The Third International Soy Symposium, held in Washington, DC November 1999, sparked a new round of promotional hype for soy products, particularly soy milk–now sold in supermarkets–and soy isoflavones–now sold as supplements for everything from cancer to hot flashes. But the symposium was not an unqualified success for its organizers.
Three papers presented at the conference sent a chill through the industry. Dr. Lon White of Hawaii presented evidence that consumption of two or more servings of tofu per week in midlife is associated with accelerated “brain aging” in late life, resulting in Alzheimers and dementia. Dr. Daniel Doerges’ research confirmed that soy causes thyroid problems by interfering with enzymes that produce thyroid hormones. And Dr. Claude Hughes reported that rats born to mothers that consumed genistein (an estrogen-like compound found in soy) had decreased birth weight and earlier onset of puberty compared to controls. A study just released indicates that vegetarian mothers are five times more likely to give birth to a boy with a birth defect of the penis than non-vegetarian mothers. Researchers blamed the problem on the estrogenic effect of soy consumed during pregnancy.
The evidence in both human and animal studies all points in the same direction–that soy consumed in more than small amounts contributes to problems with the nervous system, the thyroid gland and the digestive system. High levels of soy consumption can cause infertility and premature sexual development in girls and birth defects of the reproductive system in boys.
Yet in spite of all the new evidence on soy’s toxicity, the FDA approved a health claim for soy in October of 1999. The original petition, submitted by Protein Technology International, requested a health claim for isoflavones, the estrogen-like compounds found plentifully in soybeans. In 1998, the FDA made the unprecedented move of rewriting PTI’s petition, removing any reference to the phyto-
estrogens and substituting a claim for soy protein, a move that was in direct contradiction to the agency’s own regulations. The FDA is authorized to make rulings only on substances presented by petition.
The abrupt change in direction was no doubt due to the fact that a number of researchers, including scientists employed by the US government, submitted documents indicating that isoflavones are toxic. The FDA had also received, early in 1998, the final British government report on phytoestrogens, which failed to find much evidence of benefit and warned against potential adverse effects.
Even with the change from isoflavones to soy protein, FDA bureaucrats were forced to gloss over concerns about mineral blocking effects, enzyme inhibitors, goitrogenicity, endocrine disruption, reproductive problems and increased allergic reactions from consumption of soy products. One of the strongest letters of protest came from Dr. Dan Sheehan and Dr. Daniel Doerge, government researchers at the National Center for Toxicological Research. Their pleas for warning labels were dismissed as unwarranted.
Products that are low in fat and cholesterol, and that contain at least 6.25 grams of soy protein per 100-gram serving, may now carry a label stating that the product will help prevent heart disease. This claim is based on studies showing that soy lowers cholesterol. In approving the claim, the FDA relied largely on a 1995 meta-analysis by Dr. James Anderson, sponsored by Soy Protein International and published in the New England Journal of Medicine. A meta-analysis is a review and summary of the results of many clinical studies on the same subject. Use of meta-analyses to draw general conclusions has come under sharp criticism by members of the scientific community. “Researchers substituting meta-analysis for more rigorous trials risk making faulty assumptions and indulging in creative accounting,” says Sir John Scott, President of the Royal Society of New Zealand. “Like is not being lumped with like. Little lumps and big lumps of data are being gathered together by various groups.”
There is the added temptation for researchers, particularly researchers funded by a company like Soy Protein International, to leave out studies that would prevent the desired conclusions. Dr. Anderson discarded eight studies for various reasons, leaving a remainder of 29. The published report suggested that individuals with cholesterol levels over 250 mg/dl would experience a “significant” reduction of seven to twenty percent in levels of serum cholesterol if they substituted soy protein for animal protein. Cholesterol reduction was insignificant for individuals whose cholesterol was lower than 250 mg/dl. In other words, for most of us, giving up steak and eating vegeburgers instead will not bring blood cholesterol levels down. The health claim that the FDA approved fails to inform the consumer about these important details.
Research that ties soy to positive effects on cholesterol levels is “incredibly immature,” said Ronald M. Krauss, MD, head of the Molecular Medical Research Program and Lawrence Berkeley National Lab. He might have added that studies in which cholesterol levels were lowered either through diet or drugs have consistently resulted in a greater number of deaths in the treatment groups than in controls, deaths from stroke, cancer, intestinal disorders, accidents and suicide. Cholesterol lowering measures in the US have fueled a sixty-billion-dollar-a-year cholesterol-lowering industry but have not saved us from the ravages of heart disease. And it’s hard to see how a substance that causes thryoid problems can protect us against heart disease when low thryoid function makes us prone to heart disease.
Meanwhile, soy products are being promoted with increasing vigor to all age groups. The USDA has just said yes to unlimited use of soy in school lunches and soy infant formula now takes up 25% of the infant formula market. How soon will consumers get wise and “just say no” to soy?
This article appeared in Wise Traditions in Food, Farming and the Healing Arts, the quarterly magazine of the Weston A. Price Foundation, Spring 2000.