Positive Health Online

Last year I was asked by the Bristol Cancer Help Centre to investigate what effect agriculture might have on cancer risk for consumers of agricultural products. The question is whether conventional agriculture has an important influence on cancer risk in the UK and whether organically grown food reduces the risk of cancer or can play a significant role in the promotion of health of those with cancer.

The question arose out of the Centre's concern about the effect of the intensive, industrial-style agriculture on food quality and health, particularly on cancer. Food contamination by agrochemicals is popularly believed to be an important cause of the assumed increase in cancer rates over the last 45 years. Quite apart from contaminants, proponents of organic farming argue that organic food is more vital and better able to promote health than intensively-grown food. Many in the alternative health movement advocate eating organic food in order to prevent cancer or the recurrence of cancer and to promote health. Those people who buy or grow organic food are mainly motivated by health reasons.

What Is the Evidence?

How do we know whether eating organic food reduces the risk of cancer? Scientific studies should be able to confirm or refute the common sense view held by so many in the alternative health and organic farming movements. But scientific research in this area is limited. In the UK and the USA, very little research into the nutritional value of organic food has been funded. Fortunately, there is more interest and more money on the continent, particularly in Germany. So there is some evidence around but it is not particularly easy to interpret this evidence and come up with solid conclusions.

Although it is tempting to select just those studies which confirm one's beliefs, it is more honest to base conclusions on an evaluation of the quality of the evidence and on trends emerging from many studies. No single study on its own, no matter how well designed, will give a conclusive result that can be quoted as proof that all organic food is healthier than conventional food. Proof arises when the same result is reproduced in a number of studies and not often contradicted.

To find the evidence, I searched as widely as possible, looking in agricultural and medical databases, review papers (particularly a German review of 150 papers by Woese and colleagues published in 1995), PhD theses, conference reports, organisations such as the Elm Farm Research Centre and the Soil Association, libraries, books and World Wide Web pages.

The strongest, most direct evidence of the health benefits of organic food would come from studies of health outcomes in people. No such studies have been done. If they were, they would have to be thoughtfully designed and carefully implemented to demonstrate the effects of the food being organic rather than the effects of a healthy diet, healthy lifestyle, smoking or social class. No one has yet done a study comparing the health of people who eat a healthy diet of organic food with the health of people who eat an equally healthy diet of conventional food. For some time, there has been a general scientific consensus that diet is responsible for 30 to 60% of cancers in developed countries. A large body of research evidence has shown that a healthy diet means eating a lot of fruit and vegetables and unprocessed cereals, not eating a lot of fatty foods, red meat, salt, pickled foods and alcohol, and keeping a balance between calorie intake and energy expenditure. As most people who eat organic food also eat a healthy diet, it would not be easy to disentangle the effects of eating organic food from eating a healthy diet.

Less direct evidence comes from studies on animals where the diet can be controlled more precisely than with people. There are at least 35 studies on mammals or birds comparing the effect of organic and conventional feed on growth rates and reproductive performance (such as litter size, percentage of babies born dead, interval between litters, egg production and hatchability in chickens). Although there were a few studies about disease resistance, none were about cancer. Some of the studies demonstrated improved health on organically grown feed for some of the health outcomes measured but such different measures were used that it's not possible to find a consistent result. In any case, it is always tricky to extrapolate from animals to people. An interesting result from a few studies is that experimental animals clearly prefer organic feed when given a choice, regardless of their previous diet. This isn't proof that organic feed is healthier but is an interesting observation.

Indirect evidence comes from comparisons of measurements of nutrients and contaminants in organic and conventional food. The evidence is indirect because it is not based on measurements of cancer rates in people but on the assumption that food contaminated with carcinogens could increase the risk of cancer and that food containing high levels of certain nutrients could reduce the risk of cancer. Although reductionist, we cannot afford to dismiss such evidence, because it is all we have to go on and because it makes sense to think that the levels of these chemicals in food are related to cancer rates. The rest of this report is about studies based on levels of nutrients and contaminants in organic versus conventional food.

Does organic food have more cancer-preventing nutrients than conventional food? (see Table 1)

A commonly voiced concern is that conventional agriculture might produce crops with a lower content of these health-protecting nutrients.

