Discussion Paper 

Risks, Rights and Regulation
Communicating about Risks 
and Infant Feeding

INFANT FEEDING PRODUCTS

Mothers have always had alternatives to maternal breastfeeding. Historically, many of these options resulted in high rates of infant death. Since the 1900s, proprietary milks or infant formulas have been commercially available in industrialized countries. The last few decades have seen them aggressively promoted in developing countries as well. Cow's milk, milk based infant formula and soy based infant formula are usually given to an infant through a feeding bottle. For this reason, some of the literature on infant feeding compares the risks of bottle feeding to the risks of breastfeeding, without
adequate consideration of the contents of bottles or the common practice of mixed feeding (breastfeeding combined with bottle feeding). While cow's milk is not an adequate substitute for breastmilk, it is considered here because it is part of an infant's diet in many parts of the world; in addition some of the literature on contaminants considers cow's milk as an ingredient in infant formula. Bottle feeding also takes place in a polluted environment using an industrially produced product subject to contamination and accidents. The product is usually reconstituted using tap or surface water, which may contain toxins and diarrhoea?causing pathogens. Since many products are delivered in feeding bottles, the risk from feeding bottles should also be considered. Hormone-disrupting chemicals such as phthalates, nonylphenols, and bisphenol-A have been found in plastic feeding bottles. While complementary foods are not recommended for use before six months of age, many infants are introduced to such foods much earlier; foods are also a source of contamination. Therefore these products are considered here in order to emphasize that breastmilk is just one of many sources of possible contamination, and that other infant feeding products and methods are far from risk-free.

Cow's Milk
Cow's milk, often promoted as "nature's perfect food", is a key ingredient in infant formula, and plays an important role in the diets of infants older than six months of age in many parts of the world. Cow's milk is not advised for feeding of infants below six months of age as it is a nutritionally unsuitable and easily contaminated species-specific product. It contains excessive protein and sodium, and inadequate iron,
Vitamin C, and taurine, among other problems. It may contain too little or too much Vitamin D, depending on the kind of milk. Fresh milk may be contaminated with pathogenic organisms if stored
unpasteurized at room temperature. While cow's milk can be processed so that it contains fewer PCBs and pesticide residues than breastmilk, concerns have been raised about chemical and hormone residues, radioactivity, and antibiotics in milk. Following the accident at Chernobyl in 1986 cow's milk (and probably milk?based infant formulas) contained 300 times more radioactive iodine and caesium than was found in breastmilk (Baumslag and Michels 1995:97).

Non-fat dry milk (NFDM) is a product often used in emergency feeding and humanitarian aid. Only recently has the distribution of this product been made to comply with the International Code of Marketing of Breastmilk Substitutes, since the product may be locally perceived as a breastmilk substitute.

Cow's milk usually contains more pesticides than breastmilk since cows eat both the plants and the soil they grow in (Baumslag and Michel 1995:101). Accidents have resulted in the contamination of the feed provided to cows. For example, insecticide from pineapple that was mixed in animal feed resulted in contaminated cow's milk (Rogan 1996:986). Other concerns about cow's milk have been raised over Bovine Growth Hormone (rBGH) approved for use in the United States, although recently rejected for use in Canada. There was a moratorium on its use in Europe until 1999. Monsanto, through the United States, asked the WTO to rule the European ban illegal (Nottingham 1998:32). Although Monsanto claims that no growth hormone can be detected in the milk, antibiotic residues show up from the large doses given to cows for mastitis following prolonged use of rBGH. Another concern is that rBGH induces an increase in IGF?1, the insulin growth factor related to human breast cancer (Baumslag and Michel 1995:102).
Milk-based Infant Formula
The risks associated with cow's milk discussed above may also be relevant to milk?based infant formula. Milk?based infant formulas are much improved from the early mixtures developed in the early 1900s. The manufacturers are constantly seeking ways to emulate some of the hundreds of components of breastmilk. For example, Wyeth, the makers of SMA, is developing a milk product." where you would milk a cow and almost human milk would come out. We now have a mini-herd of transgenic cattle that are making human alphalact-albumen.in their milk;" according to a spokesperson from PPL Therapeutics in Virginia. Additions also include long chain polyunsaturated fatty acids to imitate those found in breastmilk and modified lactoferrin. Such efforts, while commendable, are doomed to failure since breastmilk is an unprocessed living substance, constantly changing in relation to the needs and environment of the breastfed infant, and cannot be commercially reproduced.

