The Case against Financing Dairy Projects in Developing Countries

Comments/corrections to

Draft last printed: 22 September, 2000

1. Introduction

The case that development assistance finance should leave large-scale grain-fed livestock production to the private sector is presented in Goodland (1997, 1998a,b). However, that case largely excluded dairy production. Development officials are starting to accept the case against financing grain-fed livestock, but reservations remain about milk production. This addendum extends the case against livestock to dairy production.

Dairy projects cause major and negative environmental and social impacts, almost the same as detailed in the 'Livestock EA' paper (available on request), so are not repeated here. This paper focuses on the impacts more specifically attributable to dairy. This paper begins with the strongest argument against dairy consumption, namely its strong association with heart disease. It then discusses the nutritional claims for of dairy, including its role with children, how most non-Caucasian adults cannot digest dairy, and the role of dairy in breast cancer.

The paper concludes that dairy projects are usually inequitable, nutritionally questionable, and risky for health. The few benefits of dairy (see below) are best obtained elsewhere. Scarce development resources are better allocated to promoting less expensive foods that are more accessible to the poor, as well as lower impact foods and healthier diets, which also are more equitable. Dairy is best left to the private sector.

1.1 Exceptions for Dairy Projects

The ruminant's extraordinary ability to digest cellulose can be valuable. When dairy cattle are not fed, but rather scavenge for most or all of their food -- as in much of India, their impact is lower. In such cases, if cattle recycle what would otherwise not be used, they often have much lower impact, while providing manure that can reduce soil depletion and provide fuel. However, these cattle produce very little milk. In addition, recent evidence suggests that India's dairy herd is exacerbating India's forest loss. This is being documented.

Cattle grazing on natural range produce little milk, and none prior to weaning, albeit possibly more than scavenging cattle. Range cattle are not useful for milking commercially, in part because they are distant from markets for dairy products, although they may supply their herders' families with dairy products, and occasionally adjacent villages. Range cattle usually do not provide attractive opportunities for dairy investments for international development assistance. However, in view of the economic inefficiencies of resource use, the environmental risks (e.g., overgrazing), the fact that the science is relatively new, and the risks of dairy consumption, these projects should be EA category A. (1)

The poor in developing countries often consume insufficient amounts of fat, so small amounts of fat in limited whole-milk consumption can be beneficial to lactose tolerant people. As this is a poor choice of fats, this should be an exception only. Small-scale projects promoting the family milch cow, mainly for domestic consumption, thus might be exceptions that could be financed. Development assistance can boost fat or oil consumption for the poor in developing countries at lower financial, social and environmental cost -- and at lower risk -- by investments in plant-based oil production, such as soy- and corn-oil. The use of the family cow for plowing or traction, and the use of its manure for fertilizer, can be useful niche roles. The widespread use of manure for fuel accelerates agricultural and environmental decline.

1.2 The Livestock EA Categorization also applies to Dairy

As the dairy industry uses only female cattle during their most productive milking years, both older female cattle and most males from birth are destined for beef or veal production -- with all the consequent adverse impacts outlined in the Livestock EA case. Therefore, environmental assessments of large-scale dairy projects or components should be categorized as "A" for the same reasons. Dairy products such as butter, cheese, ice cream and yogurt would be included. Full cost pricing of all food products, would take into account the levels of adverse impacts, including inequity (dairy is consumed mainly by the rich or foreign in poor countries). Such pricing would likely not favor dairy products.

1.3 What are Dairy Products?

Dairy products here include: non-human mammalian milk, cheese, butter, ghee, buttermilk, cream, and ice-cream. Yogurt and the many local varieties of curdled milk (kefir, koumiss, dadih) are included. These products differ in their health risks. Milk here refers to primarily cow's milk, unless otherwise specified, because buffalo, goat, camel and sheep milks are less well known scientifically and are relatively rare on a world scale. Many food products contain milk or milk products, including margarine, custard, biscuits, cakes and many 'processed foods'. Non-animal milks, such as soy milk and rice milk are unrelated to animal milk, and do not pose the risks associated with animal milk consumption.

1.4 Magnitude of the Issue

The consumption of dairy products is rising worldwide. The best data are from a major consumer, the US. Cohen (1998) calculates that the average American consumed 932 pounds of milk and dairy products in 1995. This sum was distributed among the top five dairy products (by converting the product back into the weight of milk that went in to its production) as: (1) Cheese 277, (2) Ice-cream etc 247, (3) Liquid Milks 207, (4) Butter 95, (5) Nonfat milk powder 42. The balance was in cottage cheese, cream, condensed milk, dry whole milk powder and dried whey to total 932 lbs./capita. That translates into 2.55 lbs. per person per day from the US's 10 million cow herd producing 153 bn.lbs of milk in 1995 according to USDA.

Because of perverse subsidies, there is an enormous surplus of US milk, but the option of exporting such surplus is being foreclosed. Europe has banned the import of rBGH dairy products, which means essentially all US dairy products. Other OECD governments, such as New Zealand, Japan and Canada are enacting similar bans. Therefore, most of US dairy surplus is shipped to developing countries, many of which lack a dairy industry. The World Bank, for example, invested $1.938 billion in 7 livestock projects and 58 livestock components in financial year 1999. About 65% of these investments were in Africa, East Asia and China.

2. Dairy Consumption and Coronary Heart Disease

Heart disease is the leading cause of death in the United States. It is the most important public health issue of most OECD nations, and is rapidly intensifying in many developing countries. Of the emerging public health priorities in developing countries with enormous long-term consequences, coronary heart disease (CHD) is at or near the top. The history of CHD is interesting and recent. Before the 1939-1945 War, atherosclerosis was recognized, but very rare, even in the US. This changed dramatically. Most (77%) of autopsied American soldiers killed in the 1951-1953 Korean War had atherosclerosis; equal-aged Korean soldiers showed no such damage. When many Koreans were put on the US Army diet, their cholesterol levels rose sharply, and atherosclerosis became unmistakable. Shocked by the results of the Korean War autopsies, the International Atherosclerosis Project examined the arteries of 20,000 autopsied bodies worldwide between 1963 to 1965. The findings corroborated those of the Korean War: people consuming diets rich in saturated fats and cholesterol had markedly more atherosclerosis, more CHD, and more strokes.

The meat, dairy and egg industries disputed these results, and still do. These industries pointed out that animal products are not the only source of saturated fats. They blamed coconut oil, palm kernel oil and chocolate, all of which are high in saturated fats. In fact they are practically the only three plants with saturated fat, and do not account for large proportions of most people's diets. Cholesterol cannot be found in any plant food, (2) but is very high in dairy products.

The meat, dairy and egg industries then claimed that heredity is more important than diet in such cases. This was promptly disproven by many studies of people of different heredity living in different countries, such as Chinese and Japanese in California, Indians and Pakistanis in England. In 1970 and again in 1980, Ancel Keys published compelling correlations between saturated fat and cholesterol in the diet, cholesterol in the blood and death rate from heart disease. In general, for every 1% reduction in total blood cholesterol, there is a 2% reduction in heart disease risk. By 1961, the American Medical Association editorialized: "A vegetarian diet can prevent 97% of our coronary occlusions". By 1979 the US Surgeon General affirmed these results; so did USDA by 1980. These industries persist in advertizing the health of their products, sponsoring much research seeking to disprove the science, and financing powerful lobbyists.

The case that consumption of animal products over a certain low minimum greatly increases the risk of CHD is well documented. It has become the consensus that animal protein intake is strongly associated with CHD (e.g. <>; <>; Hu & Willett, 1998; Campbell et al. 1998; Geissler 1999). It is well established that serum LDL cholesterol levels are very important predictors of CHD.

