Bottom line
In the world of supplements, the firmest class of evidence is replace what is missing. Vitamin D, vitamin B12, and iron are examples. A blood test can say directly whether there is a shortfall, and a marker can show whether adding it did anything 27.
The weakest class is take it to prevent disease and to prevent aging. For vitamins and minerals taken to prevent cardiovascular disease and cancer, the evidence report of the U.S. Preventive Services Task Force concludes that the benefit is small or absent. Beta-carotene is harmful in people at high risk of lung cancer 13.
A multivitamin sits in the middle. Large randomized trials have seen a small improvement in cognition in older adults 710, and a small reduction in cancer incidence in a cohort of male physicians 11. They have not seen a benefit for cardiovascular disease or for mortality 713.
One line is that a multivitamin is nutrition insurance. It is not a longevity drug, and it cannot stand in for a varied diet 7.
Another line is to define the problem before talking about a supplement. The U.S. regulatory system for over-the-counter supplements itself produces unreliable products, so they are judged by a different standard than drugs 12.
A third line is that behavior is the foundation and a supplement is extra support built on it. If something is added, use single ingredients and layer them, so it is clear what is doing the work 45.
What it is
A dietary supplement is a vitamin, mineral, herbal extract, amino acid, or other physiologically active ingredient sold as a capsule, a tablet, a powder, or a similar form.
In the United States, the line between a supplement and a drug is drawn by regulation. A Q&A episode has a section whose title is a question: why does this regulatory system produce unreliable products 2?
Products fall roughly into three classes. One is the nutrient type, replacing molecules the body needs. Vitamins and minerals are the typical case.
One is the extract type, a component pulled from a food or an herb and concentrated past what food readily provides.
One more is the product of a gray market, peptides being the typical case. The risk discussed in that episode is exactly that source and quality are not guaranteed 3.
One account calls the widest-coverage version of the first class a foundational supplement: water-soluble and fat-soluble vitamins, minerals, digestive enzymes, adaptogens, probiotics, and prebiotics, meant to hold the floor of overall nutrition 5.
Nutrition insurance is a similar idea. Diet is supposed to be the main source, and in reality many people do not reach an ideal pattern 7.
Why it matters
Use is extremely common, and the expectation is often pointed the wrong way. About one third of U.S. adults report taking a multivitamin 7.
Some treat it as insurance against disease. At the level of preventing cardiovascular disease and cancer, the evidence is basically empty 13.
Others treat it as a longevity drug, and that layer is even emptier. A 2024 study pooling about 390,000 people specifically corrected for healthy user bias — people who take a multivitamin already smoke less, and they often have more education 9. After that correction, taking a daily multivitamin was not associated with lower mortality from any cause 79.
There is risk beyond the money. The failure modes of over-the-counter supplements listed in that episode include an unclear effect, poor quality control, contamination, interactions, toxicity, and marketing-driven overuse 2.
The Task Force evidence report also flags several harm signals, all marked as limited evidence: vitamin A and hip fracture, vitamin E and hemorrhagic stroke, and vitamin C and calcium with kidney stones 13.
For someone on a prescription drug, an interaction is not a small matter. A supplement can change how a drug is absorbed and metabolized, and it needs to be told to a clinician as something entering the body 2.
Different perspectives
Main positions
When supplements come up in one account, the talk is first about standards and process, not about brands 12.
Longevity strategy is organized into five domains. Exogenous molecules — drugs, supplements, hormones — are only one of them, and the point is made plainly that the aim is not to tell other people to take X and skip Y. A frame for thinking is what is offered 1.
In that episode the risk of over-the-counter supplements is split into six failure modes: effect, quality, contamination, interactions, toxicity, and marketing 2. The argument is that drugs and supplements are judged with different standards of trust and of evidence 2.
