Supplements

How supplements are regulated#

In the US, supplements are regulated more like food than like drugs. The Dietary Supplement Health and Education Act of 1994 (DSHEA) set the rules that still apply, and under them the FDA does not have the authority to approve supplements before they are sold. Manufacturers are responsible for making sure their products are safe and accurately labeled. The FDA acts mostly after problems appear, through inspections, warning letters, import checks, and recalls.

Three consequences matter for you:

• No proof of benefit is required. A label can make a "structure/function" claim ("supports immune health") as long as it carries the disclaimer that the statement "has not been evaluated by the Food and Drug Administration" and that the product is not meant to diagnose, treat, cure, or prevent disease. Only drugs may make disease claims.
• New ingredients get a light review. A company selling an ingredient that was not in the food supply before 1994 must notify the FDA with safety information at least 75 days before marketing it. That is a notification, not an approval.
• Enforcement is reactive. By the time the FDA issues a warning letter, a product may have been sold for years.

The US Preventive Services Task Force, the independent panel that sets screening and prevention guidelines, looked at the whole field in 2022. It found insufficient evidence to recommend multivitamins or any single or paired nutrient for preventing heart disease or cancer in healthy, nonpregnant adults, and it recommended against beta-carotene and vitamin E. That is the baseline: for disease prevention in people who are not deficient, the burden of proof is on the supplement.

Quality and contamination#

Even when an ingredient works, the product in the bottle may not match the label. There are three separate problems: contaminants, wrong doses, and hidden drugs.

Heavy metals#

In October 2025, Consumer Reports tested 23 protein powders and ready-to-drink shakes for lead. More than two-thirds had more lead in a single serving than its experts consider safe for a whole day (0.5 micrograms, based on California's Proposition 65 limit), and some exceeded it more than tenfold. Plant-based powders averaged about nine times the lead of dairy-based ones. CR advised against daily use of most powders and recommended choosing products that publish test results, with dairy-based options as the default when results are not available. There is no federal lead limit for protein powder as of October 2026.

The industry argues that CR's threshold is stricter than any government limit, which is true. It is also true that lead has no known safe level and that daily exposure adds up. If you use a powder every day, the brand matters.

Wrong doses#

Labels are often wrong in both directions. When researchers tested melatonin gummies sold in the US, the actual content ranged from 74% to 347% of the labeled dose, and some products also contained CBD that was not on the label. For a sleep aid that is already sold at many times the dose the body makes, that error is not trivial.

Hidden drugs#

The most dangerous problem is deliberate spiking. Between 2007 and 2016, the FDA identified 776 adulterated supplements from 146 companies. Nearly all were sold for sexual enhancement, weight loss, or muscle building, and the hidden ingredients were real drugs: sildenafil (the active ingredient in Viagra) in sexual-enhancement products, sibutramine (a weight-loss drug withdrawn for heart risks) in diet products, and synthetic steroids in bodybuilding products. Some products reappeared after warnings with new hidden ingredients. The FDA keeps a searchable Health Fraud Product Database of products it has flagged.

What third-party seals mean#

A seal from an independent tester is the single most useful thing on a supplement label. It does not mean the product works. It means someone outside the company checked that it contains what the label says and is not contaminated.

The sport seals were built to protect drug-tested athletes from accidental doping, which is exactly why they are useful for everyone else: they look hardest for the hidden drugs that cause the most harm.

The evidence at a glance#

The table grades each supplement for the reason most people take it, in people who are not deficient unless noted. "Dose studied" is what trials used, not a recommendation. The Interventions Testing Program (ITP), the National Institute on Aging's rigorous multi-site program that tests compounds for lifespan effects in genetically diverse mice, comes up often; its full results table is public. For how the badges are assigned, see Reading the evidence.

What to take and what to test#

An optimal supplement plan has three parts: a short default list for adults who train, a few nutrients to test and correct, and everything else only for a specific reason. It is deliberately smaller than most longevity stacks, because for people who are not deficient, more pills add cost and risk without adding health.

The testing goes beyond the guidelines. Guideline minimum: the Endocrine Society's 2024 guideline suggests against routine vitamin D testing in healthy adults, the USPSTF calls the evidence for vitamin D screening insufficient (neither for nor against), and guidelines test B12 and iron only in specific groups. This guide suggests testing anyway because the tests are inexpensive, low levels are common (about 23% of Americans were below 20 ng/mL of vitamin D in 2011–2014), and the trials that found no benefit mostly enrolled people who already had enough Weak. How to order the tests is in Blood tests and biomarkers.