One theory is that conventional methods reduce the amount of organic matter in the soil by the continued use of synthetic fertilisers, pesticides, heavy tractors, deep ploughing, leaving land fallow and applying excessive amounts of phosphorous. When there is less organic matter in the soil, soil micro-organisms such as the mycorrhizal-forming fungi are less numerous and less active. These fungi are particularly effective at increasing the uptake of minerals such as selenium from the soil. They may also influence the plant's ability to synthesise vitamins.

However, the level of nutrients in plants is also under the influence of other environmental and genetic factors, only some of which can be related exclusively to organic or to conventional agricultural practices. An example of the importance of genetic factors is the content of beta-carotene in different varieties of carrots. Commercial carrot varieties vary nearly three-fold in beta carotene, ranging from 110 mg/kg in a Savoy variety to 270 mg/kg in the Beta III variety. There is a long list of environmental factors other than the type of fertiliser which could raise or lower the level of nutrients in the food:

1. the nitrogen application rate (how much nitrogen is in the fertiliser) – for example, with one variety of spinach, the amount of vitamin C was the same when the lower rate of fertiliser was applied, whether in the form of compost from cattle manure, from plant wastes or as synthetic fertiliser. At a higher nitrogen application rate, the vitamin C content remained the same with the synthetic fertiliser and increased with both composts.

2. the type of organic fertiliser. What is true for bushwood compost may not apply to organic fertiliser made from swine manure or horn shavings. The composition of organic fertilisers varies from one type to another.

3. the time of year when fertiliser is applied,

4. geographical location and mineral content of the soil. In a USA study, average values for iron in tomatoes ranged from the lowest in Indiana (52 parts per million) to the highest in Colorado (265 parts per million). Calcium values tended to increase and magnesium values to decrease from south to north while sodium and manganese tended to decrease from east to west.

5. the climate (light intensity, rainfall and temperature) – In a study done in Vermont, USA, six week old identical tomato plants were distributed to organic and conventional growers who planted them in their farms and treated them with the same cultivation techniques used on their own plants. Higher vitamin C and carotene contents were found in organic tomatoes from two farms which supported the plants by wire cylinders, thus exposing the fruit to sunlight.

6. seasonal effect – In the same Vermont study, there were significant differences in vitamin C between organic and conventional tomatoes of two varieties. In the 1974 season, conventional tomatoes had the higher values but in the 1975 season, the higher values were from organic. If the study had stopped after one season, it would have been presented as evidence of the nutritional superiority of conventional tomatoes. When repeated the following year, the results imply a seasonal effect, rather than the method of fertilisation.

7. plant age at harvest – Many kinds of fruit such as apples, peaches and apricots contain very little vitamin C when unripe. For example, unripe apples contain no vitamin C, half-ripe apples contain vitamin C at a concentration of 18mg/100g and when fully ripe, vitamin C increases to 60 mg/100g. A study comparing unripe organically grown apples with ripe conventionally grown apples would not be a fair comparison.

8. management practices other than fertilisation. These include greenhouse cultivation, ploughing, type of plant cover, and extent of tillage.

9. Post-harvest handling. Once a crop is harvested, nutrient losses are inevitable, especially of some of the more sensitive vitamins such as vitamin C. These losses can be minimised to some extent if the food is stored under appropriate conditions. For example, in one study, kale stored at room temperature (21°C) had 60% of the vitamin C content after two days as did kale stored in the refrigerator (0°C) in humid conditions. Temperature, humidity, light, oxidation, and alkalinity affect vitamins differently during storage. The vitamin C content of potatoes drops as much as 70% even when the temperature and humidity are optimal while riboflavin and niacin levels hardly change. Losses in riboflavin are increased by exposure to light and thiamin by alkalinity. To detect differences in vitamin content resulting from the type of fertiliser, it is crucial that the study be designed to have the same storage and transport conditions.

When I first started the research project, I wasn't sure I would find any relevant studies but after considerable delving, I finally identified a total of 355 references. These studies mostly compared levels of minerals and vitamins in organic versus conventional foods. I didn't find any comparative studies of the phytochemicals. Despite the encouraging number of studies, I found that it is not possible to determine whether there is a difference in vitamin and mineral content between organic and conventional crops. The quality of the studies was often poor, usually because the researchers did not taken into account the other environmental and genetic factors. Unless all of these factors are accounted for, it is not possible to know what caused the results. A higher content of vitamin C in organic potatoes may have nothing to do with the use of organic fertiliser but could be due to the fact that the organic and conventional farmers planted different varieties of potato or to any of the environmental factors listed above. If organic cultivation did cause a significant increase in vitamin and mineral content, you would expect such an increase to show up in most of the studies. Since it doesn't, it is doubtful whether the type of fertiliser, meaning, whether the fertiliser is organic or synthetic, is as important a factor as other agricultural and post-harvest influences on nutritional value for which there are consistent and nutritionally significant differences.

