Which supplements actually work, and which the research does not support
Only a short list has strong human evidence: creatine for strength and lean mass, caffeine for performance, vitamin D for people who are genuinely deficient, and protein powder as food. Most of the rest sits between weak and no human data. Three deserve real caution — high-dose biotin distorts blood tests, St John's wort interferes with many medicines, and yohimbine has a narrow safety margin.
The supplement aisle is not sorted by evidence, so the first useful thing to know is that being on sale is not a finding. Supplements do not have to prove they work before they are sold.
Which ones have solid evidence?
- Creatine monohydrate — the most studied sports supplement there is, with consistent effects on strength, power and lean mass at 3 to 5 grams a day. It is also the one most often avoided for reasons that turn out to be myths.
- Caffeine — a real, measurable effect on endurance and perceived effort, which is why it appears in almost every pre-workout ever formulated.
- Vitamin D — well established for bone and for correcting a genuine deficiency. In people who are already replete, the large VITAL trial of 25,871 people found no reduction in fractures, which is the part the label never mentions.
- Protein powder — works as food, not as a drug, and that is reason enough.
- Omega-3 — solid for triglycerides, much weaker for everything else it gets sold for.
Where does the marketing part company with the research?
Garcinia cambogia — a randomised trial in JAMA (Heymsfield 1998) with 135 people found no difference from placebo in either weight or fat mass.
Raspberry ketone — cell and rodent data at doses that do not convert to anything a person could eat. No human trials worth citing.
Collagen for skin — digested into amino acids like any other protein; the trials that report benefits are mostly small and industry-funded.
Detox blends — the liver and kidneys already do this, continuously, and no supplement has been shown to make them better at it.
Why does an evidence badge matter?
Because "helps with sleep" and "helps with sleep, measured once in fifteen people by the company selling it" are not the same sentence, and on most labels they look identical.
Grading every claim is the only way to let someone judge, without reading a paper, whether the sentence in front of them is a finding or a hope. It is also why some entries in this app say plainly that a popular belief does not survive the literature — a page that only ever lists benefits is an advertisement, not a record.
Three that need real care
High-dose biotin is not toxic, but it interferes with laboratory assays. It makes troponin — the heart attack test — read falsely low, and a patient has died because of it. Stop it several days before any blood test and tell the lab you take it.
St John's wort induces the enzymes that clear many medicines, which can quietly reduce the effect of contraceptives, anticoagulants, transplant medication and antiretrovirals. This is one of the few supplement interactions that has caused documented harm.
Yohimbine has a narrow margin between a dose that does something and a dose that causes a racing heart, anxiety and raised blood pressure, and the amount in a capsule frequently does not match the label.
Reading something you can act on beats reading something you cannot.