CHD risks can be disaggregated into the two principal sources of animal protein, namely dairy and meats. Eggs are a close third. Of the dairy risks, part stems from the high levels of saturated fats in most dairy products. The other dairy risk factor contributing to CHD is milk protein. Davies (1980) found myocardial infarction patients had elevated antibody levels to milk protein compared with healthy persons. Dairy consumption correlates positively with blood cholesterol, as well as with coronary mortality. In 17 countries, Law (1994) found a good correlation between national cholesterol levels and ischaemic heart disease. The saturated fat and cholesterol contributions to CHD risk are now inescapable. Heart (1997) monitored 11,000 overtly health-conscious adults for 13 years. Death rates from CHD were higher in those consuming relatively more meat, and significantly higher in those also consuming high levels of cheese and eggs. This implies that replacing meat in the diet with other foods rich in animal fats and cholesterol, such as cheese and eggs, will not reduce CHD risk

Gordon (1999) synthesizes 428 peer-reviewed scientific papers on this topic, and concludes that milk drinking is a health hazard, 'even a threat of death' in later years. Most of the world has become urban and more sedentary than fifty years ago when the world was primarily rural and active. Because of this change in activity, milk for adults has become risky (e.g., Cohen 1998). As long ago as 1977, Segall had correlated milk consumption and coronary mortality rates in 43 countries, and with myocardial infarction in 19 regions of Europe (Segall 1977, 1994). Public policy has not yet caught up with this switch. CHD is not an inevitable consequence of high milk diets; physical activity and dietary factors (such as antioxidants and omega-3 fatty acids) lower CHD risk. Figure 1 shows the correlation between coronary heart disease (CHD) mortality and milk consumption in 23 countries.

Figure 1:

Correlations between CHD mortality rates and the consumption of unfermented milk protein

(Source: Gordon 1999; Seely 1981)

Figure 2:

Milk consumption in 15 countries and serum cholesterol levels

Source: Gordon, 1999 (with kind permission of Harcourt Brace)

Figure 1 shows the correlations between CHD mortality rates and the consumption of unfermented milk protein. Figure 2 shows the correlation between milk consumption and blood serum cholesterol level in 15 countries. Seely disaggregated the strong dairy/mortality correlation. From WHO data, Seely (1981) found mortality most closely correlated with dairy consumption, second with animal protein, third with animal fat, and fourth with sugar.

Migrants from India to other countries suffer from higher CHD and CHD-mortality rates than those of the host population, whether in Singapore, South Africa, England or elsewhere. Such Indians consume much more milk and butter than the host population does. Dairy products in India are consumed more by the rich (who are sedentary and have high risk of CHD) than by the rural poor who are active and have lower CHD risk. This suggests that the health effects of dairy consumption are part of the health damage of affluent lifestyles. Much milk and curds in India - the world's largest producer of milk --are from a mixture of cow and buffalo milks, which have different compositions. In the wet season, rural Indians sell milk and its products to more affluent urban Indians if possible, and consume any surplus that they are unable to sell.

For coronary heart disease, liquid cow's milk itself seems to be the main risk of dairy consumption. CHD risk seems to be somewhat reduced in cheese consumption. The very high saturated fat, cholesterol, and salt contents of most cheeses impose their own risks. The difference in CHD risk between milk and cheese gave rise to the 'lactose hypothesis', as lactose is higher in milk than in cheese. As milk sugars are water soluble, they are retained in the whey fraction.

Fig. 3: The Maasai, Fermented Milk, and Atherosclerosis

Gordon (1999) addresses the paradox of the Nilo-Hamitic Maasai nomad's high milk diets (3-5 lt./person/day in the wet season) in N. Tanzania and S. Kenya. Despite very high Zebu milk intakes, the Maasai do not suffer from hypertension, nor from high cholesterol levels. However, they develop high rates of atherosclerosis, similar to the rates in old US males, but its ill effects seem to be counteracted by their exceptionally active lifestyles and unusual coronary arteries (Mann et al. 1972, Isono et al. 1994, Ho et al. 1971, Biss et al. 1970). Mursik is the milk fermented (mainly by Lactobacillus confusus and Lactococcus lactis) in a smoked, long Kibuyu gourd before consumption, which may alter the properties of the diet, such as by reducing lactose. Some Maasai occasionally add cow's blood to two-day old Mursik, or drink it on its own as a tonic, especially for the infirm or pregnant. More blood is added to the mixture during the dry season when milk production is low. Mursik seems to lower the risk of Entamoeba, Escherischia coli, and Staphylococcus aureus, which are common in pastoral societies with poor sanitation. As Maasai intake of dietary fiber is so low, the binding properties of lactic acid bacteria may help maintain health. After fermentation, the gourd is washed out with wood ash (a base?), and cattle urine (ammonia?), which probably alters the fermentation. The length of fermentation varies from a few hours, to a couple of days, to one to two weeks. Zebu milk contains significantly higher lipid and cholesterol levels, and less sugar than ‘standard' cow's milk. Maasai consumption of tea and mead also may be protective.

Zatonsky (1997) and Zatonsky et al. (1998) graphically document major declines in mortality from heart disease (and some decline in stroke) soon after saturated animal fats (e.g., butter, lard, suet, tallow) decreased in Polish diets, and were substituted with unsaturated plant oils. Before the switch to unsaturated oils, CHD mortality had doubled from 1961 to 1991. Curiously, cholesterol levels did not fall as dramatically during this dietary switch from animal fats to plant oils. Even the margarines were manufactured by a new method, which kept the risky trans-fatty acids very low. The Polish experience is being repeated in the Czech Republic as its diets are switched from animal fats to plant oils.

3. The Nutritional Value of Dairy

3.1 Dairy Fats and Calories

Milk is about 87% water. Of the solids fraction, about 50% of milk's calories come from fat. One cup of whole cow's milk contains about 8g of fat. (3) Around two-thirds of the fatty acids in milk are saturated. Polyunsaturated fatty acids make up less than 4% of milk fat. 'Low fat' milk is not low fat. It contains from 24% to 33% fat as calories. '2%' milk contains about 5g of fat per cup. The '2%' figure also is misleading as it refers to weight, but milk is mainly water. '1%' milk contains 2-3g of fat, and skim milk about half that. "Non-fat" milk contains 0.4g of fat per cup. As the link between elevated LDL cholesterol levels and CHD are well established, milk fats are risky.

The fat content of many cheeses varies from 7g to 10 g per standard (1 oz or 28g) serving, and rises to 38g in some brands. 'Low fat' cheese contains 2g to 4g of fat. Most cheeses provide 50% to 80% of calories from its fat, of which most is saturated. The hyperbole of cheese as 'solid cholesterol' is exaggerated as cheese often contains 20-30mg of cholesterol per serving. Saturated fats, and to a lesser extent cholesterol intake, raise blood cholesterol levels.

Butter's calories are 100% from fat, which is high in cholesterol and saturated fat. The U.S. Department of Agriculture's latest dietary guidelines (1995 add 1999) recommend low- and non-fat dairy products -- but reduced fat dairy products (e.g., skim milk) are generally not available in poor developing nations.

3.2 Dairy protein

Protein deficiency was a grave problem worldwide until fairly recently. Fortunately it has become less widespread, but still intolerable, and certainly for development agencies. Protein deficiency occurs worldwide, largely in poor people living in harsh remote areas with little education and subsisting on cassava or yams. Even 'Irish' potatoes can supply enough protein to prevent protein deficiency unless overly peeled. (4) If calorie deficiency is solved by grain-based diets, protein deficiency is unlikely to be a problem. Development agencies and governments should tackle both protein- and calorie-deficiency as top priorities. But the most economic solution, with the least environmental damage, will not include dairy or animal protein (see exceptions of the family cow or pig, and occasionally rodents). The solution will vary from place to place, and will hinge on legumes and grains, and possibly supplements. Leafy vegetables are useful sources of protein (2-5 gms/serving) and especially legumes, without the risks associated with dairy protein.

Cow's milk is rich in protein; human milk has the least protein (0.9%) of all milks. The amount of milk calories as protein is 15% cow, 5% human, 17% goat, 49% rat, and 11% mare. In comparison, the percentage of calories from plant protein: legumes 25-50%, vegetables 20-49%, grains 10-30%, fruits 5-15%, seeds 10-20%. The protein in milk is a weak reason to consume milk. Milk proteins are among the worst protein choices, as they are implicated in diabetes, allergies, migraines and some cases of arthritis (N. Barnard, Pers. Comm. 11/99).

Most OECD member country inhabitants consume an excess of protein. Protein overload is related to osteoporosis, atherosclerosis and kidney damage. OECD octogenarians commonly lose 30% of their kidney function caused by excess protein intakes (Rowe et al.1976). Low protein diets (4%-8%) are commonly used to treat patients with kidney failure. As most OECD protein is animal in origin, cholesterol and saturated fats are largely unavoidable with all the risks of cancers, stroke, and coronary heart disease (CHD). The recommended protein daily allowance has been revised downward through the years and is currently 45gm/day (1.5oz) for a 60kg person between 19 and 51. Even this is not the minimum. Humans can get by on 30gm/day if necessary.

