A daily multivitamin taken for about 11 years made no significant difference to rates of heart attack, stroke or death from cardiovascular disease among 14,641 male US physicians, while the same trial recorded 8 per cent fewer cancer diagnoses in the men who took the pill.
The results come from the Physicians' Health Study II, a randomised, double-blind, placebo-controlled trial of a common daily multivitamin, Centrum Silver. It began in 1997, with treatment and follow-up continuing through 1 June 2011. Two reports published in JAMA in 2012, one led by Sesso and colleagues on cardiovascular outcomes and one by Gaziano and colleagues on cancer, set out the findings.
The video below, on the VegOut channel, is titled "THE VITAMIN SCAM: Why Most Supplements Are Worthless (Or Worse)". The trial's two main results point in different directions.
The heart results
Over a median follow-up of 11.2 years, with an interquartile range of 10.7 to 13.3 years, the trial counted 1,732 confirmed major cardiovascular events. The primary outcome was a composite of nonfatal myocardial infarction, nonfatal stroke and fatal cardiovascular disease.
The rate was 11.0 events per 1,000 person-years in the multivitamin group and 10.8 in the placebo group. The hazard ratio was 1.01, with a 95 per cent confidence interval of 0.91 to 1.10 and P=0.91. On those figures the two groups were indistinguishable.
The individual components gave the same picture. Total myocardial infarction ran at 3.9 versus 4.2 events per 1,000 person-years, hazard ratio 0.93, confidence interval 0.80 to 1.09, P=0.39. Total stroke was 4.1 versus 3.9, hazard ratio 1.06, confidence interval 0.91 to 1.23, P=0.48. Cardiovascular mortality was 5.0 versus 5.1, hazard ratio 0.95, confidence interval 0.83 to 1.09, P=0.47.
Total mortality from any cause gave a hazard ratio of 0.94, confidence interval 0.88 to 1.02, P=0.13, which the paper reports as not statistically significant. Of the men enrolled, 754 already had a history of cardiovascular disease at randomisation, and the effect on major cardiovascular events did not differ between those men and the rest, with an interaction P value of 0.62.
The Sesso paper concludes: "A daily multivitamin did not reduce major cardiovascular events, MI, stroke, and CVD mortality after more than a decade of treatment and follow-up."
The cancer numbers
The cancer report covered the same men over the same period. There were 2,669 men with confirmed cancer, excluding non-melanoma skin cancer, including 1,373 cases of prostate cancer and 210 cases of colorectal cancer.
Total cancer incidence was 17.0 events per 1,000 person-years in the multivitamin group against 18.3 in the placebo group. That gives a hazard ratio of 0.92, with a 95 per cent confidence interval of 0.86 to 0.998 and P=0.044. The 8 per cent reduction in the headline figure is one minus that hazard ratio, and the upper bound of the interval sits just under 1.0.
The site-specific results were flat. Prostate cancer gave a hazard ratio of 0.98 with P=0.76. Colorectal cancer gave 0.89 with P=0.39. The paper reports no significant effect on those cancers or on other site-specific cancers.
Cancer mortality moved in the same direction as total incidence but did not clear the usual threshold. The hazard ratio was 0.88, confidence interval 0.77 to 1.01, P=0.07. The paper describes this as a lower risk of cancer mortality that did not reach statistical significance. On the trial's own figures, a daily multivitamin cannot be said to have cut deaths from cancer.
One subgroup stood out. Among the 1,312 men who already had a history of cancer when they were randomised, daily multivitamin use was associated with a reduction in total cancer, hazard ratio 0.73, confidence interval 0.56 to 0.96, P=0.022. Among the 13,329 men initially free of cancer, the hazard ratio was 0.94, confidence interval 0.87 to 1.02, P=0.15. The paper says that result did not differ significantly between the two groups, with an interaction P value of 0.07.
The Gaziano paper concludes: "In this large prevention trial of male physicians, daily multivitamin supplementation modestly but significantly reduced the risk of total cancer."
Who was enrolled
The limits of the trial start with its participants. All 14,641 were male US physicians, initially aged 50 or older, with a mean age of 64.3 years. The trial produced no results for women, and none for younger men, because neither group took part. The 8 per cent figure applies to this population and cannot be transferred to anyone else.
The trial also tested one product, Centrum Silver, taken daily. It did not test high-dose single nutrients.
Gaziano and colleagues say multivitamin preparations are the most common dietary supplement, taken by at least one-third of all US adults, and that limited observational studies had not provided evidence on total and site-specific cancer incidence or mortality. Sesso and colleagues say observational studies of regular multivitamin use had been inconsistently associated with cardiovascular disease, with no long-term clinical trials of multivitamin use. The Sesso paper also says multivitamins aim to prevent vitamin and mineral deficiency.
More than a decade of treatment and follow-up produced no significant effect on major cardiovascular events, myocardial infarction, stroke or cardiovascular death, a modest but significant reduction in total cancer diagnoses, and a cancer mortality hazard ratio of 0.88 at P=0.07 that did not reach statistical significance.
Other VegOut nutrition-research explainers cover how Joseph Goldberger traced pellagra to niacin deficiency in corn-heavy diets, Denis Burkitt's 1971 colon cancer paper and daily fibre targets, and a survey of how many nutrition hours US medical schools taught.
Photo: Multivitamin tablets (thematic stock image, not from the trial). Карма2, CC BY-SA 4.0, via Wikimedia Commons.




