The results of a study recently published in the Journal of the American Medical Association (JAMA) has the health and science columns in the print media and on the Internet buzzing. Called the Physicians’ Health Study II, it followed 14,641 male doctors over a period of 14 years, and tracked their incidences of developing cancer. In the double-blind study, the doctors were randomly assigned to either take a daily multivitamin or its placebo equivalent. At the end of the study, researchers found that the group who took the multivitamins had an 8% lower occurrence of cancer than those who took the placebo.
Because several different types of cancer were tracked, the researchers found that multivitamin use didn’t seem to have any effect on incidences of prostate cancer, but it did seem to have an effect on incidences of other cancers. Among those study subjects who developed cancer, multivitamin use was also associated with a 12% lower risk of dying from the disease.
A “significant but modest reduction of risk”
Study authors were hopeful but cautious in reporting their findings, emphasizing that the risk reductions they found were “modest,” but still statistically significant. Still, if the research can be replicated and validated, it would offer great hope in the fight against cancer. Balz Frei of the Linus Pauling Institute (who was not involved in the research but commenting on it) said that if the 8% reduction in risk turned out to be verifiable, “Given that more than 1.6 million new cancer cases are diagnosed in the U.S. each year, this translates into about 130,000 cancers prevented every year, and with it all the health care costs and human suffering.”
The authors were also careful to point out that the study group consisted of all men, all over 50, and thus that their findings represented a potential cancer risk reduction benefit only to that demographic. But the study is still important, because it was conducted over such a long period of time, and in a controlled clinical situation that would be difficult to achieve with a selection of subjects taken from the general population.
The other side of the coin
Critics who warn against reading too much into this study point out a number of flaws with it. First, and most important, this study should be viewed in the context of many other equally long-term, equally large studies that have found the opposite – that vitamin and mineral supplements don’t protect against cancer, and have either little or no effect. One study reported in the Archives of Internal Medicine that was conducted on 38,000 women found that supplement use actually resulted in a 2.6% increase in mortality risk. A similar 10-year study of 35,000 women in Sweden found that breast cancer risk was 19% higher in those who took supplements than in those who didn’t.
Second, the study group is a classic instance of selection bias, in that all of the subjects were doctors, who might be more likely than other populations to follow generally healthy lifestyles in terms of diet and exercise. The doctors were healthy to start with, only 4% were smokers, two-thirds of them exercised regularly, and all of them were male. Thus the findings in this study may have conceivably been as much a result of lifestyle choice or gender as the results of the multivitamins.
Third, almost 9% of the men in the study had “a history of cancer at the time of randomization,” and because the study was randomized and double-blind (neither subjects nor researchers knowing who received a real multivitamin and who received a placebo), there is no way of knowing whether those with a previous history of cancer were placed in the vitamin group or the placebo group. The random selection may have placed an inordinate number of the cancer-history subjects into either group, and no one would know.
Fourth, study subjects were prohibited from taking additional vitamin supplements “containing more than 100% of the recommended dietary allowance of vitamin E, vitamin C, beta carotene, or vitamin A,” but were not prohibited from taking additional supplements, period. Also, the study authors admit that the daily use of aspirin by study subjects was very high (77.4%), due in part to being recruited from a previous study on aspirin usage (Physicians’ Health Study I). So we cannot be certain that the results were due entirely to the single multivitamin pill taken by subjects in the vitamin group.
I am pointing out these criticisms because I’ve found that many of the popular articles I’ve found in the media about this study have not. It’s an interesting study, but even the study authors express doubts in their JAMA article about whether their findings can ever be replicated, due to the “high prevalence of multivitamin use in adults and difficulty of conducting a long-term, blinded clinical trial.” So in the end the study remains what it is – a somewhat anomalous study that may be difficult to ever replicate. We all hope for definitive research that points to something that will lower our risk of contracting cancer, but this study may not constitute that definitive research.