Randomized human trial · Evidence assessment
Did a daily multivitamin prevent cancer in COSMOS?
COSMOS did not establish a reduction in total invasive cancer during its follow-up. This result does not settle every use of a multivitamin.
Longevity Lab Report · Assessed · Main study published 2022-03-16
Claim being examined
A daily multivitamin prevents cancer in generally healthy older adults.
This page assesses the named study and relevant context. It is not a complete systematic review or an individual medical assessment.
What the study measured
- Population
- 21,442 US adults: women aged 65+ and men aged 60+, without major cardiovascular disease or recently diagnosed cancer.
- Study design
- Randomized, double-blind, placebo-controlled factorial trial.
- What was tested
- One daily multivitamin-mineral product (Centrum Silver); participants were also randomized to cocoa extract or its placebo.
- Comparison
- Matching multivitamin placebo.
- Duration
- Median 3.6 years.
- Measured outcome
- Total invasive cancer, excluding nonmelanoma skin cancer.
- Observed result
- 518 of 10,720 versus 535 of 10,722 participants developed cancer. Hazard ratio 0.97 (95% confidence interval 0.86–1.09): the interval includes no difference.
- Safety limits
- This efficacy result is not an individual safety assessment.
- Funding and disclosures
- NIH and Mars Edge support; Pfizer supplied study products. The paper reports no sponsor role in analysis or interpretation.
What limits the conclusion?
- A few years of follow-up cannot rule out a longer-term effect.
- A result for one product and older participants does not establish effects in younger people or people with a documented deficiency.
Prior and related evidence
The earlier Physicians’ Health Study II randomized 14,641 male physicians and found a modest reduction in total cancer over roughly 11 years (hazard ratio 0.92). Its population and follow-up differed from COSMOS.
The source list below distinguishes the main study from context.
What this evidence added
COSMOS added a larger trial that included women, with a null primary cancer result.
What it did not establish
Cancer prevention, cognition and treatment of deficiency are separate questions. This explainer assesses the cancer endpoint only.
Still unknown
- Whether a longer trial would find a different cancer result.
- Whether any individual would benefit from correcting a specific nutrient deficiency.
Does this study address your question?
Optionally compare the population, intervention and outcome. These are study comparisons, not medical advice or a treatment recommendation. Your answers stay on this page and are not saved or included in sharing.
Keep the evidence available
Open the public card to print or shareSaved in this browser only. No account or email required.
Assessment history
· Baseline assessment
First entry in this evidence history. This is a dated assessment of existing research, not a newly published study or a change to a library grade.
Conclusion changed: not applicable — no earlier version in this history.
Sources: cosmos · phs · cosmos-results
Sources
- COSMOS multivitamin trial (Sesso et al., 2022)
Main study · Published 2022-03-16 · Checked 2026-11-15 · PMID 35294969
Full primary publication via indexed PMC text
- Physicians’ Health Study II (Gaziano et al., 2012)
Context · Published 2012-11-14 · Checked 2026-11-15 · PMID 23162860
Primary publication abstract and indexed results
- COSMOS investigators’ results summary
Context · Published Undated results page · Checked 2026-11-15 · Trial NCT02422745
Trial team results page; indexed text
Follow the research in the newsletter
Read the week’s longevity reporting and evidence assessments. Your watchlist and comparison answers are separate from your subscription.
Free. Unsubscribe anytime.
Every grade follows a published rubric. Sponsors do not influence findings or grades.