Science, uncertainty and policy: food for thought
11食工1
John R.Krebs * Food Standards Agency, PO Box 30080, Elephant & Castle, London SE1 6YA, UK
Abstract The organisation and work of the Food Standards Agency are described.The Agency is a new non-Ministerial Government department with responsibility for protecting the health of the public and other interests of consumers in relation to food.Its roles encompa aement of risk (through scientific expert committees) as well as risk communication and management.Among the many changes that the Agency has brought about is a new commitment to openne.All policy discuions and decisions take place in public.Diet and bovine spongiform encephalopathy are used as examples to illustrate the Agency’s approaches to dealing with risk and uncertainty.Crown Copyright © 2001 Published by Elsevier Science Ireland Ltd.All rights reserved.
Keywords: Food policy; Risk; Uncertainty; Phytoestrogens; BSE 1.Introduction The Food Standards Agency is a new UK government department with responsibility for all aspects of food safety and standards.Its job is to give useful advice and make policies that work, based on the best available scientific evidence.Some of that comes from the Agency’s own in-house staff, but much is provided by outside experts, such as toxicologists.While decisions based on science are useful, one of the crucial aspects of the scientific approach is that results are often uncertain.As scientists, we may not like the way that uncertainty is portrayed in the media or exploited by vested interests, but the Food Standards Agency has to interpret the scientific evidence — uncertainties and all — to propose sound policy options, despite uncertainties.The Agency also does as much as poible to reduce uncertainty by funding research.The Agency’s two main concerns are diet and health and food safety.While the boundaries between the two are often blurred, they can also be distinguished.On the one hand, general aspects of diet — what people eat, how much, how often have a profound influence on general aspects of health such as longevity, particular symptoms, and what is coming to be called wellne.On the other hand, specific foods, ingredients and the chemicals they contain can have a very direct impact on health.Those are the broad questions the Food Standards Agency is interested in.How does it operate?
2.The Food Standards Agency The Agency is not an agency at all, in the old government sense of the word, but a new UK wide Government department with offices in England, Northern Ireland, Scotland and Wales.Unlike most other departments, however, it is non-ministerial; there is no Minister for Food Safety, but the Food Standards Agency is accountable to the Westminster Parliament and the devolved authorities through their Health Ministers and Secretaries.The Agency’s powers are vested in its Board, which consists of 14 independent members, and are exercised by the staff of the Agency, one-third of whom are scientists by training.The Board is completely independent.Its members were chosen to reflect the national responsibilities of the Food Standards Agency and the expertise needed to do the job properly — public health, consumer champions, food production and proceing, catering, communications, and so on.Most importantly, all the members of the Board were appointed after an open competition, held under the rules and principles established by the Commiion of Standards in Public Life.There is a publicly available code of conduct for Board Members, and a public register of members’ interests, so the Agency can set itself high standards , which is important if it is to gain the trust of the many different stakeholders with an interest in food.These standards, and the appointment and register of interests of the Board, reflect another novel aspect of the Food Standards Agency: it is committed to openne.To this end, all Board meetings, including debates and discuions of key policy iues, take place in public.In addition to being open and providing information to the public, the Agency has also set out deliberately to listen to the public.Openne and public participation are very important.Traditionally, people in government have regarded consultation as an add-on, the final step in the decision-making proce.At the Food Standards Agency, we are determined to make consultation an integral part of the work, and to consult those affected by any decision as early in the proce as poible.However, this will not be merely an exercise in giving those with power more of it.Relevant industries, consumer organizations and the like will have their say, but the Agency is also very keen to reach out to people who have traditionally not had a voice — the underprivileged and the deprived — and is working to develop better ways to discover their views than the traditional formal procees.In addition to a new approach, the Food Standards Agency also has new powers.It will publish the advice given to ministers, so that people can judge whether the Agency is doing its job effectively.It has the authority to collect information and samples at every point in the food chain from the field to the table.This is in addition to the normal good practices required of the food industry, and will be used particularly when there is reason to believe that public health is threatened.Furthermore, the Agency will be monitoring the performance of local authorities as they carry out their duties to enforce food law.The intention is to build close partnerships with all interested parties, working together to improve food standards and deliver benefits to consumers.The Food Standards Agency also funds its own research programmer to the tune of about £25million a year.This is not new money; it comes from the previous existing programmers at the Ministry of Agriculture, Fisheries and Food and the Department of Health.The novelty is that the Board will have the chance to set strategic research priorities, enabling the Agency to reduce scientific uncertainty where it is most important to do。
