Select Committee on Science and Technology Minutes of Evidence


Examination of Witnesses (Questions 440-459)

TUESDAY 25 FEBRUARY 2003

PROFESSOR BILL REILLY AND PROFESSOR CHRIS THORNS

Lord Oxburgh

  440. In fact we have gone into a number of the aspects of this question, which is fundamentally: which animal diseases are under surveillance and who decides what is to be done and what the priorities are? To be more specific, are the priorities primarily driven by animal husbandry or by implications for human health?  (Professor Thorns) As we have already said, the priorities first of all are dictated by certain national and international requirements. Some of these are enshrined in legislation. Examples will be TSE in cattle and sheep, bovine tuberculosis in cattle, brucellosis in cattle, sheep and goats and to a certain extent salmonella in poultry, particularly breeders and layers. The driver for this in some cases is animal disease and welfare but of course in other cases it is the fact that they are zoonoses. New emerging diseases is a very important component of our work at VLA, and I am sure at the Scottish Agricultural College. Professor Stephen Palmer reminded me that nearly 70 per cent of all new and emerging diseases turn out to be zoonotic and therefore it is very important that this particular project is given high priority. Indeed, the front line people in this are the veterinary pathologists and microbiologists at the regional laboratories in England and Wales and the Scottish Agricultural College in Scotland. I think the examples of BSE and the pig complex (PMWS/PDNS), which is thought to be caused by a viral disease, are relatively recent examples and they were picked up in regional laboratories really quite early through an unusual pathology linked with microbiology. That is a very important point. However, as Professor Reilly said, there is another large group of organisms which generally do not cause clinical disease in animals but they are of huge zoonotic significance. These include VTEC salmonella and campylobacter; these are often referred to as the big three. You could add cryptosporidium to these as well. These are of public health significance. I would say that the Department of Health, the Food Standards Agency and DEFRA identify those infections as key in public health terms, and that is where we do a lot of our target surveillance at abattoirs and on the farm.

  441. To follow that up, you have sufficient resources in fact to meet these demands imposed by these external bodies so that you do not have to exercise internal discretion to the priorities in these respects?  (Professor Thorns) Resources are not infinite, and therefore it is very important that we have continued discussions with DEFRA in particular and DEFRA has continued discussions with the Food Standards Agency and the Department of Health. That is why I mention that it seems to me very important that these are brought into a risk-based framework where the priorities are totally transparent because I do not believe that we can offer everything simultaneously.  (Professor Reilly) May I comment on one aspect of the resources because I think for clinical diagnostic work I would agree with Professor Thorns but for some of the other pathogens, particularly I give the example of E-coli O157, I do not believe the laboratories are sufficiently resourced to look and identify the prevalence of these pathogens without specifically funded projects. That is the way this surveillance work has been carried out. Essentially we have had one major study to set the base line; that was separately funded. If, over a period of time, we want to follow how the prevalence is changing, we will need to fund additional similar studies. I do not believe the routine diagnostic budget would allow sufficient samples to be examined. In any event, the sample bias for clinical material would be quite wrong for looking at that. I think for the difficult ones, as Professor Thorns mentioned, additional resources are required to fund these.

  442. Would you normally expect this to come through DEFRA through negotiation with other bodies or would other bodies have contracts with the agencies directly?  (Professor Thorns) For most of our survey work for these particular pathogens, the salmonellae and the food-borne pathogens, our contracts come through DEFRA. I would like to add a further point to Professor Reilly's points. At the moment, these sorts of projects are annual projects and the targeted surveys which are animal surveys repeated every two or three years. For our organisation, it would be extremely helpful if our projects were extended to three to five years and that the targeted surveillance would be enshrined in ongoing surveillance rather than stop-start targeted surveillance. That would help us in our resource planning.