Carcinogens

Carcinogens may contaminate food as a result of the deliberate application of agrochemicals (pesticides, synthetic fertilisers, growth promoters) and of the unintentional contamination by pollutants from industry, waste disposal, transport and mining. Since the focus of my research was on agriculture, I didn't look at studies about other sources of contaminants on food.

Not all carcinogens are synthetic chemicals. Natural pesticides are synthesised by plants as protection against pests. These are as likely to be carcinogenic as synthetic chemicals in animal carcinogenicity tests but have not been studied as extensively. Consequently little is known about the influence of agricultural practices on their content in plant-foods. Organic food may have higher levels of natural pesticides if organic farmers cultivate pest resistant varieties more often than conventional farmers. Or the absence of synthetic pesticides could force the plant to synthesise and accumulate higher levels of natural pesticides. It may be that conventional agriculture forces the plant to put more of its energy into growth and high yield resulting in less energy for the synthesis of natural pesticides. It is equally possible for the type of agriculture to have no influence on levels of natural pesticides.

Among the many chemicals used in conventional agriculture, pesticides and nitrates from synthetic fertiliser use have been of great concern and were covered in the report in detail.

Do synthetic pesticide residues on food increase the risk of cancer?

It might seem unnecessary to even ask this question as it is well known that some of the pesticides used in agriculture are carcinogenic.

Laboratory tests and experimental animal tests confirm that organochlorine pesticides such as DDT and lindane can cause cancer, at least in rats and mice. These pesticides do not initiate the cancer process but are able to promote the process once it is initiated by something else. It is theoretically possible that exposure to a single carcinogenic molecule could trigger the cancer process in people. But our bodies have considerable defence mechanisms to cope with low level exposure. The question about exposure is ultimately about how much is ingested. "The dose makes the poison" is the common sense view of toxicologists who study toxins and carcinogens.

Each year, the Ministry of Agriculture, Fisheries and Food (MAFF) monitors the food supply for pesticide residues and estimates the amount of pesticides we consume through our diet, based on national surveys of food consumption. An Acceptable Daily Intake, ADI, is determined for each pesticide. This is done by subjecting experimental animals to a range of pesticide doses, finding the dose at which no damaging health effect (including benign tumours) was observed, in the most sensitive animal species. To allow for a safety margin, this dose is then divided by a factor of 100 to 1000 to determine the ADI. The ADI is "the amount of the pesticide which can be consumed every day of a person's entire lifetime in the practical certainty, on the basis of all known facts, that no harm will result". Daily intakes of organochlorine pesticides such as DDT and lindane have always been and still are far below the ADIs in this country.

Maximum Residue Levels, MRLs, are set for a pesticide in a particular food. For example, the MRL for DDT in milk is 0.04 mg/kg, for DDT in apple juice it is 0.05 mg/kg, for DDT in wheat it is 0.01 mg/kg, for lindane in milk it is 0.008 mg/kg. The MRL is the highest level of pesticide that should be on the food when the pesticide has been used correctly. MRLs are not safety limits but are set with the intention that typical intakes will be within the Acceptable Daily Intake set for the pesticide. If the Maximum Residue Level is occasionally exceeded, it does not mean that health is at risk. The Acceptable Daily Intake is the safety limit set for lifetime exposure at the Maximum Residue Level. In 1994, residues were detected in 30% of more than 3,700 samples of food but only 23 samples were above the MRL. Fruit and vegetables in the UK are not a major source of exposure to organochlorine pesticides. None were detected in fruit and vegetables in 1994. The storage of organochlorine compounds in fat and their bio-accumulation up the food chain means that meat, fish and dairy products tend to contain higher residues than plant-foods. However, the higher exposure is still not higher than the Acceptable Daily Intakes set for those pesticides.