3.Risk Perhaps central to the work of the Food Standards Agency is risk — not only measuring risk and quantifying it scientifically, but also understanding how people perceive risks and what they are willing to do about them.Food is a particular source of worry about risks, because it is more than merely nourishment.It is also deeply embedded in culture and society.It is supposed to keep us healthy.We feed it to our children.We expect it to be safe.So it is not surprising that, whenever a food comes under suspicion — real or imagined it hits the headlines.Life, or course, is full of risks.What is interesting is that some risks, such as those aociated with food, seem to be so much more alarming than others, e.g.smoking.There has been a great deal of research into this, all of which points to a few well-established rules of thumb that relate directly to the values people hold.One of the most important of these values is how much control people think they have over the risk.Most people say that the risks of smoking and motoring, which they choose to do themselves, are more acceptable than risks such as nuclear power that are imposed on them by government or industry.The question of control may be why paive smoking and speed limits around schools have become such powerful iues.Genetically modified (GM) foods are a case in point.The public seems to feel the risk is high, although there is very little evidence for them to base that fear on.Public concerns about GM foods relate to perceptions of risk, not to its real magnitude.People consistently underestimate considerably the true risk of common causes of death, e.g.cancer and heart disease.But they greatly overestimate the risk of dying in a high-profile disaster such as a flood or tornado.If people are not very good as estimating risk, can anything be done to help them? Several researchers have tried to come up with comparative scales of relative risk.In an ideal world, scales such as these might be some use but, in reality, because people rate risk according to how the activity involved relates to their own values, scales like this probably do not help them.Although the public underestimates the risk of common causes of death, statistics reveal the ‘real’ killers, and the part that diet plays in various causes of death.Top of the list is cardiovascular disease, with more than one in three of all deaths in England and Wales, a total of 218 000 in 1999.Second is cancer, which killed 136 000 people in the UK in 1999.One in three people develop cancer at some point in their lives, and one in four die from it.Diet is an important contributory factor in both broad categories.Some epidemiologists estimate that at least one-third of premature deaths in Europe are attributable to unhealthy diet.So 73 000 of the deaths from cardiovascular disease are ‘caused’ by diet.And although one third of all cancer deaths are linked to smoking, diet is also important, with a role in about one quarter of all cancer deaths.Against this background of actual risk, what are people actually worrying about? Bovine spongiform encephalopathy(BSE), food poisoning, and GMOs.The latest figures published by the Department of Health show 82 definite and probable cases of variant CJD, about 15 per year.Food poisoning kills perhaps 50 people a year.This is not to say that the low risk worries expreed by the public and the news media are unimportant.But it is as important to get things in perspective
4.Diet and health One way to meet targets for reducing cardiovascular disease and cancer deaths would be to help people to change their diets but, before that can even begin, we need to know what they are eating.The Food Standards Agency, together with the Department of Health, funds the National Diet and Nutrition Survey.The results of the most recent survey in this series, into the diet of young people aged 4 to 18, were released in June 2000.They provide the most comprehensive picture ever of food eaten by this age group.On the one hand, there is good news: the results showed no evidence of any widespread malnourishment n this age group.But there is also depreingly familiar bad news: children were generally eating too many sugary foods, too much saturated fat, too much salt, and not enough fruit and vegetables.The UK is near the bottom of the league in Europe.Consumption of fruit and vegetables, while it has risen gradually since the 1950s, is still very low at only 250 g per day.Social cla differences are wide and getting wider.People in the lowest income groups eat about one-half of the fruit and