Soy research and expert insights
Guidelines for Healthy Soy Intake
One of the most commonly asked questions about soyfoods regards the amount of these foods that is considered a reasonable intake, particularly for children. There is also a desire to know the amount of soy required to derive potential health benefits with respect to reducing risk of chronic disease or alleviating hot flashes. While in most cases it isn’t easy to provide specific answers, there is a strong basis for providing intake guidelines.
The Food and Drug Administration (FDA) concluded in 1999 that 25 g/d of soy protein as part of a diet low in saturated fat may reduce risk of coronary heart disease.1 This figure may not be relevant to all consumers since it is aimed specifically at reducing cholesterol. Also, the FDA ruling doesn’t provide any guidance on an ideal intake of isoflavones – soybean components that have been intensely investigated for the potential health benefits.
Soyfoods are not part of the traditional Western diet so cultural norms provide little guidance. In fact, legumes in general play a limited role in North American diets, making daily consumption of any bean product unusual for many.2 The dietary guidelines recommend consumption of three cups (six servings) of beans per week; recent data suggest Americans come close to meeting this modest recommendation2 although this intake still represents only about 10% of total protein, assuming ½ cup provides ~8 g and total protein intake is about 80 g/d.3
The soy intake recommendations that follow are based on three considerations: Asian soy consumption, clinical and Asian epidemiologic studies that have examined the relationship between soy intake and a variety of health outcomes, and standard principles of dietary practice.
Asian Soy Intake
Estimates of Asian soy intake are often misrepresented in both the online and print media. There is no reason for confusion because intake data are widely found in the peer-reviewed literature. Since part of the initial enthusiasm about the potential health benefits of soy was based on disease rates in Japan, and more is known about soy intake in that country than any other, it’s a good place to start.
A recent review4 that included five studies involving older adults in Japan,5-9 found soy protein intake in women ranged from a low of 6.0 g/day8 to a high of 10.5 g/d,7 whereas the range in males was 8.0 g/day9 to 11.3 g/day.7 Soyfoods contributed from 6.5%8 to 12.8%7 of total protein intake. For comparison, one serving of a traditional soyfood provides anywhere from about seven (one cup soymilk) to as much as 20 g (3-4 ounces of some types of tofu) protein per serving.
These data and other more recently published concurring data come from epidemiologic investigations that involve hundreds to thousands of study participants who responded to food frequency questionnaires that comprehensively evaluated soy consumption.10,11 These studies also indicate that isoflavone intake ranges from about 30 to 50 mg/d.4,10,11 When isoflavone intake is not directly measured, it is easy to estimate because each gram of soy protein from traditional soyfoods is associated with approximately 3.5 mg isoflavones.4 (This ratio does not hold in more Westernized soyfoods because isoflavone loss can occur with processing).
Interestingly, according to food disappearance data from the Food and Agricultural Organization, per capita soy protein intake has remained constant during the past 40 years in Japan. However, as a percentage of total protein intake it has decreased from about 13 to 10%.4 This is because of the increased protein content (mostly from animal sources) of the Japanese diet. Since soy intake is decreasing among younger Japanese, absolute per capita intake may slowly begin to decline.
In comparison to Japan, soy intake of Hong Kong Chinese is about only half as much.12 Korean intake appears to be between that of Japan and Hong Kong.13
In mainland China, estimating intake is more difficult since the population is heterogeneous with regard to dietary behavior. There are, however, excellent data from Shanghai, where soy intake appears to be higher than in other parts of China. The Shanghai Men’s Health Study (SMHS) and the Shanghai Women’s Health Study (SHWS) are prospective epidemiologic studies each involving approximately 50,000 subjects.14-16 These studies indicate that daily mean soy protein and isoflavone intakes are similar to Japan14,15 or somewhat higher.16 For example, in the SMHS, for protein and isoflavones, the mean ± SEM were 12.5 and 7.94 g/d, respectively and 36.2 and 24.4, respectively.16 There are also excellent data on the upper range of soy intake in Shanghai. In a report from the SHWS, about 10% of women reportedly consumed about 20 g of soy protein and about 85 mg/d isoflavones whereas about 2% consumed ≥ 25 g/d soy protein (mean isoflavone intake in this group was 145 mg/d).14 In another report from Shanghai, among those women consuming a more plant- rather than meat-based diet, fourth quartile soy protein intake was 17 g/d.17 Finally, in a study involving almost 3,000 Shanghai men, the overall and fourth quartile soy protein intake means were 7.82 and 16.34 g/d, respectively.18
These upper intake data are consistent with that reported in some but not all Japanese studies. For example, in a case control study involving 1400 participants, Ahkter et al.11 found that the mean fourth quartile (n=174) isoflavone intake was 78.5 mg/d (estimated soy protein intake, ~20 g). The major soy products consumed in Shanghai are soymilk and tofu whereas in Japan, about 90% of soy is consumed via four foods. About half of Japanese intake is in the form of the fermented foods such as miso and natto and half from the unfermented products such as tofu and dried soybeans.
