• 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

Questions/Comments to the Panel Assessment of Risk: Contaminants


DR. CONDON: Robert Condon again. I am just amazed sitting here and it just reminds me of a group of blind people grabbing a hold of the elephant, trying to describe what is going on. And I think maybe the best thing CVM can do with this whole process is to sit down and describe what is the legal criteria.

It is not 409. It is not adulteration. It is 512, okay. And 512 is basically all tests reasonably applicable. It is any substance formed in or on food. So the question there is Campylobacter resistance, is that a substance formed on food due to the use of the product.

I am not sure of the legal definition, if a bacteria is a substance. But that is the basis. Unfortunately, the adulteration issue is different between FSIS and FDA. FDA has a very easy standard. The food is deemed, deemed adulterated if it bears or contains an unapproved new animal drug. We don't have to show any harm. We don't have to show anything. Only that the substance is present.

Under adulteration under the food additives in USDA, you have other standards you have to meet. It is different. So that is -- the same thing, adulteration of the food, is different depending on which standard you do it.

And I think to make progress on the risk assessment, we need to define what standards we are working with and make that clear. Just what do we have control over What can be regulated Because it is just everybody has their own idea and everybody is using their own standard.

Is it uncertainty of harm Is it, you know, if you demonstrated there actually is harm There are all different kinds of standards for food safety. Unfortunately, CVM can't choose which one it wants to use. These are animal drugs. It is 512. And I think it might be very helpful of CVM would lay that out right up front. Maybe you might have to do a little work on it.

But this is the standard. And then you can start looking to see how things are going to fit into that standard because you may arrive at different conclusions depending on the standard. So I think there is a lot of people's interpretation. And that is something we need to get squared away right at the beginning.

DR. STERNER: I'm not sure we have any panel members really qualified to respond to that, Robert

DR. CRAWFORD: I am not qualified, but I will respond.


Well, I have known and admired Dr. Condon for over 45 years. But I think, Bob, I am sure you are not saying that just because we are not gifted in the law that we shouldn't be addressing the problem of antibiotic resistance. And unless you are saying that, you should sit down and not say anything else.


DR. CONDON: No, it's not -- but it is what the standard -- do you use the standard of, you know, is there harm Is it the standard that keeps showing that there is no likelihood of harm

DR. CRAWFORD: Yes, but give us a break. I mean, we were asked to come up here and comment on the thing without worrying about having a lawyer sitting on each shoulder. So we don't need that.

DR. CONDON: No, it is important. Okay, because going down the panel --

DR. CRAWFORD: Well, maybe you and I should go outside or something.


DR. CONDON: But, no. People have made their comments in interpreting and based it on different standards. We've got to try to get people together so we are thinking of the same part. As long as somebody is using one standard and it is different -- and you might go off and develop a risk assessment that is great for this other standard.

But it is no good to CVM because it doesn't apply to their section of the law. That is all I am saying. And whether it's -- you know, I am not saying it is your job to do it. But that is something that CVM has got to do because just in discussion of the panel, it points out there was at least three different interpretations of what the standard was.

DR. STERNER: I don't see anybody at a microphone right now. And I am going to give panelists a chance. But before we do, it may cut to the chase just a bit if I offer at least somebody from CVM the opportunity to respond to Dr. Condon's comments. And I think it is very germane to the task of the panel here if you would like to do that, anybody.

They are conferring right now. In the meantime, in the interest of keeping the proceedings moving, Dr. Angulo has a comment.

DR. ANGULO: Well, I just -- maybe it is tangential. But I like the image of an elephant. And I think it paints a good picture. But please recognize the elephant is moving and perhaps going down the hill. And it has been going down the hill since 1995 when serafloxacin was approved. And it gained speed when enterofloxacin was approved for poultry use.

And in the meantime, we were hoping to slow it down. And we see eventually slowing it down through the framework process which was announced a year ago. And it has been a year. And we are at this destination which is wonderful step progress, a wonderful step forward. But we really need to gain momentum and address this issue.

