Cambridge University Press; 1986.

3. Schaumburg H, Kaplan J, Windebank A, et al. Sensory neuropathy from pyridoxine abuse: a new megavitamin syndrome. N Engl J Med. 1983;309(8):445-448.

4. Dalton K, Dalton MJT. Characteristics of pyridoxine overdose neuropathy syndrome. Acta Neurol Scand. 1987;76(1):8-11.

5. Vrolijk MF, Opperhuizen A, Jansen EHJM, et al. The vitamin B6 paradox: supplementation with high concentrations of pyridoxine leads to decreased vitamin B6 function. Toxicol In Vitro. 2017;44:206-212.

6. Hadtstein F, Vrolijk M. Vitamin B6-induced neuropathy: exploring the mechanisms of pyridoxine toxicity. Adv Nutr. 2021;12(5):1911-1929.

7. Brasky TM, White E, Chen CL. Long-term, supplemental, one-carbon metabolism-related vitamin B use in relation to lung cancer risk in the Vitamins and Lifestyle (VITAL) cohort. J Clin Oncol. 2017;35(30):3440-3448.

8. Fanidi A, Muller DC, Yuan JM, et al. Circulating folate, vitamin B6, and methionine in relation to lung cancer risk in the Lung Cancer Cohort Consortium (LC3). J Natl Cancer Inst. 2018;110(1):57-67.

9. Ebbing M, Bønaa KH, Nygård O, et al. Cancer incidence and mortality after treatment with folic acid and vitamin B12. JAMA. 2009;302(19):2119-2126.

10. Cole BF, Baron JA, Sandler RS, et al. Folic acid for the prevention of colorectal adenomas: a randomized clinical trial. JAMA. 2007;297(21):2351-2359.

11. Figueiredo JC, Grau MV, Haile RW, et al. Folic acid and risk of prostate cancer: results from a randomized clinical trial. J Natl Cancer Inst. 2009;101(6):432-435.

12. Melhus H, Michaëlsson K, Kindmark A, et al. Excessive dietary intake of vitamin A is associated with reduced bone mineral density and increased risk for hip fracture. Ann Intern Med. 1998;129(10):770-778.

13. Feskanich D, Singh V, Willett WC, Colditz GA. Vitamin A intake and hip fractures among postmenopausal women. JAMA. 2002;287(1):47-54.

14. Miller ER 3rd, Pastor-Barriuso R, Dalal D, et al. Meta-analysis: high-dosage vitamin E supplementation may increase all-cause mortality. Ann Intern Med. 2005;142(1):37-46.

15. Klein EA, Thompson IM Jr, Tangen CM, et al. Vitamin E and the risk of prostate cancer: the Selenium and Vitamin E Cancer Prevention Trial (SELECT). JAMA. 2011;306(14):1549-1556.

16. Taylor EN, Stampfer MJ, Curhan GC. Dietary factors and the risk of incident kidney stones in men: new insights after 14 years of follow-up. J Am Soc Nephrol. 2004;15(12):3225-3232.

17. Farkouh A, Seibly E, Buell MI, Jhang D, Amasyali AS, Okhunov Z, Baldwin DD. The effect of vitamin C supplements on urinary stone risk: a systematic review and meta-analysis. J Endourol. 2026. PMID: 42535313.

18. Galior K, Grebe S, Singh R. Development of vitamin D toxicity from overcorrection of vitamin D deficiency: a review of case reports. Nutrients. 2018;10(8):953.

19. Holick MF. Vitamin D deficiency. N Engl J Med. 2007;357(3):266-281.

20. WHO. Global prevalence of vitamin A deficiency in populations at risk 1995–2005. WHO Global Database on Vitamin A Deficiency. Geneva: World Health Organization; 2009.

21. Goldberger J. Pellagra: causation and a method of prevention. A summary of some of the recent studies of the United States Public Health Service. JAMA. 1916;66(7):471-476.

22. Sechi G, Serra A. Wernicke’s encephalopathy: new clinical settings and recent advances in diagnosis and management. Lancet Neurol. 2007;6(5):442-455.

23. Vollset SE, Clarke R, Lewington S, et al. Effects of folic acid supplementation on overall and site-specific cancer incidence during the randomised trials: meta-analyses of data on 50,000 individuals. Lancet. 2013;381(9871):1029-1036.

24. Zhang SM, Cook NR, Albert CM, et al. Effect of combined folic acid, vitamin B6, and vitamin B12 on cancer risk in women: a randomized trial. JAMA. 2008;300(17):2012-2021.

25. Fortmann SP, Burda BU, Senger CA, et al. Vitamin and mineral supplements in the primary prevention of cardiovascular disease and cancer: an updated systematic evidence review for the U.S. Preventive Services Task Force. Ann Intern Med. 2013;159(12):824-834.

26. Rautiainen S, Manson JE, Lichtenstein AH, Sesso HD. Dietary supplements and disease prevention — a global overview. Nat Rev Endocrinol. 2016;12(7):407-420.

27. Biesalski HK, Tinz J. Multivitamin/mineral supplements: rationale and safety. Nutrition. 2017;36:60-66.

28. Desai CK, Huang J, Lokhandwala A, Fernandez A, Riaz IB, Alpert JS. The role of vitamin supplementation in the prevention of cardiovascular disease events. Clin Cardiol. 2014;37(9):576-581.

29. Hamishehkar H, Ranjdoost F, Asgharian P, et al. Vitamins, are they safe? Adv Pharm Bull. 2016;6(4):467-477.

30. Ames BN. Low micronutrient intake may accelerate the degenerative diseases of aging through allocation of scarce micronutrients by triage. Proc Natl Acad Sci USA. 2006;103(47):17589-17594.

31. Baxmann AC, De Mendonça CO, Heilberg IP. Effect of vitamin C supplements on urinary oxalate and pH in calcium stone-forming patients. Kidney Int. 2003;63(3):1066-1071.

32. Zhang YF, Shi WW, Gao HF, Zhou L, Hou AJ, Zhou YH. Folate intake and the risk of breast cancer: a dose-response meta-analysis of prospective studies. Br J Cancer. 2014;110(9):2327-2338.

© 2026 Yoon Hang Kim, MD, MPH. All rights reserved.