• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services

Animal & Veterinary

  • Print
  • Share
  • E-mail

Panel Discussion: How Should CVM Evaluate Risk from Resistant Pathogens

DR. STERNER: We will begin with the session's introductions while people will filter into the room. I tried reading biographical sketches this morning and find those dreadfully boring and they don't really add a whole lot. What I have a lot more fun with is hearing people tell who they are and where they are from.

And I would challenge each of you to tell us one thing about yourself in addition to your professional interests that nobody might have ever guessed about you, just in case somebody wants to strike up a conversation after the panel discussion is over. So we will start with A) Dr. Apley on the end. Would you introduce yourself and give us a small biographical sketch.

DR. APLEY: I am Mike Apley. I am Assistant Professor of Beef-production Medicine, Iowa State University. My advanced training is in clinical pharmacology with my Ph.D. in boards. My interests are risk assessment. I work primarily in feed lot, clinical pharmacology and other animal species. And something nobody knew, semi-serious truck puller.

DR. LIEBERMAN: Well, I think this part has me more intimidated than my comments now. I am Patty Lieberman from the Center for Science in the Public Interest. I have been there as a staff scientist for about three years.

What people might not know about me is that my grandpa was a cattle dealer. And my father's biggest client was a pork producer/processor people. So I think people would probably not expect that I have some -- although nothing immediate in my life is revolving in agricultures at my work, but that I have some background in and appreciate for it.

DR. McCLURE: Hi. My name is Kent McClure. I am with the Animal Health Institute. I am both a veterinarian and a lawyer. I practice both veterinary medicine and law. And I practice law in a regulatory context. As far as something that someone might not know about me, I enjoy brewing beer at home.

DR. ANGULO: Hi. My name is Fred Angulo. I am the Chief of the FoodNet and the NARMS activities in the Food-borne and Diarrheal Diseases Branch at CDC, where I have been since 1993. And I am the proud father of three children. And I think that is the thing I am most proud of.

DR. MORRIS: I am Glenn Morris. I am on the faculty as a professor at the University of Maryland School of Medicine. I am a physician/epidemiologist. I was at CDC. I spent several years with FSIS with Mike Taylor.

And I am now happily back at the university working primarily in the area of emerging pathogens, with a particular focus on emergence of multi-resistant pathogens within a variety of environments. I will follow up with Fred. I am the father of three daughters. And that alone I am sure has had a major impact on my psyche, probably more than anything else in my life.

DR. CRAWFORD: Thanks. I am Les Crawford. I am Director of the Center for Food and Nutrition Policy at Georgetown University. I used to be at FDA, University of Georgia and also FSIS and about 12 other places. And I enjoy drinking beer at home.

(Laughter.)

DR. McEWEN: I am Scott McEwen and I knew that sitting beside Lester Crawford was going to be a problem, trying to follow that act. I am a professor at the University of Guelph in epidemiology, focusing on food safety. I have two boys and I like woodworking. I make Windsor chairs. And like all Canadians, I have built a log cabin.

(Laughter.)

DR. STERNER: Kenneth Petersen is not here yet. When he comes, we will go ahead and pin him with the same task that the rest of the panel has. I am Keith Sterner, moderator for the first part of this afternoon's session.

I am a graduate of Michigan State University in 1969. After that, I did two years on active duty in the United States Army Veterinary Corps, served in Seattle, Washington and Pouson, Korea. And I got an in-country discharge. And I found that if you go far enough in that direction, you will come back in that direction.

And I took about six months to go around the world. And due to the vagueness of my discharge papers, they entitled me to trans-oceanic transportation within one year of discharge. They failed to specify which ocean. And it is amazing when you show up at duty stations late enough on a Friday afternoon what you can get done and hopping military hops.

(Laughter.)

I am a co-owner of a ten-person mixed practice veterinary clinic in the central part of Michigan. We do all creatures except for horses and have -- and in the practice that my father started 52 years ago this month as a matter of fact.

And I have been active in organized veterinary medicine, having served as an Officer in the American Association of Bovine Practitioners. And I have been active in the National Mastitis Council and I am the past president of it, as well.

I am currently the Chair of the American Veterinary Medical Association's Council on Education. And I think I am here because Sharon has a grudge against me. I serve as the current Chair of VMAC, as well.

With those introductions, I would like to go ahead. And I am going to randomly move through the panel and ask them to address the questions here. And because Glenn Morris reminded me that he has real patients to see this afternoon and may have to leave as the discussion begins to wind down a bit, he has the prerogative of speaking first and trying to address these questions that the panel has been posed. So, Glenn, the floor is yours.