Clinical and Epidemiologic Studies Evaluating Health Outcomes
Clinical studies on the effects of soy consumption on a variety of health outcomes (including hot flashes, endothelial function, bone mineral density and blood pressure) have generally used soy protein and isoflavones doses between 20 and 50 g and 50 and 100 mg/d, respectively (see references for reviews).19-24 Very few human studies have used more than one exposure level, so estimating the threshold amounts theoretically needed for efficacy is difficult. Furthermore, the clinical data are inconsistent and some endpoints are likely affected by different soybean components – for example, isoflavones are the soybean components assumed to be responsible for the alleviation of hot flashes but hypotensive effects are generally attributed to the protein or the peptides formed upon digestion of the protein.
The protein amounts used in the clinical trials markedly exceed the amounts associated with reductions in risk of chronic disease observed in Asian epidemiologic studies. For example, in the SWHS, fracture risk was reduced by about one-third in the fifth intake quintile which had a mean daily soy protein of 18.5 g.25 In another prospective study from Singapore, fracture risk was also one-third lower but in this case the fourth quarter soy protein intake cutoff was only 7.6 g/d.26 In a case-control study conducted in Southern China, men and women in the third soyfood intake tertile, where the highest intake level was around three servings of soyfoods per week, had more than a 70% reduction in stroke risk.27 Finally, in Japan, a large 12-year prospective epidemiologic study found that among postmenopausal women a mean isoflavone intake of only 41 mg/d (range, 38-71 mg) was associated with more than a 60% reduction in the risk of cerebral and myocardial infarctions.28
Whereas the amount of soy protein used in clinical studies is much higher than usual Asian consumption, isoflavone levels tend to more closely reflect Asian intake patterns although even this varies quite a bit from study to study. The reason isoflavone intake more closely matches Asian intake than does protein intake is because the intervention studies have primarily used Westernized soy protein products which have a lower isoflavone concentration than traditional Asian soyfoods. Thus, to achieve a certain isoflavone intake, higher amounts of soy protein are needed.
Why the protective effects reported in epidemiologic studies occur at soy intake levels lower than the amounts needed to produce results in clinical studies isn’t clear. Perhaps it is because traditional soyfoods have different physiological effects than more processed ones, although there is little scientific support for this explanation. Alternatively, Asians may respond to soyfoods somewhat differently than non-Asians, although again, there is little support for this hypothesis. Of course, it could be that the Asian studies are really identifying a healthy-user effect rather than a direct effect of soyfoods. This also appears to be an unlikely explanation because most of the epidemiologic studies extensively controlled for confounding variables (although there are limits to the extent to which this is possible). Furthermore, among older Asians, the age group evaluated in most of the epidemiologic studies, soy consumption is much less of an indicator of lifestyle than it is in Western countries. Finally, it is possible that the epidemiologic studies show more robust findings because these studies are actually identifying the effects of lifelong intake whereas the intervention period in the clinical trials is relatively short and begins in adulthood.
Identifying Optimal Soy Intake: Consideration of Standard Dietary Practice
In theory, no single food should play too big a role in the diet. However, it is not uncommon for certain foods to be play an especially important role in meeting individual nutrient needs. For example, orange juice29 and dairy products30 provide approximately one-third and two-thirds of the total intake of vitamin C and calcium, respectively. Also, Americans who consume the recommended daily servings of dairy foods would consume about one-third of their protein from this single food group.
