Response to the publication “Multivitamin Use and Risk of Prostate Cancer in the National Institutes of Health–AARP Diet and Health Study” by Lawson K.A. et al.

May 2, 2007


Prof. Dr. med. Hans K Biesalski, Institute for Biological Chemistry and Nutritional Science, University of Hohenheim, Stuttgart, Germany

In a study (1), published in the Journal of the National Cancer Institute (JNCI) on ‘Multivitamin Use and Risk of Prostate Cancer in the National Institutes of Health–AARP Diet and Health Study’ researchers found "an increased risk of advanced and fatal prostate cancers among men reporting excessive use of multivitamins (more than seven times per week) when compared with never users."

However, there are limitations in the research design that preclude any firm conclusions about the observed association. To their credit, the study authors acknowledge most of these limitations. Unfortunately, this is buried in the “Discussion” section and has been largely ignored by the media reporting the study to the public. Overall, the researchers found no link between multivitamin use and prostate cancer risk. Only when they grouped (stratified) the data by comparing “excessive use” (defined as greater than 7 times per week) vs. “no use” with the outcome being advanced prostate cancer they did observe a significant positive association. The definition of “excessive use” must be questioned, since the survey the researchers used asked if subjects used multivitamins from “never” up to “every day”. Thus, it is unclear how the researchers derived a value for “greater than 7 times per week”.

The observed associations for some individual nutrients must be questioned as well. These findings were based on subjects’ recall of supplement use over the past 12 months. No information was collected or analyzed vis-à-vis nutrient dose or duration of use; the absence of time course or dose-ranging data weakens any notion of causality. Most importantly, those at high risk for advanced prostate cancer (i.e. early diagnosis, family history, etc.) are more likely to exhibit certain self-selected behaviors in an attempt to intervene in their situation, including increasing their use of alternative therapies (such as taking more supplements), and proactively seeking care (such as more frequent screening). Likewise, supplement users tend to exhibit aspects of a healthy lifestyle, such as more frequent visits to their physician, which can lead to more screening and early detection. The bottom line is that those diagnosed with advanced prostate cancer are more likely to use supplements, and that increased use is the likely source for the observed association.

There is no evidence in this report or elsewhere in the scientific literature that multivitamin use causes prostate cancer. Indeed, the authors acknowledge these as a major confounder, and when they controlled for this factor in their analysis, the association disappeared. This study adds little to the evidence base for the use of multivitamins by the general population to maintain health. It does suggest however, that those individuals who are at high risk for or have been diagnosed with prostate cancer should be in close contact with their healthcare professional to determine what the best course of action is for them.

May 2007


  1. Lawson KA et al. Multivitamin Use and Risk of Prostate Cancer in the National Institutes of Health–AARP Diet and Health Study. J. Natl. Cancer Inst. 2007; 99(10):754–764.