The ten that matter most#

These are the supplements with either the best evidence or the widest use. Each entry covers what the trials show, the dose they used, and the safety points that matter.

Creatine#

Creatine is the best-supported performance supplement there is Strong, and for adults who strength train it is the one supplement worth taking by default. Combined with resistance training, it reliably increases strength and lean mass, and the International Society of Sports Nutrition's position stand rates it safe for healthy people at the usual dose. The benefit holds with age: in a meta-analysis of 22 trials in 721 adults averaging 57 to 70 years old, creatine during resistance training added about 1.4 kg more lean mass and more strength than training alone. A 2023 meta-analysis of 23 randomized trials found it improved memory, with the largest effect in adults aged 66 to 76 and no clear effect in younger people Moderate.

• Dose studied: 3–5 g of creatine monohydrate a day. Loading phases are optional. Other forms cost more and have less evidence.
• Safety: Expect a few pounds of water weight in muscle. An expert review in 2021 found no evidence of kidney harm in healthy people at usual doses. Creatine can modestly raise creatinine, a blood marker of kidney function, which can be mistaken for kidney trouble, so tell your doctor you take it before a kidney panel. If you already have kidney disease, ask first.
• Pairing: The benefit for muscle depends on training. See Exercise.

Protein powder#

Protein powder is food in a convenient form, not a supplement with special properties. Prevention-minded protein targets sit well above the minimum RDA, and they matter more as you age; the numbers live in Nutrition. Powder is the easiest way to close a gap when food falls short Moderate.

• Dose: 20–40 g a serving is typical.
• Quality: Choose a brand with a third-party seal or published heavy-metal results. In the 2025 Consumer Reports tests, dairy-based powders generally had the least lead and plant-based powders the most.
• Perspective: Whole foods (eggs, dairy, fish, meat, beans, lentils) carry no such risk. If you use powder daily, rotate brands or limit servings.

Psyllium and fiber#

Psyllium husk is a soluble fiber that forms a gel in the gut. In a meta-analysis of 28 trials, about 10 g a day lowered LDL cholesterol by about 13 mg/dL, with similar drops in non-HDL cholesterol and ApoB Strong. It also treats constipation and modestly blunts glucose after meals. Food fiber is better still; see Nutrition. The optimal use is as a top-up: add psyllium if your food fiber falls short of target or your ApoB or LDL is above goal.

• Dose studied: 5–10 g a day, stirred into a full glass of water. Start low to limit gas and bloating.
• Safety: Psyllium can bind medicines. Take other drugs at least two hours apart, and thyroid hormone (levothyroxine) about four hours apart.

Omega-3 fish oil#

The fish oil story is a lesson in how much the dose and the product matter Mixed. In the VITAL trial of 25,871 healthy US adults, about 1 g a day of fish oil did not reduce major cardiovascular events or cancer. In people with high triglycerides and heart disease or diabetes, 4 g a day of a purified prescription form of EPA (icosapent ethyl) cut cardiovascular events by about a quarter, but a similar trial of a different high-dose omega-3 against a corn oil placebo found no benefit. The prescription question belongs in Heart and blood vessels.

• Dose studied: About 1 g a day of EPA plus DHA in prevention trials; 4 g a day for the prescription drug.
• Safety: Across seven trials with 81,210 people, omega-3 supplements raised the risk of atrial fibrillation (an irregular heart rhythm) by 25% overall and 49% at doses above 1 g a day. Tell your doctor if you take blood thinners.
• Default: Two servings of fatty fish a week, which covers most people without a pill.
• Omega-3 index (optional): A finger-prick test measures EPA and DHA in red blood cells. In a pooled analysis of 17 cohorts with 42,466 people, those with the highest blood omega-3 levels had 15% to 18% lower mortality than those with the lowest, an observational link. The index's developers proposed 8% or more as the target Weak. If you rarely eat fish and your index is low, more fish or about 1 g a day of EPA plus DHA is a reasonable choice; skip high doses if you have atrial fibrillation.

Vitamin D#

Vitamin D is essential for bone, and people who are deficient benefit from replacing it Strong. Taking more when your level is already adequate does not prevent disease Negative. In VITAL, 2,000 IU a day for about five years did not reduce cancer or cardiovascular events, though it was linked to 22% fewer autoimmune diseases in a secondary analysis. In the D-Health trial of 21,315 Australians aged 60 and older, a monthly 60,000 IU dose did not reduce deaths; the placebo group already averaged about 31 ng/mL, and cancer deaths trended slightly higher with the vitamin.

That pattern favors testing and correcting over taking a large dose blindly.