Milk?based infant formulas, like all industrially produced food, are subject to industrial accidents. (Two occurred during the preparation of this report. On March 19, 2000, Nestlé recalled batches of its ready-to-eat formula, because of "temperature fluctuations" during the manufacturing process. On June 9, 2000 a woman in Hamilton, Ontario found a key in a tin of formula.) Industrial accidents include the addition of excessive amounts of Vitamin D, aluminum, lead and iodine; insufficient amounts of chloride, zinc, Vitamins D, K, A, and B6; cans found to contain broken glass, live insects and worms; in addition to salmonella and bacterial contamination which circulate diarrhoea?causing pathogens. These problems resulted in 22 FDA recalls in the United States from 1983 to 1990 (Baumslag and Michels 1995:103; Appendix C, Walker 1998). Other risk factors include preparation errors, usually under or over concentration, in making up bottles.

Phthlatates, fat-soluble chemicals used in the production of plastics were found in 15 brands of infant formula tested in the United Kingdom. They were also found in ground water, rivers, and drinking water (MAFF 1996, Jobling et al 1995). These man-made chemicals that accumulate in body fat have been linked with cancer and a lowering of sperm counts in rats. Although the levels are below TDI (tolerable daily intake), government ministries have called for
efforts to reduce phthalate levels in infant formula. Concentrations of phthalates would be higher in newborns than in older, heavier
infants. Infant formulas contain negligible amounts of PCBs and dioxins when they are made from lipids of vegetable origin. Manufacturers of infant formula remove atrazine and other contaminants from the water used to make premixed, ready to feed formula.
Expensive advanced filtration and separation processes purify the water in these products, making them considerably safer for infants than formula reconstituted with tap water. Generally, these products are much more expensive than powdered infant formulas that have to be reconstituted with water before use.

Children raised on infant formula absorb five times as much manganese as breastfed infants. Manganese lowers the levels of serotonin and dopamine, neurotransmitters associated with planning and
impulse control. Masters et al (1997) suggest possible links between altered brain chemistry and violent behaviour, an association that deserves careful consideration in combination with sensitive sociocultural research. Babies fed exclusively on infant formula have up to twice the risk of developing diabetes, compared with breastfed babies. The link between milk-based infant formula and type 1 juvenile diabetes is clear (Steingraber 2001:227). Recently, concern has also been expressed over the use of peanut oil in infant formula, and the growing number of children with peanut allergies.

Soy-based Infant Formula

Infants who are sensitive to milk-based infant formula are increasingly being fed with soy-based formulas. (In Canada, 20% of infants consume soy-based infant formula). As with milk-based infant formulas, consideration of risks must include the risks related to the water used to
reconstitute the formulas. Soy-based products are also subject to industrial accidents and have been recalled for salmonella contamination, bacterial contamination, and Vitamin A and B6 deficiencies (Baumslag and Michel 1995, Appendix C).

However, the risks most recently associated with these soy products concern the use of genetically modified (GM) soybeans. Monsanto's Roundup Ready soybeans contain gene sequences from a virus, a bacterium and a petunia ? none of which are recommended for
infant feeding. Isomil, Carnation Alsoy, Similac Neocare and Enfalac Prosobee all tested positive for GM soy, although suppliers did not know the products contained GM soy (Burros 1997). All soy-based infant formulas sold in Canada are manufactured with GM soybeans and are unlabelled as such. (Canada has a system of 'voluntary
labelling').

