Over the past several decades, many infants have received soy-based formulas. According to the American Academy of Pediatrics, “the use of soy-based formula has nearly doubled during the past decade to achieve 25 percent of the market in the United States.”1 Many mothers say that they have fed soy to their babies on the basis of their pediatrician’s recommendation. “High utilization rates,” which could not be accounted for due to the vegan lifestyle or galactosemia, were “presumably because of . . . perceived or real allergy to cow’s milk protein (CMP) and perhaps due to lack of lactose-free CMP formulas, of which there are now plenty.”2
The American Academy of Pediatrics’ (AAP) policy on feeding soy formula (2008) states that that there are “few indications for the use of soy protein-based infant formula in place of cow milk-based formula.” According to the AAP, the only real indications for soy formula use are for infants with congenital galactosemia, for use by families who are strict vegans, or infants who are truly lactose-intolerant.3
Problems with Soy Formula
The main concern regarding use of soybased infant formula is the amount of phytoestrogens, “plant-derived substances with estrogenic activity,” which are present in “relatively large amounts in the formula,” and may “mimic the actions of estradiol or alter estradiol metabolism, and modify the processes influenced by estradiol.” Infants who consume soy formula have isoflavone plasma concentrations approximately 13,000 to 22,000 times greater than those infants fed breast milk or cow’s milk formula.2 Kaayla Daniel, PhD, author of The Whole Soy Story: The Dark Side of America’s Favorite Health Food, describes calculations indicating that these babies are consuming the estrogenic equivalent of three to five birth control pills per day.5
A new 2011 paper on infant feeding, “Developmental Status of 1-Year-Old Infants Fed Breast Milk, Cow’s Milk Formula, or Soy Formula,” is another attempt to equate formula from soy beans with human milk and milk-based formula. It contains data from the on-going study on child development, “Beginnings,” originating at the Arkansas Children’s Nutrition Center (ACNC) in Little Rock, Arkansas, one of six Human Research Centers in the U.S. The study is funded by public monies in the form of two USDA Agricultural Research Service grants from the United States Department of Agriculture (USDA), Agricultural Research Service (ARS).6-8
The Badger study was highly anticipated, reported in the major media and lauded at soy symposia, even as far back as 2005. Often a speaker at these symposia, Badger was always presented as the great white knight who was in the process of clearing up all the confusion about soy formula, hopefully proving soy infant formula to be safe and good. Even supporters of soy spoke quite candidly at the symposia about the wretched quality of the one earlier study on soy formula, by Strom and others, and the need for well-designed quality work.
However, Thomas M. Badger, director of the ACNC Center in Little Rock, Arkansas, and professor in the College of Medicine at the University of Arkansas, wears several hats. In addition to his position at the USDA, he is “on staff” at the Department of Pediatrics at the University of Arkansas. Much like Mark Messina, another soy promoter, whose name appears in many articles on soy, Badger is not a physician.
Dr. Badger reports that he is a member of the Science Advisory Board of the Soy Nutrition Institute, a corporation with assets that seeks exemption from federal taxes according to the “Articles of Incorporation.” Its mission is to conduct programs to assist the United States soybean industry and the United Soybean Board in collecting and disseminating soy-related nutrition information based on responsible scientific research.”6-10
In “The Health Consequences of Early Soy Consumption (2002),” a paper presented at the Fourth International Symposium on the Role of Soy in Preventing and Treating Chronic Disease, funded by Monsanto, Central Soya Company, Protein Technologies International, United Soybean Board, Archer Daniels Midland, Cargill, and others, Dr. Badger explains his research goals: “Recent data suggest that there are no long-term adverse effects of early exposure to soy formula through young adulthood. It is as yet unknown whether soy formula consumption by infants will result in health problems or benefits upon aging, but multigenerational animal studies with diets made with SPI have not revealed any problems.”11
The name of Thomas M. Badger, PhD, as an author of a scientific article about infant formula should be an immediate red light to any reader or reporter. Although he is an employee of the USDA Agriculture Center in Little Rock, Arkansas, a public institution, a review of his papers published in academic journals shows that Dr. Badger is focused on the agenda of promoting the use of soy infant formula through journal articles,6,11-17 many via the “Beginnings” Study as a source of data and funding, and as quoted in numerous web articles, such as “Infant ‘Smarts’ Similar With Different Types of Formula,” by Carina Storris for Healthwoman.org.18
According to Storris, the “study is very important because it shows that the growth and development of children in the U.S. who are fed soy formula is the same as children who are fed milk formulas…” She also quoted Thomas Badger: “Researchers did find a difference in brain development between breast-fed babies and those of cow’s milk or soy formula, but is was so small that it will probably not affect long-term ability. I don’t think parents should be worried at all if their kids are on formula.”
