Current issues of ACP Journal Club are published in Annals of Internal Medicine


Long-term supplementation with β-carotene did not affect the incidence of cancer or cardiovascular disease in men

ACP J Club. 1996 Sept-Oct;125:41. doi:10.7326/ACPJC-1996-125-2-041

Source Citation

Hennekens CH, Buring JE, Manson JE, et al. Lack of effect of long-term supplementation with beta carotene on the incidence of malignant neoplasms and cardiovascular disease. N Engl J Med. 1996 May 2;334:1145-9. [PubMed ID: 8602179]



To determine the effects of long-term supplementation with β-carotene on the incidence of cancer and cardiovascular disease in men enrolled in the Physicians' Health Study.


Randomized, double-blind, placebo-controlled trial with a 2 × 2 factorial design and mean 12-year follow-up.


United States.


22 071 male physicians who were 40 to 84 years of age in 1982 and had no history of cancer (except nonmelanoma skin cancer), myocardial infarction (MI), stroke, or transient cerebral ischemia. 11% of participants were current smokers, and 39% were former smokers. All participants completed 11 years of follow-up at which time 99.2% provided morbidity information and mortality follow-up was complete for 99.9%; 80% of participants were still taking study pills.


In 1982, participants were allocated to aspirin, β-carotene, both, or neither. In 1988, the aspirin component of the study was stopped. The β-carotene component continued with 11 036 physicians receiving β-carotene and 11 035 physicians receiving placebo. β-carotene was administered in a 50-mg dose given on alternate days. Physicians were mailed yearly questionnaires and nonresponders were contacted by telephone for follow-up information.

Main outcome measures

Incidence of malignant neoplasms (other than nonmelanoma skin cancer), MI, stroke, and death from cardiovascular causes.

Main results

Results are displayed in the Table. At 12 years, treatment with β-carotene did not show any protective or harmful effect on the incidence of malignant neoplasms compared with placebo (P = 0.65). No effect was shown when cancers were analyzed individually. Furthermore, β-carotene did not show any protective or harmful effect compared with placebo on the incidence of MI (P = 0.50), stroke (P = 0.60), or death from cardiovascular causes (P = 0.28). No major side effects were noted with β-carotene.


Long-term supplementation with β-carotene did not convey any beneficial or harmful effect on the incidence of cancer or cardiovascular disease among men in the Physicians' Health Study.

Source of funding: National Institutes of Health.

For article reprint: Dr. C.H. Hennekens, 900 Commonwealth Avenue East, Boston, MA 02215, USA. FAX 617-731-3843.

Table. β-carotene to prevent cardiovascular disease or malignant neoplasms in men*

Outcomes at 12 y β- carotene Placebo RRR(95% CI) NNT
Incidence of malignant neoplasms 11.5% 11.7% 1.6%(-5.9 to 8.5) Not significant
Incidence of myocardial infarction 4.2% 4.4% 4.3%(-8.3 to 15) Not significant
Stroke 3.3% 3.5% 3.9%(-11 to 17) Not significant

*Abbreviations defined in Glossary. RRR, NNT, and CI calculated from data in article.


Combination b-carotene and vitamin A was harmful in patients at risk for lung cancer

The CARET trial is 1 of a series of important randomized trials that tests the efficacy of dietary supplements. The study design is exemplary, although follow-up is relatively short (median 3.7 y). Patients in the target group (men exposed to asbestos and long-term smokers, 44% of whom were women) were at high risk for lung cancer and cardiovascular disease.

The finding that patients who were randomized to β-carotene and vitamin A had an increase in overall mortality and lung cancer deaths, plus an increase in cardiovascular mortality, is contrary to the observational studies that show a lower mortality rate with higher intake of fruits and vegetables (1). Intriguingly, the baseline serum β-carotene levels in this study were inversely correlated with subsequent development of lung cancer. Because supplements had no beneficial effect on survival, this finding needs to be elucidated. One explanation for the effect is that β-carotene is a marker for a different protective component in fruit and vegetables or for another separate confounding factor. Further, the supplementation may have been pharmacologically inappropriate.

The study by Hennekens and colleagues is a double-blind placebo-controlled trial of β-carotene in over 22 000 male physicians. Only 11% were current smokers, although 39% were former smokers. The result showed no benefit for men who were randomized to β-carotene. As with the CARET trial, and again in contrast with observational studies, no decrease was observed in the mortality rate from lung cancer or heart disease.

Unlike the CARET trial and 1 other study (2) that showed increased death rates for patients who received β-carotene and other supplements, no apparent detrimental effects from β-carotene occurred. This could be the play of chance, or perhaps the choice of patients at high risk for these diseases in the other studies was important. High serum levels of β-carotene, retinyl-palmate, and vitamin E may have interacted negatively in the process of carcinogenesis and formation of atheroma. No evidence existed, however, indicating that those with the highest β-carotene levels were at greatest risk.

The fact that trials have failed to show any advantage for nutritional supplements underlines the dictum that positive associations in observational studies must be tested in randomized controlled trials. The observation that a high intake of fruit and vegetables and serum levels of β-carotene are correlated with a low risk for lung cancer and cardiovascular death remains unexplained. The only study that showed a modest benefit for β-carotene, vitamins, and selenium only recruited patients who were vitamin and mineral depleted (3).

If all the studies are taken into account, no support exists for the use of supplementation, although a diet high in fruits and vegetables is still recommended. Stopping smoking remains the best way to decrease the risk for lung cancer and cardiovascular disease. Research is needed to find out why high intake of fruits and vegetables is apparently beneficial.

Chris J. Williams, DM
Institute of Health SciencesOxford, England, UK


1. Block G, Patterson B, Subar A. Fruit, vegetables, and cancer prevention: a review of the epidemiological evidence. Nutr Cancer. 1992;18:1-29.

2. The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. The Alph-Tocopherol, Beta Carotene Cancer Prevention Study Group. N Engl J Med. 1994; 330:1029-35.

3. Blot WJ, Li JY, Taylor PR, et al. Nutrition intervention trials in Linxian, China. J Natl Cancer Inst. 1993;85:1483-92.