vegetables of people in the highest income groups, and the difference is especially marked in fruit, fruit juice and fresh green vegetables.In fact, when it comes to anti-oxidants such as vitamin C, consumption among the poorest households has actually been falling since 1980.In children, consumption is even lower.The survey found that one in five ate no fruit at all.Just over one-half did not drink any fruit juice.And 4% did not eat any vegetables over the course of an entire week.The amount of fruit and vegetables that young people are eating has fallen since 1983.This is a major cause for concern, and the Food Standards Agency is committed to doing more than just being concerned.Patterns of consumption reflect patterns of acce to fresh fruit and vegetables.In 1999, the Social Exclusion Unit highlighted low-income areas where corner shops have closed, where healthy food is expensive and in short supply, and where the nearest source is three bus rides away and taxis are unaffordable.Under those circumstances, exhorting mothers to buy their kids a bag of apples or carrots is not very helpful advice.Nutritional needs and health inequalities are a high priority, and already there are several initiatives to deal with these problems.In deprived areas, communities are working to identify and map their food needs, e.g.the distance from food shops and the types of food they would like to have, with the aim of improving acce, availability and affordability.Elsewhere, residents are being encouraged to grow food on housing estates, and co-operatives have been set up to buy healthy food in bulk and distribute it to residents.The food is often actually cheaper than it would be from the supermarket.There are breakfast clubs in schools, a new emphasis on cooking for kids, and even tuck shops selling fruit to children, who often say they prefer fruit to sweets when they have cheap and easy acce to it.This is no more than a taste of the various programmers that are underway to improve diet, especially of children, around the UK.Why so many of them? Because one of the greatest uncertainties in this area is that it is by no means clear which interventions are best at changing be-haviour.Is price more important than availability? Is it true that if children are given a taste for fruit and vegetables when they are young they will develop good eating habits for the rest of their lives? At the moment, the Food Standards Agency is supporting different approaches in different places and hoping to adapt and develop the most succeful into programmes targeted to the needs of different groups acro the UK.A further point about diet and health is that there is still a shortage of information about the proximate mechanisms by which fruit and vegetables protect against various diseases.Which compounds are important and how they operate are obviously still major areas of uncertainty, with a great deal of research behind them.Making the link from epidemiological correlations to biochemical pathways is also fraught with difficulties, and many of the dietary supplements that purport to offer the benefits of fresh fruit and vegetables have yet to prove their value.Knowing what people eat is important in setting overall priorities for the national diet.This information is also crucial to ae the chemical risks of various foods.Armed with a complete record of everything people eat in a week and information on the levels of some given chemical in food, it becomes poible to estimate the amount of that chemical in the diet.This is a key piece of evidence used by the Food Standards .Agency’s expert advisory committees when they consider food safey
5.Food safety Food safety is the remit of six divisions within the Food Standards Agency headquarters and its offices in Scotland, Wales and Northern Ireland.Of the six HQ divisions, it is the Chemical Safety and Toxicology division that is responsible for aeing the health implications of chemicals in food.These ‘chemicals’ can be of several different kinds.Some are added deliberately for one reason or another, e.g.food ingredients or proceing such as enzymes, some are contaminants, and some are ‘natural’ in the sense that they are neither added deliberately nor accidental contami-nants but are simply present in the food.The Food Standards Agency aees the impact of these various potential risks through two main channels: in-house scientists and outside experts serving on expert advisory committees.The Agency also receives advice from the European Union Scientific Committee for Food and the FAO:WHO Joint Expert Committee on Food Additives.The Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment (COT) faces a particular challenge, because food is more complex and more variable than almost every other substance that humans are exposed to.The level of exposure drives the engine of risk aement.But people’s lives change, and shifts in food preferences can caused marked changes in exposure.Soya is one of those foodstuffs whose consumption patterns have changed as a result of changing food preferences. stories have recently emphasised the potential drawbacks of a diet high in soya.One article, for example, quoted American scientists who decided to break ranks with the US Food and Drug Administration over its decision to approve a claim that