Food Safety: Elusive Goal and Essential Quest IUFoST Founders Lecture for the 10th World Congress of Food Science and Technology Sidney, Australia, October 4, 1999 Consumers are fearful of involuntary risks such as those from food additives and pesticide residues, but such risks are low because of past effort towards evaluation and control. Microbiological and nutritional risks are the largest risks but are primarily under individual control and hence are voluntary risks. They are dealt with effectively by three words: sanitation, variety and moderation. Preventive measures require continuous education and training, and thus time, money and expertise, and often require present effort and investment for uncertain future gain. Safety is simply the absence of risk. In our imperfect society in a probabilistic universe, risk can never be entirely absent. Therefore, the practical meaning of "safety" is that a particular risk, or group of risks, is at an acceptably low level. That phrase implies 1) an objective, reproduceable method of defining and measuring the level of a risk, 2) the practical means to achieve that level of risk, and 3) a broad social consensus on the acceptability of that level. Thus, at the end of three sentences, we are already out of touch with reality, an deep into acrimony. It simplifies discussion of food risks to categorise them by shared characteristics of causation, impact and correction. A slightly modified ranking of food risks first used by Virgil Wodicka (Anon 1971) is still the most useful classification. Based on available data, in first place by several orders of magnitude are the microbiological and nutritional risks; next are those from natural toxicants and environmental contaminants; and last, again by orders of magnitude, are the risks from food additives and legal pesticide residues on food. Throughout the world, this relative ranking remains generally valid, although the absolute size of these risks in developing areas is vastly larger. These comments will attempt, all too briefly, to describe current
trends in the nature of each risk category, and contemporary methods of
dealing with each. Consistent with what we would like regulatory agencies to do,
this Microbiological Risks An outbreak of foodborne illness occurs whenever a susceptible population without effective interceptive measures, encounters a foodborne pathogen, or its toxin, at sufficient levels. If someone takes enough notice of the outbreak to identify the pathogen as one not previously known, we say it "emerged." Those outbreaks of foodborne illness of microbiological origin that affect only one or a few people, and that last only a day or two, seldom receive the attention of a physician. Still less frequently are they reported to public health authorities. As a result, even in the most industrialised nations, the true extent of foodborne illness can only be estimated. Current estimates for the United States are in the low tens of millions of cases per year (CAST 1994), or roughly 10,000 per 100,000 of population. Estimates for Western Europe, Japan and Australia are generally comparable. Next to the common cold, foodborne illness is the most frequent cause of lost time from work. Fatalities in the USA, for example, are estimated at 9,000 annually (CAST 1994), typically among the elderly, the very young, and the immunocompromised. Quite aside from the costs of temporary illness and death from acute infections, chronic sequelae may occur in 2-3% of cases of foodborne disease (Lindsay 1997). We have already noted that the extent of foodborne illness in the developing world is far greater still. Unobscured even by incomplete reporting, are two striking trends in outbreaks of foodborne illness in the last quarter-century. First is the emergence of pathogens previously unknown, as well as newly-recognised pathogenic strains of organisms previously considered harmless. Table 1 presents a list of these. Second is a general increase in the incidence of foodborne disease. Table 1.Foodborne and predominantly foodborne pathogens newly recognised in the past 20 years (adapted from Tauxe 1997) *Corresponding tables & figures found at www.foodaust.com.au Campylobacter jejuni
It is impossible to overemphasise the importance of clean water for drinking, food
processing and watering of Mishandling is a pervasive and difficult problem because it occurs in many different ways at many different points in the food chain. Inadequate refrigeration and cross-contamination are common problems. They are particularly likely to occur where trained personnel, quality control measures and the capability of prompt intervention are most lacking. That is why more than 90% of all US outbreaks of foodborne illness are due to mishandling of food in the home or in food service establishments. For the same reasons, street foods have received much attention throughout the world by national governments, FAO and WHO. Personal Insanitation Primarily because of inadequate handwashing, this remains as much a factor as it did in the day of the notorious "Typhoid Mary." We rely on education and training. We must, because, as Joshua Lederberg (1997) has observed, it is not clear that, in this malignantly litigious age, we could put the public health ahead of individual freedom as was finally done when "Typhoid Mary" was confined to an island in New York city. Changes in the Environment and the Food Supply The environment in which transmission of foodborne illness takes place has changed in this century, and almost entirely for the worse. As Professor EM Foster has noted; early in this century, populations were largely rural, and most food was grown and produced at home or nearby. Refrigeration was unavailable, but food leftovers were fed to livestock, not stored (Foster 1997). Foodborne illness was virtually unknown or unrecognised, and when it did occur, it affected only the few within a family who had eaten the contaminated food. Conditions today in industrialised nations could hardly be more different. The food trade is a complex global web. The United States, for example, imports more than 50% of its seafood from 172