J. Glenn Morris, Jr., M.D.

DR. MORRIS: I am not sure that this is the appropriate award here because I am not sure I can necessarily completely answer these questions. It would have nice to hear somebody else first. However, I may slip out here in a minute. It's nothing personal. It's just unfortunately because of my scheduling. I have attending responsibilities this month and have patients waiting for me. So I am going to slip out in a little while.

But to deal with the issues that are raised here -- and, again, as I said, I will admit as to some uncertainty as to how to address these. The question first is what is an appropriate risk standard to apply to resistant pathogens.

And I guess as a physician, I struggle with trying to understand the concept of standards with regard to resistant pathogens. I am not overly fond of resistant pathogens. And I would prefer to see the numbers minimized. I recognize from a regulatory standpoint, there is a need to try to put this in a, you know, better framework.

I think if you begin to look at some of the impacts associated -- human health impacts associated with resistant pathogens, one of the things that has been mentioned for Campylobacter resistant to quinolones has been prolongation of diarrheal illness. I realize that there is probably of a sense of, oh, okay, so they have a few more days of diarrhea, so what.

I can tell you as a physician who has seen a lot of patients with Campylobacter infections that a couple of extra days of diarrhea is not a so what. If you had Campylobacter, it is a really -- you are pretty sick. Actually, the sickest patients I see in terms of occurrence of diarrheal disease are those with Campylobacter. It makes young adults really sick. And those couple of extra days of diarrhea are not trivial.

Nonetheless, I think there should also be a recognition that the approach that has been taken in this risk assessment has not really looked carefully at specific population breakdowns. And, again, this sort of gets into the question number two.

As an epidemiologist, clearly I think what happens with these types of data is that you are skewed more towards the high risk populations. Those are the people who come to see doctors. Those are the people who seek medical therapy. And it is the high risk populations that are at greatest risk for serious illness.

And, again, particularly from my vantage point in a, you know, large university medical center, the population that comes to mind most often is the HIV positive patients. Those patients are at clear risk for significant Campylobacter infections. For those patients, if they come in with serious illness, we need to be able to use medication empirically. We need to have a high degree of confidence that the drug we are using is going to be efficacious.

With the rising rates of resistance to quinolones of Campylobacter, suddenly what is our first-line drug We are beginning to have doubts about it. And so I realize that this doesn't give a quantitative level, but I will tell you from a clinical practice standpoint, the fact that we are beginning to have questions about the ability to use certain drugs empirically is a fairly significant problem.

I think that a second concern that arises in looking at this is that this model is, if you will, a static model. It recognizes things at one point in time in 1998. And unfortunately, this is not a static process. And I think what we are seeing is very much the dynamism of it which are rising rates of quinolone resistance to Campylobacter.

And I would like somehow to be able to see this concept of the dynamism of the process incorporated into the model, both in terms of the dynamism that reflects physician responses and physician use of drugs which I talked about earlier, but also the dynamism of, you know, potential amplification of resistant organisms within animal populations. It is a dynamic process.

So although you are talking about -- I believe the number was three percent in general population numbers, my understanding is that those numbers actually probably are fairly low compared to what is happening in 1999. And so the dynamic element of this is something that I think has to be taken into account.

And, again, I think there needs to be an ability to deal with the concept of amplification, that things tend to get worse fairly rapidly. And my concern in a hospital setting is that I am seeing things get worse fairly rapidly with multiple pathogens. And it is very clear to us when we are dealing with this on the front lines of medicine that we have a substantive problem across the board with rising resistance rates to all of our pathogens.

As has already been noted, there are multiple factors that drive that process. But I think that dynamic element needs to be considered when you look at the standards and the establishments of standards. And I think I will stop there.

DR. STERNER: Dr. Lieberman.

Patricia Lieberman, Ph.D.

DR. LIEBERMAN: Well, I think I jumped the gun yesterday and stated some of my views. So some of this is going to be a little repetitive. I would have to say that consumers feel that the only legal or scientific standards acceptable is the standard of reasonable certainty of no harm. And I guess I would have to express some concern about these -- looking into what other standards are with the thought that it is possible that CVM is considering trying to change these standards and how they would go about doing it.

It seems to us that a discussion of it or a guidance document or the discretion of the FDA Commissioner would not be an acceptable way to do that if that is what is going to happen and that that would have to be done either by rule-making or by Congress.

As to the appropriate populations on which to base the standards, I think we need to take into account the most susceptible members of the population, not the entire population of the United States, but thinking about children, the elderly and immunocompromised people for whom the disease is more likely to be harmful and, in fact, is more likely. And I think Dr. Morris already spoke to that.

With the issue of children, it seems like at this point, treatment with fluoroquinolone -- that the risk assessment undertaken by CVM which looks at fluoroquinolones wouldn't consider children at higher risk. But I think we need to keep in mind what I don't remember who said about how it seems likely that fluoroquinolones will be used in children whether or not they are approved for use and if they will in the future be approved for use in children.

