Vitamins and Supplements: Assessing Benefits and Risks, from Vitamin D and Multivitamins to Folic Acid

An Evidence-Based Reading of Vitamins and Supplements

The conclusion in one sentence

For healthy adults without a known nutritional deficiency, vitamins and supplements generally cannot serve as a universal means of preventing cardiovascular disease, cancer, fractures, or death. Confirmed deficiency, the periconceptional period, and specific clinical needs, however, must be evaluated separately (PMID: 35727272)

First ask: who is taking it, for what purpose, and what is being supplemented?

“Do supplements work?” is not a question that can be answered without context. The risks and potential benefits differ among healthy people with no known deficiency, people with a diagnosed deficiency, those preparing for pregnancy, and those undergoing treatment for a disease. The central concern in Professor Lin’s original article was the portrayal of supplements as a shortcut to health that everyone needs. That warning is consistent with the overall direction of current evidence for generally healthy adults. Research shows little or no clear overall benefit when vitamins, minerals, or multivitamins are used broadly to prevent cardiovascular disease, cancer, and death. This finding, however, must not be stretched into the claim that “supplementation is ineffective for everyone in every circumstance” (PMID: 35727272)

Feeling better after taking a product also cannot, by itself, establish a specific therapeutic effect. Symptoms may fluctuate naturally, while expectations, the care setting, and placebo effects may all shape how a person feels. Determining efficacy still requires comparing outcomes among people receiving the product with those in an appropriate control group (PMID: 11372012)

Vitamin D: it acts like a hormone, but that does not mean everyone should supplement

Professor Lin used the statement “vitamin D is a hormone” to remind readers that it is not merely an ordinary nutrient tablet of the kind people often imagine. Modern literature describes vitamin D as a substance with hormonal actions and confirms that its effects are mediated by nuclear receptors. In more precise biochemical terminology, it is often classified as a secosteroid; this refinement does not alter the core physiological concept conveyed in the original article (PMID: 41849024) (PMID: 41897332)

The skin can synthesize vitamin D under ultraviolet radiation from sunlight, but saying that “a little sun every day guarantees that deficiency will not occur” remains an oversimplification. Season, latitude, time of day, skin pigmentation, age, clothing, and sun protection all affect synthesis, and deficiency can occur even in regions with abundant sunshine. Moreover, ultraviolet exposure itself carries risks of sunburn and cancer (PMID: 34580202) (PMID: 42280308) (PMID: 34698034)

Among generally middle-aged and older adults who were not selected for deficiency, low bone mass, or osteoporosis, routine vitamin D supplementation did not reduce fracture risk or prevent cancer or major cardiovascular events (PMID: 35939577) (PMID: 30415629)

In generally older populations whose vitamin D status was largely adequate, monthly supplementation likewise did not reduce acute respiratory infections or the overall risk of falls (PMID: 33444565) (PMID: 34337905)

However, “no preventive benefit was observed in the general population” cannot be extrapolated to mean that people with a confirmed deficiency do not require treatment. Nutritional rickets has a clear relationship with vitamin D deficiency, and treatment in that context differs from self-directed supplementation by healthy adults (PMID: 40859384)

Dose also requires precise interpretation. Research supports the direction of Professor Lin’s warning that high doses may be harmful; in particular, intermittent high-dose regimens have produced signals of increased falls or fractures (PMID: 20460620)

Yet it is inconsistent with the trials catalogued by ECU to label a particular total monthly amount mentioned in the original article as categorically dangerous. When the same total monthly amount was divided into daily doses, the research did not find hypercalcemia or other adverse events due to excess intake. Dose, administration schedule, and population must therefore be distinguished. This is a boundary that needs to be added as science advances; it does not rewrite Professor Lin’s historical position in the original article (PMID: 30415629)

True vitamin D toxicity can cause hypercalcemia, kidney injury, and tissue calcification. Neither “there is no universal benefit” nor “every dose is dangerous” is an accurate statement (PMID: 41001862) (PMID: 39775139)

“Natural,” quality seals, and product labels are not evidence of efficacy

Key precursors of industrial vitamin C can be produced through microbial fermentation; the product is not necessarily extracted directly from fruit (PMID: 36073939)

The name given to a source, however, cannot substitute for human outcomes. Human steady-state comparisons of equal doses of synthetic and food-derived vitamin C have not demonstrated a medically meaningful advantage for either source. The label “natural” therefore does not itself prove superior efficacy (PMID: 24169506)

Likewise, seals relating to quality, safety, or manufacturing do not independently establish efficacy in treating disease. ECU did not identify an applicable PMID for the specific Taiwanese certification discussed, so only its administrative scope can be left open for further verification here. “No medical literature was found” must not be misrepresented as meaning that the product is necessarily ineffective.

The actual contents of a supplement may also differ from its label. Adulteration and labeling problems are indeed documented in the literature on sports supplements, but proportions observed in a specific market cannot be extrapolated directly to all products (PMID: 37841887)

Some dietary supplements and herbal products can cause liver injury. People who also take medication, have a chronic illness, or develop symptoms should proactively tell their healthcare professionals about every product they are using (PMID: 38528750)

Common ingredients: the evidence cannot be treated as all-or-none

For the general population, evidence does not support routine vitamin C use to prevent the common cold. The literature has, however, observed a modest reduction in illness duration and possible preventive benefit among people under high physical stress. It is therefore inaccurate to say that “all evidence is against it” (PMID: 23440782)

For melatonin, the literature has been updated: a modest sleep benefit is seen compared with placebo, so it is no longer appropriate to claim broadly that all sleep-related benefits lack evidence. A limited benefit also does not mean that self-directed long-term use is suitable for everyone (PMID: 23691095)

For oral collagen, the literature has been updated: newer pooled data from randomized trials indicate that some skin measures may improve. This does not mean that intact collagen is deposited into the skin unchanged. It means that the assertion that “no measurable improvement is possible” is no longer consistent with current evidence (PMID: 41924746)

An observational association between elevated blood B12 and mortality risk cannot directly establish that B12 supplementation causes death. Data linking high intake with hip fractures also concern combined intake of B6 and B12 and cannot be attributed to B12 alone (PMID: 31940038) (PMID: 31074816)

Clear exceptions: folic acid and treatment of deficiency

Periconceptional folic acid is an important counterexample to the claim that “supplements never work.” Folic acid should be started before conception and continued through the first trimester to reduce the risk of fetal neural tube defects (PMID: 41754157)

For regions that have implemented folic acid food fortification, the literature has been updated: current recommendations still support additional folic acid supplementation for people who could become pregnant. The older conclusion that “no supplementation is ever needed after food fortification” can therefore no longer be maintained (PMID: 37526713)

The most practical principle is to identify the goal and indication first, then ask whether the population, dose, administration schedule, and outcomes studied in the evidence match one’s own circumstances. Do not substitute personal testimonials for controlled trials, or “natural” claims and seals for evidence of efficacy. Likewise, when a study finds no benefit in the general population, do not independently stop treatment prescribed for a deficiency, pregnancy, or disease.