3.3 Dairy vitamins

Vitamin A is not strictly essential. It is synthesized from the provitamin beta-carotenoids, which are widely distributed in many fruits and vegetables. Whole milk and butter contain some retinol, or preformed vitamin A (37g, 15g, traces per 100g in whole, partially skim, and skim cow's milk respectively), whereas a small serving (½ cup) of pumpkin contains 2691 mg. Vitamin A deficiency is a problem in some developing countries, causing 500,000 new cases of corneal lesions each year. These nutritional components are difficult to acquire from dairy without also ingesting much animal fat. Vitamin A needs are healthier met from the carotenoids common in carrots, squash, sweet potato, tomato and many other vegetables. Many carotenoids, such as beta-carotene and lycopene, are antioxidants, especially inhibiting the oxidation of lipoproteins, thus probably reducing the risk of heart disease.

Dairy products are rich in riboflavin (vitamin B2, 0.18mg/100g whole cow's milk). Riboflavin is easily available at low risk from soybeans, green leafy vegetables, and especially sea vegetables, such as kelp, dulse (Wales), lavabread, arame, kombu, alaria (Maine), kim (Korea), Nori, Wakame & Hijiki (Japan). Riboflavin deficiency is rare worldwide, except in the institutionalized elderly.

Only microorganisms, fungi and algae synthesize Vitamin B12; other plants and animals cannot produce any B12. It is ingested in animal products if the animal ate foods containing B12, and in plant products contaminated with B12-producing bacteria. Human gut bacteria produce much B12 although it is not clear how much is absorbed. The WHO recommends 1.0 g daily, but 0.1 g is probably enough for most people. Because so little B12 is needed, B12 deficiency due to inadequate intake is rare; 95% of today's B12 deficiency worldwide occurs in individuals genetically unable to absorb it. Even so, all need to be careful to obtain adequate B12, especially vegans. One tablespoon of yeast provides 4g; a cup of milk contains 0.9 g. Most modern and cheap multivites provide more than enough B12 from one pill a week, although there is no standardization as yet, especially in developing countries. B12 supplements and fortification have started to spread in developing countries.

As vitamin D is manufactured less by dark-skinned people; they need more sun than light-skinned people do. Rickets are now rare except in some highly polluted northern cities where children do not get enough ultraviolet light for their bodies to manufacture enough vitamin D. Many commercial cereals a rich sources. OECD milk has been fortified with vitamin D from the 1930s; soymilk and ricemilk are often fortified.

3.4 Essential Fatty Acids:

Cow's milk differs greatly from human milk in its marked deficiency of essential fatty acids (e.g. possibly linoleic acid, although milk's fatty acid tables are not yet complete). Skim milk has little or no linoleic acid either. Milk is not rich in the often limiting essential fatty acid -linolenic acid. Milk contains the most myristic acid of any food except coconut oil. Myristic acid is very high in butter and is four times as effective as palmitic acid in raising cholesterol levels. This is much of the reason that milk consumption raises blood cholesterol levels.

3.5 Dairy Calcium

The science of calcium nutrition remains controversial. As the science improves, it seems as if improving its availability also improves the assimilation of adequate calcium, not only its total dietary amount. The USA's Recommended Daily Allowance (RDA) (800 mg) is almost twice that of the UN WHO RDA. Calcium is indeed needed for bones, but as one of several factors. After weaning, cows do not consume milk; they get enough calcium from their grass and feed. The curious fact is, though, that bone fractures are highest where animal protein intake (e.g., milk) is highest. Milk, cheese and yogurt contain relatively rich levels of calcium (119mg/100g in whole cow's milk), yet substantial milk consumption does not seem to reduce bone fracture rates significantly. Most dairy calcium neutralizes the acidity brought on by milk protein (Fescanich 1996). When protein (especially animal protein) metabolizes to acidic products, calcium is needed for buffering. This renders such calcium unavailable for nutrition.

Osteoporosis is a systemic disease in which fractures later in life result from gradual deterioration of bone mass and architecture (Kanis 1997). It will become an increasingly serious problem as the population ages and as a result of a poorly understood secular trend. The contribution of dietary milk and meat to the risk of osteoporosis also remains unclear, but it seems that increasing milk intake is not advisable to reduce such risks. Hip fracture rates are correlated with high calcium intakes (the Harvard Nurses data support this argument). Bone fracture rates in China, for example, are among the lowest in the world. Chinese dairy intake is near zero, and animal protein intake was very low until the last few years.

Does milk and dairy consumption tend to weaken bones or to strengthen them? The global problem of osteoporosis suggests answers to this controversy. Osteoporosis is complicated by the fact that dairy calcium intake and hip fracture rates are directly correlated. Thus milk may not to be as helpful against osteoporosis as calcium in a more available form, without the acids which sequester the calcium, such as in green vegetables, and as supplements if necessary. If increased calcium availability is needed, milk may not be a good source as it is difficult to consume calcium from dairy without also consuming much fat and animal protein. A further complication is that calcium may be assimilated only in the presence of magnesium, which is very low in milk.

The increasing evidence is that milk is not helpful in osteoporosis, and can be counterproductive. There are a few, but declining, numbers of papers that claim the opposite. For example, Heaney et al. (1999), financed by the dairy industry, claim that drinking three 8-oz servings of milk a day could improve older adults skeletal health. Nearly half (46%) of the 132 women in the milk-drinking control group of the study were on conventional menopause therapy to prevent bone loss by taking steroid hormones. But only 31% of the control group women - those not drinking milk -- were on steroid therapy, so the whole study is questionable, according to Cohen (1999; The study fails to show decreases in fracture rates, nor increases in bone density.

Animal foods are generally acidic; plant-based foods are primarily alkaline. The body balances pH so that pH-dependent biochemical reactions take place normally. Phosphates and calcium buffer acids, which are derived from the diet or from bones if necessary. This is the first stage of osteoporosis. The second stage in osteoporosis consists of changes in kidney physiology caused by such acids, the sulfur-containing amino acids (high in meat and dairy), and the increased solute load, all resulting in loss of large amounts of bone, including calcium, into the urine. This bone-material in the kidney system also lays the foundation for the formation of kidney stones.

The use of calcium in the management of osteoporosis is reviewed by Kanis (1999). Cutting calcium loss is probably all that is needed. Exercising more, reducing animal protein consumption, and reducing sodium intake reduce calcium loss. Green vegetables are rich in calcium and magnesium in the proportions that enable both to be assimilated. Calcium is obtainable at much lower risk from broccoli, kale and other green leafy vegetables, and especially tofu. Staples are fortifiable with calcium at low cost, so can orange juice, although consensus has not been achieved on this for developing countries. Development assistance can help more people at lower cost by promoting plant foods high in assimilable calcium. Investing in calcium fortification of staples, legislation concerning calcium fortification and supplements also have their place.

3.6 Dairy Iron

Iron is considered to be the most common nutritional deficiency. About half a million people may be iron deficient worldwide. Parasites and repeated pregnancies, both common in developing countries, exacerbate iron deficiency. The US RDA for women and men is 15mg and 10mg respectively. Decrease in parasites, iron fortification of flour, and oral contraceptives have decreased iron deficiencies in developing countries. Cow's milk products are very low in iron. Vegan diets are generally higher in iron than diets rich in eggs and dairy. Sea vegetables, tofu, molasses and many species of beans are rich in iron. Iron absorption is inhibited by consumption of tannic acids, such as in tea, chilies, and milky coffee. The calcium in dairy products greatly inhibits iron absorption. Vitamin C increases iron absorption from the diet. Low cost iron fortification of flour and salt seems to be the least-cost method of reducing iron deficiency in developing countries.

3.7 Dairy Zinc

As it is so difficult to determine zinc status, the US zinc RDA (12mg for women, 15mg for men) should be viewed with caution. Canada's RDA is 8 or 9 mg, and UN WHO's RDA is one third to three quarters of the US level. Milk provides about 1 mg per cup; many species of bean provide quadruple that per cup.

4. Milk and Children

Human milk is essential for infants, although it is possible (but not at all advisable) for newborns to survive on formula. All infants possess lactase, but it decreases rapidly following infancy in most humans except Caucasians. Western medical nutritionists (e.g., Sanders & Reddy 1994) confirm that even the growth and development of vegan children appears normal, especially if the well-known pitfalls are avoided. Despite protestations to the contrary, understanding the true role of cow's milk for children is neither heresy, nor sacrosanct, nor taboo. Cow's milk is not necessary for growing children. Most children thrive without consuming a drop of cow's milk.

