• 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

Roundtable Discussion

DR. McDERMOTT: -- Alessandra and I are the last speakers for the last sessions today and I know you are all exhausted. You must be. I’m tired. I had originally in putting together the agenda had put together, because I knew that we weren’t have the time for questions that many people may have wished to have during the course of the presentations, I thought, you know, at the end we could do a couple of things. We could spend time asking questions we didn’t have an opportunity to ask, or we could have what I called a Roundtable Discussion.

But really to me is to share with you some of our thinking in trying to address the Science Board recommendations for NARMS, because there are a lot of questions to consider and a lot of possible -- let’s see, who do I work this.

We’re faced often with a lot of different alternatives and how we make use of limited resources in the program. We are the members of NARMS in the Federal Government, public health servants, running a public health monitoring program that we want to be as good as it can be to serve its public health purposes.

And it’s, you know, it’s hubris to think that we can always come up with the best answers on our own, and often what, the input of others who have thought about different aspects of monitoring, we just heard about harmonization of terminology and how many challenges there are. There are just simply faced with agreeing on language or -- it took a long time as Tom and Jeff and others can say, just to get people to use methods that were comparable. What a huge step forward that was. At least we knew if people were using the same methods we could do something with the MIC values even though we may fight over how we draw a line in the population.

So, there are always these challenges. And a large one that we’re faced with now, is what do we do in the face of limited resources in light of the Science Board recommendations for NARMS.

One of the purposes of bringing together people from different parameters with different experiences is to sort of do this lesson-learned. Well, you know here’s the way we approached it. And sometimes it’s different depending on the public health burdens in different regions or countries.

But so for example we have -- well, let me see if I could -- I typed a few slides during Peter’s presentation. Sorry, Peter, I was -- somebody might have to help me. I don’t use this version and I’m trying to figure out how to get to -- there is no My Computer icon. Hang on. I have a Mac so forgive me, I don’t -- I don’t know my way around it too well. How do I get to -- oh, here we go. Sorry.


DR. McDERMOTT: And it’s not necessary that we answer all these questions, but I thought it would be worth taking a few minutes to at least air these questions. So that you know what we’re dealing with.


So, looking back it was mentioned, Peter mentioned it, at the start about the different activities of OIE, well, back in 1999 her’s what OIE said about what you should do integrated surveillance. And here are the considerations that you should weigh.

The animal species and categories to be sampled, the food samples, whether they should to go abattoir or retail outlet, how much weight you should put on domestic versus imported, the sampling strategy to be employed. This as we all know is a huge thing as Peter just pointed out in his way about how -- where you get your samples makes such a big difference, what types of samples, what species to be isolated.

Now, some of these we’re pretty clear on. We know that for example in NARMS we have to have Salmonella and Campylobacter in there. But aquiculture was brought up before. Well, do we want -- we’re in 10 FoodNet sites in NARMS, do we want to test more samples from those sites, do we want to expand those sites and do -- and get a better geographical representation or do we want to start doing Tilapia.

So, it’s always this bouncing act because drug use may be high in aquiculture. I’m not saying it is. But let’s assume that it is, is that important if the burden of illness from foodborne disease from fish sources isn’t as high as it is from poultry or some other source for example.

So, these are the sort of things that we’re trying to weigh in the balance as we look forward in the program in how to make best use of our limited resources.

Debby Cho said in 2001 an ideal in-grade surveillance program would be isolates from human clinical cases, retail meats, supplier, food supply and then animals.

You know, I think about that sometimes and I think well, do we still believe that because we know, and I’ll show you some slides in a minute, well, let me just skip ahead to them.


We know when we look at animals, and Paula, this is Paula’s data, it matters whether you get them as slaughter or on-farm. There are huge differences between slaughter and on-farm. There are huge differences in serotypes between --- and the carcass at the end of processing. And so which is the sentinel site. I agree you want to be consistent so your data is consistent. And you can say here’s what we did, we did it year to year, we did it in the abattoir, we did it after transport, and at least you have a stable sampling area. But all of them have weaknesses, all of them have challenges. The resources to get on-farm are considerable despite maybe advantages of sampling on-farm. So there are lots of different things we try to grapple with in setting priorities.