Lord McColl of Dulwich

  443. We have dealt with some of this already. Is the current system effective in picking up diseases that might have consequences for human health; if not, how could it be improved?  (Professor Reilly) I can start on this question. I think we have already touched on some of these. I see two difficulties with the comprehensiveness of the surveillance programme. First of all, there is targeting the organisms that do not cause clinical disease, and then, secondly, there is the economic cost of actually testing clinical samples, and the economic effects on agriculture of that. I can give one example of this. We have recently been trying to develop a genotyping system for Cryptosporidium Parvum. In Scotland we have a large number of water supplies that are not filtered and therefore cryptosporidium can come through the water. As a way to identify the source of many of these cryptosporidia, it would be useful to be able to fingerprint them to say: is it an animal source; is it a human source, and sub-divide the animal isolates. We had specific funding to look at this work but we found it impossible to get any significant number of samples from sheep. It is not economically worthwhile for the farming community to pay for laboratory investigations in sheep in many instances. This was particularly harmful to the study because we feel that sheep are probably the major contributor to contamination of water. In terms of the comprehensive nature of how we can identify infections that are likely to come through to humans, that is done through identification of the organisms: does it cause clinical disease in animals or not? If it does not, we need to target specifically the organisms. Then how do we address the bias in clinical submissions on which most of the surveillance is based? How do we get samples from those species where economics preclude against that initiative being taken? That would go across the age ranges: perhaps young lambs are more important than adult ewes. But sample bias will be age-dependent. I think there are ways we could target sample collection that would help us understand the prevalence of specific organisms in our livestock communities.  (Professor Thorns) I would agree with everything Professor Reilly has said. I would like very briefly to add, going back to the risk question, that to me it is very important we target the appropriate population of animals in terms of zoonoses, those that pose greatest risk. For me, one of the questions that has not been answered in this country, or in very few countries in fact, apart perhaps from Denmark, is: what is the relative contribution of livestock species and their products to human infections via the food and the environment? If we could begin to answer that question, and I think quite a lot of the data is out there to be able to answer that, we could perhaps then prioritise some of the targeted surveillance towards healthy populations that may harbour or pose the greatest risks for the zoonotic pathogens transferring through the food chain.

  444. I suppose one of the problems is that if these diseases from animals produce diseases in humans that are very mild, they are not likely to be investigated, are they? For instance, with diarrhoea, most cases are not investigated.  (Professor Reilly) I think if we are going to start getting into surveillance of human disease, which is what we are talking about here, we must recognise that any figures that any surveillance centre produces are heavily biased again towards the sample collection. Most surveillance systems are based on clinical sampling. For example, I would be very surprised if we have much of an under-estimate of cases of E-coli O157 in children because it is serious clinical disease. I think we will capture most of those, but I am absolutely certain we capture less than maybe 0.1 per cent of the cases of (NORD) virus infection that also causes diarrhoea because the diarrhoea itself may be less serious, but many laboratories do not have the ability to make a diagnosis. Yes, I agree with you that perhaps the less serious infections are not being captured by the human surveillance system for some of the infections. I could make another point that I think is important when we are talking about zoonotic disease, which is to actually keep a sense of proportion in that. I think we tend to think of zoonotic disease as a nightmare scenario. That is not how I would describe it. Professor Thorns has mentioned the top three. Perhaps I could mention the top four: campylobacter, salmonella, cryptosporidium and E-coli O157. These are all zoonotic infections but we really do not know how many human cases actually come from animals, what the role of environmental contamination is, which probably from an animal reservoir, and also person-to-person spread. When we get away from these four, and in Scotland for campylobacter we are taking about 5,000 or 6,000 laboratory confirmed infections, for E-coli O157 we are talking about 250 confirmed infections. For others, such as leptospirosis, in an average year we might see two cases. For toxoplasmosis, we might be seeing 15. The other infections are in a different league in terms of numbers. That is not to say that they are unimportant to the individual, but it helps to put the whole question of zoonotic disease into perspective. The most important of these four is enteric infection, which may be food, water-borne, animal contact, environmental or person-to-person. I think the point Professor Thorns was making is that we really do not know for most of these what proportion is attributable to livestock.

Chairman

  445. Can I ask what, in your opinion, is the role of the private veterinary practitioner, especially the large animal practitioner, in surveillance and reporting because he is in private practice and he has not got a national health service to back him up? Does he or should he have an important role in identification of infection and health maintenance?  (Professor Reilly) I mentioned earlier that I teach public health to veterinary students and the one message that I hope to get across to the students is that general practitioners are at the front line of veterinary public health. This is the group of individuals who will see, recognise, hopefully identify, and treat disease, which will include the zoonotic disease. These individuals should be at the front line of passing that risk assessment on to the other appropriate authorities. I think there are some difficulties because, with the exception of notifiable disease and the reportable infections, such as salmonella, there may be an adverse effect on veterinary practice to become too closely involved in that. There is an economic disincentive to take samples for public health reasons. Who bears the cost of investigation if someone suspects E-coli O157 in a livestock unit? We, as a society, must address that. Fortunately, we have agreement with the Scottish Agricultural College, and I think the same exists within the Veterinary Laboratory Agency, that if we have a specific incident, the cost of that investigation is picked up through public funding, but the cost of the private practitioner is not. The private practitioner is at the front line, I stress, of zoonotic disease investigation.