And the way to address this issue is we believe that we really could diffuse much of the consternation on this issue of the framework if we could categorize the drugs. If the drugs were categorized, at least the strawman to allow people to comment on whether they think the categorization is appropriate, then those people that are concerned that the categorization would be over-stringent or under-stringent could begin -- we could then start that discussion.

So we need to categorize the drugs in the near term. We need to have near term discussions on where the appropriate thresholds are at -- on the one hand, it has been a Herculean effort to have this risk assessment. It was called for almost unanimously for it to be done. But at the same time, there was optimism when this meeting was first announced that this meeting would be talking more substantively about establishing thresholds.

And although -- not to comment negatively about the progress that has occurred, but recognize it is just a -- the delay is frustrating. And in the meantime, the elephant is still going down the hill. And we need to mitigate the emerging fluoroquinolone resistance.

DR. MORRIS: Actually, if I could add to that, I would also second this idea -- this concept. Again, speaking as someone who is taking care of patients on a regular basis, we have an elephant who is sort of rolling down hill. Our resistance rates are rapidly rising. What was the rate for Campylobacter in the '99 data

DR. ANGULO: Well, we don't have December data yet which might dilute it. But it is 21 percent. And last year, it was 13 percent. And that is not final data yet. But it is going to be two to three to four to five percent higher than 1998. And it is likely to increase at a rate of two to five percent a year.

DR. MORRIS: I have this vision of Rome burning as Nero fiddled sort of thing. And I can say that because I am not in the government. But we have a very substantive clinical problem on our hands. We have gone -- you know, there has been a substantive jump in resistance over the past year.

And we get into -- we could argue about thresholds for years. And as I said, it comes back to my concept, this is not a static process. We can't argue over the thresholds in the '98 data because they are already out of date. The process is moving much too rapidly for this.

And there must be a sense of urgency in this because although I am concerned about the resistance in Campylobacter, I am scared, you know, I won't say what, about the emergence of fluoroquinolone resistance in Salmonella. And, again, that is not quite yet on the radar screen.

But the data that are coming out of Denmark says that even when you are not seeing at the clinical breakpoints, you are beginning to see clinical effects. And I don't want to be arguing three years from now about quinolone resistance in Salmonella when we are up at a rate of ten or 20 or 30 percent. I think there is a sense of urgency which needs to be instilled in the process. And I will stop at that point.

DR. STERNER: Dr. Beaulieu, it is your opportunity to respond for CVM to Robert Condon's comments.

DR. BEAULIEU: Yes, and I don't want these comments taken in any way as a response to what we have just heard since --


DR. BEAULIEU: -- Bob's question came up. I would have to agree with Kent McClure's assessment I think. There is no safety standard per se established in 512. There is a lot of legislative history that would argue that it ought to be reasonable certainty of no harm. That I think is debateable to some extent. Some Courts have found that to be a reasonable argument. Some Courts have not.

Even if we accept reasonable certainty of no harm as the standard, we still have to define what harm means in this context. And what we asked the panel for was their judgement about what they thought harm might mean in this context.

I agree that CVM, the Agency has to define at some point, has to try to quantify what is an acceptable risk in this context and start working from there. That is not an easy thing to do. We are charting new territory here. And I thought it was very important this morning that we heard how other federal agencies are dealing with this same issue and the kinds of standards that they are establishing to deal with some of the risks.

We will take all that information under advisement and we will certainly try to come up maybe in further processes like this with a standard that hopefully we can all live with. I appreciate having said that. There is some urgency to get on with this. We are concerned about the issue as you folks are.

There are things we can do in the meantime to try to mitigate this risk. And some of them are already ongoing now in terms of increasing judicious use of drugs in animals and so on. And we will certainly continue to do all that as we seek to try to quantify the level of risk that we deem acceptable.