Whether foods containing pesticide residues below the MRLs can be considered safe depends on confidence in MAFF's monitoring programme and on their basis for calculating ADIs and MRLs. Short of saying that any detectable residue is unacceptable, some method of risk assessment has to be used. The method used by MAFF has been validated and revised in consultation with the World Health Organisation.

A diet of organic food should eliminate the main source of exposure to pesticides. Although rare, pesticide residues have been detected in organic food in some MAFF surveys. Organic food can be contaminated with pesticides by cultivation on previously contaminated soil; by unauthorised use of pesticides by the organic farmer; by spray drift from nearby conventional farms; by application of contaminated sewage sludge or by contamination during transport, processing and storage.

As pesticide levels in conventional food have been either negligible or below the Maximum Residue Level in most studies, it is not strictly correct to claim that conventional food is more contaminated than organic food. There are studies showing a lower pesticide content in organic food and others which show no difference in pesticide content. In a German review paper of 150 studies, the authors conclude that there are no striking differences in pesticide content between organic and conventional cereals, potatoes, vegetables, fruit, wine, bread and milk but the poor quality of many of the studies makes generalisations difficult.

Any residue is undesirable but the extent of exposure from pesticide residues on food has to be seen in comparison with other sources of exposure to pesticides. I focussed on organochlorine pesticides because they are persistent and degrade very slowly and are thus more likely to remain as residues than other kinds of pesticides.

Comparison with other sources of exposure

The amount of exposure we have to organochlorine pesticides from eating contaminated food is insignificant compared to exposure from public health programmes, to occupational exposure and to exposure in developing countries.

1. Public health programmes – DDT, chlordane and other organochlorine pesticides are used to control malaria mosquitoes, ticks carrying Chagas Disease, and termites, exposing people to much higher levels of pesticides than people in the UK have ever received. Application rates in public health vector control are more than ten times higher than agricultural applications. Yet cancer rates in developing countries where these programmes have been in operation for decades are lower than in developed countries.

2. Occupational exposure – The body burden of DDT in pesticide manufacturing workers is 10 – 1000 times greater than in the general population with no obvious increased risk of cancer. Agricultural workers and their families are also exposed to higher levels of pesticides than are members of the general public. Yet agricultural workers have a lower rate of cancer than the general population.

3. Developing countries – Exposure to organochlorine pesticides is substantially higher in developing than in developed countries but the health consequences of such exposure is not well understood due to poor monitoring and lack of research. Exposure is higher because public health vector control programmes are more common; use of pesticides is less well regulated and controlled and because organochlorine compounds that have been banned in developed countries are often legally produced and used in developing countries.

Comparison with exposure to other pesticides

The largest source of dietary exposure to toxic and carcinogenic chemicals is to natural pesticides. Food preparation and cooking reduce toxicity but do not purify the diet. Only 52 of the thousands of natural toxicants present in plants have been tested for carcinogenicity. Of these, 29 are carcinogenic in rodent tests. Professor Ames and colleagues produced a quantitative ranking of both natural and synthetic carcinogens. For each compound, they calculated the ratio of lifelong daily human exposure (HE) to the potency of the carcinogen in rodents (RP). The higher the HERP (the Human Exposure/Rodent Potency), the greater the risk of cancer. Their calculations were based on US exposure data. Examples are listed in Table 2.

This ranking is not evidence that apples, lettuce and parsnips are an important cause of cancer. Indeed, a diet rich in plant-foods reduces the risk of cancer. But it gives an indication of the extremely low risk posed by exposure to pesticide residues in food, putting the risk of cancer from synthetic and natural compounds into perspective.

Cancer-preventing nutrients in food

Cancer risk is not simply dependent on the presence of carcinogens but on interactions between carcinogens and anti-carcinogens in foods and the interactions between these compounds and the body's defence systems. Very little is known about the amount of natural toxins and natural cancer-preventing nutrients in different foods or in the overall diet. Laboratory studies may show that a chemical extracted from a plant is carcinogenic but epidemiological studies may show that eating the plant protects against cancer. Other natural constituents may be present in the same plant or in the same meal which may offer protection from the toxin, be it natural or synthetic.

Whether a compound initiates or inhibits the cancer process may depend on how much you are exposed to. Some chemicals, such as caffeic acid found naturally in many fruits and vegetables, are carcinogenic at high doses and protective at low doses. At a concentration of 2% caffeic acid, rats develop stomach tumours. At a concentration of 0.0005%, caffeic acid inhibits cancer from developing.