Rarely mentioned is the fact that currently more than 90 percent of soy grown in the USA is genetically modified.19 Soy formula contains soy protein isolates, produced by processing the soy bean (likely genetically modified) using chemicals like hexane.20 According to the Occupational Safety and Health Administration, hexane is a neurotoxin, causes embryotoxic effects and is cytotoxic in mammals.21 The FDA considers GE soy (and other food-like GE products) “substantially equivalent” to conventionally produced products.22
Even though a few infant formulas are “organic” they still may contain laboratory produced GMO oils, which resemble but are not the same as the DHA and ARA needed by the infant brain for growth and development (see Sidebar, opposite).23
In addition to isoflavones, soy beans contain many “anti-nutrient substances” which interfere with digestion, reproduction and thyroid function. Over two hundred fifty citations for articles describing these anti-nutrient functions are listed in the FDA’s Poison Plant Data Base.24
The “Beginnings ” Study
In the “Beginnings” study, the plan is to follow human infants for six years. But according to the National Institute of Environmental Health Sciences (NIEHS), at the National Institutes of Health (NIH) website, “the health effects resulting from a soy-based diet may not be apparent until years later.”25 Animal studies indicate that health effects of possible concern include early onset of puberty in females and alterations in the basic development of breast tissue. The phytoestrogens in soy are listed in the NIEHS website as “endocrine disruptors (ED)” where ED is defined as “naturally occurring compounds or man-made substances that may mimic or interfere with the function of hormones in the body. They may turn on, shut off, or modify signals that hormones carry, which may affect the normal functions of tissues and organs. Many of these substances have been linked with developmental, reproductive, neural, immune and other problems in wildlife and laboratory animals.”26
Dr. George Kent, author of Regulating Infant Formula, wonders about cohorts after the six-year termination point. “What happens to them after six years when health concerns with a soy diet become more apparent? Assessing the impact of diet on development at one year of age, or even younger makes little sense. Development refers to changes over time.”27
A Look at the Study
In this 2011 paper, infants fed breast milk (BM), cow’s milk formula (MF), and soy-based formula (SF) are compared at three, six, nine and twelve months to assess developmental status in mental, motor, and language areas. Milk formulas included Similac Advance or Enfamil Lipil; and soy formulas, Similac Soy Isomil or Enfamil Prosobee formulas. All formulas were supplemented with synthetic docosahexaenoic acid (DHASCO) and synthetic arachidonic acid (ARASCO) made from algae and fungi.6
Infant developmental status was assessed by the Bayley Scale of Infant Development (BSID) (second edition) using two “derived” sections, the Mental Development Index (MDI) and Psychomotor Development Index (PDI). A separate instrument, the Preschool Language Scale-3 (PLS-3), supplied the “expressive communication and auditory comprehension subscales.”6
Problems with Measurement
Tests such as the BSID-2 are usually carefully scrutinized for reliability and validity by pre-testing with various groups, as a whole test, or sections of tests in relation to the whole. In this case two “derived” sections from the BSID, the MDI and PDI were used with the babies. Were these derivations pilot-tested and can they actually be used independently of the main instrument? What other sections were not used and why not? How was the scoring for these “derived” sections determined?
Unfortunately, how the measures were scored is not made clear in the study. According to the text, “Results are presented as standard scores with a mean of 100 (SD = 15).” Is t he mean for each test 100 (in points)? The MDI, PDI, and PLS-3 (expressive communication and auditory comprehension subscales) were assessed by group with results shown in Tables 3-5 in the paper. 28 The scores given are “estimated means” so there is no way to determine the range of actual scores for each cohort at each time interval.