Plant geneticists have succeeded in adding cystine, an amino acid found in human breastmilk to Prolina, a soybean bred for animal feed. The body uses cystine to make taurine, a compound that promotes eye and brain development. The makers say the ingredient is sought after as an ingredient in soy based infant formula. When these soybeans, or the GM herbicide?resistant soybeans, are used to make infant formula, the risks are completely unknown. Transgenic ingredients pose the risk of introducing novel toxins, new allergies and increased antibiotic resistance in infants (Walker 1998:1). The risks from consuming a product made from transgenic ingredients may be small, but they are unknown, unpredictable and irreversible. Consuming a product containing GM soy is much more likely to be significant for premature or rapidly growing infants than for adults. Infants fed soy formula can have circulating phytoestrogens 13,000?22,000 times higher than normal levels (Walker 1998:1). High levels of phytoestrogens in soy have been linked to premature
puberty, and a number of reproductive problems for both men and women. One maker of soy formulas in Australia (Abbott) said that no adverse effects have been identified in infants fed soy?based infant formula. "A warning about phytoestrogens would raise public concerns," small comfort to mothers who have chosen soy formulas to avoid the perceived risks associated with cow's milk infant formulas or toxins in breastmilk. In critiquing an article that found no difference in maturation and growth between infants fed soy-based vs milk-based formula, Goldman, Newbold and Swan argue that "there is now ample reason to question the safety of soy proteins in the diet of infants" (2001).

While the U.S. FDA says GM soy is safe, a recent press conference on this topic in Canada (1999) generated little news coverage and no reassurances with regard to soy-based infant formula from government spokespersons who assure the public that GM foods are safe. During the press conference a reporter asked, "is this scare tactics or are you saying that these kids [fed soymilk or soy-based infant formula] are going to have some sort of warped adolescence"? Unfortunately, no one could answer the question. As with DES, a synthetic estrogen marketed in the 1960s to prevent miscarriages, and later found to increase reproductive cancers in the next generation, problems only become visible many years after the use of a particular product. Is this a lesson that may be applicable to soy and other GM foods used to feed infants?
Water
Water is also relevant to the discussion as it is used to reconstitute milk and soy based infant formula, and sometimes given unnecessarily to breastfed infants. (Breastfed infants do not need any water, even in hot climates). Of particular concern is lead-containing water used to reconstitute infant formula (Shannon et al 1992), and any lead in the water may be further concentrated by boiling. Lead builds up in the body over many years and can damage the brain, kidneys, and red blood cells (Nutrition Action 2000:4). Chemicals leaching from fertilisers produce nitrate?contaminated water; if this water is used to reconstitute infant formula, the result is "blue baby syndrome" (Baumslag and Michels 1995:104). Mothers who use expensive bottled water to be "safe" may find that tap water has a greater range of valuable minerals (Baumslag and Michels 1995:105); bottled water may contain bacteria, lead, and mercury or excessive fluoride, chloride, nitrate or sodium. Many impurities can be removed from tap water or well water with filters but the filters need to be changed regularly to be effective.

The dangers of nitrates in water used to reconstitute infant formula is seldom referred to in the media. As fertilizer use intensifies, there have been high levels of nitrates found in the drinking water in France, Belgium and Switzerland. In Lithuania, a three week old baby died of acute nitrate poisoning after nitrate polluted water was used to prepare infant formula (Lietuvos Rytas 2001).

The Washington-based Environmental Working Group (EWG)
reported that the Environmental Protection Agency underestimated the health risks of mixing infant formula with water contaminated by the weed killer, atrazine, by a factor of 15. The EPA set maximum safe levels of atrazine at three parts per billion (ppb). One study found that in 798 Midwestern communities tested by EWG, 10.4 million people were drinking tap water contaminated with atrazine. By 8 months of age, bottle fed infants in Kansas City can get their entire legal lifetime dose of atrazine if the infant formula is reconstituted with tap water, creating cancer risks up to 20 times higher than federally mandated limits. The EPA and agricultural
industry representatives say the study is merely a scare tactic, and that actual risks from atrazine are much lower than the EWG report states. But other federal agencies agree that the EPA made mistakes in calculating the risks for children and infants.