I think it is very important that the threshold should be set to identify problems before people have been harmed, preferably looking at resistance in the livestock and also taking into account not just full-blown resistance, but decreases in susceptibility. And those should be dealt with as the early warnings which would necessitate mitigation strategies.

I have a few other comments that I guess I will make now. I guess they could also be done during the public comments. But about the issue of if this process is supposed to be transparent, what does that mean. What is the impact of having a public meeting when we have no real sense of how this information is going to be used and how the people who are the decision-makers -- you know, whether it is just so that we can vent our feelings to you and so you can say, "We listened", or whether or not these -- our input is shaping the decisions that are going to be made.

In other words, have we gotten riled up for nothing And I have only been in this field for a little while. But I feel like I have done this a lot already which is okay I guess. But it is hard to tell the impact of it.

And with this particular risk assessment on fluoroquinolones, now that there has been this risk assessment that shows there has been harm to humans, what is going to be done about How can the use remain permissible

And looking at how other regulators look at risk, seeing how the people who regulate food additives, if they had a food additive that harmed about 5,000 people a year, would they feel that they had to take action And how does the situation differ from that of a food additive because the prevalence of resistant Campylobacter is likely to increase and fluoroquinolone-resistant Salmonella are beginning to emerge

So those are some things that have me concerned about the process. And I guess I will stop.

DR. STERNER: Dr. Crawford.

Dr. Lester Crawford

DR. CRAWFORD: Thank you. I am going to address most of my remarks -- I will cover these three subjects. But the framework I will use will be the concept of threshold. As one who was involved in earlier initiatives with respect to antibiotic resistance at CVM and elsewhere, I think this is a concept that we could certainly have used in dealing with those problems. And am thinking primarily of penicillin and tetracycline.

I believe that the risk assessment is obviously well done and it is an enormously good tool for dealing with this. And it leads then naturally into what I will say about thresholds.

I believe they must be based in regulation and not in a gentlemen's agreement. I mentioned earlier the Regulatory Improvement Act of 1999. And when I testified on that, I mentioned this particular aspect.

The second thing is that products that are known to rapidly engender resistance or that are known to have dangerous cross-resistance profiles should not be eligible for approval. And I believe the Framework Document addresses that quite adequately.

Presumably, these would wash-out in the pre-approval risk assessment process. And whoever made the comment earlier about pre-approval risk assessment I think was right on target.

Thirdly, I think that post-approval monitoring should be performed for all approved antibiotics from animal isolates. I recognize that human isolates would help to some extent. But animal isolates are primarily within CVM's purview. And so that should be sufficient.

I think they should be tailored, the thresholds, for each antibiotic, but consistent in magnitude and based on the minimum inhibitory concentration of the target organisms. For example, when ten percent of isolates from human and veterinary isolates require a significant concentration increase over the pre-approval level, action should be taken. The question is what is the action. And that is the $64,000.00 question. Something like a moratorium with the approval still in place might be a good idea.

We were -- we actually did discuss this in the London Conference on Antibiotic Resistance in 1981 where we presented university figures from the University of Georgia which show that over the many years of use in the veterinary teaching hospital there, that we had a natural selection process for antibiotics because when an antibiotic -- when bacteria became so resistant to certain antibiotics that clinicians stopped ordering those from the pharmacy.

So over a period of time, their resistance profiles declined and susceptibility improved. And I thought it was a very powerful testimony based on fact and based on thousands and thousands of isolates. Unfortunately, we did not translate that into regulatory action. But it was nonetheless interesting.

And I think that if a moratorium or some sort of ameliorating action is initiated, I think that monitoring should continue. And if there is no improvement, then perhaps the moratorium or whatever the remedial action is should continue.

And I also think that you have to be very careful -- and I notice you've got a legal question here which I am not qualified to answer. But I think you have to be very careful about due process. And you need to have a carefully articulated position on what happens to administrative hearings, whether these would be truncated, abbreviated or obviated. And I would recommend some of all three in closing.

DR. STERNER: Thank you, Lester. Kent McClure.

Kent McClure, Esquire, D.V.M.

DR. McCLURE: Thank you. First of all, I want to say that I am very happy to be here today. I believe antibiotic resistance is an important issue that needs addressing. And we are happy to have input on the process with respect to this panel. We have had frankly little time to review the risk assessment. And we intend to comment on it, analyze it and do that in detail. And we will provide further comments later.

But I do want to say that one thing that I think has been missed from looking through the -- just a preliminary look through the document was everything was stated in a negative sense. And if you flip it around, one thing that struck me was that you can say that for the average U.S. citizen, there was greater than a 99.99 percent probability that they would be unaffected by a resistant Campylobacteriosis. And I think that has to be kept in mind when you discuss what standards should apply and how we should implement them.