Lord Oxburgh

  446. In other sessions, we have heard a certain amount about near-patient testing in the recognition of particular diseases. Is near-patient testing something that is coming in in the veterinary field, because there are certain possibilities for private practice. There are relatively inexpensive tests for human beings. Is there any discussion of this in the veterinary field?  (Professor Thorns) Yes, there has been a lot of discussion over a number of years about near-patient testing—simple rapid tests. It is not a great area of expertise for me. My opinion is that with most of these rapid tests that are available, at best we can see them moving towards those private practices which have small laboratories attached to them. In most cases we would not be talking about a test that you have done by the side of the animal in the pen, but the test could possibly be done in the car of the veterinary surgeon, but most likely these tests will be done, if they have a practice laboratory, in that laboratory, or they would be developed and used in a laboratory such as in VLA or a private one.

  447. I am talking about non-laboratory tests which can be done effectively at the side of the animal, unless it is pouring with rain.  (Professor Thorns) For infections?

  448. Yes?  (Professor Thorns) My opinion is that at the moment they are not used widely, but they exist and from time to time we hear of tests that are claimed to work next to an animal that is kicking about and jumping all over the place, and so on, but, as far as I know, these tests have not really come to fruition.

Chairman

  449. We have been talking about farm animals. What about the situation with companion animals? Is that a field that is adequately covered both in general practice, or by the VLA, or by public health?  (Professor Thorns) I can start firstly by saying that of course, as I mentioned, the VLA does not deal with companion animals except for specialist referrals such as rabies and at the moment the burden of responsibility is held again with the private veterinary surgeons and veterinary schools. I am aware at a recent zoonoses conference, one at which I believe you actually spoke, in Cardiff there was a lot of discussion about the likely risk of cats and dogs coming back from the Continent who have picked up zoonotic, parasitic disease, to ensure that these infections are spotted early in this country. I think there is a potential gap there that probably is being addressed, and certainly needs to be addressed.  (Professor Reilly) In Scotland, the SAC does process samples from companion animals and two veterinary schools will also receive clinical material, particularly through the developments of small animal clinics where they act as a referral centre. I think we have some good examples of how that system works. If I can give you one such, I had a call last year from a veterinary microbiologist at the Glasgow Veterinary School who reported an isolate of corynebacterium ulcerans from two cats from the same household. Out of his interest, he had sent these isolates to the PHLS and the PHLS had found these isolates to be toxigenetic and were producing diphtheria toxin. That raised two questions: what do you do about the immediate contacts of the cats; and, are cats a problem for this pathogen? Using our established lines of communication from veterinary school to SCIEH, we spoke to the PHLS, who spoke to CDSC at PHLS who spoke to the appropriate health authority in England where the cats had originated to allow the CCDC to ensure that there was no human implication in the family where these cats had come from. That allowed us to put forward a proposal to look at cats to see whether corynebacterium ulcerans was a problem in cats that had previously not been recognised. If material had not gone to this referral service at the veterinary school, we would not have been able to follow that change through. In that instance, I think the system worked very well, but I suspect that most small companion animal material will go through private laboratories and that is then lost to surveillance. If it had been a private laboratory that processed these samples, we would not have been able to take that any further. We do capture some; we do not capture as much as we do in the large animal side, but perhaps companion animals are not the major source of zoonotic disease that livestock are.