And like I said I’m not sure that we need to answer it all today, but I just want you to be aware that these are the sorts of things that we’re faced with.


The other issues in the OIE guidelines, culture method is used, that certainly matters. Temperature difference can skew species that you might isolate of Entercococcus for example. But of course certainly from Campylobacter when we’re only culturing for thermophilic species, both in the clinic and in our monitoring program. So we know culture methods matter a lot. They matter in the types of serotypes you recover.

The antimicrobial --- susceptibility testing, I’m still befuddled by streptomycin in the EASSA program for Campylobacter. There must be a good explanation for it. But we should try to be doing the same antimicrobials as well as trying to come up with the same interpretive criteria.

The quality control, quality assurance as I mentioned, at least in the laboratory portion of monitoring has been pretty well worked out. Ensuring compliance is a bigger challenge then in most cases then working on new standardized methods.

The database design, that has -- the biggest challenge of that is of course dedicated resources for it, but we’re working on that in NARMS right now.

Analysis and interpretation of data, one goal we would like to do -- that we have in NARMS is to get the data out there for anybody to look in a way that’s user friendly so they can perform their own analysis. That’s one of our ultimate goals because there are always new ways of looking at data and more eyes are helpful, and try to get it there faster as well. So reporting considerations, et cetera.


So, I just want to quickly rehash these. They’re just a couple of these. So the database, like I said, is really just getting resources dedicating to it, and getting smart people to work on it. I think this is probably the one where in my mind we have the biggest challenge, which is making decisions about sampling.

So, in retail meats, like I said, we have 10 sites. And that represents about 14.5 percent of the population. And in those 14.5 percent of the population we sample a fraction of the meats that people are purchasing and bringing home. We know it’s not geographically representative, there are big gaps, if you just look at the picture of the country and where we are. We don’t have all the populations represented. Where different ethnic groups live in different regions of the country.

So, you know, the first question is if we want to improve the retail meat sampling what’s the most rational way to do it. Is it to be more geographically represented, is it to get more samples, to get more isolates. So these are the types of things that we’re chewing on.

And I guess what I would say is whether we have time to discuss it or people want to provide input on these things now or not is not so relevant as to let you know the sort of things that we’re trying to balance.

Probably even more important in the retail meats sampling of course is the abattoir sampling as Paula eluded to. FSI has gone to a risk base inspection process where they revisit failed plants multiple times, and leading to bias in the sample set. How do we get around that. And it’s the biggest challenge because if we’re agreed that say on-farm is the ideal place, as WHO recommended, we’re looking at the cost and logistics that are just I think are prohibitive to be representative.

So then where do you go. Do you go to -- as I showed you before or -- I haven’t shown you this.


Paula showed you this slide, do you get animals right off the truck. And well, now what you got is microbiological shifts that happen during transport which we know happen quickly. Is that weakness easier to live with say than the fact that what’s on-farm is not representative in meat. So there are all these sorts of things that we’re trying to consider. And it’s not just scientific.

We don’t have the advantage that CIPARS had where they sort of build their program from the ground up. NARMS was build on the back of pre-existing infrastructure where it was -- we did it certain ways because we could, we could do it affordable, we could implement it quickly. But that doesn’t necessarily make for the best monitoring program as far as data robustness and representativeness goes.


So, I guess maybe I can leave it there and say that as we look forward in trying to implement the Science Board recommendations, they like us recognize sampling as probably one of the biggest priorities, we’re trying to weigh these different competing priorities in the light of resources and logistics and come up with a program that serves public health needs better, that’s always been our goal.

And I guess I would say that if anybody has any specific recommendations on that I hope you won’t hesitate to let us know.


And I don’t want to delve in these closing comments into the issue of harmonization. I think there are definitely ways forward there. I think a lot of it is communication and definition. And I’m a firm believer that you can’t start a conversation until you’ve agreed on the definition of your terms and so maybe that’s why that conversation hasn’t advanced as much as we would like it to.

But we certainly have to understand each other if we’re going to be able to communicate our results and understand the situation in different areas of the world.