On the contrary, there is growing evidence that milk can do more harm than good. First, the American Academy of Pediatrics (1992) discourages giving cow's milk before the first birthday because a solely milk diet is the leading cause of iron-deficiency anemia in infants. Cow's milk is low in iron. Cow's milk is not a substitute for breast milk. Second, cows milk displaces human milk. Third, milk causes micro-bleeding of the intestinal tract (A. Robinson, WHO, 2000). Fourth, cows milk contributes to type-1 diabetes in children. Fifth, milk allergies are common in children causing sinusitis, diarrhea, constipation and fatigue. Cow's milk proteins are the first foreign proteins entering the infant gut, since most infant formulas are cow-milk based. Milk allergies are related to chronic ear infections, to behavioral problems, and to childhood asthma. Soymilk causes far fewer problems. Humanitarian aid supplying infant formula (e.g., in Kosovo) causes dependency, and force families to buy cows milk when both aid infant formula supplies and breastmilk dry up.

Cow's milk combined with other foods can nourish lactose-tolerant children for a few years. But recently, many conditions previously not linked with milk now are, even constipation. Breast-fed infants are smarter than formula/cow's milk fed infants. Breast feeding was associated with significantly higher scores for cognitive development than was formula feeding in the meta-analysis of Anderson et al. (1999; see also Uauy et al. 1999). Breast feeding was felt to be inadvisable for infants with HIV+ve mothers, but recent evidence (Coutsoudis 1999; UNICEF 1999) questions this. Most children in the world thrive without consuming any dairy products. Pediatrician Dr Benjamin Spock (1992) warned of the faults of cows milk for humans: "Human milk is the right one for babies". The present paper refers mainly to milk consumption after childhood.

Low energy density diets can be a problem for some children under five years of age, which argues for delaying weaning for as long as possible. Vitamin B12 deficiency, iodine deficiency, iron-deficiency anemia are risks. For the poor, these can be provided as supplements at relatively low cost and at low risk, much lower than provision of dairy or meat in both health risks, and in costs (financial and environmental). Diets with oils with a low ratio of linoleic to linolenic acid prevent most risks. Salt fortification with iodine is spreading even to many developing countries, although this needs to be accelerated. Iron fortification of salt has only just begun, although is low cost and effective. Soy's isoflavones (estrogen-like compounds) seem to protect health in general although the risk of excessive consumption by children needs to be avoided. Soy's saponins seem to protect cells from cancer-starting activity and may slow cancer tumor growth rates while leaving healthy cells alone. Soy and soy milk wins hands down on health, economic, environmental, social and other grounds when compared with cow's milk and dairy.

4.1 Milk and Menarche

Dairy consumption accelerates menarche. OECD girls' menarche has fallen from 16.5 years in 1840, to 13 years in 1995, and to 11-12 years today. A clear picture from Japan shows that dairy helped accelerate menarche from 15.2 years in 1950 to 12.2 years in 1975. Accelerating menarche parallels the increase in degenerative diseases later in life, and is related to increased energy intake, including milk and other forms of dairy.

Late menarche is associated with: (a) decreased breast cancer rates later in life, (b) decreased coronary heart disease, (c) fewer teen pregnancies, and (d) later first pregnancy (Rees, 1995). Early age at menarche is one of the few established early life predictors of breast cancer risk (Petridou et al. 1996). In cases of early menarche, changes in diet from animal to plant-based may reduce breast cancer risks later in life.

Opinions on the 'normal' age of the onset of girl's puberty is changing too. Breast development in 8-year old girls used to be called precocious puberty. Now such development is no longer uncommon in 6-year olds. By 8 years, 48% of African American girls, and 15% of US white girls had begun to develop breasts (Pediatrics, 10/'99). These trends may also be related to diet, as obesity, Type II (senile) diabetes, and a surfeit of the hormone leptin (manufactured by fat cells) are also becoming more common in such girls (see section on diabetes and milk).

5. Dairy Products are Inequitable

Most people on earth do not drink cow's milk. Most of them can't drink milk because it makes them sick (Bertron et al. 1999a,b). The oldest major civilization, the Chinese, did not have any dairy until very recently, partly because of their widespread lactose intolerance. The main groups benefiting from dairy consumption in developing countries are rich élites or expatriate Caucasians. Most people in developing countries cannot digest dairy products. In many developing countries, dairy products are generally not consumed because most people there are lactose intolerant (Figures 4 & 5). Policy makers in the US are slowly recognizing this. For example, in September 1999, one Senator called for alternatives to cow's milk in schools because he said 95% of Asian Americans, 56% of African Americans, and 50% of Hispanics in his electoral district are lactose intolerant. US Dietary Guidelines (1999??), the basis of all federal nutrition programs, recommend two to three daily servings of dairy products. As most US African, Asian, Hispanic and Native American individuals are lactose non-persistent, such diets would harm them, hence are racially biased (Bertron et al. 1999a,b).

Fig. 4: Lactose and Lactase

Milk contains about 4.7-4.9% of lactose, the main milk sugar, which is a disaccharide. Thus, lactose is the main carbohydrate of milk. Lactose intolerance is sometimes called lactase non-persistence or lactose malabsorption. This was discovered in humans very recently, in about 1963. Lactose intolerant people lack or have low levels of the digestive enzyme lactase (b-galactosidase), produced by the epithelial cells of the small intestine. Lactase hydrolyzes lactose into two monosacharides, galactose and glucose, before it reaches the colon. If not, colonic bacteria ferment the lactose, thus producing harmful symptoms. Galactose is toxic in high doses or if insufficiently removed, and causes tissue damage above certain levels in humans. Galactose reaches high levels in human plasma following milk ingestion. Galactose increases atheromatous plaque formation in Baboons and other experimental animals, causes cataracts in rats (and possibly in humans), and is related to the onset of diabetes in humans (Gordon, 1999). Fermented milk products, such as yogurt, contain Lactobacillus bulgaricus or Streptococcus thermophilus, which split some of the milk's lactose into galactose (which remains) and glucose, which is converted via pyruvate into lactic acid.

Possibly most adults worldwide were lactose intolerant until a mutation producing lactase appeared in Northern Europe 10,000-30,000 years ago. Lactose phenotypes are genetically determined, as shown in the study of twins. Therefore lactose intolerance is mainly hereditary. This may have been an advantage when light-skinned people migrated Northwards where winter sunshine is scarcer, so some (little) vitamin D formation occurred in the small areas of skin left unprotected from the cold (e.g., the face), but this may have been inadequate. Milk contains little vitamin D before supplementation. So dairy consumption is fairly new for humans. Note that Inuit diets contain excessive protein and calcium, and their osteoporosis rates are among the worst in the world. As most people become lactose intolerant, an enormous lactose-intolerance industry sells over $200 million of products annually.

Most infants possess lactase. Lactase production normally decreases in most humans (and most mammals) after weaning and remains low throughout life. As lactase production declines with age, so lactose intolerance can develop even in lactose tolerant consumers later in life. Dairy consumption by lactose intolerant people can cause total incapacitation, extreme discomfort, or gastrointestinal problems including cramps, and diarrhea. Milk also seems to be the leading cause of childhood allergies. Thus, dairy projects cannot benefit the diets of most people in developing countries. Even if the dairy project were slated for countries composed mainly of Mediterranean ethnics, the diets of much less than half the population would benefit.

Figure 5:
The Prevalence of Lactose Intolerance in Adults of Certain Ethnic Groups
Sources: Kent DeLong (199*), Flatz 1989, Salvadori del Prato, 1990.
Ethnic GroupPercent Intolerant
'African Blacks'
'Indians' (India)
Northwestern Indians and Pakistanis
Northern Indians
Southern Indians
North American Blacks
Mexican Americans
North American Whites
Italians (overall)
Southern Italians
Europeans (overall)
  Northern Han
  Southern Han
Niger: Tuareg
Nigeria: Fulani
Saudi Arabia: Bedouins

As can be seen from Figure 5, the distribution of lactase phenotypes is highly variable. Different authors provide slightly different figures. The most detailed are those of Gebhard Flatz (1989) who summarized the information on lactase activity in 23,000 humans from over 100 national and ethnic groups (Table 122-2; p. 3003). He points out that the lactase persistence allele predominates in only two sites in the world. One is Northern and Central Europe; the other site is the ambit of the nomads of North Africa and Arabia. The only two ethnic groups he lists as having zero low lactase allele are Senegal's Peuhl or Fulbe, and the Dutch. Groups listed as 100% are Vietnamese living in USA, Taiwanese, Fijians and the Chami Amerindians of Colombia.