DR. STERNER: Other questions for the panelists Fred, you had a comment

DR. ANGULO: I think the point about is there something that can be done to mitigate the risk short of withdrawal of the drug which I agree completely, withdrawal of the drug demonstrates I guess failure might be -- I mean, it didn't work -- isn't there a way that we can figure out how to mitigate this problem short of the draconian approach of withdrawing the drug

That doesn't serve anybody's purpose perhaps except for -- well. So is there -- can you mitigate the risk And mitigate the risk, you can mitigate the risk by either decreasing drug usage or decreasing transmission. And ideas on how to mitigate -- how to reduce drug usage -- well, first, I think it would be a wonderful show of good faith, although there is no legal requirement for it -- it would be a wonderful show of good faith, now the public health has shown -- believes that there is a harm from the use of fluoroquinolone in poultry, for the industry to provide the data -- drug use data freely to show how much fluoroquinolones have been used to the public; a show of good faith that you share the equal concern the public health has, provide the data.

And it would allow us to feel more or less comfortable with this escalating resistance. If the drug usage is remaining fairly constant, we could interpret perhaps changes in resistance we are seeing with some -- have some understanding perhaps about whether it is -- whether mitigation is possible.

So I call on a show of good faith from the drug industry to provide, as they did in an excellent example in the United Kingdom when they provided fluoroquinolone use data in the United Kingdom in a similar manner. Kilograms of useable by animal species by year would help us understand the risk. If they share the concern of the risk, I think they could demonstrate good faith by providing that data, although there is no legal requirement for them to do that, obviously.

Secondly then, in terms of mitigation of the risk, decrease in drug usage, it is a wonderful development in terms of develop the judicious use programs. The one developed by the American Association of Avion Pathologists is a major step forward. And it would be very useful to show implementation of that and then to, as anything that is implemented, fine tune it according to what is or isn't working.

You heard the problems this morning with such a program for physicians. It is being fine tuned by studies that demonstrate where the barriers are, etcetera. Can such studies be done amongst the relatively small numbers of poultry practitioners and just to see -- it would help ourselves as the risk identifiers. We would be assured if we knew the extent that the poultry veterinarians were adhering to these guidelines. So maybe a self-survey of how they are adhering to the guidelines that are developing would be useful.

Other ways to decrease drug usage is there could perhaps be evidence provided on the culture sensitivity necessity of using fluoroquinolones. And an interesting development in Denmark is that Denmark is now soon to be -- or will soon be implementing a requirement that before fluoroquinolones are used -- as I understand it, before fluoroquinolones are used for a second time on a premise, they must have culture sensitivity data that demonstrates its utility.

Thirdly -- then I mentioned that you could also mitigate the problem by decreasing transmission. And one way to decrease transmission, I don't know if it is practical, but perhaps those houses where birds receive fluoroquinolones, the integrators could schedule their kill schedule or their slaughter schedule so that those houses that get treated with fluoroquinolones, that they just simply go to slaughter immediately before clean-up.

And, therefore, we might decrease the transmission at least to other houses that haven't been treated with fluoroquinolones. And then a house that is treated with fluoroquinolones, the -- doing studies to see whether it is useful to clean out the litter and spray wash the house before repopulating it with the next chicks would be very useful and might be a practical intervention.

Well, of course, all of these have practical concerns and economic costs. But they would help us in public health feel that at least the people share our concern and are beginning to address it. The reason why we feel the -- for an analogy, but the reason why we are frustrated that the elephant is going down the hill is because we have been calling for evidence of some mitigation for a number of months and perhaps years. And we are still unaware of any concrete evidence of mitigation.

DR. WAGES: Dennis Wages. I am a Professor of Poultry Health Management at North Carolina State University at the vet. school and also the Chairman of the Drugs and Therapeutics Committee of AAAP which are writing these guidelines Fred, they are not done completely.

The guidelines are I would say 75 percent complete. Ninety percent of the bacterial infections that we deal with, the guidelines are written and there are certain approvals.

And as most of you noticed, we have looked at taking the neomycins and the streptomycins and the drugs that physicians would not consider important and we would consider are older standard drugs and use them labeled or extra labeled before we would go to a fluoroquinolone. And that violates -- that is against federal law.

And until we get some regulatory direction, we will still finish the guidelines. They will be as good the paper they are written on without AAAP and AVMA backing. So that is what we are kind of waiting on. But they will still be out there.

And things are being done. You know, every time that we get into a situation -- and I can't speak totally for every integrator in the United States. But every time that we get into areas where we try to go a direction, people that don't understand the poultry agriculture and the way we produce birds, it all comes back in our face as a negative connotation versus a positive act.

