• 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

Public Comment Period

DR. FEDORKA-CRAY: I think that I have some idea of the amount of time that it took the people at CVM to put this together. And I really think that we do owe those people our thanks and a round of applause.


DR. LONG: Jim Heslin who is our FDA Training Officer is going to join me up here on the stage to facilitate the public comment period. And Dr. Sundlof is also coming up, so you can address your comments to him.

DR. HESLIN: Thank you. I am a little hesitant to say this, but good evening. I took a look outside and it is dark out there. We have come to the end of a long day. But I think it is an important part. It is an opportunity for you all to provide your comments on perspectives, on the issues that were presented and discussed here today.

I know Dr. Thompson and others felt this was an important component, that they wanted to hear from you. So this shift now is to FDA as the listener, to hear your comments. It is not a forum for debate or protracted discussion. But we are here to listen.

A couple of ground rules. When you come forward to the microphone, I would appreciate it if you would identify yourself and your organization. We don't have a lot of time and I am not sure how many people are going to be speaking. But I would ask that only one person from each organization, if there are multiple representatives, would speak for the organization.

We are going to start by limiting comments to about three minutes. If there is more time at the end, we can always come back or you can always submit comments in writing. At about two and a half minutes, if I can figure out the clock, I am going to let you know that you have about half a minute left. At that point, if anyone else wants to speak, that is a cue to move to the microphone so that we don't lose time in the transition.

Now, this is the important piece. When the three minutes is up, you are supposed to tap the person on the shoulder and tell them to move on. Okay?


Because I don't like to shut people down. I will, but I don't like to. Okay. Thank you for your cooperation in advance. And we will go ahead and get started. Who would like to begin? Would you like to step forward to the microphone, please.

DR. SHRYOCK: Tom Shryock representing the NCCLS, Veterinary Antimicrobial Susceptibility Testing Committee. I probably took 30 seconds there. I just wanted to add an assumption I think that should be included here. And that is that the breakpoints that are used to characterize an isolate as fluoroquinolone-resistant need to be assumed -- are assumed to be valid in terms of the clinical outcome.

At this point, just to reiterate from this morning's talk, I am unaware of data that has really matched MICs specifically to clinical outcomes with regard to Campylobacter. And I think that sort of data, there has been assumptions made on that. And I think it would be worthwhile to try to piece together whatever available data there is or to secure sponsored data along those lines as appropriate.

Since we have talked about gastroenteritis being treated by fluoroquinolones as well as systemic disease, there are two different pharmacologic patterns that could be involved which could affect where that breakpoint is set.

And then finally, the breakpoint itself may not be indicative of a resistance mechanism. It may be due to the pharmacology which is the achievable drug level exceeding the MIC. So there are some factors that go into what really determines a fluoroquinolone isolate. Thank you.

DR. HESLIN: Thank you.

MS. LIEBERMAN: Hi. I am Patty Lieberman from the Center for Science in the Public Interest. We represent a million consumers in the United States and Canada. And I guess I am part of the risk communication team.

Basically, we feel that FDA's responsibility in regulating animal drugs is to assure the reasonable certainty of no harm to human health due to the use of antibiotics in livestock. But CVM's own risk assessment shows harm to about 5,000 people. Therefore, we think that the fluoroquinolone approval in poultry which never should have been allowed should be revoked.

Consumers should not have to continue to be guinea pigs in this regulatory experiment. What level of harm will result in CVM action? Is it going to take 10,000 more severe illnesses? Will it take death? Or will CVM continue to not do anything in regulation by redefining what the word, "harm", means and looking for a different legal standard to apply?

Now, about using the similar risk assessments for other antibiotics and pathogens, the concern is that using a risk assessment like this for future decisions is predicated upon waiting for resistance to develop, for being transferred to people, and for causing significant human health harm before action could possibly occur.

Instead, we need a preventive strategy to apply to new drugs considered for approval that would monitor changes in susceptibility in livestock before they have a human clinical consequence. Finally, the entire process which has been initiated by the FDA is clearly very slow and laborious and controversial. And it is too slow to deal with the public health risk.

We can endlessly debate the framework, the risk assessment, the legal standard and still do nothing. Meanwhile, consumers are being harmed. Thanks.

DR. HESLIN: Thank you. Is there anyone -- yes, over here.

MR. CONDON: Robert Condon. Just a couple of issues. I want to thank you. You have done a good job. There is a lot of details in there. Don't get caught up in the details and don't put too much emphasis on certain point values.

The issue I would like to bring up is when you look at this, look at it as a probability of risk given the exposure. That is the bias that occurs in a lot of risk estimates and a lot of risk assessments. Like this data has the bias in it, it assumes 100 percent of the population is exposed. Therefore, those risk values you have under-estimate the true population risk to those that are exposed to the hazard.

