Animal & Veterinary
Interpreting and Weighing Risk
Will Hueston, D.V.M.
DR. HUESTON: My challenge, I would like you to note first my challenge is to talk about risk management. And I am going to speak to you, in fact, as an ex-risk manager. So I have donned the appropriate apparel. I have my dark suit, white shirt and power tie.
The challenge is that I am going to try to put myself in the position, in fact -- I use Steve Sundlof as an example -- as the risk manager that must consider the information that is put forward and and make the decisions or recommendations for regulatory action.
I would like to start by giving you a little clarification, Politics 101. Now, there are actually two types of risk assessments that are ongoing for every decision or every regulatory issue. There is a scientific risk assessment and then there is a politic risk assessment.
So what happens is the lower down you get into an organization, the more science becomes important and touted. The higher you get into an organization, the more important politics. So in the United States at our top tier are all political appointees. And don't ever kid yourself that politics aren't taken into the equation for making decisions.
At the same time, the government has excellent scientists. And the scientists at the other end of the spectrum are doing some very good and some very clear science. Now, in the middle rests the position of the risk manager who must manage both the political risks and the scientific risks.
I want to give you then ten or 12 points or bullets to take away about the challenge of risk management. And these are Will Hueston's personal opinions based on my personal experience and please take them as such.
Number one, the risk manager must balance science and politics. Let's start with the science part. I think it is very -- it is noble. It is appropriate to say that policy making must be science-based. Do not kid yourself for one moment that there are not politics in science.
If we take the same set of data and ten scientists, we are very likely to get multiple, if not ten, different interpretations of the data. I think it was pointed out yesterday, this is one of the advantages and benefits of the scientific method, the conjecture and refutation, the argumentation that ultimately we hope leads to the best conclusion.
Secondly, this concept of mixing in the politics is all about who stands to win and who stands to lose, and recognize that wherever two or more people are gathered together, there are politics. And decisions in the end are not necessarily predicated on their relative importance in terms of risk to the United States. And I need not go very far with discussing cigarettes and human health to make that point in terms of regulatory decision-making.
We add for the risk manager the challenging of adding a modicum of common sense which is practicality in economics. The most scientifically sound risk management strategy is worthless if compliance is low, a lesson painfully learned in some of our recent public health issues.
Point number two, risk analysis is a tool. And i firmly believe that risk analysis is a tool to support rational decision-making in the face of uncertainty. Now, this risk analysis tool incorporates hazard identification, risk assessment, risk management and risk communication.
The critical first step as emphasized I think yesterday by Louise Kelly, the critical first step is clarifying the question. Often overlooked is the importance of clarifying the question. The interpretations of the results can only be done in the context of the question that was asked.
Now, nextly -- next, the question itself may limit the need for the implementation of this risk analysis paradigm. We may, in fact, ask what are the hazards alone or ask in the case -- part of the case of the situation we are currently describing what are the consequences.
As well, and just a reality check for you, because of the involvement of politics and risk analysis, there are occasions in which political decisions are made. And the risk analysts are asked to put together a risk assessment to justify a political decision that has already been made.
Now, I am not trying to say in any way, shape or form that the scientists that are involved in these agencies are biased. I am just giving you a reality check on what actually happens at times.
Point number three, the risk assessment or risk analysis process is far more important than the final output, probability or numbers. So the beauty of this concept or this paradigm is the process. It is a logical approach to organizing data, available information to taking inputs through processes to outcomes in a systematic way. And the many benefits of risk analysis include identification of data gaps, detailing of assumptions, the redistribution of resources in terms of risk management potentially, and the targeting of educational priorities.
Point number four, risk communication is far more than simply sharing the results. Risk communication means at its heart the involvement of all of those potentially affected parties in the entire risk analysis process.
Now, I would like to clarify some points that were maybe perhaps miscommunicated yesterday from my personal perspective. The risk analyst must solicit information from both the scientists and the stakeholders.
And I think one of the great advantages of risk analysis, in fact, is that as one begins to clarify the hazards, hazard identification in and of itself is one area in which we have the greatest amount of expertise and the widest population. In other words, every American considers himself an expert in hazard identification.
At the same time, if we involve the stakeholders, as well as the scientists, in the process of hazard identification, model development, identification of data sets, they gain greater buy-in and the ultimate product or process of risk assessment gains additional credibility.