We have got companies that have purchased irradiating areas and companies that have done extensive research on pH adjustment of chillers to negate everything but Listeria at least, to reduce Campy through whole bird washes.

We have -- it is a practical -- when you -- you know, when you look at a suite and it has, you know, 100 growers or whatever and we have 100 growers in one county in North Carolina and you try to truck chickens, it's just the logistics -- it's like, you know, when people say that a ten-percent increase in a production efficiency.

And just in chickens alone, that is 32.5 million tons of grain that we need extra. And then that lack of efficiency has to go somewhere in a clean-out. You want us to clean out every time. I think a lot of people would like to do that on certain areas. I have EPA over here telling me I can't do that because I can't do the deal. You know, it is just not a simple thing.

But it is unfounded I believe in my opinion whole-heartedly to think of this issue as -- I used to have a lot of hair. And I thought it was going to turn gray and it didn't. It fell out and turned gray.


But they do and they care. And their product has their names on it. You know, if they go down the tubes on bad product, that is their livelihood. And I am not, you know, sitting up here saying that I am going to equate sick chickens with a human. And I am not going to do that and I don't think we should.

But we still have a job to do, too, to provide protein. And we still feel that consumers are choosing either a chicken or soy bean or beef or whatever. And there is a need for the consumers to own up to some responsibility.

But we intend to take as much action as we can to try to mitigate, if you will, and look alternatives prior to product removal. That may have gotten into public comment and I am sorry.

DR. STERNER: That's all right. It is germane to the discussion. Are there other comments from the floor for our panelists If not, seeing none, we are slightly ahead of schedule.

Mike Bolger is scheduled to come in sometime. And he has yet to make an appearance. But when he does, we are going to afford him the opportunity to do ten minutes just ahead of our next start time. We are scheduled for a break for 15 minutes. We will break for 15 minutes from now and start then. Thank you.

(Whereupon, a brief recess was taken.)

DR. STERNER: This morning's speaker, Dr. Mike Bolger, did finally make it down from Annapolis. And he has quite a tale of woe to tell. But that is not germane to the deliberations at hand this afternoon.

Dr. Bolger is the head of the Contaminants Branch in the Center for Food Safety and Applied Nutrition, that is CFSAN. His group is responsible for the hazard-safety risk assessment of natural and anthropogenic food-borne contaminants. Dr. Bolger.

Dr. Michael Bolger

DR. BOLGER: Well, I want to apologize for my tardiness. Unfortunately, on the way in from home, I blew a tire and was on my hands and knees about the time this presentation was supposed to be made, changing my tire. I had to return home, find my wife, get her car and start all over again. So I did have the best intentions of being here. Unfortunately, my rather dated car didn't want to cooperate this morning.

My task, as I understand it, is to give you a very brief, ten-minute overview of how we deal with contaminants in the food supply. And I -- when I talk about contaminants, as indicated in the introduction, we are talking about contaminants that are either natural origin or of human-derived origin.

Now, I know that you have had several presentations on pesticide, safety assessment, risk assessment and I believe food additives. I will try not to go over the same material. Oh, right here. Go it.


But as any true risk assessor, I always have to start off with my risk assessment paradigm. It gives me an anchor by which I can move from. And in terms of how we approach safety risk assessment in dealing with contaminants, this is the paradigm that we generally work in. We don't have pointer, right Okay.

So in terms of risk assessment, I make a fairly pronounced distinction between what I call safety assessment which is what most people are thinking about and talking about when they are talking about risk assessment, and in terms of quantitative risk assessment.

So most of the time when we are talking about risk assessment, we are really talking about safety assessment which is very much like what you have heard about in terms of the pre-market safety assessment paradigm that is practiced in terms of pesticides and food additives. Okay. Could I have the next slide, please.


And I will come back to this paradigm here. I have no way of forwarding this. And but remember that in terms of how we deal with safety and risk assessment for food-borne contaminants, the standards that we use are really dictated by what Congress has delineated in the act. And, again, I think for pesticides and food additives, those standards were already described to you.