An example, if I told you only five people were killed bungee jumping last year, you could say, well, that is way less than one million; that is one in 50 million. So bungee jumping should pretty safe. Now, if I told you there was only 20 people that went bungee jumping, you would probably have a different idea of that.

But when you spread that risk across a whole population, you end up with a bias estimate. You can get at this -- USDA has some very good intake data. I mean, they can tell you how much processed chicken, how much cooked chicken each individual had; how many people had chicken down to -- the detail that they go to is they can tell you how many people had raccoon. That is in the database.

So you can get -- to get the exposure, you can get a pretty good handle on that. The data is available. One of the things that I have a hard time reconciling here with the data is the paper on the seasonal variation. You have got about a four-fold difference in incidence seasonally due to cases. I doubt whether your chicken consumption is four-fold -- varies four-fold seasonally.

If it is truly 50 percent exposure from causing the Campylobacter, if 50 percent is coming from chicken, you should be able to track the chicken consumption in those incidence of cases fairly well. I don't know if that is in the data, whether you could do it.

I doubt it, when you've got a four-fold difference in the number of cases you are going to see a four-fold increase in the consumption of chicken. You might be able to look at that from USDA data. But I think that is something to look at to evaluate those estimates.

DR. HESLIN: Thirty seconds.

MR. CONDON: Okay. Thank you. So that is one of the things. Look at the USDA consumption data. The other thing I would like to bring up is you have got to look at the quality of the data. And just because you have a number that says it is six and it's just -- like this risk assessment -- you are probably going to see tonight on TV, 70 percent of the Campylobacter cases come from poultry.

I mean, that is a value. I mean, people pick up on the single values. They take on an intrinsic worth. I mean, I lived for years with the value of two parts per billion being safe for DES. It just -- there was no good basis, but it becomes entrenched.

And because a value is published doesn't mean it is accurate or worthwhile. And I think you need to look at little bit more at that at your own -- some of the data that CVM had collected. There is questions on some of that data. Take a look at those studies and really spend a little time looking at the studies to see whether they are worth it. Just because you've got a value doesn't mean it is better than no value.

DR. HESLIN: Thank you. Next.

DR. ANGULO: Fred Angulo, Food-borne and Diarrheal Diseases Branch, Center for Disease Control and Prevention. We agree that there is a marked seasonality of Campylobacter. The seasonality of human sporadic illness actually matches quite closely to the seasonality of Campylobacter contamination found in grocery stores and also found on farms.

We also -- there also though -- seasonality is not all explained by contamination rates. There are seasonality "mishandling" characteristics such as increased outdoor barbecuing and other factors in the kitchen that might explain seasonality.

But the seasonality is fully -- the seasonality affects are fully understood in terms of the current understanding of the epidemiology. And the conclusion is still the same, that the predominant source of Campylobacter is poultry.

There also was questions raised about the MICs of fluoroquinolone resistance and Campylobacter. Campylobacter is remarkable bacteria in that a single point mutation in the gyrase causes the MICs to be at the highest detectable or measurable level. Wherever you set the breakpoint, they are always at that level. It is not a breakpoint set point. They are all at the highest level of the MIC.

And we have done -- we have looked at Campylobacter isolates from humans that are fluoroquinolone-resistant and find the consistent base permutation and correlating the biological resistance -- correlating the laboratory resistance with that mutation.

So reiterate a point made by Dr. Bell, CDC would like to commend CVM for undertaking this risk assessment. The risk assessment clearly demonstrates that the use of fluoroquinolones in chickens is now causing harm in humans in the United States.

This harm is not trivial. Harm to people now may be somewhat greater than estimated. Harm is likely to increase each year. Steps to mitigate the harm are needed now.

A meeting to plan these steps should be held within the next three months. In particular, we need to establish fluoroquinolone-resistant threshold in Campylobacter and we need to establish a timely procedure for drug withdrawal in the event the resistance threshold has been crossed.

DR. HESLIN: Thank you. Yes?

MS. BUTLER: Good afternoon. I am Kelly Butler with the Bureau of Veterinary Drugs from Health Canada. And I would certainly like to commend the Center for Veterinary Medicine of the Food and Drug organization of the United States. It is especially gratifying to have a tool that seems to be a tool that will be used for all food-borne pathogens and resistance.

I think some people that I have spoken to here have found that their -- they feel that chicken are being targeted or a particular bug is being targeted. But in terms of regulators who have to make decisions, we need a tool based on science.

Doug Powell earlier today said the alternative is astrology. We are scientists here and we need to make decisions based on science.

I was especially pleased to have two mathematicians who could actually speak English and explain issues to us because as scientists, we know, too, that we end up speaking a language that many people can't understand. But when we speak to the public, which is my job as a regulator and the CVM's job, as well, we need to be able to speak English to communicate the risk.