In an interview the two went item by item through the research on NAD, NMN, NR, rapamycin, and resveratrol 6. In that episode rapamycin is called the geroprotective drug seen as the most promising. NR and NMN are set aside, with a look again if the data change 6. The same episode adds that it is hard to imagine any supplement or drug bringing a benefit to energy that matches good sleep, exercise, and nutrition 6.
One account is the most positive toward supplements, and the order has not changed. Behavior first, supplements after 4.
The supplement topic page is blunt. Get the amount, the quality, and the timing of food in order first. Only then can a supplement work as extra support 4.
In the episode on method, the opening says the word supplements is disliked. It implies that supplements are only substitutes for food, when some ingredients are not found in food, or the amount in food is nowhere near enough 5.
The way of working is closer to engineering. Use single-ingredient supplements, add one thing at a time, layer them, and keep a record, so it is clear which thing is working. Source, cost, and continuous use versus a periodic stop are designed ahead of time 5.
Foundational supplements are separated from targeted supplements. The first holds the floor. The second aims at a concrete goal such as sleep, focus, or hormone support 5.
In the episode above, host and guest share the same view of order. Behavior — sleep, nutrition, exercise — ranks ahead of any supplement 6.
Another account is the most willing to defend micronutrient sufficiency.
When a supplement is looked at, the first question is how many people are actually short. U.S. intake figures cited show that micronutrient gaps are common. Vitamin D, magnesium, and calcium are on the list 7.
The cost of a severe shortfall is also described. DNA damage from severe folate deficiency can be comparable to one acute radiation exposure. At the same time, a severity of that degree is not common in the population 7.
On the hope that a multivitamin prevents disease, cold water is poured anyway. For cardiovascular disease, cancer, and mortality from any cause, the evidence does not support a benefit 7.
The COSMOS cognition results cited are marked with a boundary too. A pool of three cognition substudies found a small advantage in global cognition and episodic memory over two to three years, about equal to two years of age-related cognitive change 7. That is a modeled comparison, not a brain that actually became two years younger, and it is not to be treated as three independent replications 7.
Scaling amounts from animals is another red line. An effect seen in animal research has to be converted by body surface area, and the conversion is only a starting point. Research on nutrition supplements faces the same wall of difference between species 8.
Where views agree and differ
Points of agreement.
One is behavior first. Diet, sleep, and exercise rank ahead of any supplement 46.
Two is that replace what is missing is the best supported. Supplementation that has evidence of a shortfall, and a marker that can be tracked, is a completely different thing from general wellness 27.
Three is that expectation cannot outrun the evidence. One place for a multivitamin is nutrition insurance, not a proven longevity drug, and other popular ingredients are still being tested as well 7.
Two points of disagreement.
One is the default posture. One account is willing to default to a foundational supplement to cover gaps 45. Another is more conservative, and wants every supplement to pass three questions: the goal, the biomarker, and the balance of risk and benefit 12.
Two is that a third account separately stresses the long-term cost of micronutrient deficiency itself — DNA damage, epigenetic change — and holds that insurance makes sense at a small cost 7. The more conservative account will ask whether this is a real problem, or whether it is doing something feels reassuring 2.
PearHealth's synthesis
We sort supplements into three tiers by the strength of the evidence.
The first tier is replace what is missing. Vitamin D, vitamin B12, and iron can be confirmed short by a blood test, and checked again after they are added. The evidence is direct, and it is worth discussing with a clinician 27.
The second tier is a possible small, specific benefit. The multivitamin signal for cognition in older adults comes from a randomized trial. The size is small and the mechanism is unclear. Treating it as insurance for cognition is tenable. Treating it as a longevity drug is not 710.
The third tier is prevent disease, resist aging. The evidence on preventing cardiovascular disease, cancer, and death is overall negative or insufficient, and individual ingredients are harmful 913.
We do not sell supplements. For any supplement, the first questions are what problem it is meant to solve, and what marker will show whether it worked 1.