So I think -- those are, I guess, thematically the two big issues. Harmonization ultimately so we can compare data, but on the front end a really rational design when we have the opportunity to improve it and how we might prioritize it.

So I don’t come here with answers but just to offer those to you and ensure all the stakeholders that we’re certainly dedicated to this program and we’re eager to learn from anyone we can and implement really rational changes that will improve it.

And so, I guess with that, I would open the floor to any comments or questions if anybody has any, or if you’re exhausted as I am we can always call it an evening.

DR. SHRYOCK: Tom Shryock, Elanco Animal Health. A question for you, Pat. You mentioned that it’s cost prohibitive to do a lot of these different approaches to NARMS. But it seems to me that that is exactly what needs to be done, is to redesign the entire NARMS program basically from the floor up to have a program that’s got all these things the Science Board recommended.

So, if one were to do that and play the outcomes in terms of what would that take, what would that look like, what would be the benefits as well as the disadvantages of doing that. Why not just go ahead and model that out and see what it really would cost --

DR. McDERMOTT: We’ve done that.

DR. SHRYOCK: -- versus trying to hobble that together.

DR. McDERMOTT: We’ve done that and we’ve done it in a series of iterations where sort of said this will be terrific, this is the best blue sky version, here’s something a little more digestible, here’s something actually realistic and here’s something practicable. And we went through a series of those.

And the perfect situation, as best we can vision it was -- raisable, perhaps the best way to describe it. It was laughable, it was okay, what’s your next recommendation. So, it was very, very expensive.

DR. SHRYOCK: Like how much?

DR. McDERMOTT: I want to say about $26 million.

DR. : (Voice away from microphone.)

DR. McDERMOTT: We don’t have that kind of support.

DR. SHRYOCK: If you haven’t asked for it then how do you know the answer is no? There is a lot of Legislation, there’s a lot of regulation being discussed, $26 million may be more affordable than you think.

DR. McDERMOTT: Well, we can talk later, Tom.

DR. SHRYOCK: Thanks.

DR. McDERMOTT: We’ve asked.

DR. ROACH: Steven Roach from Food Animal Concerns Trust, you know, I’m an advocate and I usually, at least in some things to do, the risk management side, I disagree with Tom. But on this I agree with him 100 percent. I also don’t think $26 million is raisable. And I think again, I’ve made offer to go and lobby with him for more money on that. And you all don’t want to hear about this lobbying for money.

But I would say one thing that I’ve requested in the past and haven’t got -- you say your budget is limited, but when I’ve requested trying to see how much money actually spend on NARMS, we haven’t been able to see that.

You know, I got some information -- I have data on what the budget, what the actual budget for NARMS from 2005, but I haven’t ever seen anything since and I thought one of Dr. Woo’s --

DR. McDERMOTT: It hasn’t changed. It hasn’t changed since. It hasn’t changed in quite a long time.

DR. ROACH: Yes. I mean if it’s still $5 million, there is a lot of room for -- increasing it if we can say okay, it’s been stuck at this level and there’s a reason to do it. And again I think all of the -- us people outside look in, you know, me and the industry people and they say well, there are problems and we all want good data.


DR. ROACH: And I think we can do better than we have been.

DR. McDERMOTT: Thank you.

DR. MÄKELÄ: Pia Mäkelä from EFSA. I fully agree that these sample strategy you have and where you take the samples it actually has a great impact on the results. And of course it depends on what you want to know, and of course you have to decide what you actually want to know out of this monitoring.

But one thing that I would like to comment on because I was following the discussion today and the presentations is actually the number samples taken or actually the number of isolates to be tested. And I kind of pay attention that in many programs it was reluctantly low. Because it is typically an issue we discuss a lot with our Member States when we make this harmonized schemes or harmonized specifications and typically of course the Member States would like to have quite low number samples because of it is a question of cost.

But from the statistical point of view at least according to our estimations you need quite a good -- quite a high number of samples to get the representative picture. So like for our harmonized schemes for antimicrobial resistance we recommend at least 170 isolates to be tested because if you really want to detect a trend within couple of years, you know, know kind of reliable way, it has to be quite high.