Lord Rea

  450. This question concerns the actions that can be taken to prevent transmission from animals to man, supposing a problem is detected by any of the methods that we have been discussing.  (Professor Reilly) I think there are two different scenarios here. One is where an infection is identified in animals and the other would be where it is identified in humans because both operate. If it is a notifiable or a reportable disease, reportable is essentially salmonella, then there is sufficient legislation to allow the veterinary authorities, that is the state veterinary service, the VLA, the (SVIO), to look at controls in that population. For example, the Zoonoses Order came about in the mid-Seventies because of an isolate of salmonella paratyphis B in cattle and it was recognised that no authority had a jurisdiction to impose restrictions on the movement of these animals or their products. For brucella, TB or salmonella, there are authorities and powers to manage that animal population to protect human health, but for the others, the zoonotic disease, that does not exist. You could not apply restrictions because of chlamydia in sheep, enzootic abortion. There would be no authority, no power, to apply restrictions to sheep populations to protect human health. One could give plenty of advice and guidance but not apply any controls. I think the power to control is only there when it is a notifiable or a reportable disease. That is something that I have found is very different to my work with my medical colleagues, where there is a greater range of power to manage all disease situations. The same is also true where the organism is found on trace back, a not uncommon scenario within the UK. I know it also applies in England and Wales. That is where we will receive some information from my medical colleagues in public health medicine about a single case or a cluster of cases of an infection of O157 that has some readily identifiable animal component; in other words, there has been a visit to a farm or some other trace back that we can go back to the farm of origin. Here we do try to investigate but more for information-gathering rather than applying controls. We are still building up our picture of the epidemiology but there are issues of confidentiality. Before I will ask the veterinary laboratories to investigate, I need to be sure that the owner of the livestock is happy to participate in that investigation because there is no power to require sampling, especially if it is a case of E-coli O157. Usually the scenario is that public health medicine will speak to SCIEH, who will ask them to obtain the permission of the farming community to investigate further. When we receive that, we will ask the veterinary laboratory to go to the farm, take appropriate samples and report back. The system really does work very well. As I said earlier, I know it also happens in England and Wales. If we identify livestock that is infected with E-coli O157, then there are no powers to apply controls in that unit. There may be a disincentive for someone to allow testing. Maybe they do not want the stigma. This is going off slightly at a tangent but this is one of the problems with the Food Standards Agency and sampling food products. Because of the openness culture we have when the source of the material is put into the public domain, it means that producers will think, "I do not want to take part in this voluntary survey because, if you find salmonella, you are going to come back and it will damage my business". So we lose information of what is out there. If it is not a notifiable disease, and it is not a reportable infection, we are looking at these investigations to increase our understanding of the epidemiology and not to apply restrictions on the farm other than good basic advice that would always be given about personal hygiene and control measures on the farm to protect the health of the people in contact with the animals.

  451. What about treatment of the herd that is found to be harbouring the zoonosis, whatever it is?  (Professor Reilly) I cannot think of a single zoonotic disease identified through that way where any treatment would be applied to the herd. If it was E-coli O157, I cannot think of any treatment that would be appropriate in those circumstances. If it were something less common like leptospirosis, then the farmer would himself be considering the use of vaccination, for example, if that was not already being used. As much as anything, that would be to protect animal health and perhaps also to meet the farmers' requirements and COSHH (Control Of Substances Hazardous to Health) regulations to protect the health of their employees. There may be a few instances like leptospirosis but for the major zoonotic diseases like E-coli O157, there would not be blanket treatment of the livestock unit.

Chairman

  452. In view of what you have said about non-notifiable diseases, is there a case for increasing the number of notifiable diseases in certain cases?  (Professor Reilly) I think there is certainly a case for increasing the number of infections in which action can be taken. We are often being directed through Europe, and the Zoonoses Directive has identified the major infections that are important. It still very largely focuses on salmonella. I would like to see the powers available under the Zoonoses Order extended to include other infections, such as E-coli O157. That does not make it notifiable. We are not talking about bringing in the full armoury of some of the control policies, but it would allow us to apply the basic, sensible hygiene measures that might be available to reduce some of the burden but also to gain more information on these infections. We are still very early in our understanding of how E-coli O157 is maintained within animal populations, so, yes, more infections coming under the direction of the Zoonoses Order.

Lord Oxburgh

  453. I may have lost this in what you said but I got the impression that without the consent of the owner of the animal, you cannot actually carry out any test at all, regardless of what you suspect to be there?  (Professor Reilly) If it is a notifiable disease, then there is the authority to require sampling. If it is not a notifiable disease, then you have no authority to require samples to be taken. For example with brucellosis, if brucellosis is suspected, there is sufficient power under the legislation to require samples to be taken; if it is E-coli O157, we do not have that power.