How strong is the evidence
The advantage of this topic is that there is a lot of research. The disadvantage is that the quality of that research varies widely.
Randomized trials are the hardest tier. COSMOS-Mind randomized 2,262 older adults to a multivitamin or to placebo. After three years, global cognition tests showed a small improvement 10. In the trial of male physicians, a daily multivitamin lowered the incidence of total cancer by about 8%. Prostate cancer and colorectal cancer did not change 11. In a large trial of vitamin D, the rates of cancer and of cardiovascular events did not differ from the placebo group 12.
Cohort studies are read at a discount. An analysis of about 390,000 people, after correction for healthy-user bias, found no mortality benefit for a multivitamin 9. Studies of this kind can say no association was seen. They cannot say proved useless.
Mechanism research and animal research can only offer a hypothesis. An effect of a nutrient in cells and in mice has to pass two walls, conversion by body surface area and a trial in people. Most never reach the second wall 8.
A personal supplement list is talk from experience. It does not even count as evidence. It can only be a reference 6.
What you can do
- Test before adding. Vitamin D, vitamin B12, and ferritin are markers a blood test can use to answer is there a shortfall directly 27.
- When something is short, food comes first. A supplement is a tool, not a substitute 47.
- If something is added, use a single ingredient, add one thing at a time, and keep a record for two weeks before judging 5.
- When prescription drugs are in use, tell the clinician and the pharmacist every supplement 2.
- Keep an extra measure of skepticism for advertising of the treats-everything kind. Marketing-driven overuse is one of the failure modes that has been named 2.
- Do not treat expensive, natural, or imported as evidence of quality 2.
- Every so often, ask what problem this is still solving. If the answer is unclear, stopping is available 1.
Questions worth raising with a clinician:
- Given the shape of my diet, which markers are worth checking first?
- The multivitamin I am taking: is there corresponding evidence of a shortfall 9?
- I have kidney disease, I am pregnant, or I take an anticoagulant. Which supplements are to be left out 13?
Common misconceptions
Misconception one: supplements carry no risk. The failure modes that have been named include contamination, toxicity, and interactions 2. Vitamins A, E, and C, and calcium, are linked respectively to hip fracture, hemorrhagic stroke, and kidney stones, though this set of signals has limited evidence 13.
Misconception two: a multivitamin prevents cancer and heart disease. Randomized trials show no benefit for cardiovascular disease or for mortality, and only a very small signal for cancer 1113.
Misconception three: effective in a cell or a mouse means effective when a person takes it. Differences between species, plus the conversion of amounts, mean most mechanism results never arrive in people 8.
Misconception four: a natural extract is mild and safe. Concentrated past what food reaches, the action and the risk are not the same thing as food 58.
Misconception five: expensive means better. Price is not among any quality signal that has been named. Poor quality control, and marketing-driven overuse, are both on the list of failure modes 2.
How it connects to your blood work
Supplements are one of the few interventions where a blood test can directly inform whether adding something is on the table. The markers below are the ones used most often.
- 25-hydroxyvitamin D (vitamin-d-25oh): the standard marker for whether vitamin D is short. It can be checked before and after 7.
- Vitamin B12 (vitamin-b12): older adults, vegetarians, and people on long-term metformin tend to run low. Adding it has a clear indication.
- Serum folate (folate-serum): read together with B12. Commonly used in pregnancy and in the assessment of anemia.
- Ferritin (ferritin): the marker of iron stores. Look at it before deciding whether iron is added.
- Omega-3 index (omega-3-index): the share of EPA and DHA in the red-cell membrane, the target marker for a fish-oil supplement.
- Red-blood-cell magnesium (magnesium-rbc): reflects long-term magnesium status better than serum magnesium, and it is the more useful look when the question is whether magnesium is worth adding 7.
- Homocysteine (homocysteine): an indirect signal of a shortfall in folate, B12, or B6. Often used to show whether the B vitamins are sufficient.