I am going to try to talk just a second about the legal standard. It is impossible in this context in this time period to have a thorough analysis of it. But I do want to kind of just give a few thoughts on it. One is -- and I will do a nutshell answer first. And that is that the Federal Food, Drug and Cosmetic Act in this context does not mandate any standard other than safe.

The statute requires that a new animal drug be shown to be safe. Safe is defined as referring to the health of animal or man. The statute gives no further guidance on the standard.

The statute does provide some factors that have to be considered. But it doesn't give you a standard to weigh those factors against. And there is a big difference in articulating factors to consider and then articulating the standard that you weigh them against. They are not the same thing.

In this context, the regulations promulgated by the FDA parallel the statute. They provide factors to consider, but no standard. In court cases in which the FDA has been a party in this context -- and that is, this context is the approval of new animal drugs in a food-producing species -- the FDA has not argued that any regulation they have promulgated sets a safety standard.

The Federal Courts that have then tried to determine what is the safety standard that applies have held that there is not one. A quote from one of the Courts that considered it is that, "The Food, Drug and Cosmetic Act does not indicate the standard an applicant must meet to demonstrate a new drug safety or the evidence upon which the FDA must base its safety determination." That was a new animal drug in a food-producing species -- that case involved that.

There are several other points that Federal Courts have made that are of interest to this discussion. One of them is that the D.C. Court of Appeals has at least twice rejected the Agency's argument that the legislative history behind the Animal Drug Amendments of 1968 set the particular standard that must be used to evaluate the safety of new animal drug in a food-producing species.

The D.C. Court of Appeals has also held that a risk benefit analysis is inherent in the process of safety evaluation in new animal drugs for food-producing species. Now, we have heard some talk today about how you can have only a risk-oriented standard.

The D.C. Court of Appeals has remanded cases back to the Agency for further consideration when that standard has been applied. That is not one that should pass the D.C. Court of Appeals.

And finally, I would say that one thing that is evident when you gather the court cases that deal with safety standards in this context, conspicuously absent from those decisions is a discussion of reasonable certain of no harm. You will not find it mentioned in any of them.

The take-away message from that is that the Agency has flexibility. They have the flexibility to craft a solution that is to a unique situation, that is workable, reasonable and protects public health.

Like I said, there is a whole lot more to that analysis than what I just articulated. But for the sake of time, I am going to move on. The -- I want to say that we agree with CVM that there are significant differences between residue-based issues and resistant-based issues.

Attempting to regulate resistance in the context of residues is like trying to put a square peg in a round hole. And that is part of the reason why we have had so much discussion over this and why CVM has had to struggle with this.

The USDA and the FDA both have standards -- or not standards, but regulations that deal with pathogens or can be interpreted to cover them. And it is imperative that they be the same. The USDA standard we believe is the most appropriate. It takes into account the HACCP Program of pathogen reduction and the fact that raw meat and poultry is intended to be cooked prior to consumption. As discussed earlier, food packaging and labeling includes warnings about how to handle food and cooking.

The USDA standard revolves around the quantity of pathogen that is present. The Poultry Inspection Act and the Meat Inspection Acts do not consider a pathogen, resistant or otherwise, to make a carcass adulterated if the quantity does not ordinarily render it injurious to health. And this standard needs to be explored by the FDA in cooperation with the USDA. And there is a huge reason for that.

And that is that you have almost identical language -- and if I had it before me to compare, I might say it was identical, but I would have to have it before me to do it -- on that particular standard. And what I am talking about is whether or not you consider a resistant pathogen to be an added substance or a -- or just a substance.

And you can't have identical language in two different regulations in the Code of Federal Regulations that is interpreted differently by the Courts, even though it comes from different agencies. A Federal Court does not say in this context, this word means this and in the same context with a different agency, this word in the same sentence means something different. It doesn't happen.

If you define a resistant pathogen to be an added substance, then every carcass that has a resistant pathogen on it is adulterated. And you can't define it one way in one place and a different way in another place in the same regulatory scheme on the same stuff when you use the same language. That won't fly. Legally that won't fly, at least that is my opinion.

The other thing that I think is important to note here is that it is important to ask where in the process is the standard applied. We have heard a lot about the risk assessment and the thresholds and things like that and post-monitoring surveillance. I want to say first of all that AHI has been in favor of risk assessments. We have helped fund one, not this particular one, but another one. And so we are glad to see them done.

We are also in favor of post-approval monitoring. We have applauded the NARMS program. However, it would be wrong and not legally justified to hold the drug approval process hostage to post-approval activities.

We talk about setting thresholds. You know, if you went back several years in time and you said let's set a threshold for fluoroquinolone with Campylobacter, you wouldn't have even foreseen that as being a problem. I guess my point is that would have never even come into the mix because it wouldn't have been considered.