We can glean that the scored ranged from eighty-five to one hundred fifteen because the authors explain that “delay” is a score below eighty-five and “accelerated” is a score greater than one hundred fifteen, and that there were several of these scores in the sample, but we don’t know in which groups. We don’t know how the tests were combined for scoring, or how “delay,” below a score of eighty-five, was determined, or how a score above one hundred fifteen for “accelerated” performance was determined.29
The BSID is basically designed to identify “developmental delay” (not developmental status as indicated in the title of the paper) and is only “modestly related to school-age cognitive development―the outcome that is most meaningful to investigators in this field,” says John Columbo, PhD, and Susan Carlon, PhD, from the Department of Psychology, at the University of Kansas. A recent study found that the BSID is not predictive of IQ of term infants at six years.30
Further, “the BSID is not an adequate indicant of specific cognitive skills that may be affected by interventions or exposures, nutrition or otherwise, and so its use to evaluate the construct of infant cognition is seriously deficient in the context of recent advances in developmental science.”30
How did the babies fare developmentally in the study based on their feeding method? According to the conclusions in the study abstract: “This unique study showed that all scores on developmental testing were within establish normal ranges and that MF and SF did not differ significantly. Further this study demonstrated a slight advantage of BF infants on cognitive development compared with formula-fed infants.”31
But exactly how slight was the advantage of BF infants over those who were SF and MF? The means for BF babies are higher for each cohort, (three, six, nine and twelve months) for each time period, in all categories of the test measures, MDI, PDI and PLS-3. Specifically BF infants scored significantly higher than formula-fed infants (MF and SF) on the Mental Developmental Index (MDI) at six, nine and twelve months; significantly higher than formula-fed infants on the PDI at three and six months; and significantly higher than formula fed (MF and SF) on the PLS- 3 at three months and six months. Overall the BF infants had a cognitive developmental advantage compared with the formula-fed infants.32
We must also consider the fact that for the infants in the category for BF, 53 percent of the infants were breastfed for 12 months, 10 percent of mothers started mixed feeding after six months, and the remaining forty-eight babies were breast fed until six months of age, and then fed milk-based formulas.32 These infants were not considered in separate categories but all “clumped” together for analysis.
If they all were exclusively breastfed their “mean” scores could potentially have been higher. In a study by Slykerman and others, longer periods of breastfeeding were associated with higher intelligence scores.33
In addition, in the “Beginnings” study, complementary foods could be introduced after age four months for all three diet groups (BF, MF, SF). There is no explanation of the amounts, types and kinds of foods that were or were not supplemented or the possible effect of foods on the scores.
Soy-fed infants scored lowest on the MDI and the PDI for almost all categories. Scores at twelve months on the PLS-3 were similar among the groups. SF infants had significantly lower PDI scores compared with BF infants at six months.34
But Badger’s main message is that although the BF infants showed advantages in all areas of testing, and SF scored lower or lowest in all measures except PLS-3 scores, all three groups were “within established normal range”36 and, as he told Carina Storris, the development is about the same with the two formulas. The breast fed infants have only a very small advantage.18,29
A DIFFERENT CONCLUSION
Regarding this conclusion, Dr. George Kent, author of Regulating Infant Formula, focuses on the “alarming inadequacy of studies relating to the impacts of soy-based infant formula on children’s health and development. . . This recent study’s findings should not be taken as a reason for complacency but as a cause for alarm. A slight difference might signal the beginning of a steadily widening gap. Overall, there seems to be a skew in that study in favor of soy-based formula, at the expense of the children and the adults that they will become.”35
Dr. Kent has worked as a consultant with the Food and Agriculture Organization of the United Nations, the United Nations Children’s Fund, and several civil society organizations. He is part of the Working Group on Nutrition, Ethics, and Human Rights of the United Nations System Standing Committee on Nutrition.36
As far as design and statistical methods used in the study, this paper has several strikes against it. Firstly, “Beginnings” was not a randomized, controlled trial, which is the gold standard for testing an intervention. It originates from a “nonprobability” (not random) grouping. The feeding method was predetermined so the researchers could not “randomly” assign a human baby to one of the three feeding groups. Instead, they assigned the baby to the feeding group determined by parental choice: BF, MF, or SF. Martin Stein, MD, of Journal Watch Pediatrics & Adolescent Medicine says of the study, “These findings should be interpreted with caution because the study was not randomized.”37
Secondly, it is an observational study. As J. M. Utts and R. Heckard write in their textbook, Mind on Statistics, “the most common mistake made in reporting research studies is to imply that a cause and effect relationship can be concluded from an observational study. With an observational study, it is difficult, perhaps impossible, to separate the effects of confounding variables from the effects of the main explanatory variables of interest.”38
And lastly, the study has inherent bias because of the method of obtaining subjects. The moms with their babies were recruited by staff from the local community and thus they do not represent all babies living in the U.S., just babies living in that area. In fact, there are few African American, Asian, or babies of other races in the study. Caucasian babies predominate (131 BF, 131 MF, and 128 SF babies) with only 4 percent African American (5 BF, 8 MF and 13 SF) and 5 percent “other” race babies (6 BF, 11 MF, and 4 SF) in the group, certainly not representative of the numbers of these ethnic groups in the actual U.S. population.9 This is called “a sample of convenience.”