Lord Lewis of Newnham

  454. Do you meet much opposition of this nature? Is it a common feature for people to refuse?  (Professor Reilly) I have to say that I am sufficiently remote from animal contact now not to be able to answer that honestly.  (Professor Thorns) We get quite a lot of co-operation on a voluntary basis, I must say. There are situations where it has proven very difficult for veterinary professionals and support staff to get on to farms, and it is an issue, as Professor Reilly has said, but we do get quite a lot of co-operation. This is not my particular field but I suspect the Food Safety Act could be invoked in certain circumstances if, for example, a product like milk coming from the farm was sold directly to the public was shown to be contaminated. You would need to ask somebody else how the Food Safety Act works in that respect.

Baroness Emerton

  455. That really leads into the next question. Is the general standard of hygiene on farms a concern and does it contribute to outbreaks of infectious disease in people? May I enlarge on that and ask: are there specific standards and, if there are specific standards, what are the constraints and who exercises the constraints?  (Professor Reilly) I shall begin, my Lord Chairman. The four organisms that I have identified are all spread through the faecal route, there is a faecal component, whether it is animal to Man, animal to food to Man, or Man to Man. I think we must recognise that for these four zoonotic diseases hygiene is the all-important question. But farms are not, have never been, I do not think ever will be, sterile environments. I think that what we are looking at is a degree of hygiene. I think that this really touches on biosecurity which is an important aspect now of controlling infections like campylobacter or tuberculosis in cattle; it is about the hygiene environment of the whole farm. I have no doubt that faecal organisms can spread through farms that are less hygienic than those that have good hygiene, but they will still spread. It is impossible to stop the movement of animal faeces. It is impossible to stop the movement of human faeces also. For example, where we have young animals, children, animals that are excreting faecal material, there is an opportunity for spread, so it does contribute to the spread of infection. There is a particular problem where the food is being perhaps produced on the farm. Perhaps one of the major areas would be dairy hygiene. I think it would be generally accepted that where we get these top four pathogens in milk, it is more likely to have arisen because of faecal contamination of the milk rather than from excretion through the udder. So I do think that we need to look at the hygiene of food production on the farm. I think there have been major changes and major improvements to dairy hygiene over time, but equally I think people working in the industry are subject to time pressures that everybody else is subject to, and I think there are opportunities for lapses in dairy hygiene. We have recognised this in Scotland. We do not allow the sale of unpasteurised milk, and have not done so since 1983. So there are ways to address it. You improve dairy hygiene and recognise that there can be failures, and so we have a safety net of pasteurisation. We can look at animals coming off the farm in as clean a way as possible. This is being tackled by the Meat Hygiene Service, by having standards of cleanliness of animals coming into a slaughter house, and so the very dirty animals—and for "dirty" read faecally contaminated—are rejected and are not allowed to enter the slaughter house. I think that has been a major step forward in reducing contamination of the final product coming out of the abattoir. But even in the cleanest looking animals there will be organisms on the outside of them, and I think we must recognise that animals are never going to be sterile. All animals entering the food chain will have organisms on the outside and on the inside, and sometimes these organisms will be pathogenic. The whole thrust of the food chain—and that is starting on the farm—is to produce material going into the chain that is as little contaminated as possible.