And so when you sit down with a new drug, it is impossible that you can have all the areas ahead of time to know what you are going to consider. And so to require a manufacture to agree up front to withdraw a product from the market or do whatever simply because some arbitrary threshold is crossed, 1) is not supported by the Act, and 2) doesn't make sense.

The Federal Food, Drug and Cosmetic Act has provisions within for removing products from the market. In fact, most of the time when there is a legitimate problem with products, the manufacture and the Agency work together on a solution. But if they can't come to a solution, then the Act has provisions for removal of product from the marketplace.

It is not -- I guess what the context is, is that it would not be right for the Agency to circumvent the provisions of the Act through the standard, itself. The bottom line is that the standard for regulation has to be coordinated with the USDA. You can't do it in a vacuum as we have heard many times.

The Agency does have tremendous flexibility in dealing with this situation. And we would say that the approval process and post-surveillance monitoring are distinct activities. You can have them both going forward at the same time. You can monitor products and take action on what happens.

And I guess rather than going on and rambling, I will conclude with that. I will just say that we look forward to an ongoing discussion on the topic and working further with CVM. Thank you.

DR. STERNER: Thank you. Fred.

Dr. Fred Angulo

DR. ANGULO: I think, as many of you know, CDC is a non-regulatory agency with the mission of identifying risks and working with partnerships to try to mitigate those risks. And I think it is a matter of public record that CDC identified the potential risk of fluoroquinolone-resistant Campylobacter prior to the approval of fluoroquinolones in poultry. In fact, it is a matter of public record that we advised against -- or, that's too strong, that we had concerns about such an approval.

Nonetheless, I don't mean to go back there again, but that is in contrast to what Kent just mentioned. Nonetheless, I did want to comment perhaps on some of the questions that were raised. The first question about the appropriate risk standard to apply to resistant pathogens, I think the question really means to ask the appropriate risk standard to apply to resistant pathogens which result from the use of antimicrobials in food animals, or what is the appropriate risk standard to apply to the use of antimicrobials in food animals.

Without commenting on the current statute or policy, the risk of adverse human health consequences due to the risk of antimicrobials in food animals should be managed based on the best available data, for example, the current risk assessment, and should protect the public from harm.

Several governmental agencies manage and regulate the risk of food-borne diseases. In particular, the USDA Food Safety Inspection Service manages the risk of food-borne diseases from meat and poultry. The FDA's Center for Veterinary Medicine manages the incremental risk or the increased risk of food-borne diseases which are resistant to antibiotics as a consequence of antibiotic use in food animals.

Therefore, FSIS manages the risk of Campylobacter infections from poultry and CVM manages the incremental risk of fluoroquinolone-resistant Campylobacter from poultry. Therefore, because FSIS is already managing the risk of a person in the general population getting a Campylobacter infection, the appropriate population on which to manage the risk for -- we would say for FDA CVM is the incremental risk of fluoroquinolone-resistant Campylobacter from poultry as a consequence of fluoroquinolone use in poultry.

And it should be managed -- the appropriate population should be for those persons with Campylobacter infections. However, this risk management should consider all the potential outcomes due to that resistance. For example, the risk assessment should consider all the potential outcomes of fluoroquinolone-resistant Campylobacter which arises as a consequence of fluoroquinolone use in poultry.

This risk assessment only considers the outcome of persons who are ill enough to seek care, receive an antibiotic and are prescribed fluoroquinolone. I am sure many of you recognize the logic error that this is a self-mitigating risk assessment because if resistance to Campylobacter emerges to such an extent, physicians will stop using fluoroquinolones. And so, therefore, when the usage of fluoroquinolones reaches zero, the harm is zero.

So it is a self-mitigating model and will self-mitigate taken to its extreme. So that is a concern with the current model.

But in terms of setting then the population, we would say that the population that should -- that -- on which to base a standard, although I don't mean this to be a legal statement -- but the population to base the standard should be people with Campylobacter as the denominator and people with fluoroquinolone-resistant Campylobacter which arises from the use of fluoroquinolone-resistant -- fluoroquinolone use in poultry as the numerator.

And if your outcome is people who seek care and receive a fluoroquinolone when they seek care, they can be the denominator and the numerator should be amongst those groups, how many of them have a fluoroquinolone-resistant infection as a consequence of fluoroquinolone use in poultry.

Regardless of what population is selected, the public should not be harmed by the use of antibiotics in food animals. Finally, we caution that to prevent this harm in Salmonella infections, conservative thresholds should be established. Even modest harm with Campylobacter, which this risk assessment clearly demonstrates is now occurring in the United States, even modest harm with Campylobacter is a sentinel event indicating the potential for much greater harm when Salmonella becomes fluoroquinolone-resistant.