other countries (Garrett & others 1997). Many steps link the agricultural and fisheries producing areas to the point of consumption. These steps typically involve processing at several points, with storage and transport over long distances, often while refrigerated or frozen. Every stage presents opportunities for failure. Initial food processing is often done in large-scale centralised facilities, then distributed to, and prepared for service in widely dispersed locations, thereby distributing widely and quickly the consequences of any failures. A large outbreak of E. coli O157:H7 in Japan in 1996 illustrates that problem (Käferstein & others 1997). Widely dispersed contaminated food is often "diluted", so to speak, by the wide variety of foods available for consumption in industrialised societies. This causes many outbreaks of foodborne illness to be diffuse, barely above normal background levels. This delays recognition and consequently delays effective efforts to deal with the problem. Recent examples in the SA Most of our populations live in urban areas, far from food sources. Close human contact increases the opportunity for transmission. Particularly in developing countries, public health services lag far behind the rush from farm to city. Cities, to be sure, are centres not only of our economies, but of culture and learning as well. However, they are also massive projects in the intensive monoculture of human beings. In agriculture or industrial fermentation, monoculture is feasible if one supplies and tightly controls all necessary inputs, monitors the operation closely, and is prepared for instant and effective intervention should any problem arise. That hardly describes our cities anywhere. Epidemics of former years spread slowly, consistent with the slow pace of the few people who traveled. Today people and foods travel widely and at high speed. People can serve as vectors of food contamination resulting in foodborne illness. National borders are porous to disease (Kemel 1996) including foodborne disease, but they frequently are barriers to disease countermeasures, including the information that makes countermeasures both possible and effective. Both the unwitting and the intentional changes we have made contribute to our problems. Archer (1996) has pointed out that environmental conditions that stress, but do not kill all bacteria, not only select for hardier mutants, but can often increase virulence as well. Recent changes in cattle feeding practices that have lowered the pH in the rumen may well have produced the acid tolerant E. coli O157:H7 found in the apple juice mentioned earlier (Diez-Gonzales & others 1998). Cattle are one of the reservoirs for human enteric infections. Our current fondness for marinades that do not kill, but select for acid tolerance, may be doing the same thing. Our too common and casual use of "sanitisers" is another example of this. Still worse is the development of antibiotic resistance, certainly due to excessive and improper therapeutic use in humans, and perhaps also from agricultural use. Moreover, we are relaxing our standards. An example is our current infatuation with "minimally processed" food. That is "playing chicken" with pathogens, and the pathogens react faster than we do. Inherent Properties of Microorganisms The inherent properties of the microorganisms that cause foodborne illness continue to be a major part of our difficulties in combating them. First is their proliferative ability; a
bacterial generation can be thirty minutes a Increase in Immunocompromised People Finally, opportunities for foodborne illness have increased because we have more individuals in our societies with impaired immune systems occasionally from a genetic disorder, more frequently and tragically from malnutrition or from AIDS, and also from medication and just plain ageing (Morris & Potter 1997). Few, if any, would want to return to the social and economic conditions of the first decade of this century. But it is no exaggeration to say that virtually every major social and economic change since then; urbanisation, increased economic opportunity, greatly increased travel, a highly varied and international food supply that is independent of season, human choices that stress, but do not kill bacteria, and more immunocompromised people, all tend toward a higher incidence of an increasingly wide range of foodborne diseases. That increase has occurred. Figure 1 shows, for the USA, the decrease in typhoid fever and the
increase in *Corresponding tables & figures found at www.foodaust.com.au
We have new tools as well. There is a broad range of new, and
newly applied, techniques for food preservation. Molecular biology, improvements in microbiological technology, and modern communications provide us with the opportunity for more prompt, focused, and effective countermeasures. Rapid methods of bacterial DNA analysis permit tracing cases to a common source even in a diffuse outbreak (Majkowski 1997). Genomics will let us learn more about bacteria, and at a faster pace. It may well allow us insight into the factors that lead to virulence in a pathogen, and this could permit improved preventive and interceptive measures (Ewald 1996). Many human foodborne infections come from animal reservoirs, and that points directly to needed improvements in food production, harvest and slaughter environments (Orriss 1997). While communication and cooperation between the various government agencies and industrial organisations potentially involved in a newly recognised outbreak have improved, much more is needed. Bioinformatics will help us handle the greater volume and detail of information we will need to use. We require predesignated rapid response teams, able to co-opt the particular expertise they need, and able to act quickly on a national and international level. We must have far wider application of HACCP (Hazard Analysis and Critical Control Points) to all links in the chain of production, storage, transport and processing. Finally, as previously noted, trade in food is world-wide, and it is a shrinking world. Increased safety for the wealthier countries lies in economic growth and improved infrastructure in the developing countries. Nutritional Risks The irony of the nutritional risks is that they are greatest in both