Comparison with carcinogens formed in the body

Professor Ames of the University of California argues that by-products of our normal metabolism can damage DNA on a massive scale.

These by-products are chemicals such as hydrogen peroxide and other reactive forms of oxygen. Damaged DNA can be repaired or it may be destroyed. At worst, it remains as a stable mutation, in which case there is the possibility of it developing into cancer. DNA repair mechanisms are not always 100% effective, and become even less so as we age (which is why cancer is more common in older people).

Professor Ames estimates that the rate of attacks on DNA from by-products of metabolism is in the range of 10,000 hits per cell per day. This is so high that it is hard to imagine how DNA-damaging chemicals in the environment can substantially add to it, especially at the extremely low levels in which organochlorine pesticides are found in the diet.

If you compare the tiny amount of pesticide residues we are exposed to in food with the much larger exposure to synthetic and natural pesticides from other sources and when you consider the extensive exposure we have to natural carcinogens formed daily in the body and to natural anti-cancer nutrients in our food, it is highly unlikely that pesticide residues on food are a significant cause of cancer.

Does contamination with nitrates from fertiliser use increase the risk of cancer?

How agriculture contributes to nitrate contamination of food

The artificial, synthetic fertilisers used in conventional agriculture supply the main plant nutrients in soluble form. Organic fertilisers supply the nutrients as nitrogenous salts and organic compounds which are converted to the same soluble forms by micro-organisms in the soil. Nitrate is the main form of nitrogen supplied by the soil and supplemented by both types of fertiliser. Food-plants become contaminated with nitrates when crops take up more nitrate than they can use. The nitrate may originate from unfertilised soil or from the excessive application of either synthetic or organic fertiliser.

Are nitrates carcinogens?

Nitrates are not carcinogenic but are readily converted into highly carcinogenic compounds called nitrosamines. This process occurs during some food processing techniques but mainly occurs within the body, particularly in the mouth and stomach. Vitamins C and E can slow down or stop this conversion process. Of more than 300 nitrosamines tested, about 90% are positive in animal carcinogenicity tests. Nitrosamines are capable of both initiating and promoting the cancer process.

Are we exposed to nitrates?

Nitrates occur in many foods but vegetables are the main source. More than three quarters of the average dietary nitrate intake is from vegetables. Certain types of vegetables accumulate more nitrate from the soil than they can convert into protein. Examples are spinach, lettuce, broccoli, and beetroot. Vegetables that do not tend to accumulate nitrate are broad beans, Brussels sprouts, carrots, cauliflower and potatoes. Some plant-foods such as cereal grains contain no nitrates.

The more fertiliser that is applied to a nitrate-accumulating vegetable, the higher the nitrate content. There are large and significant differences in nitrate content between varieties. Nitrate accumulates more in some parts of the plant than in other parts. The amount of sunlight received before harvest affects nitrate accumulation. Nitrate levels are higher in plants grown in winter and in northern European countries compared to southern countries.

Two recent reviews conclude that the use of organic fertiliser does lead to significant reductions in nitrate levels in nitrate-accumulating vegetables. Organic potatoes may have lower nitrate levels but the results are not statistically significant. Seed, fruit and bulb vegetables do not accumulate nitrate and results from a number of studies are conflicting. For fruit, there are no obvious differences. As an example, in a British study, lettuces fertilised with composted farmyard manure had significantly less nitrate, 1184 parts per million, than lettuces fertilised with synthetic fertiliser, 1410 parts per million. However, significant differences in nitrate content between organic and conventional vegetables do not mean that the differences are large nor that the higher concentrations in synthetically-fertilised vegetables are a significant health risk. The natural range of nitrate concentrations is wide and can vary by as much as a factor of 40 – for example, in lettuce nitrate levels naturally range from 90 to 3520 parts per million.

Other dietary sources of nitrates are preserved meats. Although a smaller source than vegetables, they may be a more significant risk factor for cancer as they lack the protective factors found in vegetables. The diet also provides a small amount of preformed nitrosamines, resulting from food processing and preservation techniques. The most common dietary sources in western countries are cured meats, beer and salt-dried smoked fish. Other major sources of nitrosamines are occupational exposure and tobacco use. These are greater sources of nitrosamines than the diet by several orders of magnitude. Cigarette side-stream smoke contains up to 1000 mg of volatile nitrosamines.