DR. STERNER: Does that conclude your comments for now
DR. ANGULO: It does.
DR. STERNER: Okay. Scott.

Dr. Scott McEwen

DR. McEWEN: Thanks very much. I would just like to say at the outset that I -- as a foreigner, it makes me a little nervous to talk about U.S. regulatory matters. So I would just like to acknowledge that. And the comments I make are made with respect and I hope no one takes offense.

I think in terms of the appropriate risk standard, I think, obviously, that those should be quantitative where possible and using good quantitative risk assessment methods. And the outcome should be public health. That could be sort of -- that could come back to thresholds at an earlier phase in the production cycle if appropriate. But it should relate quantitatively ideally to a public health outcome.

Again, ideally I think it all hinges on the adverse effect in humans attributable to the use of the drug in question in approved species. And I think it should be drug-specific and organism-specific where possible. And I think that should include the treatment and failure issue as well as the issue of pre-existing drug use being a risk factor for infection, pathogen load in terms of spread within an animal species and the concentration on food products, the altered virulence question that is out there.

And it should include resistant, food-borne pathogens as well as commensals in the treating transfer issue, so the whole thing. Now, I realize that pragmatically, it is probably necessary to back off that and focus on specific aspects. But I think that is a regulatory political sort of decision based on priority setting and that sort of thing.

But they emphasize that it should be the portion attributable to the use of the drug in animals. And for that, I think we could look to some other examples of risks in other food safety applications. And I think we have two main sort of classifications, the naturally-occurring hazards that are already in existence including things like 0157:H7, Salmonella enteritidis, natural sex hormones in a sense.

And for those, we have a kind of background level that is out there. They are already in place to some extent. And we heard this morning in the water area, that EPA is using a one in 10,000 yearly risk of enteric disease, so that sort of background level of pathogen.

And we've got -- the other class is sort of -- well, it would be called the technology-created hazards. It would be things like antibiotic drugs, hormones, for example. I think I put drug-resistant organisms in that category. So we are in a sense creating these, not to be sort of inflammatory about it. But there is no in a sense natural background level.

Maybe you could argue that there -- just to sort of back off from that, that if the drug has been used already a lot in human medicine and we do have a background degree of resistance. And that could be sort of factored into this. But I guess in the case of food-borne pathogens, Salmonella and Campylobacter, we tend to think that resistance arises from drug use in agriculture. So that's maybe a moot point.

I think we have to consider the adverse effect, both in terms of morbidity and mortality. In the case of mortality, we have examples in carcinogenicity and so on of an estimated risk of one in a million being acceptable. And this is sort of targeting the discussion on acceptable levels.

And I think we could look at translating not to infectious agents. You will recall that National Academy in the '80s looked at this for infectious disease. I forget the number. We would have to ask the statisticians. But it could translate the one in a million to the annual risk of fatality or daily or something of that particular sort.

I think the morbidity question is a lot more problematic. And I think anything -- allowing anything more than one case is in a sense an implicit acknowledgement that we are balancing risks and benefits. I think the unique situation with the microorganisms as opposed to the xenobiotic drugs and so on is we have actual cases. They are kind of interfaced and being diagnosed. And it is not some sort of esoteric theoretical sort of risk calculation.

So I am in favor of balancing risks and benefits. And I think in western democracies, we do that all the time. We should have a mechanism for allowing that.

In terms of setting the allowable levels, we heard from Dick Whiting I think this morning that that is not being done yet in the naturally-occurring organisms. I think it might be naive to suggest it, but maybe it is time for an open symposium to try and nail down some figures for that in terms of acceptable levels of morbidity for food-borne pathogens.

I think this morning I was trying to think of a corollary to the antibiotic-resistant organisms with drug use in animals. But I couldn't think of another example. I think it is very unique in a sense that -- and without being an alarmist, we are almost creating a new type of organism from using a technology in an area that is not primarily intended to enhance public health. And so that creates some new difficulties.

I think there are figures out there of something like 78,000 people. They put variability on that figure, I think. But people dying as a result of mistakes in the health care system. And I think we would all agree that that is too high.

But, again, some of that is probably a result of treating people, for example, for life-threatening conditions and you are going to acknowledge that there is some risk to that. Again, we are trading off public health risks, not public -- one public health risk and another type of benefit.

I think in terms of looking at morbidity, we also have to scale in different types of morbidity in terms of severity perhaps, transient diarrhea at one level, pain being factored in there, long-term organ dysfunction or failure is another one and so on. So there has to be a kind of weighting of degrees of morbidity.

In terms of the appropriate population, I think obviously we have got to look at both general and high risk groups. And I think whether you go for something that has been used elsewhere and say that to protect a ninety-eighth percentile in either group, I think -- which is again an implicit sort of trade off of risk and benefit, I think we have to -- that would be a function of the costs of having a high standard.