6. Breast Cancer and Dairy Consumption

Breast cancer is the most common cause of cancer death in women, and the third most common cancer overall. Breast cancer is increasing worldwide; about one million new cases are diagnosed annually. Only half all diagnosed cases survive more than five years, even with appropriate treatment. Breast cancer is evenly balanced between developed and developing countries. Based on thousands of epidemiological research findings and empirical studies, the World Cancer Research Fund and the American Institute for Cancer Research (1997) warned of the risk of breast cancer from consumption of animal fats, saturated fats, total fats and meat. They go on to point out that breast cancer risks are reduced by consumption of fruits and vegetables (details at: <;;;>; Willett et al 1987; 1992; Epstein 1998). WCRF's warning is widely and increasingly accepted throughout the profession. But it is not well known by the American public (Barnard & Nicholson 1997), and is relatively unknown in developing countries.

The ecological evidence supports the compelling epidemiological evidence. Figures 6 & 7 show the correlation between cancer and diet (Harris, 1999). Cancer mortality statistics in 33 countries of the world were compiled and calculated from data edited from a magnetic tape copy of the World Health Organization (WHO) database of cancer mortality (Tominaga et al. 1994). Correlations between breast cancer mortality rates in the 30 countries for which data were available, and food and environmental factors were collected from various other sources (FAO 1998; Kurian 1979, 1991). The highest correlation (R=.79: Probability 'p' much less than 0.01) was between breast cancer and animal source calorie consumption (Figs. 7 & 8).

Figure 7: Correlation between Breast Cancer and Milk Production

Source: William Harris, M.D., 1999, Medical Director, Kaiser-Permanente:

Outwater et al. (1997) review the literature showing correlations between breast cancer incidence and dairy consumption.

Source: William Harris, MD, Kaiser Permanente

Figure 7: Correlation between Diets and Cancer

From: Harris, W. 1999; Multiple regression by BMDP

Variable Partial_R p value
Animal source calories/day 0.7897 1.3E-07
Animal fat/day (gm) 0.7807 2.1E-07
Total fat/day (gm) 0.7767 2.1E-07
Animal protein (gm/day) 0.7029 1.0E-05
Meat Kg/caput/year 0.6937 1.5E-05
Animal source calcium (mg/day) 0.6401 0.0001
Total Calories/day 0.6071 0.0003
Milk production (lbs/caput/day) 0.5521 0.0013
Total protein (gm/day) 0.533 0.002
GNP/cap($) 0.5268 0.0023
Female life expectancy (years) 0.4661 0.0082
Hen eggs (lbs/caput/day) 0.353 0.0514
Vegetable source calcium (mg/day) -0.1682 0.3657
Vegetable source Calories/day -0.2864 0.1182
Vegetable source protein (gm/day) -0.3607 0.0462
Infant mortality -0.4123 0.0212

7. Carcinogens in Milk

Fats seem to be the main cause of breast cancer from consumption of dairy. Milk fats act as a sink for a wide range of environmental pollutants that contaminate range, pastures, forage, and all other sources of cattle feed. Contaminants include carcinogenic biocides, industrial chemicals, and air-borne pollutants such as from coal-fired power plants, or from aerial and other crop spraying (e.g., Epstein 1998). Fat-soluble contaminants (e.g., DDT) bioaccumulate in fats. Lipophilic pollutants in both human and cow's milk (e.g., PCBs, dioxins, DDT) transfer from the air (e.g., incinerators, disease vector control, pesticides, fungicides etc) to forage, thence to meat and milks. The contaminants in milk fat reflect the pollution falling on cows feed, thus are highly variable. However, the list of such pollutants in milk is long and growing. These carcinogens are found worldwide in air, soil, sediment, fish, meat, and dairy (Roeder et al. 1998; India - Kumar et al 1996; Uganda - Ejobi et al. 1996). The residues or metabolites of carcinogenic pollutants in cow's and human milks can exceed the safe minima set by FAO/WHO (e.g., Kannen et al.1997).

8. Food Safety Risks of Milk

Unpasteurized and raw milks are the most widely available forms of milk consumed in most developing countries. This causes much morbidity. The most recent survey is discussed by Headrick et al. (1998). They found that most outbreaks of disease associated with raw milk occurred where raw milk sales are legal.

TB to be added

9. Conclusion

1. This paper does not seek to persuade readers to eschew dairy consumption. While that may be one result in readers formerly not up-to-date in the recent science, the goal is to persuade development agencies to cease promoting livestock and dairy production in developing countries for health, equity and environmental reasons.

2. When an activity raises threats of harm to public health or to the environment, precautionary measures should be taken even if in some cause and effect relationships are not fully established scientifically (Raffensperger 1999). Much of the case against dairy rests on firm epidemiological grounds. The correlations, while accepted as strongly suggestive by the medical, nutrition and scientific establishments, do not finally prove that dairy causes breast cancer. This 'Precautionary Principle' is a prudentiary measure to ensure economic development funds help rather than hinder its goals of poverty alleviation, promotion of health, and environmental conservation. The burden of proof is now on cattle proponents to show why their beef and dairy projects are not hazardous.

3. Human milk is ideal and almost essential for infants. The main exception may be where mothers may be HIV positive (but see #4 above). Weaning should be postponed as long as possible (eventually with supplements). Infants can thrive on formula. Cow's and other mammalian milks can be useful as components of infant's diets. However, many, if not most, children in the world (e.g., in China) thrive without dairy consumption. "There is no nutritional requirement for dairy products, and there are serious problems that can result from the proteins, sugar, fat and contaminants in milk products." (Barnard 1993, 1997). Milk consumption later in life, even in Caucasians, has recently been recognized to be associated with cancers, heart disease and other risks. It has become clear that animal-based diets contain no essential nutrients that are not also contained to better advantage in plant-based foods. Plant-based diets are inversely correlated with degenerative diseases.

4. Cattle and dairy production impose major environmental impacts, while being much less efficient in the use of natural resources, compared to grain and vegetable production. For these reasons, dairy production should not be financed by development agencies. In any event, dairy projects should be subjected to the highest category of environmental and other risk assessments in order to assess the impacts in the light of the new scientific findings. Dairy production is preferably left to the private sector, but only if the health, environmental and social risks are fully taken into account.

5. Development agencies wanting to fight poverty, approach sustainability, reduce social and environmental impacts, and improve efficiencies of natural resource use would preferably promote consumption of vegetables and grains, (5) and would demote dairy consumption. This could postpone or at least change the epidemiological transition. It would also improve the health of people in developing nations.

6. As many developing countries are making the 'epidemiological transition' from communicable disease to degenerative diseases, development agencies should support healthier diets, those with little or no animal fats, and by consumption of healthier foods.

7. To end on a positive note, there are many less costly and healthier alternatives to dairy. From 26 October 1999, soy product labels became permitted by the US Food and Drug Administration (FDA) to state that soy can reduce heart disease. After comprehensive studies, FDA concluded that foods containing soy protein included in a diet low in saturated fat and cholesterol might reduce the risk of heart disease by lowering cholesterol levels. Foods that may be eligible for the claim include soy beverages, tofu & tempeh. In order to qualify for the claim, foods must contain 6.25 grams of soy protein per serving; one-fourth of the 25 grams of soy protein daily that studies have shown is needed to show a significant cholesterol-lowering effect. In addition, soy inhibits blood-clotting mechanisms, maintains bone calcium levels with potentially beneficial effects in the prevention of osteoporosis, prevents menopausal symptoms, and reduces the incidence of breast cancer (Geissler 1999; Soy, 1999).


Sincere thanks for great support during the compilation of this paper to: Jeff Anhang, Neal Barnard, (President, Physicians Committee for Responsible Medicine), T. Colin Campbell (Director, Cornell-Oxford-China Diet and Nutrition Study), Samuel S. Epstein (School of Public Health, University of Illinois), Andrew Flood (National Cancer Institute), David Gordon, William Harris (Kaiser Permanente), Marco Jermini (WHO), Richard Pollard, and Aileen Robertson (WHO).

References Cited

Anderson, J.W., Johnston, B. M. and Remotely, D. T. 1999. Breast feeding and cognitive development: a meta-analysis. Amer. J. Clinical Nutrition 70(4): 525-535.

Barnard, N. D. 1993. Food for life: how the new four food groups can save your life. New York, Harmony Books 334 p.

