2019NEJM
Manson et al. (VITAL)
25,871 adults, 5 years
2,000 IU of vitamin D or 1 g of fish oil a day didn't lower cancer or heart disease overall in healthy adults.
Find on PubMed ↗Research library · Food & life
Supplements: what the evidence really supports
Supplements are a huge industry, but large trials show most do little for healthy people who eat well. A handful have solid evidence for specific needs, and testing first tells you which ones you need.
Good evidenceUpdated Oct 2026 · 4 new studies−22%autoimmune disease with 2,000 IU of vitamin D a day over 5 yearsHahn 2022, BMJ (VITAL trial)
2 yrsof memory ageing slowed by a daily multivitamin in older adultsVyas 2024, Am J Clin Nutr, 3 COSMOS trials
390kadults followed 20+ years: daily multivitamins did not help them live longerLoftfield 2024, JAMA Netw Open
01
The basics
Food first, then fill real gaps
A supplement helps most when it fixes a shortfall. Common ones: vitamin B12 in vegans, older adults and people on metformin; vitamin D in people who get little sun; iron in people with heavy periods; folic acid before and during pregnancy.
When people who aren't short of a nutrient take more, big trials usually find no benefit. Some high doses cause harm: beta-carotene raised lung cancer in smokers, and high-dose vitamin E raised prostate cancer in one large trial.
Supplements aren't checked like medicines before they're sold. Products can contain less (or more) than the label says. Look for third-party seals such as USP, NSF Certified for Sport or Informed Sport.
Which supplements have evidence?
A rough guide from large trials and reviews. 'Strong' means it helps a specific group, not that everyone needs it.
02
How it works
How to choose a supplement
1
Test first
A blood test shows whether you're low in vitamin D, B12 or iron.
2
Fix food
Many gaps close with food: oily fish, eggs, dairy, leafy greens, beans.
3
Pick one need
Choose the one supplement that matches a real gap or goal.
4
Choose tested brands
Third-party seals check purity and dose.
03
Measure it
Tests worth knowing
- Vitamin D (25-hydroxy): the most common shortfall in northern countries in winter.
- Vitamin B12: worth checking if you're vegan, over 60, or take metformin or acid-reducing medicines.
- Ferritin: your iron stores. Low iron causes tiredness well before anaemia shows up.
- Check interactions: St John's wort weakens many medicines; vitamin K affects warfarin; high-dose fish oil can add to blood-thinners. Tell your pharmacist what you take.
Try your number
DeficientLowAdequateNo added benefit
10200
Ranges from the US National Academies. Divide by 2.5 for ng/mL.
04
What the research says
The key studies
2022BMJ
Hahn et al. (VITAL)
25,871 adults
In the same trial, vitamin D cut new autoimmune diseases such as rheumatoid arthritis and psoriasis by 22%.
Find on PubMed ↗2024Am J Clin Nutr
Vyas et al. (COSMOS)
3 trials, 5,000+ older adults
A daily multivitamin improved memory and overall thinking scores, roughly equal to 2 years less cognitive ageing.
Find on PubMed ↗2024JAMA Network Open
Loftfield et al.
390,124 US adults, 20+ years
Healthy adults who took a daily multivitamin did not live longer than those who didn't.
Find on PubMed ↗2023Nutrition Reviews
Prokopidis et al.
10 trials
Creatine (usually 3–5 g a day) improved memory, with the clearest effect in older adults.
Find on PubMed ↗2025Nature Aging
Bischoff-Ferrari et al. (DO-HEALTH)
777 adults aged 70+, 3 years
1 g of omega-3 a day slowed several DNA-based ageing clocks by about 3 months. Adding vitamin D and home exercise added to the effect.
Find on PubMed ↗05
New in 2025–26
What the newest research adds
Big trials keep finding small, mainly lab-measured effects: vitamin D slowed telomere shortening and a multivitamin slightly slowed epigenetic ageing clocks. Targeted vitamin D after a heart attack and creatine for brain and muscle look promising but need confirming. Popular NMN still lacks solid human evidence.
The new studies, in order
Every study added to this page since January 2025. Green rings are the strongest evidence: large trials or meta-analyses.
Vitamin D and the chromosome caps
VITAL sub-study, 2025. How much telomeres (protective caps on chromosomes) shortened over 4 years with 2,000 IU of vitamin D a day vs placebo.
NewMay 2025Am J Clin Nutr
Zhu et al. (VITAL)
VITAL sub-study, adults 50+, 4 years
Telomeres (protective caps on chromosomes) shortened by 1.8% on placebo but only 0.2% on 2,000 IU of vitamin D a day, roughly three years less ageing. Omega-3 had no effect.
Good evidenceRead the source ↗
NewMar 2026Nat Med
Li, Sesso, Manson et al. (COSMOS)
958 adults, average age 70, 2 years
A daily multivitamin slowed epigenetic ageing clocks by about 4 months over 2 years, most in people ageing faster at the start. Cocoa extract had no effect. It is not yet known whether this means less disease.
Good evidenceRead the source ↗
NewNov 2025AHA Scientific Sessions
May et al. (TARGET-D)
630 heart attack patients
Testing vitamin D and adjusting the dose to keep levels above 40 ng/mL halved the chance of a second heart attack, though overall heart events did not differ. Presented at a conference, not yet peer-reviewed.
Early evidenceRead the source ↗
NewSep 2026J Int Soc Sports Nutr
Chun, Kreider et al.
64 untrained adults aged 45 to 65, 12 weeks
10 g of creatine a day, even without exercise, increased lean mass and strength compared with placebo, and improved some memory tests. The authors call the brain findings exploratory.