And I wondering how much attention you have paid for this and of course you have different states, so do you consider the states as a different structure. It actually means that you know in a high number of isolates to be tested. I mean, some is history I know of course it’s much about the cost, but it’s an important issue.

DR. McDERMOTT: So I think you’re suggesting that it be better to get a lot more samples to better trend resistance say in Salmonella where you would have to have 170 probably per serotype. So we’re getting two percent or less in pork and beef, so that would be quite a substantial change to get that many of say the top, even the top serotypes, or the top four serotypes.

It’s a good consideration. But I don’t want you to get the impression that where we sample makes a difference, where you sample too, where everyone samples makes a difference. And that’s probably something missing from the harmonization, even in Europe is where the samples are being collected.

DR. MÄKELÄ: Of course we pay attention to that as well. But it is -- like in our Salmonella --- I mean Salmonella survey it’s basically like the control programs. But I do agree that this is first agree with the Member States, what exactly we want to know. Do we want to know what’s happening in slaughterhouses level or the retail level and then of course it’s totally different --

DR. McDERMOTT: Well, one of my concerns is sampling inside the abattoir because if -- I mean if you think about it it’s quite a dramatic change from entering the front and going out the back. And in that process you’ve exposed the animal, if you will, to an environment that it’s never seen before, that has the residues of the selection pressure from a lot -- over a long time from a lot of sources.

And I think we need to do better on that. We need something that represents the animal rather than it’s untimely demise, I guess. And all that environment of the abattoir might impact the results. And then pile that on top of the risk based sampling and in something I think has got to be at the top of our list.

DR. MÄKELÄ: Okay. Thanks.

DR. McDERMOTT: But thank you for your comments.

DR. SMITH-DeWAAL: You’re on the hot seat.

DR. McDERMOTT: That’s okay.

DR. SMITH-DeWAAL: Caroline Smith-DeWaal, Center for Science in the Public Interest. I want to rise the question again, not just with you, but with the other countries. I was going to ask the question of Richard from Canada, but he got away before I could ask it. On the issue of what are the lessons learned, not just about the front-end data, I mean sample collection and methodology piece, but on the data sharing piece.

The complexity of this system is that you’re often working with multiple government agencies, the U.S. demonstrates that very well. But how to build a platform data sharing that is very effective not only in the intra-government level but at the inter-government government level, and what are the lessons learned there.

And Richard, if you want to go like relieve Pat, you could help out, too.

DR. McDERMOTT: Oh, come on, Richard. As I mentioned, Caroline, that’s a goal of ours is to make the data two things, more real time and more accessible. And I think we’ve got some pretty good momentum on getting a data base that will do that, that we hope to post on an internet site at some point, that would be suitable clean of any unneeded, or unneeded medi-data if you will, but out there so other people can do analysis. And even to supply the tools, at least some of those that we can incorporate into it and do exactly that. And Richard, you may want to comment on your situation.

DR. REID-SMITH: Well, as I mentioned in the structure, even though it wasn’t exactly by design, since most of the operations of CIPARS are within the public health agencies, it’s really just an internal IT issue which is still a problem. But the data sharing is -- at that level is just within different sections within one organization.

As far as the transfer out, we don’t have a good solution for that although we are working on it, for example we would like to transfer the data back to the provinces that, for example, where we collect the retail samples. And we have been running a pilot project with the Province of British Columbia where they get their data back almost immediately and we hope to extend that to the other provinces in the long run so that they get their retail information.

The human data, the National Micro Lab, you guys just -- it just goes straight back through an exiting data sharing mechanism. But it’s not unified across all of CIPARS. We don’t have that yet. Although there is actually a probable, a mechanism that is being designed to share information from the Federal government to the provinces for public health that may be, that may actually be useful for that.

DR. McDERMOTT: Nkuchia, you have a comment?

DR. M’IKANATHA: My comment that I wanted to mention is that actually for at least my understanding is that you work with state’s health departments and these are independent entities. So the cost of NARMS is not really the cost that it costs -- that the NARMS program itself spends because there is a lot of indirect cost covered by the partners. And the partners should cover those costs.