One form of that cost could be whether or not that new drug approvals actually have taken place. I think look to the residue situation for example. You are going to have a very high standard there of safety.

And one of the reasons we can do that is that it seems that the industry, animal industry, and the drug industries can actually live with that. It is not sort of ruling out drugs that -- maybe it has some, but not too many that I know of. But if we do that for microbial resistance risks, it might be a lot more difficult.

In terms of the legal standard, again, I don't have any real comments on that. And I think it is the same in my country and others, I think that this business of public health agencies being only able to look at harm and safety and not the sort of benefits is a problem that we -- that nations have to get around. We have to be able to weigh in explicitly I think the benefits somehow. Thank you.

DR. STERNER: And last but by no means least, Dr. Apley.

Michael Apley, Ph.D.

DR. APLEY: I think the Chief put me here just so I couldn't have a piece of candy in this whole deal.

(Laughter.)

Well, when this first came up and Lyle Vogel said, "Well, why don't you talk up there " And I said, "Well, I don't know anything about risk assessments." And I think as someone pointed out, that doesn't stop you from talking about cattle and pharmacology, so why don't you get up there.

But, you know, what that brings up is that I don't think you have to be an risk assessment expert to have a real meaningful part in this whole process. And AVMA would like to commend -- those of here today would like to commend the FDA CVM for the process and bringing people in. And we are glad to be here. And we appreciate the ability to comment.

But you don't have to be an expert in mathematics to have a lot of input onto things. And the comment about the importance of the process, it really dawned on me these last two days is very similar to our decision system as we go around coming up with pharmacokinetic, pharmacodynamic susceptibility data to work on that project. You end up finding a lot of holes that you thought somebody knew that. You thought we were doing something because somebody knew it. And I think we are finding out through a lot of things we didn't.

The AVMA has been involved in animal welfare and human health for about 130 years. Again, we commend the FDA CVM for the work completed on this risk assessment. And we would like to thank Dr. Sundlof and Dr. Bell and Dr. Sharon Thompson has also attended some, the Steering Committee for Judicious Therapeutic Antimicrobial Use. This dialogue has meant a lot to us on this issue, talking back and forth.

I think one of the things all the stakeholders have to avoid on this is the drunk and the lamp post syndrome where we use the process and the data for support rather than illumination. And I think we are on the way there with the dialogues that we are having.

I think of the groups represented at this meeting, there is a reason so many veterinarians are here. And that is because we are uniquely prepared to address most of these issues. And we are responsible for both human and animal health in our daily activities. And we are the people on the front lines for the antimicrobial use decisions being discussed here and made.

We will be submitting more detailed written comments later, but we wanted to take a shot at addressing the three main questions and a couple of other comments. What is an appropriate risk standard to apply to resistant pathogens Well, we are concerned that the assessment of fluoroquinolone use in poultry based only on possible adverse effects on human health is incomplete.

There has been significant process on this issue through the risk assessment presented at this meeting. However, there is a risk not being evaluated. And that is the risk of harm from increased disease and impaired health of animals going to slaughter.

Whether or not you personally believe that an adverse event could occur due to the withdrawal of a drug from food animals doesn't matter at this point. The important concept is that pathogen load of target or other organisms could be either increased or decreased by withdrawal of the drug.

The assumption that only good can result from withdrawal of an antimicrobial from use in food animals is unfounded and is a dangerous precedent on which to proceed. Dr. Shaub this morning represented a refreshing concept in addressing water treatment to control Cryptosporidium. In addressing the Crypto. contamination of drinking water, could another hazard be created and what are the risks associated with that hazard

I also noted a particularly appropriate statement by Dr. Morris this morning. "Don't forget the long-term, downstream sequelae of the lack of an appropriate first-line therapy." We should remember that enterofloxacin is an effective, improved therapy for colibacillosis in chickens. Removal of this agent would require the extra label use of antimicrobials to address this issue. We would move from a labeled drug to the uncertainties of extra label use or to no therapy at all.

It should be remembered that there is an economic disincentive to use this compound in chickens which balances any desire that would be present to use it is a precautionary measure. And I would point out that we were HMO before HMO was cool.

We should also keep in mind that in some hospital studies, we see a dramatic decrease in resistance to the drug that is pulled from the formulary or an example of the Danish data with erythromycin with streptococci that after several years, the resistance level declined.

But some of these studies also report -- not necessarily that one, but some of the hospital studies I have reviewed -- an increase in resistance to the drug that was put in its place. So while we may focus on the drug that is taken away, what will happen with others that are then needed to be used in its place.

