Is a multivitamin necessary?

What is a multivitamin, and who uses one?

The NIH divides multivitamin-mineral (MVM) products into three groups by content: “basic” products taken once daily that contain most vitamins and minerals, mostly in amounts that do not exceed the daily values; “high-potency” products that markedly exceed the daily value, and sometimes the tolerable upper intake level (UL), for some vitamins and minerals; and “special-purpose” products combined with herbal and specialized ingredients for purposes such as energy, athletic performance, weight control, immunity, or eye health. ^[7] In PHS II a daily multivitamin was used, and in COSMOS-Mind a commercial multivitamin-mineral product was used; the results do not apply directly to high-potency or special-purpose products.

The NIH reports that an estimated one-third of adults in the US use a multivitamin-mineral and that one-quarter of children and adolescents take a multivitamin or multivitamin-mineral; use increases with age. ^[7] These rates are US data. The NIH also reports that some users see a multivitamin as a “nutritional insurance,” a concept introduced in the 1940s through a manufacturer’s marketing campaign. ^[7]

PHS II: what are the cancer and cardiovascular results in male physicians?

The Physicians’ Health Study II (PHS II) is a large randomized, double-blind trial comparing a daily multivitamin with placebo. In two separate publications, 14,641 US male physicians (aged 50 and over at baseline, mean age 64.3) were followed for 11.2 years (median).

Cancer (Gaziano et al., 2012). For total cancer (excluding skin cancer), there were 17.0 events per 1,000 person-years in the multivitamin group and 18.3 in the placebo group: HR 0.92 (95% CI 0.86–0.998; P = 0.04). There was no significant difference in prostate cancer (HR 0.98; 0.88–1.09), colorectal cancer (HR 0.89; 0.68–1.17), or cancer death (HR 0.88; 0.77–1.01; P = 0.07). In the 1,312 men with a history of cancer at baseline, the HR was 0.73 (0.56–0.96), and in those without, 0.94 (0.87–1.02); the difference between the two groups was not statistically significant (interaction P = 0.07). ^[1] That is, the result was obtained in a single trial, only in male physicians, and with a borderline confidence interval showing an average reduction.

Cardiovascular disease (Sesso et al., 2012). For major cardiovascular events (nonfatal heart attack, nonfatal stroke, cardiovascular death; 1,732 events), the HR was 1.01 (0.91–1.10); for heart attack 0.93 (0.80–1.09), for stroke 1.06 (0.91–1.23), for cardiovascular death 0.95 (0.83–1.09), and for total mortality 0.94 (0.88–1.02). The effect did not differ in men with and without cardiovascular disease at baseline. ^[2] No cardiovascular prevention benefit for multivitamins emerges from this trial.

Funding. In both publications, the study was supported by NIH grants and an investigator-initiated grant from BASF; the study agents were provided by BASF and Pfizer. ^{[1] [2]} According to the publication, the NIH, BASF, and Pfizer had no role in the design, conduct, or analysis of the study. ^[1] This does not mean the results are invalid; however, it is a structure in which product suppliers also contributed to the study, and this should be known when reading it.

Limits. The participants were men and physicians; the same result was not tested in women, at younger ages, or in groups with different diets. The result belongs to 11 years of daily use of a single product.

COSMOS: what did it show for cognitive function in older adults?

COSMOS is a large trial of 21,442 US adults (aged 60 and over) examining, in a 2×2 factorial design, the effect of cocoa extract and a multivitamin-mineral on preventing cardiovascular disease and cancer; cognitive function was measured in separate substudies. ^[4]

COSMOS-Mind. In 2,262 participants (mean age 73, 60% women, 89% non-Hispanic white), cognitive tests were administered by telephone over three years of follow-up. Cocoa extract had no effect on global cognition. The multivitamin, as a prespecified secondary endpoint, produced a statistically significant difference in global cognition (mean z = 0.07; 95% CI 0.02–0.12; P = 0.007); benefit was also reported in memory and executive function. The effect was more pronounced in those with a history of cardiovascular disease (0.14; CI −0.02–0.31) and was 0.06 in those without; the subgroup interaction was reported as “nominal P = 0.01,” that is, without correction for multiple comparisons. ^[3]

