New Report on Soy Finds Limited Evidence for Health Outcomes

Press Release Date: August 24, 2005

Daily consumption of soy protein found in tofu and other soybean products may result in a small reduction in low-density lipoprotein (LDL, known as bad cholesterol) and triglyceride levels, according to a new evidence review supported by HHS' Agency for Healthcare Research and Quality. In addition, isoflavones found in soy may reduce the frequency of hot flashes in post-menopausal women. However, the available studies on the health impacts of soy were limited in number, of poor quality, or their duration was too short to lead to definite conclusions.

Overall, across the 68 studies that examined the impact of soy on cholesterol levels, consumption of soy products resulted in a 5 mg/dL (about 3 percent) reduction in LDL and an 8 mg/dL (about 6 percent) decrease in triglyceride levels in the populations studied. Among these studies, a large variety of soy products, doses of soy protein, and doses of soy isoflavones were tested. The average dose of soy protein in the studies was equivalent to about one pound of tofu or three soy shakes daily.

There was some indication that soy consumption may be more effective at lowering LDL among people with higher LDL levels. Also, larger amounts of soy protein, but not soy isoflavones, are more effective in people with abnormally elevated LDL levels. Similarly, soy consumption may be more effective at lowering triglycerides among people with higher triglyceride levels; however, there was no evidence of how much soy protein or isoflavones would be needed to affect triglycerides.

Reviews on the relationship between soy consumption and high-density lipoprotein (HDL, known as good cholesterol) levels and between soy consumption and blood pressure did not find significant effects. Among 21 studies evaluating the consumption of soy isoflavones for menopause-related symptoms, there was a net reduction in hot flash frequency ranging from 7 percent to 40 percent, however, these trials were mostly rated as poor quality. Among studies with statistically significant improvements in symptoms, the dose of soy isoflavones ranged from 17.5 to 100 mg/day.

The evidence review completed by AHRQ's Tufts-New England Medical Center Evidence-based Practice Center also found insufficient data among the 200 human studies examined as part of this analysis to suggest that soy had an effect on bone health, cancer, kidney disease, endocrine function, reproductive health, neurocognitive function, or glucose metabolism. A wide variety of soy products were studied, including foods such as soybeans, soy flour, soy milk, tofu, miso, tempeh, natto, and okara; isolated and textured soy protein that is added to foods; and soy-derived isoflavone supplements. Aside from minor gastrointestinal problems reported in some short-term studies, consumption of soy products by study participants was not associated with adverse events. However, long-term safety data are lacking.

"This report shows us that there is a lot we don't know about soy, and that more research is needed to determine if soy has any impact on a number of health conditions," said AHRQ Director Carolyn M. Clancy, M.D. "An important role for AHRQ's Evidence-based Practice Centers is to identify gaps in evidence and knowledge that can be used to develop future research agendas."

The researchers who conducted the evidence review, which was supported by the National Institute of Health's National Center for Complementary and Alternative Medicine and Office of Dietary Supplements, considered the type of soy product used, amount consumed, frequency of consumption, and safety issues in their review of health effects.

"The AHRQ report provides valuable information about the studies that have been done on soy for a variety of health conditions," said Stephen E. Straus, M.D., NCCAM Director. "It also highlights research needs in this area and will help inform NIH's research agenda."

The AHRQ report notes that future studies of the health effects of soy need to better address the complex relationship between health and food components, including how variations in the diets, lifestyles, and health of participants might affect the results. Such studies should also be of better quality, include larger numbers of participants, and be of longer duration, the researchers said. In particular, studies that substitute practical amounts of soy products into people's diets would better address the question of whether people should make the effort to include more soy in their diet.

The evidence review was prepared by a team of researchers led by Ethan Balk, M.D., and Joseph Lau, M.D., of AHRQ's Tufts-New England Medical Center EPC in Boston. The Effects of Soy on Health Outcomes can be found online at http://www.ahrq.gov/clinic/tp/soytp.htm. Copies of the summary and full report are available free of charge by calling the AHRQ Publications Clearinghouse at (800) 358-9295 or by sending an E-mail to AHRQPubs@ahrq.hhs.gov.

For more information, please contact AHRQ Public Affairs: (301) 427-1241 or (301) 427-1855.

---

Internet Citation:

New Report on Soy Finds Limited Evidence for Health Outcomes. Press Release, August 24, 2005. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/news/press/pr2005/soypr.htm

---