Multivitamins and Cancer Prevention: Using the Data

A Best Evidence Review

Charles P. Vega, MD; Megan Tan


December 14, 2012

In This Article

Description of the Study

The PHS II was a randomized, double-blind, placebo-controlled, 2 × 2 × 2 × 2 factorial study examining the efficacy of a daily multivitamin, vitamin E 400 IU every other day, vitamin C 500 mg/day, and beta-carotene 50 mg every other day. Outcomes were far-reaching and included incident cancer, cardiovascular disease, eye disease, and cognitive disease.

The current study examines the results from the multivitamin component of the trial only. All study participants were male physicians aged 50 years or older. Men with cirrhosis and other serious illness were excluded from the study protocol, although men with a history of cancer or cardiovascular disease were allowed to participate.

Participants completed annual questionnaires to update their health status, and researchers used national databases to investigate potential deaths. Overall, follow-up rates exceeded 98% over a mean of 11.2 years.

Self-reports of cancer were validated in the medical record. The main study outcome was an intention-to-treat analysis of the effect of daily multivitamins on the risk for incident cancer (excluding nonmelanoma skin cancer). Researchers also investigated the effects of multivitamins on specific types of cancer.

The study analyzed data from 14,641 men, whose mean age was 64.3 years. Of note, 77.4% of these physicians used daily aspirin at baseline, and 9.0% of all participants had a history of cancer.

Adherence to study treatment declined slightly from 77% at 4 years to 67% at the end of follow-up. Over 80% of participants avoided the use of nonstudy multivitamins during the study period.

There were 2669 cases of incident cancer during follow-up. Prostate cancer was by far the most common type reported (1373 cases), followed by melanoma, lymphoma, and colorectal cancer. There were 859 cancer-related deaths in the study cohort.

Rates of incident cancer were 17.0 and 18.3 cases per 1000 person-years in the multivitamin and placebo groups, respectively (hazard ratio [HR], 0.92; 95% confidence interval [CI], 0.86-0.998; P = .04). Multivitamins were similarly effective in an analysis limited to the prevention of all epithelial cell cancer. 

The effect of multivitamins was slightly more pronounced in preventing cancers other than prostate cancer (HR, 0.88; 95% CI, 0.79-0.98; P = .02). However, the reductions in the risk for site-specific cancer of the colon/rectum, lung, and bladder associated with multivitamin use were not statistically significant. Multivitamins also had no effect on the risk for cancer mortality or overall mortality.

There was a nonsignificant trend toward greater efficacy of multivitamins in preventing cancer among men older than 70 years. In contrast, men with no parental history of cancer experienced a cancer benefit with multivitamin therapy (HR 0.86; 95% CI, 0.76-0.98; P = .02), whereas men with a parental history of cancer did not. Multivitamin therapy was also particularly effective in preventing cancer among men with a history of cancer at baseline (HR, 0.66; 95% CI, 0.50-0.88; P = .004).

There was a 7% increase in the risk for rash among men receiving multivitamins vs placebo. Multivitamins were associated with a paradoxical increase in the incidence of epistaxis but a decrease in the incidence of hematuria. However, they had no effect on the rate of easy bruising or other bleeding.