COSMOS-Clinic and the meta-analysis of three substudies. In COSMOS-Clinic (573 people, 2 years), where detailed in-person tests were performed, the difference in global cognition was not significant (0.06 standard deviation units; CI −0.003–0.13), there was a significant increase in episodic memory (0.12; CI 0.002–0.23), and none in executive function and attention (0.04; CI −0.04–0.11). In the meta-analysis of non-overlapping participants from the COSMOS-Clinic (573), COSMOS-Mind (2,158), and COSMOS-Web (2,472) substudies, significant differences were found of 0.07 units (CI 0.03–0.11) in global cognition and 0.06 units (CI 0.03–0.10) in episodic memory. The authors interpreted the global cognition effect as “equivalent to slowing cognitive aging by about 2 years.” ^[4]

These effects are test score differences expressed in small standard deviation units; clinical outcomes such as the development of dementia are not what these analyses measured. Most participants were white and older. COSMOS was also supported by an investigator-initiated grant from the cocoa producer Mars Edge (infrastructure and donation of study pills) and by Pfizer Consumer Healthcare (now Haleon) providing some of the study pills; the COSMOS-Mind publication states that the NIH, Mars, and Pfizer did not take part in the design, conduct, analysis, or the decision to publish. ^{[3] [4]} The results should therefore be read as a signal awaiting confirmation by an independent second study.

What does USPSTF 2022 say about multivitamins?

The USPSTF evaluated the effect of individual nutrients, functionally related nutrient pairs, and multivitamins on preventing cardiovascular disease and cancer in community-dwelling, nonpregnant adults. Results: it was determined with moderate certainty that the harms of beta-carotene supplementation outweigh the benefits and that there is no net benefit for vitamin E; a recommendation against both was given (grade D). The balance of benefits and harms of multivitamins and of single or paired nutrients (other than beta-carotene and vitamin E) for preventing cardiovascular disease and cancer could not be determined because of insufficient evidence (I statement). ^[5]

The NIH reports that the evidence review prepared for the USPSTF covered 78 randomized trials (324,837 participants) and six cohort studies (390,689 participants) and concluded that vitamin and mineral supplements have little or no benefit in preventing cancer, cardiovascular disease, and death, but that multivitamins may provide a small benefit against cancer. The NIH’s own summary is also that multivitamins do not reliably reduce the risk of chronic disease. ^[7]

Is the risk of death lower in multivitamin users?

Loftfield and colleagues followed 390,124 adults in the US, free of cancer or other chronic disease at baseline, in three prospective cohorts (NIH-AARP: 327,732; PLCO: 42,732; Agricultural Health Study: 19,660 participants) for up to 27 years (median age 61.5; 164,762 deaths). Multivitamin use was self-reported; analyses were adjusted for the confounding effect of a healthy lifestyle and for people with disease starting the product (reverse causation). Multivitamin use was not associated with lower all-cause mortality risk: in the first half of follow-up, risk was 4% higher than in nonusers (HR 1.04; 95% CI 1.02–1.07), and in the second half the difference was not significant (HR 1.04; 0.99–1.08). Results were similar for the main causes of death. ^[6]

This is an observational study; it does not prove causation, but, consistent with the picture given by randomized trials (HR 0.94; 0.88–1.02 for total mortality in PHS II), it shows no benefit on mortality. The study was also conducted in the US, mostly in middle-aged and older adults.

Who may benefit from a multivitamin or a similar supplement?

According to the NIH, a multivitamin increases the amount taken in people who cannot meet their nutrient requirements from food alone and helps reach recommended levels. The same source states that the populations at highest risk of nutritional inadequacy and who could benefit most from multivitamins are the least likely to take them; it lists pregnant and breastfeeding women, people of low socioeconomic status, and underweight or overweight people among the groups at greatest risk of deficient intakes of one or more nutrients. ^[7]

• Pregnancy and pregnancy planning: The NIH reports that ACOG recommends a daily prenatal vitamin; products for children, adults, pregnant women, and older adults typically provide different amounts of the same vitamins and minerals. ^[7] Details and the official guideline in Türkiye are covered on a separate page.
• Children: There are targeted practices such as vitamin D and iron programs in infants; a general multivitamin is a separate question.
• People with specific gaps: If there is a defined deficiency (for example, B12, vitamin D, or iron), measuring and replacing the single nutrient is often the basis; risk groups are described with their sources in these guides.
• Older adults: The NIH states that products for older adults (50+) generally contain little or no iron and are richer in calcium, vitamin D, and B12 than products for younger adults. ^[7]

Detailed guideVitamins in pregnancy: what changes, what to avoid?→ Detailed guideVitamins for children: D and iron programs, gummies, upper limits→

What should be checked on the label?