Barnard, N. D., and Nicholson, A. 1997. Beliefs about dietary factors in breast cancer prevention among American women, 1991 to 1995. Preventive Medicine 26: 109-113.

Bertron, P., Barnard N., and Mills, M. 1999a. Racial bias in federal nutrition policy, Part I: the public health implications of variations in lactase persistence. J. Nat. Med. Assoc. 91(3): 151-157.

Bertron, P., Barnard N., and Mills, M. 1999b. Racial bias in federal nutrition policy: Part II: weak guidelines take a disproportional toll. J. Nat. Med. Assoc. 91(4): 201-208.

BMDP Statistical Software. BMDP New System for Windows. Los Angeles, 1994. ISBN 0-935386-30-0.

Biss, K, Taylor, C.B., Lewis, L.A., Mikkelson, B., Hussey, L.K. and Jey-Ho, K. 1970. The Masai's protection against atherosclerosis. Pathol. Microbiol. (Basel) 35 (1): 198-204.

Brouwer, A., Ahlborg, U.G., van Leeuwen, F.X. and Feeley, M.M. 1998. Report of the WHO working group on the assessment of health risks for human infants from exposure to PCDDs, PCDFs and PCBs. Chemosphere 37 (9-12): 1627-1643.

Cadogan, J., Eastell, R., Jones, N. and Baker, M. 1997. Milk intake and bone mineral acquisition in adolescent girls: randomized, controlled intervention trial. Brit. Med. Journ. 315 (7118): 1255.

Campbell, T.C., Chen Junshi, Brun, T., Parpia B., Qu Yinsheng, Chen Chumming, and Geissler C. 1992. China: From diseases of poverty to diseases of affluence: Policy implications of the epidemiological transition. Ecology of Food and Nutrition 27: 133-144.

Campbell, T. C. 1997. Associations of diet and disease: A comprehensive study of health characteristics in China. (31 pp.). Cambridge MA., Harvard University, Fairbank Center of East Asian Studies, Proc. Conf: Social Consequences of Chinese Economic Reform.

Campbell, T.C., Parpia, B., and Chen J. 1998. Diet, lifestyle and the etiology of coronary artery disease: the Cornell China study. Amer. J. Cardiol. 82 (10B): 18T-21T.

Carroll, K.K. and Khor, H.T. 1975. Dietary fat in relation to tumorigenesis. New York, S. Karger, Pharmacology: Lipids and tumors 308-345.

Cavallo, M. G., Fava, D., Monetini, L., Barone, F., and Pozelli, P. 1996. Cell-mediated immune response to beta-casein in recent-onset insulin-dependent diabetes: implications for disease pathogenesis. Lancet 348 (9032): 926-928.

Committee on Medical Aspects of Food Policy (COMA), 1997. Annual Report for 1996. London: COMA (UK Government advisory group).

Chan, J. M. et al. 1998. Plasma insulin-like growth factor 1 and prostate cancer risk: a prospective study. Science 279: 563-566.

Chen , Junshi., Campbell ,T.C., Li, J., and Peto, R. 1990. Diet, lifestyle and mortality in China: A study of the characteristics of 65 Chinese counties. Oxford, Oxford Univ. Press, Cornell University Press [and] Beijing, China People's Publishing House 894 p.

Chen, J. S., Geissler, C., Parpia, B., Li, J. and Campbell, T.C. 1992. Antioxidant status and cancer mortality in China. Int. J. Epidemiol. 21: 625-635.

Cohen, R. 1998. Milk: the deadly poison. Englewood Cliffs, NJ., Argus Publ. 317 p.

Coutsoudis, A. et al. 1999. Influence of infant-feeding patterns on early mother-to-child transmission of HIV-1 in Durban, South Africa: a prospective cohort study. Lancet 354(9177): 471-476.

Cramer, D. W. 1989. Lactase persistence and milk consumption as determinants of ovarian cancer risk. American J. Epidemiology 130(5): 904-910.

Cramer, D. W., Willett, W., Bell, D., Ng, W., Harlow, B., Scully, R. and Knapp, R. 1989. Galactose consumption and metabolism in relation to the risk of ovarian cancer. Lancet II: 66-71.

Cunningham, A.S. 1976. Lymphomas and animal-protein consumption (in 15 countries). Lancet 27 (2):7996: 1184-1186.

Davies, D.F. 1974. Food antibodies and myocardial infarction. Lancet I (7865): 1012-1024.

Davies, D.F. 1980. Cow's milk antibodies and coronary heart disease. Lancet 1(8179): 1190-1191.

De Long, Kent.199*. [lactose intolerance………...per ethnic group] tba

Ejobi, F 1996. Organochlorine pesticide residues in cow's milk in Uganda. Bull. Environ. Contam. Toxicol. 56 (4): 551-558.

Epstein, S.S. 1996. Unlabeled milk from cows treated with biosynthetic milk hormones: a case of regulatory abdication. International Journal of Health Services 26(1): 173-185.

Epstein, S.S. 1998a. The politics of cancer - revisited. Fremont Center NY., East Ridge Press, 770 p.

Epstein, S.S. 1998b. The breast cancer prevention program. New York, Macmillan (2nd. ed.): 416 p.

European Commission, 1999. Opinion of the scientific committee on veterinary measures relating to public health (SCVPH). Assessment of potential risks to human health from hormone residues in bovine meat and meat products. [In the context of WTO's 'Hormones' case, EU requested SCVPH's 'Opinion']: (30 April) 139 p.

Ferrer, J.F., Kenyon, S.J. & Gupta, P. 1981. Milk of dairy cows frequently contains a leukemogenic virus. Science 213 (4511): 1014-1016.

Fescanich, D. et al. 1996. Protein consumption and bone fractures in women. Amer. J. Epidemiology 143 (5): 472-479.

Flatz, G. 1989. The genetic polymorphism of intestinal lactase activity in adult humans (Vol 2: Ch. 122: 2999-3006) in Scriver, C.R. et al. (eds.) The metabolic basis of inherited disease (6th. Ed.) New York, McGraw-Hill, 2 vols.

Food and Agriculture Organization of the United Nations. 1997. Rome, FAO Production Yearbook.

Foy, R., So, J., Rous, E. and Scarffe J. 1999. Perspectives of commissioners and cancer specialists in prioritizing new cancer drugs: impact of the evidence threshold. Brit. Med. Journal 61 (318): 456-459.

FAO/WHO, 1999. Report of the 23rd. session of the Codex Alimentarius commission. Rome, FAO (28 June-3 July): 63 p.

Geissler, C. A. 1999. China: the soyabean-pork dilemma. In " Wheat or meat for the next millennium? Feeding the world in the future" Proc. Nutrition Society 58: 345-353.

Geissler, C.A., Parpia, B., Chen, J.S. and Campbell, T.C. 1998. Increased adiposity in Chinese adults. Proc. 16th Internat. Congress Nutrition PM 522: 0178.

Goodland, R. 1999. Livestock sector environmental assessment (239-261) in Hardtlein M. et al. (eds.) Nachhaltigeit in der Landwirtschaft. Berlin (E.Schmidt) Umwelt Stiftung, Deutsche Bundesstiftung Umwelt 421 pp.

Goodland, R. 1998a. The case against consumption of grain-fed meat (6: 95-115) in Crocker, D. and Linden, T. (eds.) The Ethics of Consumption. Lanham MD., Rowman and Littlefield 564 p.

Goodland, R. 1998b. Environmental sustainability in agriculture: the bioethical and religious arguments against carnivory (235-265) in Lemons, J., Westra, L. and Goodland, R. (eds.) Ecological Sustainability. Dordrecht, Kluwer 315 p.

Goodland, R. 1997. Environmental sustainability in agriculture: diet matters. Ecological Economics 23: 189-200.

Gordon, D.B. 1999. Milk and mortality: the connection between milk drinking and coronary heart disease. Livermore CA, Gordon Books 208 p.

Groenewegen, P.P., McBride, B.W., Burton J.H., & Elasser, T.H. 1990. Bioactivity of milk from bST-treated cows. J. Nutr. 120(5): 514-520.

Headrick, M.L., Korangy, S., Bean, N.H. and Angulo, F.J. 1998. The epidemiology of raw milk-associated foodborne disease outbreaks reported in the United States, 1973-1992. Amer. J. Pub. Health (Aug.) 88:1219-1221.

Heaney, R., McCarron, D., Dawson-Hughes, B., Oparil, S., Berga S., Stern J.S., Barr, S. & Rosen C. 1999. Dietary changes favorably affect bone remodeling in older adults. J. Amer. Dietetic Assoc. 99: 1228-1233.

