Skip to content
Longevity Record

Evidence brief

The biggest longevity trials ever run were null — and almost nobody talks about them

6 min read ·

Ask what the largest randomised experiments in the history of longevity-adjacent science found, and the answer is awkward for an industry: mostly nothing. That nothing is worth more than most of what is sold in its place.

Two enormous experiments

D-Health gave 21,315 older Australians high-dose monthly vitamin D or placebo for five years, with death from any cause as the primary outcome. No difference. VITAL gave 25,871 US adults a gram of omega-3 daily for a median of five years. It prevented neither major cardiovascular events nor cancer.

These are not obscure studies — they are among the largest, cleanest prevention trials ever conducted, testing the two best-selling supplements in the western world against the outcomes people buy them for.

Vitamin D evidenceOmega-3 evidence

Why null results disappear

Nobody's business model improves when a trial finds nothing. There is no press tour for a null result, no product launch, no before-and-after photo. So the observational studies that hinted at benefit get cited for years after the randomised trials that tested them properly came back empty. On this site the nulls are recorded as first-class evidence: both claims grade as evidence against, the strongest 'no' our vocabulary has.

What survives the nulls

Precision survives. Vitamin D for diagnosed deficiency is standard medicine. Purified high-dose EPA reduced cardiovascular events in one specific statin-treated, high-triglyceride population in REDUCE-IT. The lesson is not that supplements never work — it is that the honest claims are narrow, and the marketing claims are wide.

Every lifespan claim, graded

Sources

  1. [1]The D-Health randomised controlled trial, 2022. PMID 35026158
  2. [2]VITAL: marine n-3 fatty acids and prevention of cardiovascular disease and cancer, 2019. PMID 30415637
  3. [3]REDUCE-IT: cardiovascular risk reduction with icosapent ethyl, 2019. PMID 30415628
Educational information, not medical advice — see the medical disclaimer. Evidence grades referenced here are produced by our published methodology.

Keep reading