• Test: Add a 25-hydroxyvitamin D test to your yearly blood work, and repeat it a few months after any change in dose.
• Target: About 30 to 50 ng/mL (75 to 125 nmol/L) Weak. The National Academies judged 20 ng/mL enough for bone in 97.5% of people, found levels above 30 not consistently linked to more benefit, and warned that levels above 50 "should raise concerns" about harm. The Endocrine Society's 2011 guideline used 30 as its floor. No trial shows that the 40–60 ng/mL targets and 5,000 IU daily doses common in longevity circles do better.
• Dose: The RDA (recommended dietary allowance) is 600 IU a day through age 70 and 800 IU after. People who test low often need more, set with a doctor. The upper limit for adults is 4,000 IU a day unless a doctor prescribes more to treat a deficiency.
• Guideline minimum: The Endocrine Society's 2024 guideline advises against routine testing in healthy adults, endorses no target level, and suggests extra vitamin D only for adults 75 and older, pregnant people, adults with prediabetes, and children.

Vitamin B12#

B12 is a supplement where testing and targeting work well Strong. Deficiency causes anemia, numbness, memory problems, and fatigue, and it is common in older adults and in people who take certain drugs or avoid animal foods.

• Who is at risk: People on metformin (in the Diabetes Prevention Program, long-term users had more low and borderline-low B12, with risk rising each year of use); people who take acid-reducing drugs for two years or more, who had higher rates of deficiency in a large health-system study; vegans; and older adults. Because 10% to 30% of people over 60 have a thinned stomach lining that absorbs B12 from food poorly, the Food and Nutrition Board advises adults over 50 to get most of their B12 from fortified foods or a supplement.
• Test: Check B12 with your yearly blood work if any of these apply, and confirm a borderline result with a methylmalonic acid test; see Blood tests and biomarkers. Guideline minimum: periodic testing on metformin, as the ADA advises, and testing for symptoms.
• Dose: The adult RDA is 2.4 mcg. Treating a deficiency takes much more, given by mouth or by injection; the right dose depends on the cause, so let your doctor set it.
• Safety: B12 has no established upper limit and is considered very safe.

Iron#

Iron is the clearest example of "only if you need it" Strong. Iron deficiency is common in menstruating women, pregnant people, frequent blood donors, endurance athletes, and people with heavy periods or gut bleeding. Replacing it fixes anemia and fatigue. Iron without deficiency does nothing good, and iron overload damages the liver, heart, and pancreas, especially in people with hereditary hemochromatosis, a common genetic condition.

• Test first: A ferritin level and a complete blood count show whether you need it. Check them yearly if you menstruate, train for endurance, donate blood often, or eat a plant-based diet. Many prevention-minded clinicians include ferritin in everyone's yearly panel, since high ferritin can also flag iron overload. See Blood tests and biomarkers.
• Dose: Set by your doctor. The adult upper limit is 45 mg a day, though doctors often prescribe more short-term to rebuild stores.
• Interactions: Iron reduces absorption of levothyroxine and levodopa, and acid-reducing drugs can reduce iron absorption. Space them apart.
• New fatigue in a man or a postmenopausal woman with low iron needs a search for blood loss, not just a pill.

Calcium#

Calcium supplements are less helpful and more controversial than once thought Mixed. The Task Force recommends against low-dose calcium plus vitamin D (up to 1,000 mg and 400 IU) for preventing fractures in healthy postmenopausal women and finds the evidence insufficient for higher doses; an update is in progress. In the Women's Health Initiative, calcium plus vitamin D increased kidney stones. Some studies link calcium supplements, though not dietary calcium, to cardiovascular risk, and others do not; the NIH summary calls the question unresolved.

• Target: The RDA is 1,000–1,200 mg a day from all sources combined, and most people get much of that from food. Supplement only the gap.
• Safety: The upper limit is 2,000–2,500 mg a day for adults. Calcium supplements interact with levothyroxine, lithium, quinolone antibiotics, and the HIV drug dolutegravir.
• For osteoporosis, calcium and vitamin D are the background to real bone drugs, not a substitute. See Muscle, bone, and mobility.

Magnesium#

Magnesium is popular for sleep, cramps, and stress, and it is cheap and low risk, but the evidence for people who are not deficient is thin Weak. A meta-analysis of three small trials in 151 older adults found magnesium shortened the time to fall asleep by about 17 minutes, with low-quality evidence and no significant change in total sleep.

