This is an archive page. The links are no longer being updated.
Date: February 7, 1996 For Release: 6:00 EST Contact: NIH (301) 402-2900
Investigators participating in seven previously published cohort (or longitudinal) studies of dietary fat and breast cancer present a joint analysis of their studies in the February 8, 1996, issue of the New England Journal of Medicine. No evidence of a link was found between fat intake and breast cancer. In contrast, numerous other studies including comparisons between fat intake and breast cancer in different countries, case-control studies, and animal feeding studies support such a link. In cohort studies the dietary fat is estimated before the development of cancer, while in case- control studies it is estimated afterwards. Because cancer can change the recall of dietary habits, scientists generally regard cohort studies as more reliable. However, neither of these types of studies is as reliable as a clinical trial. In a clinical trial the dietary fat intake of one group but not the other is changed, and then the rate of breast cancer is followed in both groups. The cohort study data in the reanalysis were essentially all available when the National Institutes of Health (NIH) launched the Women s Health Initiative (WHI) in 1992. Then, as now, the widely different estimates of reductions in breast cancer risk that may follow a low fat eating pattern ranged from 50% or greater from international comparisons to little or no reduction according to the cohort studies. This uncertainty underscores the need for reliable information on the overall health benefits of a low fat eating pattern. The WHI trial is designed to answer confidently this and other important questions related to women s health. More than 16,000 women in the age range of 50-79 have already enrolled in the dietary component, and enrollment will continue until early in 1998. A total of 48,000 women will enroll in the dietary trial at 40 clinical centers nationwide, and they will be followed for an average of 9 years. In addition, women will be able to enroll in other parts of WHI, including a study of the ability of hormone replacement therapy to prevent heart disease, bone fractures, and loss of memory, and of a combination of calcium and vitamin D to prevent fractures and colorectal cancer. For more information about participating in WHI call 1-800-54-WOMEN.
Breast cancer is the second most common cancer in women. Therefore, the definitive randomized clinical trial of dietary modification being conducted as part of WHI will answer a question of considerable public health importance: does lowering dietary fat in the middle decades of life lead to a reduced breast cancer risk? Furthermore, will there also be reductions in colorectal cancer and heart disease, and what are the effects on overall health? The dietary pattern is low in fat (20% of calories) and high in fruits, vegetables, cereals and grains. This is an eating pattern widely believed to be healthful, but it has never been tested in a definitive clinical trial. The cohort study analysis uses recall of recent food intakes to calculate an estimate of total fat that is then related to subsequent breast cancer. The validity of such studies, individually or collectively, depend heavily on accurate dietary measurements and on the ability to discern the effects of fat intake from that of a myriad other components of the diet, and from other participant characteristics or lifestyles that may relate to cancer risk. Since there is no available gold standard measure of diet the reliability and interpretation of such studies remains quite unclear. Moreover, since these studies do not attempt to alter eating patterns, they cannot directly assess the health benefits that may follow adoption of a low fat eating pattern.
In contrast the formal dietary trial included in the WHI will compare the health outcomes of women randomly assigned by computer to continue their usual eating pattern (the control group) to that of women who are taught how to make and maintain a major change to their eating pattern (the intervention group). The randomization eliminates the possibility that other personal characteristics or lifestyles may explain differences in cancer risk. Also, the comparison of disease rates between control and intervention groups does not rely on accurate measurement of individual dietary habits. Dietary measurements for both groups of women are obtained only to document that sufficient change in diet has occurred. By 2005 or shortly thereafter women will have the answer: is it worth their while to make a major change in dietary pattern?