The comment I want to make is there was an excellent presentation by Doctor Woo from Korea, and I --- of bringing these international partners is that we can learn from them. And he has a program for young kids and he really has shown a very effective communication, if you look at the range between him and the Doctor Ran from China and Doctor Kariuki from Kenya where they haven’t been able to reach the policy makers. Perhaps he can elaborate on that or he can help us learn how you can communicate at the Federal level and how we can communicate even at the state level. Thank you.

DR. McDERMOTT: Yes, I think at least from FDA’s perspective on this issue, we’ve spent most of our energy communicating with practitioner groups and professionals about our thoughts on prudent use and judicious use guidelines, and haven’t gone to the level Doctor Woo has in Korea.

But -- I don’t know -- I don’t know what FDA’s role would be in that sort of out-reach exactly, but it’s been more along the lines of working with the professional groups so far.

DR. WOO: First, thank you very much for having interest in our national antimicrobial resistance management forum in Korea. And I think that this is the kind of case that the USDA invited me and Doctor ---, the Deputy Director of the Food and Microbiology Institution. She is the Codex Secretary for the task force on the task force on antimicrobial resistance.

So, I think we can collaborate between countries and also with the international organizations. As you compare the NARMS program, is a national antimicrobial resistance monitoring system. It’s a -- title is monitoring system. So it is based on the focus of the risk assessment. But in our country we focus on the title -- is National Antimicrobial Resistance Measurement Program. So we focus on the risk management, so including risk assessment. So we focus on the public campaign and education. That is very important.

In some sense I would like to say that we, all the people in this room, are the heros in the 21st Century. We cannot solve this problem within our generation. We should solve this problem in some sense is the within our next generation or for the generations to generation. So we should give these some platform and the basis for our second generation.

So, all the people working in this area is I would like to say heros in the 21st Century, we are working for us and for our children and also for our children’s children. So, that’s the -- my point of view. Thank you.

DR. McDERMOTT: Thank you, Doctor Woo.

DR. WATERS: Andrew Waters, TGEN. Doctor White touched briefly on the FDA’s draft guidance that’s coming out. I was wondering if you could give us an update on if NARMS plans to monitor this program and how that might come about.

DR. McDERMOTT: You’re speaking of the recently released one that’s open for comment or the public health action plan.

DR. WATERS: Yes, that’s correct, on antimicrobial growth promotion, I was wondering if there was going to be a monitoring system put in place.

DR. McDERMOTT: Nothing in addition to NARMS, no. That’s what we have for -- that’s all we have.

DR. DUTIL: Lucy Dutil, ---, I would just like to go back to what we were talking about, previously where sampling and -- our abattoir components probably the cheapest one we have and this is because we have participation from industry. I know that it seems like out of -- even thinking about it in the States, but it seems to me that -- given it’s a global problem and it’s everyone’s responsibility that it should make sense that the industry also participate and that it does reduce the cost quite a bit.

And when you look at it, when you have a sampling sheen that is representative and you don’t need tons of samples at the end of the year and it’s not that big of a contribution, so I would encourage industry to --- participating on a volunteer -- like the Canadian industry is doing, it doesn’t kill them, it doesn’t put them out of competition either.

And what we see when we compare farm data and abattoir data, and I can only talk about the pigs data, because we only have both in pigs, we don’t see any difference, significant difference for serotypes or -- well, we don’t have that many Salmonella in pigs, but for serotype resistance. So, because we look at cecal samples so in that case I think it -- it needs to be demonstrated without the industry.

But anyway my point was I think you could have something relatively less costly if you could get the participation of industry.

DR. McDERMOTT: That’s a good point, Lucy. And that’s a great segue to wrapping up today because tomorrow we’ll have an opportunity to hear from industry representatives and maybe that subject will be brought up.

So, tomorrow also we have reports from some of the research, focus on the research work. And then we’ll hear from our stakeholders and so I’m looking forward to another fun day tomorrow. And thank you everyone for coming and have a great evening.


(Whereupon, the meeting adjourned at 5:05 p.m.)