soya reduces the risk of heart disease.The scientists drew attention to a series of research studies showing adverse effects of isoflavones in soya.Some isoflavones mimic the female hormone oestrogen, and are thus known as phyto-oestrogens, and research claims links between these phyto-oestrogens and some types of breast cancer, abnormalities of foetal development, thyroid problems, and brain damage.Given the factors affecting perceptions of risk already mentioned, these claims are liable to worry consumers.Despite the tone and implications of the article in question, this is not a new concern.The Food Standards Agency has long been aware of the problem, and has a comprehensive programme of investigation under way, with a research budget of £1 million a year.The main interest in phyto-oestrogens stems from the fact that these compounds can bind to the claical oestrogen receptor to produce typical oestrogenic responses in animals.They can also act as agonists and antagonists of endogenous oestrogens.A further twist in this complicated tale was added by the discovery of another oestrogen receptor, the oestrogen b receptor, to which many phyto-oestrogens bind preferentially, compared with the oestrogen a receptor.The two kinds of receptor have different tiue distributions, and the response to phyto-oestrogens depends on both the receptor type and the tiue.Most of the research to date has concentrated on the a receptor but, to unravel the effects of phyto-oestrogens, we need to know what is happening at both receptors.The Food Standards Agency is research in this area too.There are also other concerns, such as the validity of data on the dietary intake of phyto-oestrogens, much of which is 15 years old and does not reflect recent changes in diet.The Agency is looking into new and improved estimates of dietary exposure.Another recommendation from the COT was for research to determine ‘whether ingestion of soya-based formulae carries any risk for infants’.In response, MAFF launched a programme of research in 1997, which the Food Standards Agency is continuing today.The Agency is also supporting work to develop reliable analytical methods and reference standards for phytooestrogens, including research to synthesise pure stable standards.The interaction between real, objectively aeed risks and the risks people perceive also underlies approaches to BSE.
6.Bovine spongiform encephalopathy BSE is just one of a whole family of diseases called transmiible spongiform encephalopathies(TSEs).Its distinguishing factor is that it can jump the species barrier.The symptoms vary slightly from species to species and disease to disease, but all TSEs share a pattern of changes in the brain.Changes in be-haviour (which resulted in cows with BSE being described as ‘mad’) are not nearly as straightforward or uniform — one of the reasons why it is so difficult to be certain of a diagnosis of BSE until a postmortem has been carried out.All the TSEs are thought to be caused by prions.These are unusual infective agents that contain no nucleic acid.Apparently, the prion protein can fold in two different ways, one normal and the other characteristic of disease, and the defective version can somehow subvert normal versions and cause them to fold wrongly too.Prions are remarkably resistant to breakdown; they withstand heat, enzymes and many other treatments that would normally destroy proteins.The link to humans was first noted in 1996, when the CJD surveillance unit reported a new pattern of CJD in people (variant CJD (vCJD)); in particular, it seemed to be affecting younger people.Evidence pointed to BSE as the source.The pathology of the disease is the same; the chemical signature of the protein in a Western blot is the same: and BSE and vCJD both have the same incubation signature when injected into the ‘mouse panel’.By the end of August 2000, 69 people had died of vCJD, five more are ‘probable’ cases, awaiting postmortem confirmation, and a further eight were suspected of having the disease but are still alive.The latest analysis suggests that the trend is still increasing, although the authors stre that they ‘‘cannot tell how long the current increasing trend will continue and, therefore, have not predicted future number of cases’’.Scientific caution, however, is not what newspaper readers want to see; they want to know how many people are going to die from vCJD.The dilemma is that nobody really knows.At the beginning of the epidemic, estimates varied hugely because, with so few cases, there was maive uncertainty about the natural history of the disease.Without accurate information, it is difficult to build useful models.For example, one of the key aspects of a disease’s natural history is the rate at which one case infects new hosts, effectively the rate of reproduction of the disease.Early in the outbreak of vCJD, the European Union’s Scientific Steering Committee thought that up to 500 000 people might be exposed to BSE from a single infected animal.This fuelled speculation that millions of people were at risk of the disease.The evidence is now a little le uncertain.A recent study from Neil Ferguson and his colleagues at the Wellcome Trust Centre for the Epidemiology of Infectious Disease in Oxford concluded that fewer than two cases arise from a single infected