And Dr. Hueston in his presentation addressed these issues in point number six which was you can't do a risk assessment in a vacuum. And we fully realize the need to narrow down the risk assessment which is valid to as little variation as possible or as little complicating factors. But then we have to put that back into the larger picture.

What is the appropriate population on which to base a standard Singling out one population or sub-population to me gives the impression that one population provides a more accurate estimate than another which we are not aware of. The percentage of the population that actually consumes poultry products should be taken into account in estimating the affected cases.

And I think an approach that is relevant to this was illustrated, again, by the EPA this morning and the strategy of evaluating the risk of the entire population that could be affected and then also considering groups with special attributes. I think what we come down to is the fact that we are going to look at all these groups and evaluate them individually. So I don't see the need to just say this one is it.

What is the appropriate legal standard to apply to the evaluation of resistance pathogens I had that exact same thing, all those things fixed up there to refer to and I left them at home.

(Laughter.)

Taken at the expense of Dr. McClure there. I have got about this much on that. We don't believe the food additive standard for reasonable certainty of no harm was conceived with a concept of the complexity of microbial issues in mind. I think we are into a new area that is much more complex than that was originally conceived to address.

Again, Dr. Shaub from the EPA has recognized -- said they recognize the futility of a zero risk approach in the areas they regulate. And their areas are similar and they involve complex interactions of source, environment, exposure and individual susceptibility.

I would have a couple of other things that I -- that also illustrate some cautions. In looking at attributing all fluoroquinolone resistance in Campylobacter to poultry, it does make the model more convenient and is an assumption that it is based on. In looking at that, one of the things that is said is the assumption that no fluoroquinolones were used in food animals prior to fluoroquinolone approvals in poultry. And that is incorrect.

Until the FDA CVM ban on extra label use in food animals, extra label use of fluoroquinolones in food animals was allowed under Compliance Policy Guideline 712506. And I think I have that number correct. While this does nothing to quantitative the use and does not propose that its use was widespread, it does point out that using the argument that no fluoroquinolones were used in food animals prior to poultry as a support for the assumption that the all resistant Campy comes from poultry is unfounded.

The issue becomes balancing between simplifying a model and weighing some of those assumptions. So that is just a point I wanted to bring forward on that.

And the other thing I have noticed today, and this comes sitting there looking at it from the angle of a pharmacologist, is hearing speculation on potential links between virulence and susceptibility, I am not aware of concrete links.

And I luckily have a graduate student sitting back at Iowa State, affectionately referred to as the "Web Goddess" who I called up and sicked Virginia on this. And she came up with about 400 articles and searched through. And we found one with some type of link between -- or they thought potentially in an organism between susceptibility states and perhaps its ability to survive out in flora. And then we saw two others that were related to penicillin therapy of pneumococci that showed no correlation on the virulence.

So I am not saying that the data isn't maybe out there being developed. But as we talk about risk communication, I would caution us to some things like that as we speculate on whether or not it is true, or that maybe data gives a preliminary idea that that could be true, that we are very cautious in stating that because that is a -- coming from a pharmacologist's point of view again, to link virulence with changes in susceptibility is a tremendous leap. And I would like us to have our ducks really in a row before that hit the press.

And I think that is a strategy we should all adhere to because that as we discuss these -- and I think you have been clear in that this is a speculation, but that we are very, very clear because this will be filtered through -- excuse me -- through, in my politically correct part of switching on now, through a scientifically challenged press to the public. So thank you again.

Keith Sterner, D.V.M.

DR. STERNER: I am scheduled to be a panel member, too. But after listening to this discussion, the only thing that I can do is shed darkness where they have attempted to bring light.

I do have one comment. And it stems from my experience in serving our country against all enemies, foreign and domestic. And it was characterized as the Ten Percent Rule. And it is more popularly know or enshrined as the Darwin Award with reference to risk standards and which populations we should look at.

And for those of you unfamiliar with the Darwin Award, these are individuals who have gone above and beyond the normal things that people do to thwart mechanisms that are designed to protect themselves and ensure that their genes do not pass on to future generations.

(Laughter.)

And I think that in all candor when you are looking at your risk model and risk assessment, there comes a point at which there are individuals who will no matter what -- no matter what efforts you make at protecting them from themselves, they will bring great harm to themselves and others around them.

And that gets back to Dr. Hueston's eloquent comments this morning about the need for a cost benefit analysis when you take a look at this. And I would echo Dr. Apley's comments that I always remind veterinary students who ride with us, "You have corrected one problem. And you may have created an entirely worse set of problems as a result of your corrections."

So -- and I realize that there is a limit to which you can do this. You are charged with enforcing the law and I don't envy you the task. You have done a yeoman's job thus far in trying to get us to some point that we all can live with it. With that, the panel is open to questions from the audience. If you would step to the microphone, identify yourself and ask the question, we will do the best we can to respond.