Recently, First Nations' communities have been made aware of the contamination in their environment and many groups have emphasized the need to publicize and organise around this issue. Women in a Sagkeeng First Nations community in Manitoba recognised this problem in relation to bottle feeding. One woman stressed the convenience of breastfeeding, particularly in polluted environments:

We worried more when we were using bottles. We were worried about germs, water. We always had to get water from other places, we couldn't drink this water here. You know this river's polluted, so why would we want to give our child that, right? So we were getting water from springs. Without that water we'd have to buy the stuff that's all ready made and that's really expensive. (Martens 1994:lines 2553?2573)

Following the contamination of Milwaukee's municipal water supply with the parasite, Crytposporidiym in 1993, the American Academy of Pediatrics now recommends boiling water for infant formula to get rid of bacteria and parasites. The recent outbreak of E.coli in Walkerton, Ontario is a reminder that the safety of the water supply is not just a "third world" issue. Mothers who were breastfeeding would not have the added concern over the safety of water used to clean feeding bottles and teats, or to reconstitute infant formula.

Food and Risk

Placing food and risk in the same sentence is the unfortunate consequence of the paradox of food that food nourishes and comforts as well as sickens and kills. It is particularly distasteful to think of eating and feeding as risky behaviour. Food safety is a growing concern in Canada and elsewhere. Contaminated food is generally recognised as the main source of persistent environmental chemicals in the human body (Jensen and Slorach 1991:202). Aflatoxin, produced by moulds, may also be present in foods such as peanuts, and has been found to be carcinogenic in animals.

Nutrition Action, the newsletter of the Washington-based Center for Science in the Public Interest, published their October 1999 newsletter on food safety. The issue opens with a reminder when "food safety simply meant no egg salad, coleslaw, or other mayonnaise? based dishes at your picnic". Increasing knowledge and changes in food processing techniques mean that almost any food could cause food poisoning. They comfort with the qualification: "Most food is safe, and most people don't get sick from eating low levels of contamination" (Nutrition Action 1991:1). Consideration of food and risk is relevant for discussions of women's diet during pregnancy, mother's diet when breastfeeding, and complementary foods for
infants and toddlers.

There is a sizeable literature in anthropology on food prescriptions and proscriptions for pregnant women. This literature was not explored here since it focuses largely on symbolic pollution and social control rather than physical contamination. However, it is important to keep in mind that the discourse on symbolic pollution and social control may reinforce the scientific discourse on chemical pollutants and breastfeeding, a point explored further in the conclusion. Recently, the public has assumed the right to comment on pregnant women's behaviour with regard to alcohol, drugs, smoking and diet. Women's diet while breastfeeding used to be closely monitored by elders. Currently, women who are breastfeeding are encouraged to eat a balanced diet, drink plenty of liquids, and avoid any foods that seem to upset the baby. This trial and error process is unique to each mother-infant dyad. However, that does not stop others from drawing assumptions about diet and the quality of a woman's breastmilk. Martens cites a husband berating his wife in a Sagkeeng First Nations community in Manitoba:

  And he says, 'Look at all the things you eat. You're just giving her junk. It's healthier just to give her the bottle.' But my mom says your body does everything for you. It purifies that milk before you give it to your baby. (1994: lines 1317?20)

... he believed that everything that I'd eat she'd eat. Like if I was eating a greasy egg or greasy french fries, then the baby was getting greasy french fries and she was getting junk food. (Martens 1994: lines 1425?8)

