Multivitamins raise nutrient levels, but what do they really do for health?

Millions of people use multivitamins to try to protect their health, but what have decades of clinical trials shown about memory, pregnancy, heart health, and aging?

A recent review in the Journal of Dietary Supplements assessed the efficacy and safety of multivitamins in adults. The findings differed according to the supplement formula, the population studied, and how long the products were used.

Overall, the most consistent changes were seen in blood levels of folate, vitamin B6, and vitamin B12. Supplements that included iron, zinc, or iodine appeared to help in some settings, but the benefits depended on the context. Some trials reported gains in pregnancy-related outcomes and cognitive test scores, although the results were not uniform across groups or clinical endpoints.

Roughly one-third of adults in the United States take multivitamins, and use rises with age. Even so, differences in formulations and in individual health profiles make it hard to balance possible benefits against potential risks. Better understanding of how products differ by market, manufacturing period, and stage of life could help shape more focused health strategies.

About the review

The review examined clinical trial evidence on whether broad-spectrum multivitamin supplementation may provide benefits and whether it is safe across different populations.

The authors searched the Cochrane Library, Scopus, and PubMed for peer-reviewed English-language studies through September 9, 2026. The review mainly focused on adults, though it also included prenatal studies from low- and middle-income countries involving participants younger than 18. Two reviewers screened the records independently, and one reviewer verified the data extracted by the other, using Joanna Briggs Institute methodology and resolving disagreements by consensus.

To qualify, each single-arm, placebo-controlled, or active-controlled study had to include more than 10 participants who took multivitamins for at least two weeks. The products had to be tablets or capsules providing at least 50% of the US National Academy of Medicine's life-stage-specific Recommended Dietary Allowance or Adequate Intake for 14 or more vitamins and minerals. They also needed to contain at least two vitamins and two minerals, with no nutrient exceeding its applicable upper intake limit per daily serving.

Non-oral products, enteral nutrition, lipid-based formulations, fortified foods, and delivery forms such as syrups, powders, or sprinkles were excluded. The researchers did not assess risk of bias or certainty of evidence, and they did not conduct a meta-analysis, which limits how confidently the size and reliability of the reported benefits can be judged.

Results

From 5,263 records, the authors removed 1,537 duplicates. After title, abstract, and full-text screening, 153 articles met the criteria. Four more were found through citation searching, bringing the total to 157. The review included 103 primary intervention reports and 54 related publications, some based on the same trial populations. Most of the studies were carried out in North America, Europe, and Asia.

Blood testing showed the most reliable increases in circulating vitamin B6, vitamin B12, and folate. Responses for other vitamins depended on participants' starting nutrient levels and the clinical setting. For instance, surgery can affect nutrient absorption, while specialized products may provide different amounts of individual nutrients.

Mineral changes were less consistent, and responses to iron, zinc, and iodine varied by circumstance. Some studies reported benefits for pregnancy-related outcomes and cognition, but results differed between study groups. Higher nutrient levels did not always lead to better clinical outcomes.

Multivitamin use was generally well tolerated during pregnancy, in older adults, and after surgery. Some products caused mild digestive symptoms, especially those containing zinc or iron. The Physicians’ Health Study II reported small increases in rash and nosebleeds, while the COcoa Supplement and Multivitamin Outcomes Study reported more gastrointestinal bleeding.

A cataract trial found negative changes in some lens-opacity subtypes. Safety reporting was inconsistent across trials. Vitamin K may also interfere with anticoagulants such as warfarin, so anyone using these medicines should speak with a healthcare professional before taking multivitamins.

In the COSMOS biospecimen substudy, total carotenoids, lutein, beta-carotene, and α-tocopherol increased. Results from the Comprehensive Assessment of Long-term Effects of Reducing Intake of Energy study suggested that anemia could continue during supplementation, although both randomized groups received multivitamins, making it impossible to isolate the supplements' independent effect. Secondary analyses from the Maternal and Infant Nutrition Interventions, Matlab and JiVitA-3 trials showed selective biochemical responses rather than full correction of deficiencies.

In the JiVitA-3 trial, multiple micronutrients lowered the risk of low birth weight and preterm birth compared with iron and folic acid, but they did not significantly reduce infant mortality through six months. The SUMMIT trial reported an 18% lower relative risk of early infant mortality. In a two-week trial involving women with preeclampsia, adding multivitamins to intravenous magnesium sulfate improved proteinuria and selected maternal and neonatal outcomes, but these results do not prove that multivitamins prevent preeclampsia or provide broad health benefits in uncomplicated pregnancies.

In a 12-week study of older adults, multivitamins did not improve the main well-being outcome, although some secondary measures favored supplementation for mood and stress. COSMOS-Web, COSMOS-Mind, and COSMOS-Clinic found benefits for selected memory and global cognition outcomes, but all were analyses of the same parent trial. PHS-II found no overall cognitive benefit, and dementia prevention remains unproven.

Some studies reported better lipid profiles in women with obesity, but PHS-II and COSMOS did not establish overall cardiovascular prevention effects.

Other trials hinted at possible lung and liver benefits, yet infection findings were inconsistent, and a COSMOS analysis found no significant overall reduction in COVID-19 incidence. PHS-II reported modest reductions in cataracts and total cancer incidence, while COSMOS found no significant overall cancer reduction. COSMOS analyses also suggested modest slowing of certain epigenetic aging markers, but these results did not prove lower disease risk or longer survival.

Conclusion

The review suggests that multivitamins can help support nutritional status, but their effects depend on the formulation, the population, and baseline nutrition. Their value may differ between adults with adequate diets and those facing food insecurity, higher nutrient needs, medication-related risks, or malabsorption.

Future trials should determine whether any cognitive gains translate into better day-to-day function or less cognitive decline. Researchers also need clearer clinical goals, validated outcomes, and more consistent reporting of formulations, adherence, nutritional status, and side effects. Including underrepresented groups and measuring real-world effectiveness could strengthen the evidence and support more targeted guidance.

The review received funding from an unrestricted educational grant from Haleon to Think Healthy Group. Senior author Taylor C. Wallace disclosed an advisory role with Haleon and serves as Editor-in-Chief of the journal that published the review. The authors said the funder had no role in the research or manuscript preparation.

Source:

Journal reference:

• Fan, Z., Nelson, J. M., & Wallace, T. C. (2026). Multivitamins – A Scoping Review of Clinical Trials Examining Their Efficacy and Safety. Journal of Dietary Supplements, 1–48.