While all nutrients and many phytonutrients can be provided by a variety of foods, soyfoods are essentially the only sources of isoflavones. Thus, any recommendation for isoflavone intake is a recommendation to consume soyfoods. Currently there is no general formal recommendation for either a minimum or maximum level of isoflavone intake. If median Asian intake is used as a guide, than approximately 8 or 9 g of soy protein and about 30 to 35 mg/d could be recommended. However, as briefly discussed, for a variety of endpoints, a considerable number of epidemiologic studies show higher soy and isoflavone intakes are associated with lower disease risks. Thus, a persuasive argument can be made that using these higher values makes more sense than mean or median intakes. (Certainly the mean US intake of many nutrients is less than optimal.) Furthermore, clinical studies that have reported health benefits have used amounts that equal or exceed even these higher values.
Recently, Biesalski et al.31 defined the physiologic intake of a dietary compound as a daily consumption that is no more than two to three times the average intake of that compound in a respective population. If one accepts this definition, then an upper intake of about four servings of soyfoods per day for adults, or approximately 100 mg isoflavones and 25 g soy protein, the latter figure matching the amount set by the FDA for cholesterol reduction, still falls within this category. There is no clinical evidence to suggest exceeding this level is harmful, but there is also no historical precedent for doing so. For most adult Americans, consuming 25 g soy protein would mean that soyfoods would provide about 25%+ of total protein intake.3 For young children, whose overall caloric and protein intake is reduced (the latter by at least one-third) in comparison to adults, a downward adjustment in soy makes sense. Thus, as a general rule, for young people two servings of soyfoods per day seem reasonable.
Mark Messina, Ph.D. is the co-owner of Nutrition Matters, Inc., a nutrition consulting company, and is an adjunct professor at Loma Linda University. His research focuses on the health effects of soyfoods and soybean components. He is chairperson of The Soy Connection Editorial Board.
1) Food Labeling: Health Claims; Soy Protein and Coronary Heart Disease. Federal Register: (Volume 64, Number 206)]; 1999. p. 57699-733.
2) Mitchell DC, Lawrence FR, Hartman TJ, Curran JM. Consumption of dry beans, peas, and lentils could improve diet quality in the US population. J Am Diet Assoc. 2009 May;109:909-13.
3) Fulgoni VL, 3rd. Current protein intake in America: analysis of the National Health and Nutrition Examination Survey, 2003-2004. Am J Clin Nutr. 2008 May;87:1554S-7S.
4) Messina M, Nagata C, Wu AH. Estimated Asian adult soy protein and isoflavone intakes. Nutr Cancer. 2006;55:1-12.
5) Nagata C, Shimizu H, Takami R, Hayashi M, Takeda N, Yasuda K. Soy product intake is inversely associated with serum homocysteine level in premenopausal Japanese women. J Nutr. 2003 Mar;133:797-800.
6) Nagata C, Takatsuka N, Kawakami N, Shimizu H. Association of diet with the onset of menopause in Japanese women. Am J Epidemiol. 2000;152:863-7.
7) Nagata C, Takatsuka N, Kawakami N, Shimizu H. A prospective cohort study of soy product intake and stomach cancer death. Br J Cancer. 2002 Jul 1;87:31-6.
8) Takata Y, Maskarinec G, Franke A, Nagata C, Shimizu H. A comparison of dietary habits among women in Japan and Hawaii. Public Health Nutr. 2004 Apr;7:319-26.
9) Nagata C, Takatsuka N, Kurisu Y, Shimizu H. Decreased serum total cholesterol concentration is associated with high intake of soy products in Japanese men and women. J Nutr. 1998;128:209-13.
10) Nagata C, Nakamura K, Oba S, Hayashi M, Takeda N, Yasuda K. Association of intakes of fat, dietary fibre, soya isoflavones and alcohol with uterine fibroids in Japanese women. Br J Nutr. 2009 May;101:1427-31.
11) Akhter M, Inoue M, Kurahashi N, Iwasaki M, Sasazuki S, Tsugane S. Dietary soy and isoflavone intake and risk of colorectal cancer in the Japan public health center-based prospective study. Cancer Epidemiol Biomarkers Prev. 2008 Aug;17:2128-35.
12) Ho SC, Chan SG, Yip YB, Chan CS, Woo JL, Sham A. Change in bone mineral density and its determinants in pre- and perimenopausal Chinese women: the Hong Kong Perimenopausal Women Osteoporosis Study. Osteoporos Int. 2008 Dec;19:1785-96.
13) Kim MK, Kim JH, Nam SJ, Ryu S, Kong G. Dietary intake of soy protein and tofu in association with breast cancer risk based on a case-control study. Nutr Cancer. 2008;60:568-76.