Another area of research explores whether toxins in breastmilk could be related to mother's diets. Koppe studied the diets of Dutch lactating mothers as a possible route for lowering the concentrations of dioxins and PCBs in their breastmilk. He concluded that short?term dietary measures would not be effective since the lowering of contaminant intake must take place years before pregnancy to be effective. Dairy products were responsible for half the mothers' exposure to PCBs and dioxins, and he thus recommends efforts be made to lower the level of contaminants in dairy products (Koppe 1995). While contamination from DDT is unlikely to be found in food grown in North America, food imported from developing countries may contain traces of DDT. For both mothers and infants, food is a significant source of contamination: "more than 90% of the total daily human exposure to PCBs and dioxins is made up of oral intake from food, whereas other routes, e.g., water, air and soil, contribute to less than 10% of total exposure" (Patandin et al 1999:45). However, what a mother eats while breastfeeding is not a serious source of contaminants in breastmilk. About 30% of fat in breastmilk comes from the maternal diet; about 60% from maternal fat stores; and about 10% from new synthesis in the breast.

By six months of age, most infants receive other foods in addition to breastmilk. Some groups in Canada and elsewhere in the world provide small amounts of food to infants almost from birth, greatly
increasing the opportunities for infections and the introduction of contaminants. Infant foods are of particular concern since infants and children consume fewer foods in proportionately larger quantities than adults, differ in their ability to detoxify and excrete contaminants, and face greater risks of damage when exposed at critical periods of development early in life. Baby foods therefore have more stringent requirements with lower tolerance levels for contaminants. Nevertheless, researchers found traces of sixteen pesticides in eight different baby foods purchased in U.S. grocery stores, five of which were possible human carcinogens. Pesticide residues are particularly high in pears, peas and apple juice (Steingraber 1997:163-165), three foods that are commonly given to infants.

GM foods and ecological risks have become a topic of scientific and advocacy debate. Following the CBC evening news Oct 4, 1999, callers were asked to express their opinions about GM foods;130 expressed concerns about GM foods, while 9 were supportive. One supportive caller referred to the foods as genetically enhanced, not genetically modified. Policy makers may call for strict regulations for field trials, or suggest they be carried out in third world countries. But accidental releases of GM organisms have occurred even when attempts are made to "control" risk by the creation of small?
scale environments in greenhouses, and planting in caged
areas. Once spread to the wider environment they may be impossible to eradicate. They are considered dangerous because of their rapid reproductive rate combined with difficulty of detection.
Genetically engineered food crops may become weeds and their genes may be transferred to wild relatives whose hybrids may be detrimental to the environment. It is likely that transgenic foods are
being interpreted as contaminated foods (Nottingham 1998:177).
Experimental evidence suggests that GM foods are unlikely to pose direct risks to human health (Nottingham 1998:91). However, both allergic reactions and resistance to antibiotics may pose risks to some individuals. Allergies are imbalances in the immune system. Transferring Brazil nut genes to a soybean to make a transgenic product causes an allergic reaction in those allergic to nuts. We have not had enough experience with these foods to assess their risks ? both
predictable and unpredictable. For example, marker genes in
GM foods have been used to track risk, but what if marker genes themselves are the risk? However, people's reactions to these foods are based on perception of risk, not just risk. For example, food irradiation was rejected by the public on perception of potential risk, not actual risk.

Our foods offer different levels of risk; most risk is in foods with live microorganisms such as lactic acid, with lower risks for cooked foods, and lower still for processed foods. Breastfeeding advocates complain that when government officials send industry representatives to negotiate food safety regulations at meetings such as the Codex Alimentarius where international standards on breastmilk substitutes and baby foods are set, commercial interests will influence food policy. With regard to GM foods, we are faced with the need to develop policy for dealing with unpredictability. Unfortunately, American newspapers and magazines are shutting out criticism of GM foods and crops from their opinion pages at the very moment when policy with regard to the production, sale, and labelling of GM foods is being made (Parker 2002). People may accept risks and tradeoffs in composing their own meals. However, many are not prepared to accept the same risks for their children when mistakes have irreversible consequences.

Risks, Rights and Regulation: Communicating about Risks 
and Infant Feeding