the most and the least affluent countries. The risks of malnutrition are well understood, but far exceed the resources we have devoted to them. In any society, those with access to the resources of the society are well-fed; only the poor go hungry. Persistent malnutrition is thus a consequence of economic development that is inadequate for, or poorly distributed within the population of the area. One cannot separate the issue of an adequate or "secure" food supply from the issue of population size and growth rate; how much food is adequate for how many people? Emergency relief is essential, but for those areas with widespread current and prospective malnutrition, appropriate economic development combined with restraint in population growth offers the only long-term prospect for improvement. Most of our attention to long-term measures has gone toward improving agricultural productivity, and this has had spectacular success; witness the "green revolution." But there are indications that we are beginning to press the practical limits of photosynthetic capability with the crops we now have, and that further increments of growth will be purchased at a higher and eventually unacceptable price in resource use and environmental degradation (Mann 1999). We need to pay far more attention than we have paid in the past to appropriate post-harvest technology, in order to avoid the waste of 20% to 50% of agricultural production that now occurs in less developed areas. Finally, of course, only economic growth and infrastructure development within developing countries can achieve the standards their food supplies must meet in order to support the export growth that economic development requires. In stark contrast, overnutrition characterises not only the most affluent countries, but higher income populations generally. Food intake in excess of caloric needs, sedentary occupations, inadequate physical exercise, and resulting obesity are epidemic in the much of the world (Popkin & Doak 1998). We are becoming a race of "couch potatoes." The increased health risks attendant upon obesity include coronary heart disease, several sites of cancer, and adult-onset diabetes. Our eating habits are enmeshed in multiple personal and social patterns, most of them pleasurable. It is unlikely that any measures except consistent and persistent consumer education, combined with more narrowly targeted steps for specific population groups, will improve this situation. An example of the latter is the substitution of 1% fat milk for whole milk in school feeding programs. This is a reduction of 40 to 80 calories per day for the average child. In a school year of 180 days that is one or two kilos of body weight not gained (Caballero 1999). The overnutrition of the affluent also includes unnecessary and obsessive intake of micronutrients. The adverse consequences of only moderately excessive intakes of Vitamins A and D are well-known. Consequences for the other micronutrients are less well-studied, and in some cases, largely unknown. Micronutrient supplements are hugely profitable and remarkably resistant to regulation. As this paper will later emphasise, by far the largest food risks we face are those that are, to a very great extent, within our own control. That is obviously true of vernutrition. Risks from Natural Toxicants Most consumers think of natural toxicants only in terms incautious and uninformed mushroom gatherer, although the taxonomy of mushrooms is more complex, and the risks are greater than most enthusiasts realise. The risks of natural toxicants, however, reach much further than mushrooms. Even today, virtually all of our foods, excepting only the major cereal grains, contain a variety of toxicants that no regulatory authority would ever permit through intentional addition (Hall 1977). In the past, these have been responsible for widespread outbreaks of human illness and death, and in times of food shortages, and in less developed areas, that still occurs. Fortunately for the more affluent countries, plant genetics, food processing, toxicological evaluation, and regulatory oversight have reduced this class of risks in traditional foods to far below primitive levels. Equally important has been the variety of foods available, which reduces over-reliance upon any one food. We are now, however, consuming a less familiar set of
substances called "dietary supplements" in the United States, and
"nutraceuticals" in many countries. Their appeal is enormous. They promise
relief from the ills that Ironically, these risks of natural toxicants are higher for those who wish to get closer to nature by gathering their own food. In the USA, several people die each year from amateur-gathered herbal teas; about the same number that die of botulism, which is also now a consequence, almost entirely, of food that has not been commercially processed. Risks from Environmental Contaminants So far as current knowledge goes, in the past few decades, food risks from environmental contaminants have occurred either incidental to serious environmental damage, eg the
mercury poisoning in Minimata Bay in the 1960s, or as a result of accident or
ignorance, eg the radionuclides from the Chernobyl disaster or the polybrominated biphenyls in cattle feed in the USA in 1974 (Fischbein & others 1992). As with the other food risks, these problems are greatest in developing areas, and in some countries that formerly had
command Risks from Food Additives A useful operating definition of a food additive is that it is any minor ingredient added to food to achieve a particular technical effect. Such effects may be to improve nutritional value, as iodate added to table salt to protect against goitre, a preservative to delay spoilage, an emulsifier to prevent separation, or a flavour to enhance gustatory appeal. We have always added substances to food for such useful
purposes, and in the past, for other ends far less defensible. Much of the
popular pressure behind the first food control legislation arose from