animal.One year ago, with le data, this estimate was over 100.More importantly, the study used this and other information to predict that the number of cases will lie between 63 and 136 000 cases.The true outcome depends crucially on the incubation period of the disease, with very high levels aociated with very long incubation periods.If the average incubation period drops to le than 20 years, the number of projected cases drops to between 63 and 630.Numbers like 63 or 136 000 are perhaps spuriously accurate, but they demonstrate that some of the wilder early guees are becoming more and more unlikely.As more is learned about the disease, so estimates of its course become le uncertain, but even when very little was known action had to be taken to protect consumers.The ban on feeding meat and bone meal to ruminants was the first measure introduced to protect the public.It was followed by the SRM ban, which removes from the food chain so-called Specified Risk Material; brain, spinal cord and intestines, tiues in which infectivity seems to be concentrated.The Over Thirty Month rule since 1996 has kept animals more than 30 months old out of the food chain.The age limit of 30 months was set because the clinical disease does not generally develop until the animal is about 35 months old, with a mean of 60 months, and because in the 2 years before it was introduced there had been only three cases of clinically identified BSE in animals younger than 30 months.UK animals younger than 30 months are allowed into the human food chain.Older animals are not used for food after slaughter.None of the control measures eliminates risk.Since the Southwood report, this has not been the intention of the scientists advising government.All the proposed measures have been designed to reduce risk, not to eliminate it entirely.The Over Thirty Month rule, for example, aims to minimize the poibility of an infected animal being eaten.The number of animals younger than 30 months of age, slaughtered for human consumption, that are infected with BSE and are within 12 months of developing BSE this year is somewhere between 0 and 4, with a mean of 1.2.That is one animal out of roughly 2 million slaughtered for food.Reduced risk, not no risk.Another question is whether BSE exists in sheep and is masked by scrapie, another TSE.The first reported case of scrapie in sheep was noted in 1732.As far as is known, scrapie poses no risk to humans — there is no evidence of any TSE in people that can be linked to eating sheep, even among people who relish the tiues most likely to be infective.There are 40 million sheep in the UK, of which perhaps 10 000 have scrapie.It is not poible to be absolutely certain that BSE is not present in sheep so, on the principle of reducing risk, specified risk materials are removed from sheep as a precaution.Looking to the future, in July 2000, the UK Government iued a consultative document for a national scrapie plan to eliminate the disease in Great Britain.One prong of this attack will be to test rams and breed only from resistant animals.In summary, dealing with BSE involved — and still involves — many unknowns.Some of these are becoming le unknown as time goes by, and are influencing the response of the Food Standards Agency.There is still no good reason to relax any of the control measures, although the Agency is already thinking about the future, and about the kinds of evidence that will be needed in order to rethink the controls currently in place.
7.Conclusion As has been indicated throughout, much of what the FSA has to deal with is uncertain.And yet the Agency needs to offer reasonably straight forward advice.At the same time, we need to reduce uncertainty by supporting useful scientific research.On the horizon are two scientific developments that might offer new insights.Different people clearly react differently to different foodstuffs.There is the question of allergic reactions: some people are allergic, others not.As information from the human genome project comes in, it may be poible to uncover the differences that make some people react to a particular chemical and others not.We may also learn how different biochemical pathways affect the metabolic fate of potentially dangerous com-pounds in food.This kind of information will enable toxicologists to go well beyond the traditional rodent bioaay to come up with entirely new ways of aeing the safety of chemicals in food.In the same realm of learning how to deal with people as individuals, scientists may soon have afar better understanding of the links between specific genes and elements of food safety and dietary health.If there is genetic heterogeneity for an important pathway, e.g.the interaction between isoflavones and free radicals, then advice that is entirely appropriate for one group of people, defined by their genome, may be wrong, and even poibly dangerous, for another.The FSA is looking forward and supporting work to explore ways in which these powerful technologies might help us to ae food toxins.Exactly how the new sciences of functional ge-nomics and applied proteomics will be put to work to reduce uncertainty in the science of food safety is anybody’s gue.But it is certain that they will, and the Food Standards Agency will embrace new findings and adjust its advice accordingly