And I think this represents a tool that we can use, that we can make decisions based upon using this tool. A small issue, I must say, too -- and I am trying to explain this issue of antimicrobial resistance to policy-maker. I am a policy person and a scientists, a published scientist. Some of the issues that we need to make clear are things like this debate or comments on the seasonality.

It is not just chicken that people eat and they get microbes I explain to other policy people. The other piece is the chickens that are in the grocery store on the little turn-around. And that contaminates other vegetables -- or vegetables that the vegetarians may eat.

And additionally, poultry manure, swine manure, all sorts of manure are used in vegetables. So this issue isn't just one bug, one species. There are a lot of issues to look at. And I think this tool represents an excellent start. Thank you very much.

DR. HESLIN: Thank you. Yes, sir?

MR. BIOWATER: Robin Biowater, Consultant to Pfizer Animal Health. I would come back to this word, "harm." And I think it is debateable still what degree of harm can really be hung under fluoroquinolone use in poultry. And, indeed, that is what we are here for today.

But I think we shouldn't forget that the harm from Campylobacter infection in man is not the harm predominantly and overwhelming -- not the harm due to the resistance to fluoroquinolone, whether through treatment or increased virulence.

The harm from Campylobacter is the shear volume of cases, the shear prevalence of the disease and the number of people who suffer from it. And we should keep that in mind. And that should be the main target. And obviously, we should make any other targets we can identify at the same time.

I would like to just make a brief comment on the model which has I think been a very interesting exercise. But I, like Louise Kelly, I am concerned that the idea can be easily applied elsewhere will firstly fall down because other organisms don't behave in the same way as Campylobacter and in particular because for other organisms and other connections, we are going to have a great deal of difficulty finding as much data to support the model as has been found for this one.

And unfortunately, I am afraid the extrapolation will be much more difficult than I think Fred implied, at least for food-borne organisms. Thank you.

DR. HESLIN: Thank you. Any other comments? And the race is on.


DR. CLOPP: My name is Buzz Clopp. I am a veterinarian. I work in the chicken industry and have for a number of years. And my intention is not to stand here and ridicule the model and the development of the model. I think it has been a lot of work. It has been very well done.

But I do have to say that, you know, I have some concerns about. And some people have already said that they don't believe that these are concerns. But I guess I have to say that I don't agree. The number one concern being the -- to somehow factor in the level of treatment that is done in the field.

I think as I understand this model, the way it is done right now, you are making a direct assumption between resistant Campylobacter on the carcass and chickens. Well, guess what. There is another factor in there. And it was mentioned about the environment and about manure going on fields and going to vegetables.

There is a huge amount that we obviously don't know about the epidemiology of Campylobacter. Does Campylobacter go from chickens to people? I suspect it does. No question.

Does it go from people to chickens? I suspect it does because we have -- chicken houses are not isolated vehicles. Processing plants are not isolated vehicles. There is cross-contamination and it goes both ways.

Now, my intention is not to stand here and to be defensive. As a person working in this industry, it is our intention to make a quality food product that people can not only enjoy tasting, not only get good nutrition from it, but feel good when you eat it. And obviously, people do.

And there is a lot of circumstantial aspects to all this. And I don't disagree with public health. You know, hey, I am a human obviously. I have children. I have grandchildren, the whole works. But, you know, slow yourselves down a little bit and think.

You know, number one is chicken consumption is increasing. But it doesn't appear that the level of all this is increasing at the same rate. You know, what else is going on? You know, study, but please don't overreact. And I am going to sit down.

As an industry, agriculture and food production in this country is under an assault from many, many, many factors. And what people had better start to realize is -- and the inference was made this morning about chicken at $1.49 a pound. That is not the same price per pound as what you bought chicken 20 years ago.

Twenty years ago, you were buying predominantly a whole chicken for probably 35 cents a pound. The average consumer today does not want to buy a whole chicken. They want to buy a boneless breast or they want to buy de-boned thighs or they want to buy buffalo wings, all of this. And that is why you see these costs going up.

So all I am saying is, you know, we need to move forward with this because it is an issue. But slow down and keep a little bit of science on the whole thing because there is a lot of factors involved in this.

DR. HESLIN: Thank you.

DR. KRISHINSKY: My name is Dr. Elizabeth Krishinsky. I am with Wompler Foods. I am also with the broiler industry. And I am not going to philosophize anymore. I just have a simple question.

I think the model itself even -- I congratulate you on the model because even I can understand how it was put together and I am not an epidemiologist. I have had some statistics, but it is not my area.

And I can clearly see how you have modeled the clinical progression of the disease and extrapolated backwards to the number of people in the population that are affected with Campylobacter illness in a year and then divided by that the consumption of poultry.