For contaminants, we deal with a section of the Act called 402(a)(1). And there are two standards that apply here. One refers to it may render injurious to health. And that is for substances that are added.

And when I mean added, in other words, there has to be the hand of man evident. It doesn't have to be completely responsible for the presence of the contaminant. And a good example is aflatoxin where part -- aflatoxin is found because it occurs naturally. But, also, you have elevations of the levels of aflatoxin because of the storage conditions under which the grain is kept and therefore the hand to man in part dictates the total level of aflatoxin you would find in the grain.

Then the other standard for contaminants is the ordinarily rendered injurious to health. Now, Congress doesn't really tell us in a quantitative sense what is the difference between these two standards. And when I say ordinarily rendered, this is for contaminants where there is no obvious hand of man present, okay, or acting.

Now, what I usually describe the ordinary rendered injurious to health standard is I call it the body bag of evidence. And what I mean by it is that we actually have information of adverse reactions at the exposure levels that we are concerned about to that particular contaminant.

Now, we could go with evidence based on laboratory animal work. But generally, when you look at the dose range used in laboratory animal work, they are quite a bit higher than what you normally would find, okay, in terms of exposure levels to the contaminant of concern. So you are always making an extrapolation of dose from the animal work to the exposure levels that you are concerned about. And they are many-fold different.

And it is rare that we actually have effects in animals in the dose range that is equivalent to the dose range that we are concerned about in terms of human dose. So generally it will come down to we really need evidence of adverse effects in humans.

Another standard that I just want to briefly mention is that -- that applies to dietary supplements which you haven't heard about and I don't really have time to go into. And there Congress identified the standard as the dietary supplement presents a significant or unreasonable risk. Okay.

But I just wanted to point this out, that within the Act itself, you have these different standards of risk that Congress has identified as to whether you are talking about contaminant, a dietary supplement, a pesticide, a food additive or whatever. Can I have the next slide


All right. Thank you. Now, one of the -- there are some key issues that we have to deal with in terms of contaminants in terms of setting a formal standard which we call a tolerance under Section 406 of the Act. And one of the distinctions which I have already alluded to is this distinction, is the hand of man evidence.

Another is avoidability. In other words, whatever standard we set, there has to be a reasonable expectation that you can avoid that level of exposure and that standard will meet that. In other words, if you set a standard so low, okay, that it is -- no matter where you look you can't avoid it, then you failed the standard as defined by the Act.

Another one is detectability. You could go through the safety assessment paradigm. You could identify an acceptable daily intake, a tolerable daily intake, a reference dose, a minimal risk level, all of the same terms for a safe level. If it is well below what you can actually measure, then the Act says, no, again, you have failed the detectability standard as delineated in the Act.

Then you also have to consider multi-source and pathway analysis. In other words, with lead, we couldn't just consider lead from the diet. We had to consider lead from all the other sources and pathways that humans are exposed to in terms of realizing their body burdens.

And then another factor that we have to take into account is the competing dietary risks. In other words, if you set a standard and you eliminate a certain portion of the food supply by that standard, what are the resulting competing risks that you have to take into account in terms of the nutritional loss and risks, in terms of the fact that if you remove this source of protein and the population has to go to another source of protein, have you considered the competing risks

A good example is if you are concerned about a chemical contaminant risk, you do something that -- in other words, you come up with a risk management decision that results in someone consuming less of this source of protein that you are concerned about, you go to another source of protein where there is a great microbiological risk. So you have to weigh these competing risks in terms of the standard that you finally decide on in terms of a chemical contaminant.


I have already gone over that. Just briefly in terms of, again, when I talk about safety assessment, in terms of what you heard about food additives and pesticides, I mean, this is a paradigm that was set up by Arthur Layman and Fitzhue in 1954. And basically, it comes down to the use of what we call, for instance -- in food additives, it is called safety factors.

At EPA, the reference dose is called an uncertainty factor. But they are basically -- you know, they are. They are the same thing. Okay. They are a different term for the same thing. You are trying to account for really two issues in terms of the ten-fold safety factor, to account for inter-species extrapolation -- in other words, going from laboratory animals to humans. And then also to account for human sub-population sensitivity, you use another ten-fold factor.