“If the sample does not represent a larger population for the question of interest, and randomization to treatments was not used, no inferences can be drawn.”40 What this means is that you can’t apply the results of this study to all babies in general, just the Caucasian babies that were volunteered for the study.
A serious flaw is the fact that we don’t get a real look at the groups, BF, MF and SF, because gestational age, parity, birth weight, mother’s age and IQ data are presented as means, standard deviations and percentages represented as “Cohort Characteristics.”41 We don’t actually see the range of scores for the MDI (Table 3), PDI (Table 4), and PLS-3 (Table 5), but are given “estimated” computer generated scores using “fitted mixed models,” adjusted for variables such as socio-economic status, mother’s age, mother’s IQ, gestational age, child’s race, child’s gender, child’s age and so on.34
The mean is the average of all scores for babies in the group. For socioeconomic status (SES) for example, scores ranged from eight to sixty-six, so the mean of 49.8 (BF), 45.6 (MF), and 45.9 (SF) presented in Table 1 of the study tells us little. Perhaps use of a “median,” the middle of the range of scores for babies in that group, or the socioeconomic levels broken down into categories with scores ranking babies in each category, controlling for specific variable (quintiles), would have been much more useful tool. But the authors tell us that “the quintile analysis showed no significant effect of diet on PLS-3 and no significant difference between SF and MF infants on any of the three developmental measures.” Interesting, but no data or quintile tables are shown.34
To better look at the distribution of the cohorts, a scatter plot showing data points for each baby in the cohort on the chart would be necessary, to see what the distribution actually looks like for the individual cohorts and for comparing cohorts, and the general grouping. Where the points are falling (located) on the chart will describe the distribution of the data. A scatter gram plots points for one, two, or more groupings of interest on the same chart or singular charts to show visually how the data relate (correlate), be it positive, negative, or nonexistent.43 But that is not given in the study. Instead, Figure 1 (a composite of Tables 3-5) merely shows standardized scores of BF, MF and SF during the first year of life.
THE BELL CURVE
A true parametric distribution is random and usually fits a bell curve, with most data falling within two standard deviations (SD) from the mean. The mean is the center point (average) and the standard deviations that are calculated using the data, fall on each side of the mean.44
A most important piece of the puzzle, data on how the three groups fared “under the curve” compared to one another, is not provided. Area under the curve” refers to the symmetrical normal “bell-shaped curve” associated with random distributions.44 The “areas under the curve were similar for all infants, BF, MF and SF, on all three behavioral measures,” say the authors, but that data is not shown in this paper.45 “If this statement is true it means that the three feeding groups don’t vary too much from one another on developmental scores. But we don’t know much about the distribution unless we see the bell curve.
Most data that are similar fall under a bell curve (within three standard deviations). The bell has normal tails (the left and ride sides of the curve are identical in shape). . . . Whereas in skewed data (asymmetrical), the curve “leans” which does not represent the random population, or left or right sides of the tails can be elongated indicating extreme scores, or shortened, in respect to the distribution of the scores.44,46
Due to the lack of data, analysis and problems with study design and subject selection, the findings of the “Beginnings” study cannot be applied to infants who are fed with breast milk, milk-based formula or soy formulas. We agree with the authors that this study is “unique,” unique because so much data is missing. We request that all the information be supplied, not just cherry-picked or extracted, so that careful conclusions can be made regarding the actual role of soy formula in development. We also recommend that “Beginnings,” in order to become a true “longitudinal” study of some value, extend their “longitudinal” analysis of their subjects not to six years as is currently planned, but to twenty-five years of age or older, in order to actually capture the effects of soy formula on the subjects during and after puberty and into early adulthood. Until there are some coherent results which track the safe use of soy formula from infancy until adulthood, the use of soy formula is not recommended.