A multivitamin label can be used to question the content and amount rather than to read the product as an “insurance.”

1. Product type. The NIH reports that high-potency and special-purpose products, other than basic products, can exceed the daily value, and even the UL, for some nutrients. ^[7]
2. Nutrients that fall short. Basic multivitamins mostly contain amounts that do not exceed daily values; however, minerals such as calcium, magnesium, and potassium are in small amounts in most products. ^[7]
3. Age, sex, life stage. The NIH states that products are designed to suit a person’s age, sex, and life stage (for example, pregnancy); most children’s products provide age-appropriate amounts. ^[7]
4. Total intake. The NIH reports that people who use a multivitamin together with other supplements and fortified foods and beverages can take in some nutrients above the UL; in a study it cites (a cohort of 90,771 people over age 45 in Los Angeles and Hawaii), excess intake was seen in 10–15% of multivitamin users for vitamin A, iron, and zinc, and in 48–61% for niacin. ^[7] Separate products carrying the same ingredient (for example, a separate vitamin D, B6, vitamin A, or iron supplement) are therefore counted in the total; the upper limits that vary by organization are described on the relevant vitamin pages.
5. Biotin before tests. B-complex and some multivitamin products contain biotin; high amounts of biotin can distort some blood tests such as thyroid hormones and troponin. Sources and the FDA warning are described in the biotin guide.

Detailed guideVitamin A: retinol, beta-carotene, excess, and pregnancy→ Detailed guideVitamin B6: deficiency, nerve damage, and upper limits→ Detailed guideIron: functions, daily requirement, excess, and interactions→

In Türkiye, is a multivitamin a drug or a food supplement?

According to the definition in Law No. 5996 (5996 sayılı Kanun) as cited by the Ministry of Agriculture and Forestry, a food supplement is the presentation of nutrients such as vitamins and minerals, for the purpose of supplementing normal nutrition, in forms such as capsules, tablets, lozenges, single-use powder sachets, liquid ampoules, and dropper bottles, with a defined daily intake dose. The same Ministry text states that offering food supplements for sale as if they were weight-loss, height-boosting, disease-preventing, treating, and curing products is contrary to legislation. ^[8] Within this framework, vitamin-mineral products with a defined daily intake dose in forms such as capsules and tablets are food supplements, and claims of the “protects against disease” type are not legal; whether a product is a drug or a supplement is distinguished from the statements on its label. Label requirements and the drug-supplement distinction are described in detail on the vitamin D products page.

Detailed guideVitamin D medicines: differences among drops, capsules, ampoules, and injections→

Misunderstandings about multivitamins

• “A multivitamin makes up for diet.” The NIH frames the purpose of a multivitamin as supplementing amounts that cannot be met from food; it reports that it does not reliably reduce the risk of chronic disease. ^[7]
• “One pill covers all vitamins.” Even in basic products, minerals such as calcium, magnesium, and potassium are in small amounts; products differ by life stage.
• “More ingredients is better.” High-potency products can exceed the UL; USPSTF, for its part, does not recommend beta-carotene and vitamin E supplements for cancer or cardiovascular prevention.
• “If one study showed it, it is certain.” The cancer and cognitive results come from single families of studies, the effect is small, and the groups are limited.

What is known and what is not?

Known: In PHS II, the multivitamin did not reduce cardiovascular events ^[2]; USPSTF finds the evidence for multivitamins insufficient ^[5]; in three US cohorts, use was not associated with lower mortality risk ^[6].

Not known: Whether the small cancer reduction seen in men holds in women and other groups; the clinical meaning of the small increase in cognitive test scores (for example, dementia risk) and whether it would be confirmed in an independent study; the difference between products with different contents.

This guide does not recommend personal products or doses; in pregnancy, childhood, chronic disease, or medication use, supplement choice is made together with a physician or pharmacist.

Detailed guideWhen to take vitamins? Meals, time of day, and medicine intervals→