• Dose: The upper limit for magnesium from supplements is 350 mg a day for adults (food magnesium does not count toward it).
• Forms: In small studies, citrate, chloride, lactate, and aspartate forms were absorbed more completely than magnesium oxide. Glycinate is popular for sleep but has no trial showing it beats other forms.
• Safety: High doses cause diarrhea and cramping. People with kidney disease can accumulate dangerous levels and should ask first.

Multivitamin#

A daily multivitamin is the most common supplement in America and one of the least consequential Mixed. In 390,124 US adults followed for up to 27 years, regular multivitamin users had no lower risk of death than nonusers. On the other side, the COSMOS trials in older adults found a small benefit for memory, equivalent to about two years less cognitive aging when the substudies were pooled. The Task Force calls the evidence insufficient for preventing heart disease or cancer.

• Reasonable uses: Restricted diets, older adults with small appetites, people after weight-loss surgery, or anyone whose doctor finds several marginal nutrients.
• What to buy: A basic product with about 100% of daily values and a seal. "Mega" formulas add risk without benefit. Avoid products with large amounts of preformed vitamin A, and skip iron unless you need it.

Longevity supplements#

A set of supplements is sold specifically to slow aging. Their logic usually starts with a real biological pathway and a striking result in worms, flies, or mice. Few have human trials with meaningful outcomes, and several failed in the ITP, which tests compounds in genetically diverse mice at three sites at once. For the drugs that target aging (rapamycin, metformin, senolytic drug combinations), see Longevity drugs.

A few others have real but narrow mouse results. Astaxanthin, a red carotenoid from algae, extended median lifespan 12% in male mice but not females. Glycine extended lifespan by a few percent in both sexes, and 3 g at bedtime improved subjective sleep in a small human study Weak. Neither has human longevity data.

Lithium and dementia#

Lithium got a burst of attention in 2025. A study in Nature found that lithium, a trace element in the diet and water, was the only metal significantly reduced in the brains of people with mild cognitive impairment, and that it was further trapped by amyloid plaques in Alzheimer's disease. Depleting lithium in mice accelerated Alzheimer's-like changes and memory loss, and low doses of lithium orotate, a salt chosen because it binds less to amyloid, prevented them. Danish population data had earlier linked higher lithium in drinking water to lower dementia rates, an observational and nonlinear association Weak.

Harms and interactions#

Supplements feel safe because they are sold next to vitamins. Several of the largest supplement trials ever run ended early because the supplement caused harm.

Antioxidant trials that backfired#

In the 1990s, antioxidants looked like an easy win: people who ate more fruits and vegetables had less cancer, so isolated antioxidant pills seemed likely to help. The trials said otherwise Negative.

Later analyses of SELECT found that selenium nearly doubled high-grade prostate cancer in men who already had high selenium levels, and an earlier trial found selenium raised type 2 diabetes risk by about half. These are the reasons the Task Force gives beta-carotene and vitamin E a "D" grade: recommend against.

Liver injury#

Herbal and dietary supplements now cause a growing share of serious drug-induced liver injury in the US, rising from 7% to 20% of cases in a national registry over about a decade. Bodybuilding products made up about a third of the supplement cases in that registry. Three popular single ingredients stand out:

• Green tea extract. Concentrated extracts, unlike brewed tea, have caused hepatitis and liver failure. The European Food Safety Authority concluded that 800 mg or more a day of EGCG, the main catechin, from supplements may harm the liver, and a US Pharmacopeia review found case reports at 140 mg to about 1,000 mg a day, with risk higher when taken on an empty stomach. USP's quality standard for green tea extract now includes a warning to take it with food and to avoid it with liver problems.
• Turmeric and curcumin. In the US Drug-Induced Liver Injury Network, turmeric cases have risen since 2011; of 10 cases, five were hospitalized and one died. Some products added black pepper extract (piperine), which increases absorption. Most cases carried a specific immune gene variant, which suggests some people are far more susceptible than others.
• Ashwagandha. This Ayurvedic herb has modest evidence for stress and sleep, and the NIH notes it may be safe for up to about three months but has a number of reported liver injuries. In a case series from Iceland and the US, patients developed jaundice after 2 to 12 weeks, which resolved over one to five months. It can also affect thyroid hormone levels and should be avoided in pregnancy. Denmark banned ashwagandha in food supplements in 2023 after its food safety agency could not identify a safe intake, the Netherlands' public health institute advised against it in 2024, and as of October 2026 the UK and the European Union are reviewing it.

Drug interactions#

Some supplements change how the liver clears medicines, and others add to a drug's effects.