The author would like to thank Summer Blackwell, Administrative Assistant at Hale Publishing, Amarillo, Texas, for providing a complimentary copy of Regulating Infant Formula by George Kent; and Sandy and her colleagues at the Free Library of Philadelphia for research assistance (they provide me with free journal articles through the mail―a service to library patrons in PA).
JUST ANOTHER REASON TO BREASTFEED .
POTENTIALLY HARMFUL DHA AND ARA ARTIFICIAL OILS IN BABY FORMULA
The infant formulas fed to the babies in the “Beginnings” study are Similac Advance, and Enfamil Lipil, or Similac Soy Isomil or Enfamil Prosobee. All these formulas contain added artificially-derived docosahexaneoic acid single cell oil (DHASCO) and arachidonic acid single cell oil (ARASCO) which attempt to model the DHA and ARA (arachidonic acid) in human breast milk. They are advertised as “closest to human milk.”
DHA (docosahexaneoic acic), an omega-3 fatty acid, and ARA (arachadonic acid), an omega-6 fatty acid, naturally found in human breast milk, are extremely important components of the human brain and eye development and function. However, these products in formula are not natural; they are made in the U.S. by Martek Biosciences Corporation and Mead Johnson Nutritional, from hexane- extracted laboratory grown-algae and fungus that have been linked to many serious side effects. Residues from hexane have been found in edible food products.
On the label the DHASCO shows up as “crypthecodinium cohnii oil,“ and ARASCO as “mortierella alpine oil.” The Alps, however, are not the origin of the ARASCO. Fungus is the actual source. These oils contain only 40-50 percent DHASCO and ARASCO respectively, with the balance being sunflower oil, diglycerides, and nonsaponifiable materials. Some of these components are not found in human breast milk, and the triglycerides carrying DHASCO and ARASCO are not identical to those found in human breast milk and thus have not been part of the diet for human infants. Despite the fact that these ingredients are grown in the laboratory, they have been given the USDA “Organic” classification. The patent application states that Martek may genetically engineer the algae and fungus to make more oils, while the website tells consumers that the oils are not GMO.
Organic baby formulas such as “365 (Whole Foods),” contains Martek Biosciences oils. For a list of all other products, including baby foods, formulas, dairy and dairy alternatives, that contain Martek DHASCO and ARASCO, see www.cornucopia.org/dha-guide.
Martek has not affirmed the safety of its DHASCO and ARASCO oils added to infant formulas. When fed to rats, these oils resulted in statistically significant increased liver weights compared with rats that were fed a high fat diet without the oils, as well as increased spleen weight, a decrease in albumin levels and/or total protein levels. The FDA is aware of the safety concerns and has never approved the oils as GRAS (generally recognized as safe). But, the FDA’s position on additives is that “companies that want to add new additives to food bear the responsibility of providing FDA with information demonstrating that the additives are safe.”
These ingredients are possibly implicated in the digestive problems of infants consuming these formulas. “When I worked in the hospital’s neonatal ward, the nurses all called it ‘the diarrhea formula,’” says Sam Heather Doak, LPN, IBCLC, from Marietta, Ohio. “We’ve seen infants, tiny little humans, with diarrhea that just wouldn’t stop after being given this formula.” For infants, long-term diarrhea is a serious and life-threatening condition involving dehydration and malabsorption or non-absorption of nutrients.
“This report presents a disturbing look at the addition of novel ingredients into infant formula,” says Marsha Walker, Executive Director of the National Alliance for Breastfeeding Advocacy. “The FDA has received scores of reports on the adverse effects of these ingredients, but, to date, the public’s only access to these is through Cornucopia’s Freedom of Information Act request. This report will help alert the health care community and federal agencies to some of the adverse effects of added DHASCO and ARASCO in infant formulas.”
“Hundreds of reports of adverse reactions to these oils have been filed, and the Cornucopia Institute obtains a sampling of these reports through the Freedom of Information Act request. Reports of severe gas, diarrhea, vomiting, constipation, colic, gastric reflux, and bowel obstruction have been recorded, yet scientific studies show little or no benefit in cognitive development from the DHASCO and ARASCO in formula.”