Estimates of the average nitrate intake from the diet is below the calculated Acceptable Daily Intake of nitrate.

Evidence of cancer in humans

There is a strong link between stomach cancer death rates in various countries and the amount of nitrate ingested in the diet. Both nitrate intake and stomach cancer rates are high in Japan, Colombia and Chile while both are low in Denmark, USA, Sweden and the UK. These observations led to concern that a high nitrate intake could lead to high stomach mortality. However, these observations have not been confirmed by other kinds of studies. Looking at historical trends, stomach cancer mortality has been declining steadily in most developed countries and in many developing countries since 1950 while fertiliser use and nitrate levels have increased. Looking at epidemiological studies, there is no evidence from these studies that nitrates in vegetables cause cancer.

One of the epidemiological studies was a comparison of the nitrate levels in the saliva of people from two areas of the UK with different rates of stomach cancer. Those at low risk for stomach cancer had higher nitrate levels in their saliva – 162 nmol/ml in Oxford and Southeast England versus 106 nmol/ml in Wales and Northeast England. As the main source of nitrates were vegetables, it was assumed that the vitamin C in the vegetables must have been protective.

From the studies linking nitrate intake from vegetables with digestive tract cancers, there is a consistent and clear picture. Vegetable consumption protects against digestive tract cancers regardless of the presence of nitrates in the vegetables. A high consumption of nitrate contaminated vegetables is unlikely to increase the risk of cancer because vegetables contain cancer-preventing nutrients which block the conversion of nitrates into nitrosamines.

Conclusions

It is clear from this review that there are large gaps in the research and many uncertainties about the health effects of organic food. The best way to fill these gaps would be to carry out well-designed epidemiological studies of health outcomes in people. The next best way is to improve the quality of the nutrient content studies by agreeing a protocol which takes all the important variables into account and which avoids the biases of many of the existing comparative studies.

As long as there is no direct evidence from studies on people of the health benefits of organic food, we have to base our conclusions on indirect evidence from studies on nutrient and contaminant levels in foods. From the research to date, it does not look as if conventional agriculture increases the risk of cancer in UK consumers. Exposure to carcinogenic agrochemical residues in conventional food is insignificant and the evidence for a reduced content of anti-cancer nutrients in conventional food is not conclusive nor consistent. This is my interpretation of the evidence. Future studies may yet present a different picture but as it stands, the common sense view that organic food is healthier is not substantiated by the evidence and there is no compelling reason to recommend the consumption of organic food in order to reduce the risk of cancer. However, there is strong and consistent evidence to recommend eating a healthy diet in order to reduce the risk of cancer. A healthy diet of conventional or organic food could reduce the risk of cancer in the developed world by 30 to 60%. If our main concern is how to reduce the risk of cancer, we have the answer† but it does not seem as if organic food has a major part to play in that answer.

Acknowledgements

This research was funded by the Bristol Cancer Help Centre. The paper is a condensed version of a more detailed 40,000 word report citing 345 references which was prepared for the Centre.