• St. John's wort, used for mild depression, weakens many important medicines, including birth control pills, the transplant drug cyclosporine, warfarin, some heart, seizure, cancer, and HIV drugs, and simvastatin. Combined with antidepressants it can cause serious serotonin-related side effects.
• Goldenseal and berberine. Goldenseal, whose best-known compound is berberine, may change how the body processes many medications; in one NIH-funded study it lowered metformin levels by about 25%. Berberine itself lowers glucose. If you take either alongside diabetes drugs or other prescriptions, your doctor and pharmacist should know.
• Vitamin E at doses above about 400 IU a day can add to the bleeding risk of warfarin and antiplatelet drugs. Fish oil and ginkgo are often flagged for the same reason.
• Calcium, iron, magnesium, zinc, and fiber bind thyroid hormone and some antibiotics in the gut. Separate them by several hours.

Too much of a good thing#

• Vitamin A. Preformed vitamin A (retinol and retinyl esters, found in liver, fish liver oils, and many supplements) has an upper limit of 3,000 mcg, or 10,000 IU, a day. Excess causes liver damage and birth defects. The upper limit does not apply to beta-carotene from food.
• Zinc. Doses of 50 mg a day or more for weeks block copper absorption and can cause anemia and nerve damage. The upper limit is 40 mg. For colds, zinc lozenges may shorten symptoms by about two days, with low-certainty evidence, but regular zinc does not prevent colds.
• Vitamin C. Regular vitamin C did not prevent colds in the general population and shortened them by about 8% in adults. In a Swedish cohort of men, vitamin C supplements were linked to about twice the risk of kidney stones. The upper limit is 2,000 mg a day.
• Selenium. In 2008, a liquid supplement with 200 times the labeled selenium caused severe reactions in 201 people. The upper limit is 400 mcg a day; Brazil nuts are unusually rich in selenium, so a few a day plus a supplement can add up.

Biotin and lab tests#

Biotin (vitamin B7), sold in high doses for hair, skin, and nails, can interfere with common lab tests that use a biotin-based chemistry, producing falsely normal or abnormal thyroid results; even a single 10 mg dose has done so within 24 hours. The FDA has warned that biotin can cause falsely low troponin results, the blood test used to diagnose heart attacks. Tell the lab and your doctor if you take biotin, including in a hair-and-nails blend or a multivitamin, and ask whether to stop it a few days before testing.

Before surgery#

Several herbs affect bleeding, blood pressure, sedation, or anesthetic drug levels. A JAMA review identified garlic, ginkgo, and ginseng (bleeding), St. John's wort (drug interactions), kava and valerian (sedation), and ephedra (heart rhythm and blood pressure) as concerns around surgery. The practical rule: bring a complete list of everything you take, including supplements, to your pre-surgery visit, and ask which to stop and when. In that review, suggested stopping times ranged from about a day to a week depending on the herb; your surgical team should decide.

How to decide and how to buy#

A good decision about a supplement looks like a decision about a drug: a specific reason, the best available evidence, a dose, a way to tell whether it is working, and a plan to stop.

Questions to ask first#

How to buy#

• Look for a seal. USP Verified, NSF Certified for Sport, or Informed Sport or Informed Choice. Search the certifier's own database to confirm the specific product is listed.
• Prefer single ingredients. Blends and "proprietary formulas" hide doses and make it impossible to tell which ingredient caused a problem.
• Match the dose to the trials. More is rarely better, and many products far exceed what was studied.
• Avoid the high-risk categories: sexual enhancement, weight loss, bodybuilding, and testosterone boosters.
• Be skeptical of subscriptions and "clinically proven" blends whose only trials were run by the seller.
• Report problems. Tell your doctor, and report serious reactions to the FDA's MedWatch program.

Why stacks backfire#

A "stack" is a daily combination of many supplements, sometimes a dozen or more. Stacks are popular in longevity circles and are a poor bet for several reasons. Each added product brings its own contamination risk, so total exposure to heavy metals and mislabeling grows with every bottle. Interactions multiply. Overlapping ingredients push some nutrients past their upper limits; zinc, vitamin A, and selenium are common culprits. And when something goes wrong, from a rash to a raised liver enzyme, nobody can tell which product caused it.

Stacks also cost real money that could buy things with stronger evidence: a gym membership, a blood pressure cuff, yearly blood work, better food, or a visit with a registered dietitian. In the DO-HEALTH trial of older adults, vitamin D, omega-3, and a simple home exercise program together slowed some epigenetic aging clocks by a few months over three years, a small effect on a measure that is not yet validated. The exercise was the part with the strongest independent evidence.

Action plan#