FDA officials were aware of studies that reported diarrhea, flatulence, jaundice, and apnea in infants fed the Martek oil-supplemented formula, but allowed the addition of the oils to baby formula in 2001. Martek says that 90 percent of U.S. formulas contain these DHA and ARA oils.
The Martek “Life’s DHA” TM is not restricted to baby formula. DHA supplements, especially those marketed as “vegetarian products,” “vegetarian DHA liquid for children and adults,” and “ EPA/DHA vegetarian,” contain Martek DHA. Fish oil concentrates, highly processed products, which are “solvent free,” contain natural forms of DHA and EPA but are manipulated through processing (“Pure Encapsulations September 2012 Product List”).
SOURCES: The Cornucopia Institute. Replacing Mother: Imitating Human Breast Milk in the Laboratory. January 2008. pp. 1-15. http://www.cornucopia.org/replacing-mother-infant-formula-report/ accessed November 2012; Regulating Infant Formula by George Kent, Hale Publishers, 2011;
Genetic Roulette: The Documented Health Risks of Genetically Engineered Foods by Jeffrey Smith. 2004. Chelsea Green Publishers.
1. Bhatia, J., Greer, F., & the Committee on Nutrition, American Academy of Pediatrics. “Soy Protein-based Formulas: Recommendations for Use in Infant Feeding.” Pediatrics, 2009; 101: 1062-1068. doi: 10.1542/peds.2008-0564.
2. Canadian Paediatric Society. (2009). “Concerns for the use of soy-based formulas in infant nutrition.” Paediatr Chld Health. February 2009; 14(2): 109-113; PMC2661347, http://www.ncbi.nlm.nih.gov/pmc/articles (Free full text article).
3. Bhatia, J et al. 1063-1066.
4. Daniel, K.T. The Whole Soy Story. The Dark Side of America’s Favorite Health Food. Washington, DC: New Trends Publishing; 2005. 303-309.
5. Daniel, K.T., 331.
6. Andres, A. Badger, T.M. et al. “Developmental Status of 1-Year- Old Infants Fed Breast Milk, Cow’s Milk Formula, or Soy Formula.” Pediatrics. 2011; 129 (6): 1134-1140. doi: 10.1542.
7. Core, J. “Study Examines Long-Term Health Effects of Soy Infant Formula.” Agricultural Research Magazine. January 2004; 8-10.
8. USDA Agricultural Research Service. http://www.ars.usda.gov/research/projects.htm?ACCN_NO=420270; and http://www.ars.usda.gov/research/projects/projects.htm?ACCN_NO=416455.
10. Soy Nutrition Institute. Bylaws of Soy Nutrition institute, Inc. http://www.soyconnection.com/soynutritioninstitute/bylaws.ph.
11. Badger, T.M. et al. “The Health Consequences of Early Soy Consumption.” J. Nutr. 2002; 132: 559S–565S. http://jn.nutrition.org/content/132/3/559S.
12. Badger, T.M., et al. “The health implications of soy infant formula.” Am J Clin Nutr. 2009; 89 (suppl):1668S–72S. doi: 10.3945/ajcn.2009.26736U.
13. Pivik, R., Badger, M. “Effects of diet on early stage cortical perception and discrimination of syllables differing in voice-onset time: a longitudinal ERP study in 3 and 6 month old infants.” Brain and Language. 2012; 120 (1): 27-41. doi.org/10.1016/j. bandl.2011.08.004
15. Ronis MJ, Shankar K, Gomez-Acevedo H, Hennings L, Singhal R, Blackburn ML, Badger TM. “Mammary Gland Morphology and Gene Expression Differ in Female Rats Treated with 17β-Estradiol or Fed Soy Protein Isolate.” Endocrinology. 2012, Oct 1. (Epub ahead of print). doi : 10.1210/en.2012-1591.