Ho, Kang-Jey, Biss, K., Mikkelson, B., Lewis, L.A. & Taylor, C.B. 1971. The Masai of East Africa: some unique biological characteristics. Arch. Pathol. 91 (5): 387-410.

Hu, F. B. and Willett, W. C. 1998. The relationship between consumption of animal products (beef, pork, poultry, eggs, fish and dairy products) and the risk of chronic diseases: a critical review. A report for the World Bank. Cambridge, Massachusetts: Harvard School of Public Health, Nutrition Department (ms): 29 p.

Isono, Y., Shingu, I. & Shimizu, S. 1994. Identification and characteristics of lactic acid bacteria isolated from Masai fermented milk in Northern Tanzania. Japan Society for Biosci. Biotech. Agrochem. 58 (4): 660-664.

Kanis, J.A. 1997. Osteoporosis: a view into the next century. Neth. J. Medicine 50(5): 198-203.

Kanis, J.A. 1999. The use of calcium in the management of osteoporosis. Bone 24(4): 279-290.

Kannen, K. Tanabe S Giesy J P & Tatsukawa R. 1997. Organochloorine pesticides and polychlorinated biphenyls in foodstuffs from Asian and oceanic countries. Rev Environ. Contam Toxicol. 152: 1-55.

Karjalainen, J. 1992. A bovine albumen peptide as a possible trigger of insulin-dependent diabetes mellitus. New England J Med 327 (5): 302-307.

Keys, A. (ed.) 1970. Coronary heart disease in seven countries. American Heart Association, Monograph 29, Circulation, 41 Supplement: 211 p.

Keys, A. (ed.) 1980. Seven countries: a multivariate analysis of death and coronary heart disease. Cambridge MA., Harvard Univ. Press 381 p.

Kimbrough, R.D. 1995. Polyclorinated biphenyls and human health: an update. Crit. Rev. Toxicol. 25(2): 133-163.

Kurian, T. 1991. The New Book of World Rankings. NY., Facts on File ISBN 0-8160-1931-2.

Law, M. R. et al. 1994. An ecological study of serum cholesterol and ischaemic heart disease between 1850 and 1990. European J. Clinical Nutrition 48 (5): 305-325.

Mann, G.V., Spoerry, A, Gray A., and Jarashow D. 1972. Atherosclerosis in the Masai. Amer. J. Epidemiology 95: 26-37.

Mettlin, C. 1989. Milk drinking and other beverage habits, and lung cancer risk. Int. J. Cancer 43 (4): 608-612.

Mettlin, C., Selenskas, S., Narayan N and Huben R. 1989. Beta carotene and animal fats and their relationship to prostate cancer risk: a case-control study. Cancer 64 (3): 605-612

Outwater, J. L., Nicholson, A. and Barnard, N. 1997. Dairy products and breast cancer: the IGF-1, estrogen, and bGH hypothesis. Medical Hypotheses 48 (6): 453-461.

Petridou, E., Syrigou E., Willett W. and Trichopolous D. 1996. Determinants of age at menarche as early life predictors of breast cancer risk. Int. J. Cancer 68 (2): 193-198.

Potter, S. M. 1998. Soy protein and cardiovascular disease; the impact of bioactive components in soy. Nutrition Reviews 56: 231-235.

Raffensperger, C. (ed.) 1999. Protecting public health and the environment: Implementing the precautionary principle. Washington DC., Island Press 350 p.

Rees, M. 1995. The age of menarche. Orgyn 4: 2-4.

Roeder, R.A., Garber, M.J. & Schelling, G.T. 1998. Assessment of dioxins in foods from animal origins. J. Animal Sci 76(1): 142-151.

Rogan, W.J. 1996. Pollutants in breast milk. Arch. Pedriatr. Adolesc. Med. 150(9): 981-990.

Rowe, J. W., Andres R., Tobin J.D., Norris A.H. & Shock N.W. 1976. The effect of age on creatinine clearance in men: a cross sectional and longitudinal study. J. Gerontology 31(2): 155-163.

Salvadori del Prato, O. 1990. Trattato di tecnologia casearia. Milano, Ed. Agricole (1070): 279 p.

Sanders, T.A. and Reddy, S. 1994. Vegetarian diets and children. Amer. J. Clinical Nutrition 59 (5 suppl; May): 1176S-1181S.

Scott, F. W. 1990. Cow milk and insulin-dependent diabetes mellitus: is there a relationship? American J. Clinical Nutrition 51(3): 489-491.

Segall, J.J. 1977. Is milk a coronary health hazard? Brit. J. Preventive and Social Med. 31: 81-85.

Segall, J.J. 1994. Dietary lactose as a possible risk factor for ischaemic heart disease: review of epidemiology. International J. Cardiology 46: 197-207.

Seely, S. 1981. Diet and coronary disease: a survey of mortality rates and food consumption statistics of 24 countries. Medical Hypoth. 7: 907-918.

Sonawane, B.R. 1995. Chemical contaminants in human milk: an overview. Environmental Health Perspectives 103 (Suppl) 6: 197-205.

Soy, 1999. Third International Symposium on the role of soy in preventing and treating chronic disease, 10/31-3/11, Washington DC;

Sun, Z. and Cade, J.R. 1999. A peptide found in schizophrenia and autism causes behavioral changes in rats. Autism 3(1): 85-95.

Tominaga, S., Aoki, K, Fujimoto, I. & Kurihara, M. 1994. Cancer mortality and morbidity statistics, Japan and the World-1994. Boca Raton FL., CRC Press, [&] Japan Scientific Societies Press. ISBN 0-8493-7748-X

Uauy, R. and Peirano, P. 1999. Breast is best: human milk is the optimal food for brain development. Amer. J. Clin. Nutrition 70: 433-434.

UNICEF/WHO/UNAIDS, 1999. [statement on breastfeeding and HIV transmission]

Willett, W., Stamfer, M., Colditz, G., Rosner, B. Hennekens, C. & Speizer, F. 1987. Dietary fat and the risk of breast cancer. N. Engl. J. Med. 316: 22-28.

Willett, W. & Stamfer M. 1992. Dietary fat and fiber in relation to risk of breast cancer: an eight year follow-up. JAMA 268: 2037-2044.

Willett, W., Giovannucci, E., Rimm, E., Colditz, G., Stampfer, M., Ascherio, A., and Chute, C., 1993. A prospective study of dietary fat and risk of prostate cancer. J. National Cancer Inst. 85 (19): 1571-1579.

World Bank, 1997. Sector strategy: Health, nutrition, & population. Washington, D.C.: World Bank: 112 p.

World Cancer Research Fund (WCRF) and American Institute for Cancer Research (AICR), 1997. Food, Nutrition and the Prevention of Cancer: A Global Perspective. Washington, D.C.: WCRF & AICR: 670 pp.

Yrjanheikki, E.J. (ed.) 1991. Levels of PCBs, PCDDs, and PCDFs in human milk and blood: second round of quality control studies. Copenhagen, WHO Regional Office 75 p.

Yu, H., Spitz, M.R., Mistry J., Gu,J., Hong WK, & Wu XF. 1999. Plasma levels of insulin-like growth factor-1 and lung cancer risk. J. National Cancer Institute 91(2): 151-156.

Zatonski, W. 1996. Evolution of health in Poland Since 1988. Warsaw: Maria Sklodowska-Curie Cancer Centre and Institute of Oncology: 38 p.

Zatonski, W.A., McMichael, A.J. and Powles, J.W. 1998. Ecological study of reasons for sharp decline in mortality from ischaemic heart disease in Poland since 1991. Brit. Med. J. 316: 1047-1051.

Annex: Other Risks Associated with Dairy Consumption

These notes are annexed because consensus has not been totally reached, and the evidence is still coming in.

1. The Risk of Milk from Cows Treated with Genetically Modified Hormones

This section applies mainly to US milk and milk derivatives where artificial hormones are commercially used on most cow herds to boost milk production. The scientific evidence is increasing that bovine somatotropin (BST) is not safe for humans. Epstein (1996) has thoroughly documented these risks. In January 1999, Health Canada, in cooperation with the Royal College of Physicians and Surgeons, and the Canadian Veterinary Medical Association ruled that recombinant bovine growth hormone (rBGH) increased the incidence of mastitis, (6)

lameness and reproductive problems, and disapproved future sales of rBGH. Australia, New Zealand and Japan have banned BGH. The European Union banned rBST in 1994, Canada also refuse to authorize use of BGH. The US FDA has been petitioned to ban rBGH milk.