14) Yang G, Shu XO, Jin F, Zhang X, Li HL, Li Q, Gao YT, Zheng W. Longitudinal study of soy food intake and blood pressure among middle-aged and elderly Chinese women. Am J Clin Nutr. 2005 May;81:1012-7.
15) Yang G, Shu XO, Li H, Chow WH, Cai H, Zhang X, Gao YT, Zheng W. Prospective cohort study of soy food intake and colorectal cancer risk in women. Am J Clin Nutr. 2009 Feb;89:577-83.
16) Lee SA, Wen W, Xiang YB, Barnes S, Liu D, Cai Q, Zheng W, Shu XO. Assessment of Dietary Isoflavone Intake among Middle-Aged Chinese Men. J Nutr. 2007 Apr;137:1011-6.
17) Cui X, Dai Q, Tseng M, Shu XO, Gao YT, Zheng W. Dietary patterns and breast cancer risk in the shanghai breast cancer study. Cancer Epidemiol Biomarkers Prev. 2007 Jul;16:1443-8.
18) Pan Y, Anthony M, Watson S, Clarkson TB. Soy phytoestrogens improve radial arm maze performance in ovariectomized retired breeder rats and do not attenuate benefits of 17beta-estradiol treatment. Menopause. 2000;7:230-5.
19) Ma DF, Qin LQ, Wang PY, Katoh R. Soy isoflavone intake increases bone mineral density in the spine of menopausal women: meta-analysis of randomized controlled trials. Clin Nutr. 2008 Feb;27:57-64.
20) Ma DF, Qin LQ, Wang PY, Katoh R. Soy isoflavone intake inhibits bone resorption and stimulates bone formation in menopausal women: meta-analysis of randomized controlled trials. Eur J Clin Nutr. 2008 Feb;62:155-61.
21) Williamson-Hughes PS, Flickinger BD, Messina MJ, Empie MW. Isoflavone supplements containing predominantly genistein reduce hot flash symptoms: a critical review of published studies. Menopause. 2006 Sep-Oct;13:831-9.
22) Messina M, Lane B. Soy protein, soybean isoflavones, and coronary heart disease risk: Where do we stand? Future Lipidology. 2007;2:55-74.
23) Rivas M, Garay RP, Escanero JF, Cia P, Jr., Cia P, Alda JO. Soy milk lowers blood pressure in men and women with mild to moderate essential hypertension. J Nutr. 2002 Jul;132:1900-2.
24) Hooper L, Kroon PA, Rimm EB, Cohn JS, Harvey I, Le Cornu KA, Ryder JJ, Hall WL, Cassidy A. Flavonoids, flavonoid-rich foods, and cardiovascular risk: a meta-analysis of randomized controlled trials. Am J Clin Nutr. 2008 Jul;88:38-50.
25) Zhang X, Shu XO, Li H, Yang G, Li Q, Gao YT, Zheng W. Prospective cohort study of soy food consumption and risk of bone fracture among postmenopausal women. Arch Intern Med. 2005 Sep 12;165:1890-5.
26) Koh WP, Wu AH, Wang R, Ang LW, Heng D, Yuan JM, Yu MC. Gender-specific associations between soy and risk of hip fracture in the Singapore Chinese Health Study. Am J Epidemiol. 2009 Oct 1;170:901-9.
27) Liang W, Lee AH, Binns CW, Huang R, Hu D, Shao H. Soy consumption reduces risk of ischemic stroke: a case-control study in southern china. Neuroepidemiology. 2009;33:111-6.
28) Kokubo Y, Iso H, Ishihara J, Okada K, Inoue M, Tsugane S. Association of dietary intake of soy, beans, and isoflavones with risk of cerebral and myocardial infarctions in Japanese populations: the Japan Public Health Center-based (JPHC) study cohort I. Circulation. 2007 Nov 27;116:2553-62.
29) Block G, Sorenson A. Vitamin C intake and dietary sources by demographic characteristics. Nutr Cancer. 1987;10:53-65.
30) Nicklas TA. Calcium intake trends and health consequences from childhood through adulthood. J Am Coll Nutr. 2003 Oct;22:340-56.
31) Biesalski HK, Dragsted LO, Elmadfa I, Grossklaus R, Muller M, Schrenk D, Walter P, Weber P. Bioactive compounds: safety and efficacy. Nutrition. 2009 Nov-Dec;25:1206-11.