Baroness Finlay of Llandaff

  456. I wondered how much the range of organisms you think is being affected by the use of antibiotics, and whether you see a different range of organisms on animals that are reared on organic firms versus those that are reared by perhaps more mass-production methods?  (Professor Reilly) Perhaps I could start, and I am sure Professor Thorns would want to comment. Cryptosporidium I think is taken out of the equation altogether, because as an antibiotic it is not very relevant, so we are left with the top three. The E-coli O157 is not antibiotic-resistant, so I do not think there has been any selection for resistant strains with O157. We are probably more concerned about antibiotic resistance in salmonella, and it is in salmonella infection where there has been the most dramatic decline in infection over the last three years—we have seen a 60 per cent fall in the number of salmonella cases in humans throughout the UK. So I do not think the very fact that these organisms are resistant is giving them a biological advantage on the farm to allow them to come through. With campylobacter the proportion of isolates that are resistant to a small number of antibiotics is perhaps slightly increasing, but the numbers of campylobacter isolates are fortunately plateau-ing again and falling. So I see no evidence that antibiotic usage is selecting the populations on farms that are going to come through into the food chain more easily. I suspect the argument with organic farms is much more complicated than simply the use of antibiotics, and it is very often tied in with extensive farming as against intensive farming. I personally am not convinced that the fact that animals are reared extensively outside makes the food from those animals any safer. I think if we take campylobacter as an example campylobacter is widespread in our wild birds, and if our animals are outside then they are more likely to be exposed to these campylobacters from the wider environmental sources. I do not think it is a simple issue. I think the two are related.  (Professor Thorns) I think everybody in the veterinary profession—and Lord Soulsby has led this very much—recognises that the prudent use of antibiotics on farms is a prerequisite now. In terms of the range of organisms, this is a very complex business. I can see no evidence that antibiotics actually have a significant contribution to selecting one particular organism from another. Clearly the relationship between sensitive bacteria and resistant bacteria through use of antibiotics is related to how much antibiotics are used. In recent years there is clear evidence that the more prudent use of antibiotics in some livestock species is being fed into data which indicates that certain organisms or groups of organisms are less resistant than they were, say, five or ten years ago. It is a little bit more complicated than that now, because we realise that antibiotic resistance can be selected through, for example, the injudicious use of disinfectants perhaps, which is a little bit of a concern. There is no doubt, though, that the use of antibiotics does cause, if they are used improperly, certain selection of resistant organsims. There is indication that some resistant bacteria are actually less virulent than their parent strains, but that is not the reason for using antibiotics. I would like to point that out. So I think it is a very valid point. We do not see, and I have not seen, any real change in the ranges of bacteria, it is just the patterns of resistance.

Lord McColl of Dulwich

  457. I believe one of the reasons why campylobacter spread so widely was to do with the way that chickens were killed. Chickens, I believe, were fixed upside down and killed by electrocution, which ensured that the contents of their bowels were spread throughout their feathers. Is that still the practice?  (Professor Reilly) Again, if I can start with that one, my Lord Chairman, I think the system of modern poultry slaughter and processing cannot prevent the spread of organisms that come into the processing plant with the chickens. The processes of stunning them and de-feathering, followed by evisceration, do facilitate cross-contamination. This is very much a feature of a food product that has been processed at the rate of between 10,000 and 15,000 birds an hour; it is the sheer scale of poultry processing that leads to contamination. So if we are going to put birds carrying campylobacter, or salmonella for that matter, into the processing plant, we will spread these organisms during the process. I think the poultry industry has tried very hard to minimise this spread, but the very nature of the process—taking the feathers off and removing the intestines—does lead to cross-contamination. So yes—and this comes back perhaps to Baroness Emerton's question about hygiene on the farm—we would be striving to prevent infection of the birds with campylobacter on the farm, so that they are not carrying the organisms when they go to the processing plant. If there is a society requirement for inexpensive poultry meat we must expect that not only do we have intensive production on the farm, but we have intensive processing in the plant, and despite the best wishes of the industry, we do still get cross-contamination of the product. The latest survey from the Food Standards Agency, which I do not think has yet been finally published, is going to show somewhere around about 70 per cent of broilers carrying some campylobacter at the end when they come out of the processing plant.

  458. 70 per cent?  (Professor Reilly) Yes, 70.

  459. Is there something to be said for putting on the packets of bits of chicken that not only do you have to cook them, but you must not put them in any place where they can spread to other bits of food? What they do is they chop the chicken up on a wooden board, then they cook it, but then they chop the salad up on the same board, and that is where it gets into the food chain.  (Professor Reilly) I could not agree more. The message for food hygiene is not simply to cook food properly, but to observe good hygiene in the kitchen with regard to raw food. I must stress, it is not just poultry. We tend to talk more about poultry than any other food, but perhaps that is because we know more about poultry. There has been more testing of poultry than any other food. So perhaps some of the criticism we are facing is out of proportion to the problem in other foodstuffs. So for any raw food you should not allow cross-contamination. As a society I think we tend to have lost understanding of good basic kitchen hygiene.


 
previous page contents next page

House of Lords home page Parliament home page House of Commons home page search page enquiries index

© Parliamentary copyright 2003