Early evidenceRead the source ↗
What this changes for you
Vitamin D (if you're low or get little sun), creatine and a basic multivitamin are cheap and have the most new support, but the effects are small and mostly on lab markers. Spend first on sleep, training and food.
06
Take action
A sensible supplement routine
Most people need few or none. These are the ones with the best case.
- 1
Test before you take
Ask for vitamin D, B12 and ferritin at your next check-up.
- 2
Vegan? Take B12
There's no reliable plant source. A daily or weekly B12 supplement is essential.
- 3
Little sun? Vitamin D in winter
UK advice is 10 µg (400 IU) a day from October to March; people with low levels may need more.
- 4
Creatine if you train
3–5 g a day of creatine monohydrate is the best-studied sports supplement, and safe in healthy people.
- 5
Fish twice a week
Food beats fish-oil pills for most people. Prescription omega-3 is for people with high triglycerides.
- 6
Look for test seals
USP, NSF or Informed Sport labels mean the product was checked for what's inside.
07
Myth or fact
Common myths, checked
✕ MythNatural means safe.
✓ FactGreen tea extract, turmeric and some 'liver detox' blends have caused liver injury. Natural products can still be strong medicines.
✕ MythMore vitamins are better.
✓ FactBeta-carotene pills raised lung cancer in smokers, and too much vitamin A weakens bones. Above your needs, extra rarely helps.
✕ MythA multivitamin makes up for a poor diet.
✓ FactFood contains fibre and thousands of plant compounds no pill copies. Multivitamins didn't help people live longer.
✕ MythCreatine harms your kidneys.
✓ FactIn healthy people, studies have found no kidney harm at normal doses. It can raise a lab test (creatinine) without meaning damage.
08
Questions
Frequently asked
Should everyone take vitamin D?
Not everyone. It's worth it if you're low, get little sun, have dark skin in a northern country, or are older and housebound.
Is magnesium good for sleep?
Small trials show modest benefit, mostly in people who are low. It's cheap and generally safe at normal doses.
What about collagen?
Some trials show small improvements in skin elasticity, often funded by makers. Getting enough total protein matters more.
09
Test yourself
True, partly true or myth?
Read each claim and pick an answer: the card flips to show the verdict and why. Not sure? Use “Just reveal”.
Card № 01True or myth?
Creatine is one of the most studied and effective supplements for strength.
True
Creatine is one of the most studied and effective supplements for strength.
Hundreds of trials show it boosts strength and muscle gains with training, and it's safe for healthy people at 3–5 g a day.
Card № 02True or myth?
Multivitamins help healthy people live longer.
Myth
Multivitamins help healthy people live longer.
Large studies, including a 2024 analysis of almost 400,000 US adults, found no lower risk of death.
Card № 03True or myth?
Vitamin D supplements help people who are actually low in it.
True
Vitamin D supplements help people who are actually low in it.
In people with deficiency, supplements improve bone health; in people with normal levels, trials like VITAL found little benefit.
Card № 04True or myth?
Fish oil prevents heart attacks.
Partly true
Fish oil prevents heart attacks.
Ordinary doses showed little benefit in big trials. A high-dose prescription form (REDUCE-IT) cut events in high-risk patients.
Card № 05True or myth?
Supplements aren't tested like medicines before they're sold.
True
Supplements aren't tested like medicines before they're sold.
In the US and many countries, makers don't need to prove they work or contain what the label says before selling.
Card № 06True or myth?
Natural supplements can't interact with medicines.
Myth
Natural supplements can't interact with medicines.
St John's wort, for example, weakens the pill, blood thinners and many other drugs.
Card № 07True or myth?
Very high doses of some vitamins can cause harm.
True
Very high doses of some vitamins can cause harm.
Too much vitamin A, D or B6 over time can damage the liver, kidneys or nerves.
Card № 08True or myth?
Magnesium improves sleep.
Partly true
Magnesium improves sleep.
It may help people who are low in magnesium; evidence in people with normal levels is weak.
Card № 09True or myth?
Folic acid before and in early pregnancy reduces neural tube defects.
True
Folic acid before and in early pregnancy reduces neural tube defects.
This is one of the strongest supplement findings, which is why it's recommended for anyone who might become pregnant.
Card № 10True or myth?
If a little is good, more is better.
Myth
If a little is good, more is better.
Most vitamins have a sweet spot. Beta-carotene in smokers raised lung cancer risk in trials.
Nothing here yet. Try another filter.
10
Sources
References and related topics
- Manson et al. (VITAL) (2019). NEJM. 25,871 adults, 5 years. PubMed ↗
- Hahn et al. (VITAL) (2022). BMJ. 25,871 adults. PubMed ↗
- Vyas et al. (COSMOS) (2024). Am J Clin Nutr. 3 trials, 5,000+ older adults. PubMed ↗
- Loftfield et al. (2024). JAMA Network Open. 390,124 US adults, 20+ years. PubMed ↗
- Prokopidis et al. (2023). Nutrition Reviews. 10 trials. PubMed ↗
- Bischoff-Ferrari et al. (DO-HEALTH) (2025). Nature Aging. 777 adults aged 70+, 3 years. PubMed ↗
- Zhu et al. (VITAL) (2025). Am J Clin Nutr. VITAL sub-study, adults 50+, 4 years. Source ↗
- Li, Sesso, Manson et al. (COSMOS) (2026). Nat Med. 958 adults, average age 70, 2 years. Source ↗
- May et al. (TARGET-D) (2025). AHA Scientific Sessions. 630 heart attack patients. Source ↗
- Chun, Kreider et al. (2026). J Int Soc Sports Nutr. 64 untrained adults aged 45 to 65, 12 weeks. Source ↗