Key References

Clancy K, 1986, The role of sustainable agriculture in improving the safety and quality of the food supply, American Journal of Alternative Agriculture, 1, 1 p11-18.
Clarke R, Merrow S, 1979, Nutrient composition of tomatoes homegrown under different cultural procedures, Ecology of Food and Nutrition, Vol 8, p37-46.
Colborn T, Myers JP, Dumanoski D, 1997, Our Stolen Future – are we threatening our fertility, intelligence and survival – a scientific detective story, Abacus, London.
Conway GR, Pretty JN, 1991, Unwelcome Harvest – agriculture and pollution, Earthscan Publications, London.
Crisp TM et al, 1997, Special report on environmental endocrine disruption: an effects assessment and analysis [EPA/630/R-96/012], EPA, Environmental Protection Agency, Risk Assessment Forum, Washington DC, USA.
Davis DL, Bradlow HL, Wolff M, Woodruff T, Hoel D, Anton-Culver H, 1993, Medical hypothesis: xeno-oestrogens as preventable causes of breast cancer, Environmental Health Perspectives 101(5): 372-77.
Finesilver T, Johns T, Hill SB, 1989, Comparison of food quality of organically grown versus conventionally grown plant foods, Ecological Agriculture Projects, McGill University (Macdonald Campus), Quebec, Canada.
Forman D, 1987, Dietary exposure to N-nitroso compounds and the risk of human cancer, Cancer Surveys: Advances and Prospects in Clinical, Epidemiological and Laboratory Oncology, Vol 6, Number 4, Diet and Cancer, Oxford University Press.
Gold LW, Slone TH, Stern BR, Manley NB, Ames BN, 1992, Rodent carcinogens: setting priorities, Science 258 9 Oct: p261-265.
Griffiths, K; H Adlercreutz, P Boyle, L Denis, RI Nicholson, MS Morton, 1996, Nutrition and Cancer, Isis Medical Media Ltd, Oxford.
Hornick SB, Parr JF, 1989, Effect of fertilizer practices on the nutritional quality of crops, In Agricultural Alternatives and Nutritional Self-Sufficiency. Proceedings of the 7th IFOAM Conference. International Federation of Organic Agriculture Movements, Tholey-Theley, Germany.
IEH, 1995, Environmental oestrogens – consequences to human health and wildlife, Institute for Environment and Health, University of Leicester, Leicester, UK.
Kaloyanova FP, El Batawi MA, 1991, Organochlorine compounds, Chap 4 in Human Toxicology of Pesticides, CRC Press, Inc, Boca Raton, Florida, p59-100.
Knight T, Forman D, Al-Dabbagh SA, Doll R, 1987, Estimation of dietary intake of nitrate and nitrite in Great Britain, Food and Chemical Toxciology 25, 277-285.
Krieger N, Wolff MS, Hiatt RA, Rivera M, Vogelman J, Orentreich N, 1994, Breast cancer and serum organochlorines; prospective study among white, black and Asian women, JNCI, 86 (8): 589-99.
Linder MC, 1991, Food quality and its determinants from field to table – growing food, its storage and preparation, Chapter 10 in M Linder (ed) Nutritional biochemistry and Metabolism with clinical applications, Elsevier, New York, London, p239-254.
Repetto R, Baliga SS, 1996, Pesticides and the immune system: the public health risks, World Resources Institute, Wash DC, USA.
Safe SH, 1995, Environmental and dietary estrogens and human health: is there a problem?, Environmental Health Perspectives Vol 103, 4, April, p346-351.
Schuphan W, 1974, Nutritional value of crops as influenced by organic and inorganic fertilizer treatments – results of 12 years' experiments with vegetables, Qualitas Plantarum – Plant Foods Human Nutrition. XXIII, 4; p333-358.
Smith AG, 1991, Chlorinated hydrocarbon insecticides, Chapter 15 in Hayes WJ & Laws ER, eds Handbook of Pesticide Toxicology Vol 2, Academic Press, London, p731-915.
Stopes C; Woodward L, Forde G, Vogtmann H, 1988, The nitrate content of vegetable and salad crops offered to the consumer as from organic or conventional production systems, Biological Agriculture and Horticulture, 5 (3) p215-221.
Tomatis L, editor in chief, 1990, Cancer: Causes, Occurrence and Control, IARC No 100, Lyon, France.
van't Veer P, Lobbezoo IE, Martin-Moreno JM, Guallar E, Gomez-Aracena J, Kardinaal AFM, Kohlmeier L, Martin BC, Strain JJ, Thamm M, van Zoonen P, Baumann BA, Huttunen JK, Kok FJ, 1997, DDT (dicophane) and postmenopausal breast cancer in Europe: case-control study, BMJ 315, 12 July, p81-5.
WCRF, 1997, Food, Nutrition, and the Prevention of Cancer: a Global Perspective, World Cancer Research Fund (WCRF) in association with American Institute for Cancer Research.
Welch RM, House WA, 1984, Factors affecting the bioavailability of mineral nutrients in plant foods, In RM Welch and WH Gabelman (eds) Crops as sources of nutrients for humans. Pub No 48. Soil Sci Soc Amer, Crop Sci Soc Amer and American Soc Agronomy, Madison, Wisconsin, p37-54, chapter 3.
Woese K, Lange D, Boess C, Bogl KW, 1995, A comparison of organically and conventionally grown foods – results of a review of the relevant literature. Federal Institute for Health Protection of Consumers and Veterinary Medicine, Division 2: Berlin, Heft 04/95- 05/95.