The European Commission's Science Committee (EU, 1999) concludes that the naturally occurring insulin-like-growth factor 1 (IGF-1) (7)

reaches excess levels in rBGH milk, where it poses major risks of cancer, particularly of the breast, lung cancer, melanoma, and pancreatic cancers. IGF-1 also is a potent risk factor for prostate cancer (Chan 1998). BGH in milk increases IGF-1 levels and is not destroyed by pasteurization. IGF-1 stimulates or initiates the growth of human breast cancer cells, and acts synergistically with BGH.

The UN Codex Alimentarius Commission (FAO/WHO, 1999) representing 101 nations worldwide did not reach consensus on the maximum 'residue' limits of Bovine Somatotropins (BST) or Bovine Growth Hormones (rBGH). It therefore decided to stick to the previous ('Step 8') residue limits, until a later date at which consensus might be reached (para.75-78). Thus, the 1993 European Commission's moratorium on genetically engineered milk (rBGH) has been extended. This appears to run counter to the 1998 findings by FAO's Food Additives Committee (JEFCA), which claimed that maximum residue limits for BST milk are unnecessary (Para. 75). The EC's decision to uphold its ban on rBGH milk suggests that the EC also finds rBGH milk to be unsafe.

2. Other Diseases Associated with Dairy Consumption

2.1 Prostate Cancer and Dairy Consumption: Prostate cancer is the most common cancer in US men and the second leading cause of cancer mortality. The scientific evidence for a link between prostate cancer and dairy consumption has been building in recent years, but has only just become confirmed. Bovine growth hormone (BGH) was first synthesized in the early 1980s using recombinant DNA biotechnology. In 1993, the US FDA approved the commercial sale of milk from cows injected with rBGH. When Bovine Somatotropin (BST) is injected into cows, their blood-IGF-1 levels rise by about 80%; so do the levels of IGF-1 in milk. In 1996, Epstein had warned that high levels of IGF-1 contained in milk from cows injected with synthetic bovine growth hormone would be risky. Even the dairy industry (Heaney et al.1999) admits that IGF-1 increases in the human body after ingestion of milk. Serum IGF-1 levels increased about 10%, at a highly significant level, above baseline in the Heaney et al. (1999) study. In other words, one glass of milk almost doubles IGF levels in human blood. Epstein (1998) found that the IGF increase can reach nine times normal levels. FDA's ruling that IGF-1 is destroyed by digestion now appears to be incorrect. (8)

Recently, the evidence for a strong link between prostate cancer risk and a high level of IGF-1, the naturally occurring insulin-like-growth factor 1, became indisputable. IGF-1 stimulates the growth of normal and cancerous cells, as well as prostate cells. Harvard Medical School researchers concluded that IGF-1 is a potent risk factor for prostate cancer (Chan et al. 1998). Pasteurization may actually increase IGF-1 concentration in milk (Cohen 1998, Groenewegen et al. 1990). IGF-1 can crossthe intestinal wall in humans and enter the bloodstream. The journal 'Cancer' (1989) concludes that milk drinking increases prostate cancer risk.

2.2 Leukemia: Bovine leukemia virus is found in three of five cows in the US, involving 80% of herds. As milk is pooled, 90%-95% of US raw milk is contaminated. (Ferrer et al.1981). German and Swiss milks are similar. Most animals exposed to bovine leukemia virus develop leukemia. Ditto human cells in vitro. Human leukemia rates are highest in those states with the most dairy.

2.3 Cancer of the Lymphatic Organs: Milk increases such cancer (Foy et al.1999). Cramer (1989) found that ovarian cancer correlates with milk consumption. Cramer & Willett (1989) published on ovarian cancer and galactose.

2.4 Cancer of the Lung from Milk Consumption: Mettlin (1989) found drinking milk three or more times a day doubled lung cancer risk compared to those who reported never drinking milk, when controlled for smoking, age, education and other variables. Yu et al. (1999) identified milk IGF-1 as the key factor in the growth of lung cancer.

2.5 Attention Deficit Disorder: May be caused by the milk hormone beta-casomorphin-7 (Sun & Cade 1999) which is elevated in the blood and urine of patients with schizophrenia and autism.

2.6 Dairy Consumption and Kidney Cancer: WCRF/AICR (1997) is clear that daily milk consumption increases the risk of kidney cancer. Ecological studies found associations between both animal protein and animal fat with cancer of the kidney.

2.7 Asthma and Diabetes: (insulin dependent diabetes mellitus): Scott (1990) found a significant positive correlation between consumption of unfermented milk and child diabetes in various countries (9)

(as well as heart disease, arthritis, leukemia, & lymphoma in adults).

Corrections and comments to: <>

1. Environmental Assessment (EA) of all livestock and dairy projects under consideration by international development should be EA category "A". Such agencies normally categorize their projects into A, B, and C. Category A projects need a thorough and comprehensive EA. These are typically large-scale projects with major impacts, such as a big reservoir, or major uncertainties. B projects need only an environmental mitigation plan. Category C projects (education, telecommunications) need no EA at all. As livestock and dairy projects are so risky on environmental, social and health grounds, the EA category A should be automatic.

2. Avocado and palm oils, for example, contain much fat, some of which is saturated, but zero cholesterol.

3. Goat's and human milks contain about 10% fat, sheep's milk 17%, and reindeer 48.6% fat.

4. Peeled potatoes contain about 0.6g of protein per standard (1 oz or 28g) serving. Whole potatoes contain more than double that, about 2g of protein. Peeling also reduces the micronutrient content.

5. This paper does not focus on the best balance between vegetables, grains and fruit in diets. However, Dr William Harris ( 1999) notes that while both are needed for global nutrition, vegetables offer more balanced nutrition. Averages of 93 vegetables turned out to be deficient only in vitamin B12. Strictly cereal grain diets can occasionally be slightly low in vitamins A, B2, B12, C, E, folic acid, and calcium. While grains may not be optimal nutrition, they are good enough to support most of the 6bn of us as of October 1999. As shown by Harris (1999, see graph and table), vegetables and grains offer healthier, more complete and far less risky diets than do animal products. The production and distribution of vegetables also merits investment by development agencies.

6. BGH increases milk yield. In 1800 the average yield was two quarts per cow daily. Today's US average is 24 quarts a day (18,000 lbs./year) and a maximum of 50 quarts a day. Mastitis and other infections result in 80% of herds given rBGH. That is the main reason BGH cows have to be injected with antibiotics, such as sulfa drugs, and penicillin; US FDA had to raise permissible limits for antibiotic residues in milk from one part per 100 million to 1 ppm. Antibiotic residues end up in the milk and can cause another range of problems, such as resistance. USDA permits milk to contain up to 1.5 million leukocytes (pus) per ml (1/30th oz). Many infectious diseases transmitted by raw milk, such as Brucellosis and Yersinia, are controlled in OECD, but less so in LDCs.

7. IGF-1 is a powerful hormone produced in the liver and body tissues of all mammals. It was discovered in 1979, found to be identical in bovines and humans in 1989, and is causing US FDA's biggest controversy from 1994 to the present. IGF-1's structure is very similar to that of proinsulin. IGF-1 is regulated by the human growth hormone and peaks at puberty. IGF-1 declines to half the levels of puberty at 70 years of age. Mature men with high IGF-1 levels are eight times more likely to develop prostate cancer than men with low levels. BST-treated cow's milk contains from twice to ten times the IGF-1 levels in normal milk, and is concentrated during pasteurization.

8. The American Council on Science and Health (3/'98) used to dispute first, that IGF-1 concentration increases in milk; and second that IGF-1 blood levels rise following milk ingestion. The answer to both now is affirmative, according to the US Dairy Industry (Heaney et al. 1999).

9. Finland has the one of the world's highest rate of dairy consumption per capita and the world's highest rate of insulin dependent diabetes. Juvenile diabetes (Type 1) occurs when the immune system attacks pancreatic -cells that produce insulin for the body. Antibodies produced against the milk protein during the first year of life destroy the pancreas during the auto-immune reaction. Bovine serum albumen is the milk protein responsible for the onset of diabetes (Karjalainne, 1992). Cavallo (1996) found -casein antibodies present in over one third of IDDM patients, but relatively non-existent in healthy individuals. There is no cure for juvenile diabetes, and failure to control blood sugar levels can cause blindness, limb loss and early death.