But to me, I think we have overstated what the consequence or what implications this has for fluoroquinolone use in poultry. There is nothing in the model -- it is a little bit of the "emperor's clothing" analogy.

There is nothing in the model that addresses fluoroquinolone use at all. It starts with the assumption that the use of fluoroquinolones causes resistance -- fluoroquinolone-resistant Campylobacter on chicken.

So the model says if Campylobacter from chicken have fluoroquinolone resistance, then what is the impact on human health. And I agree, it is an excellent model for that. There may be some people that quibble with the data, etcetera. But it is very simple. It is easy to understand. And even the statistics are easy to follow.

My question to you is how can you comfortably and really with good conscience extrapolate and draw any conclusions or suggest any interventions on the live side and tie this to fluoroquinolone use when there is nothing in the model that at all addresses that. Thank you.

DR. HESLIN: Thank you. Yes?

MR. : I guess I would be remiss if the Animal Health Institute didn't make some comments at this meeting today. First of all, let me congratulate CVM on undertaking a very difficult and complex task.

AHI has certainly supported the idea of risk assessment as a way to get a handle on this whole area of antibiotic resistance for a number of years now. And we appreciate the difficult job that has been undertaken in trying to tackle this problem.

Now, we haven't obviously had this risk assessment for very long. So we can't obviously give you very many detailed comments today. We will be providing more detailed comments in writing, of course, in the future. But let me just make some general comments on the model itself and then touch on some of the assumptions.

In a way, I guess I wouldn't characterize this as a true risk assessment. I think for what it was intended for was really a retrospective case prevalence estimate based on the FoodNet data. So then we back calculated some probabilities. But this is really based on what has occurred in the past, extrapolating to what my occur in the future.

A true risk assessment in my mind would factor in the consumption aspect of poultry, what happens in the process through the handling and cooking, estimation of the infectious dose and what cooking and handling procedures can do to reduce that risk as a prospective estimate of what the risk to the population could be.

So I think there is a little bit of difference here between this risk assessment and what I would view as a true prospective risk assessment. So that would be the first comment I would make.

On the assumptions, there is one assumption that is in the document that states that the level of resistance -- or the level of risk is assumed to be the same as it was in the 1980s -- it is the same today as it was in the 1980s.

And I think that is probably a fairly flawed assumption if you look at what has happened in the 1990s with regard to the changes in meat inspection, the implementation of HACCP, the implementation of Salmonella performance standards, safe meat and poultry handling labels which is on every single package of raw meat and poultry that is in the supermarket case today telling the consumer there can be potential pathogenic organisms in the product, that they must handle it, they must cook it properly and they must take proper precautions.

There has been an incredible amount of money put into the President's Food Safety Initiative. The FDA has their own "Bite Bac" program which I think is being considered a success. So what I am saying is that you cannot assume that the risk from the 1980s is the same as it is today. And I think that assumption is one that really needs to be addressed in the model.

I won't make any further comments other than, again, I appreciate the opportunity to comment on this today. We will be having more in-depth remarks in writing on this particular process.

One other thing I would like to close with before I forget is that, you know, we came here today -- the industry I think came here today expecting maybe a little more progression in where we were headed with this whole thing. We do support the idea of risk assessment. But we are afraid that this quite hasn't connected the dots to our satisfaction. Okay.

What do we do from here? Exactly where does the industry go from here and how does the industry deal with the drug approval process? I know it is a complex issue, Steve, and I understand you are trying to work through it. But I guess we expected a little more definite program to be laid out for us today. Thank you.

DR. HESLIN: Thank you. Yes?

MR. WAGES: I am Dennis Wages. I am a teacher at the Veterinary School at North Carolina State University in poultry medicine. And, Steve, as always, your group has done an excellent job putting this model risk assessment together. The one thing I would like to emphasize is the pathogen load and the number of organisms on these carcasses.

From my standpoint, looking at mitigation intervention strategies from our end in the industry, you are, are the affecting numbers either by pH adjustment in chillers, the cold pasteurization, irradiation, etcetera, etcetera, how that is going to affect public health impact by reducing numbers.

And that load that is on that carcass, because we know the infection is a result of some kind of a dose relation there, that is going to be important for us to try to go in either through research or whatever and intervene to try to decrease numbers, if not eliminate numbers, of bacteria on the carcass.

So I think the pathogen load on those carcasses are a very important tool for the mitigation intervention strategies to try to employ -- to prevent the contamination from occurring at all in the poultry.

DR. HESLIN: Thank you. Is there anyone else? Okay. Keeping with the pledge that we would be out by 6:00, we are pretty close to it. So thank you very much and see you all tomorrow.


(Whereupon, at 5:50 p.m. on Thursday, December 9, 1999, the Workshop on Risk Assessment and the Establishment of Thresholds was recessed, to reconvene at 8:30 a.m. on Friday, December 10, 1999.)