Now, you know, since Layman and Fitzhue set this up in '54, there have been further modifications to this. One is the additional use of another ten-fold factor that is used for the reference dose where you are taking a sub-chronic, in other words, a less than lifetime study. And you are extrapolating to a reference dose which is intended for chronic exposure. You would then apply another ten-fold factor to account for that.

And then there are other factors that are called modifying factors that are applied sometimes to account for uncertainties that are surrounding the severity of the response. You have preliminary information on particular end points, immunological or developmental. But it is very sketchy, highly uncertain, but somewhat suggestive. And depending on how conservative you want to be, an additional modifying factor could be applied.


Another important distinction here though, in terms of the safety assessment paradigm that I have been talking about, and I am sure you have already heard about this, but bear in mind, in terms of the safety assessment paradigm, there is a distinction in terms of when we are talking about a non-carcinogen versus a carcinogen. The methodologies are different.

I have just told you about the safety assessment paradigm that applies to non-carcinogens. Now, for carcinogens, basically, what the process involves is the extrapolation generally using dose information from a bioassay. And it is a downward extrapolation because, again, the dose range that you are studying in a cancer bioassay is many-fold higher than the dose range or exposure range that you are concerned about.

So it is an extrapolation downward. And it could be linear or it could be sub-linear. It could be super-linear. It could be, you know, any way you want to model it. Now, generally the default way to do it is through a simple linear extrapolation, through zero. But I just wanted to point out this distinction in terms of safety assessment in terms of these two general categories of end points.


It is important to bear in mind though that this safety assessment paradigm is really a first step in an iterative process. And I showed you that model, that paradigm in the beginning. And as I pointed out, there are many terms that have been coined to -- that really do mean the same thing.

And I think sometimes this lends some confusion that when somebody hears the term ADI, TDI, reference dose or minimal risk level -- this is the Agency for Toxic Substances' terms -- that these are different paradigms. They are not. All right It all goes back to Layman and Fitzhue in 1954. So I think you need to bear that in mind.

And it is a very useful screening paradigm for rooting out or eliminating trivial public health problems. And that is that by and large it serves us very well. It provides us with the answer to say this answer is sufficient to assure us of a level of safety and we need to go no further.

And as I said, by and large, when you are talking about pesticides or food additives or contaminants, that is as far as we have to go. Now -- but there are problems and there are instances where it doesn't always serve us that well.

And that is those are the cases that you hear a lot about and that is the leads and the dioxins and the PCVs because when you go through this safety assessment paradigm where you end up looking at a whole data set of information, you select one study.

You identify one dose level called the no observed adverse effect level or the lowest observed adverse effect level. And you apply your uncertainty safety factors. You end up with an ADI, TDI, whatever term you want to call it. And you compare that to your estimates of exposure. And lo and behold, your estimates of exposure are over this safe level.

And so -- okay. And so you reach the conclusion that it is unsafe. Well, from a contaminant standpoint, going back to what the act mandated to us in terms of avoidability, detectability, competing dietary risks, we need to think about risks above the safe level because we have to weigh our risk assessment at the end of the day against these other issues.


So -- and just to point out that in some minds and in some circles, the uncertainty safety factor issue is deemed to be not a science issue, but a risk management issue. In other words, it is -- and the size of that uncertainty safety factor range is dictated by your level of ignorance. In other words, the less you know, the bigger it is. Okay And so some people look upon that at the end of the day as a risk management tool. And just let me --


So just getting back to this paradigm, for contaminants, many -- most of the time, we operate very well within the safety assessment consideration of paradigm. But there are issues like lead, PCVs, methyl mercury, where we really need to move to the next level of the paradigm and deal with issues of the degree of adversity, the variability and uncertainty of dose response.


DR. STERNER: Questions for Dr. Bolger on contaminants That was very clear and very understandable. After the day you have had, we appreciate you just showing up. It is just good to have you here. You can go down here. We are all set. We are moving here to the public comment period. So we have an hour scheduled for this. Dr. Sundlof, did you --

DR. SUNDLOF: No, I am just going to sit up here.