16. Gilchrist JM, Moore MB, Andres A, Estroff JA, Badger TM. “Ultrasonographic patterns of reproductive organs in infants fed soy formula: comparisons to infants fed breast milk and milk formula.” J Pediatr. (Epub 2009 Oct 21). 156(2):215-20. doi:10.1016/j.jpeds.2009.08.043
17. Badger, T.M. et al.” Lactation and Neonatal Nutrition: Defining and Refining the Critical Questions,” J Mammary Gland Biol Neoplasia. 2012; 17(2):167-88. doi: 10.1007/s10911-012-9261-5. Epub 2012 Jul 1.
18. Storris, Carina. “ Infant ‘Smarts’ Similar With Different Types of Formula. “ 2012 HealthWoman. org. http://www.healthywomen.org/content/article/infant-smarts-similar-different-types-formulastudy.
19. Smith, J. Genetic Roulette: The Documented Health Risks of Genetically Engineered Foods. VT:Chelsea Green; 2005.
20. Daniel, K.T. 80, 97,109, 131-132.
21. Occupational Safety and Health Administration. Occupational Safety and Health Guideline for n-Hexame.http://www.osha.gov/SLTC/healthguidelines/n-hexane/recognition.html
22. Food and Drug Administration (FDA). “Statement of policy: Foods derived from new plant varieties.” Federal Register, May 1992. (57 FR 22984).
23. The Cornucopia Institute. “Replacing Mother: Imitating Human Breast Milk in the Laboratory.” January 2008. pp. 1-15. http://www.cornucopia.org/replacing-mother-infant-formula-report/ accessed November 2012.
24. Food and Drug Administration. (FDA) Poison Plant Database. http://www.accessdata.fda.gov/scripts/plantox/index.cfm
25. National institute of Environmental Health Science (NIEHS). “The Soy Infant Formula. Questions and answers about Soy Formula.”http://www.niehs.nih.gov/health/topics/agents/sya-soy-formula/index.cfm
26. National institute of Environmental Health Science (NIEHS). “Endocrine Disruptors.” http:// www.niehs.nih.gov/health/assets/docs_a_e/endocrine_disruptors.pdf
27. Kent, G. “Skewed Soy Studies,” Pediatrics. 2011. E-letter.http://pediatrics.aappublications.org/content/129/6/1134.long/reply#pediatrics_el_53967; and Kent, G. Regulating Infant Formula. Amarillo,TX: Hale Publishing; 2011.
28. Andreas, A. et al. (2011). 1135-1138
29. Ibid. 1134
30. Columbo, J. & Carson, S.E. “ Is the Measure the Message: The BSID and Nutritional Interventions.” Pediatrics; 2012: 129 (6). 1166-1167. doi: 10.1542/peds. 2012-0934.
31. Andreas, A. et al. (2011). 1134.
32. Ibid. 1136-7.
33. Slykerman, R. F. et al. “Breastfeeding and intelligence of preschool children.” Acta Paediatrica. 2005; 94(7): 832-837 DOI: 10.1080/08035250510031601S
34. Andreas, A. et al. (2011) p. 1138.
35. Kent, G. Skewed Soy Studies, Pediatrics. 2011. E-letter. (Epub 2009 Oct 21).
36. Kent, G. Regulating Infant Formula. 2011 Book Review. www.ibreastfeeding.com/newsletter/2011/12/regulating-infant-formula
37. Stein, M. “Cow, Soy or Breast Milk: Does it Make a Difference in Developmental Outcome? “ Journal Watch Pediatrics and Adolescent Medicine. June 20. 2012.
38. Utts, J. M & Heckard, R. Mind on Statistics, 3rd edition. Belmont, CA: Thomson-Brooks/Cole; 2007. p.136.
39. Andreas, A. (2011). 1148
40. Utts, J. M & Heckard, R. Statistical ideas and methods. Belmont, CA: Thomas Brooks/Cole; 2006. 669.
41. Adreas, A. et al. (2011) 1137.
42. Andreas, A. et al. (2011) 1138.
43. Utts, J. M. & Heckard, R. “Looking for Patterns with Scatterplots.” Mind on Statistics, 3rd edition. Belmont, CA: Thomson-Brooks/Cole; 2007.152-156.
44. Ibid. 2.7. “Features of Bell-shaped Distributions,” 49-57.
45. Andreas, A. et al. (2011) 1137-6.
46. Utts, J. M. & Heckard, R. (2007) 32-41.
This article appeared in Wise Traditions in Food, Farming and the Healing Arts, the quarterly journal of the Weston A. Price Foundation, Winter 2012