justifiable public There can be no absolute guarantee of safety under all conceivable circumstances. Highly distorted diets may result not only in nutritional imbalances but in some degree of toxicity from either naturally occurring constituents or from intentionally added ingredients. Allergies and other idiosyncratic reactions to particular food constituents occur. The principal allergens, such as tree nuts, shellfish, peanuts and eggs must be specifically declared on food labels in order to warn susceptible consumers, but rare adverse reactions are difficult to trace to a cause and thus difficult to avoid. In many respects, the current concern over foods and food ingredients derived from genetically modified organisms (GMO) parallels the concern a generation earlier over irradiated food. In the past decade, a series of reports from private, national and international agencies has spelled out in consistent detail the steps that must be undertaken to assure, in each case, the safety of such foods and ingredients (IFBC 1990, FDA 1992, OECD 1993, 1996). These carefully written documents, however, rarely persuade the unconvinced, for reasons to be addressed in a moment. It seems likely that, as with irradiated food, consumers will need to be shown products with clear advantages for them, obtainable from the use of GMO, before opposition will crumble away. Risks from Pesticide Residues These comments will address only
the risks to consumers from legal pesticide residues on food. The risks
to Proper use of pesticides requires their careful evaluation, statutory and regulatory controls, informative product labeling, trained applicators, and an elaborate and effective monitoring system by both industrial food processors and government agencies. That infrastructure operates in industrialised nations; it is largely lacking in many developing areas, from where, therefore, come most of the egregious examples of pesticide misuse. Again, economic and infrastructure development is the only long-term answer. The current levels of risk from food additives and pesticide residues are so low because of the past effort devoted to their evaluation and control, and because they are usually used by persons expert in their use. The microbiological and nutritional risks are high because they depend heavily on how all consumers choose, use and abuse their food, and all consumers are not expert. Concluding Comments The risk categories discussed here
are not mutually exclusive, and often overlap. The most common and
tragic In particular areas, certain specific risks may be unusually prominent. In and near Qidong, Jiansu Province, China, the incidence of hepatocellular carcinoma is so high that it accounts for up to 10% of all adult deaths a rate five to ten-fold higher than in the United States. This is a consequence of the high rate of hepatitis B infection enormously potentiating the effects of a high aflatoxin intake (Qian & others 1994, Wang & others 1999). Ironically, when first proposed by Wodicka, and to some extent, even today, the six risk categories were usually seen, in the more developed nations, in inverse– and perverse–order, with the risks from food additives, pesticide residues and environmental contaminants as highest, and microbiological and nutritional risks as lowest. This undoubtedly was because those from food additives and pesticide residues were seen as beyond individual control, and therefore, as involuntary risks, imposed on the individual by "others." In contrast, nutritional and microbiological risks are, to a much larger extent, the result of the way we individually choose and handle our food. To that extent, therefore, they are voluntary risks. Chauncey Starr (1969) demonstrated years ago that we accept voluntary risks about a thousand times larger than involuntary risks. By providing information through which consumers can make a choice, labeling can convert an involuntary risk into a voluntary one. It is a pity that the food industry has often been reluctant to take advantage of that fact. Consumers and competitors frequently object strongly to particular products or technologies based on economic, social or aesthetic concerns. Because the mandate of regulatory agencies typically is restricted to safety issues, safety frequently becomes a surrogate issue raised by opponents because that is the only basis on which they can force regulatory attention and delay. Because safety is the absence of risk, and absence, as with any negative, cannot be "proved," this leads to an effective strategy. Opponents ask for the impossible in order to delay the inevitable. Nothing is more important that confidence in the food supply, and in the competence and integrity of those who produce and safeguard it. It is bad enough when people are actually harmed, and almost as bad when they only fear they are. However, our worst disasters are those we create for ourselves by compounding failures in agriculture, processing, and distribution by more serious failures in industry and government to deal candidly with the public in the aftermath. Such behaviour reinforces the anti-industry, anti-government and anti-science attitudes we so often deplore. A final irony is that Western countries have the luxury of worrying about trivial or nonexistent risks while much of the rest of the world must worry about simply getting enough to eat. Our food supply and its safety is a complex web. Fortunately the principles important in controlling it are simple–not ABC, but the three Ks. There are also three words–Sanitation, Variety, and Moderation that deal effectively with microbiological and nutritional risks. They are the two largest risks, and are primarily within our individual control. This entire topic is filled with contrasts and ironies, but these will fade away if we can distribute equitably the will, the knowledge and the resources to resolve them. The goal will remain elusive, but the quest is essential. Acknowledgements The author is grateful to Professors Fergus M. Clydesdale, Michael P. Doyle, Michael W. Pariza, and Stephen L. Taylor for their valuable comments on this manuscript. References Anonymous. 1971. Wodicka rates food additive hazard as
low. Food Chemical News 12(49): 11 Caballero, B. 1999. Personal communication. Child, J. 1813. American Frugal Housewife (Publisher unknown). Diez-Gonzales, F, Callaway, TR, Kizaulis, MG & Russell, JB.
1998. Grain feeding and the dissemination of acid |