At the end, we need most to avoid what I have heard called the "dad fallacy." And that is you do your analysis in private, in secret. You decide on your regulatory approach or your risk management approach. You announce your approach and you hold up the risk assessment as your justification. To me, that is an abuse of the process.
Point number five, a systems approach or a pathway analysis supports risk management. So as a former risk manager, it was very critical for me to be able to differentiate between the options and a pathway analysis that allows then not only an evaluation of the magnitude of various impacts -- of various inputs, but also allows the evaluation of their effect on the outcome of concern.
As was presented in this risk assessment, a very interesting use of these spider diagrams, sensitivity analysis, and that can be used very effectively from a risk manager's point of view to look at the relative impact of different strategies of risk management on the final outcome.
At the same time, risk analysis can help us to identify the attributable fraction. In other words, that part of the risk that can be attributed to specific practices. Our ultimate decision may want to incorporate that because we may want to target those behaviors or those actions that have the greatest contribution.
At the same time, the reality check is as just was eloquently presented and discussed by Glenn. And that is the challenge that some of the risk management procedures that one might -- options that one might lay out in the idea situation may not be available to us because the population and the risk -- the change in risk behaviors cannot be accomplished.
Point number six, evaluation of one risk cannot be accomplished in a vacuum. We need to develop holistic approaches. Back when I was studying physics, about the only thing I remember from physics is a concept that says for every action, there is an equal but opposite reaction. And an actual fact in the world in which we live and the ecology of the world in which we live, to take an action has implications in other areas.
And one of our challenges in doing -- as risk analysts is to begin to incorporate this into a holistic approach to risk analysis. We must consider the impacts of proposed risk management on other risks.
An interesting example, the DPT vaccine. Corporate America made a risk -- an economic risk assessment that said that the risk of a lawsuit for the sequelae to DPT vaccine was greater than the benefit, the profit that they made from selling the vaccine. It led to in a sense at one point if I understand it correctly the lack of a company to produce the vaccine. We ended up coming with a risk management strategy to address those concerns to get the vaccine back into place to meet the public health need.
Point number seven, effective risk management must consider economics, cost effectiveness and practicality. Now, I realize -- and this morning was presented I think some very important concepts. If you follow the legislation, it is very clearly stated in the legislation as it regards the evaluation of some risks, that benefits cannot be considered.
I also recognize that one area of -- let's see, I won't use the word, "friction" -- of difference between the way in which the U.S. Department of Agriculture evaluates risks or implements risk analysis and the way in which the human health services has implemented risk analysis is the question of whether or not economics are incorporated in the risk analysis.
In the human health side, the public health side, we tend to shy away and say that we cannot put a value on a human life. We cannot translate a human life into a value. I would contend that, in fact, we do that on a daily basis. We may feel more comfortable to suggest or to say let's look at the public health measures that have the greatest impact in reducing illness or length of illness or number of deaths. But much of that translates very clearly into economics.
We have finite resources for public health. Therefore, we must look at the opportunity cost. In other words, what are we not doing if we put more money into a risk management strategy.
At the same time, we need to reiterate that the most logical and ideal solution may not be the most effective. Again, we can't rely, if you will, on scientists alone. We have the issues of sociology, of behavior. Now, a lesson that I learned quite painfully is that regulation alone does not accomplish risk mitigation.
If you would like proof for that, then drive home with me tonight on the Beltway at 4:30, a beltway on which the speed limit is 55 miles an hour. And I would argue, and anyone who lives in the D.C. area, that if you drive 55 miles per hour, I think that you are probably at a greater risk than if you drive somewhere between 65 and 70.
I would like also to end this risk management -- in the challenge of risk management to point out and to emphasize the very difficult position in which the Food and Drug Administration finds itself. I have great empathy.
I think of all the federal agencies with which I have had experience, the Food and Drug Administration is in the unique position of having the greatest numbers of unfunded mandates and the least increase in terms of their budget while at the same time carrying with them the greatest impression of being the bad guys. So it is a real challenge for my colleagues in FDA I think.
Point number eight, reasonable and acceptable risk are fluid concepts and they vary according to a couple of -- a number of factors. So we have already established or we discussed the challenge of defining safe. Safe is a subjective term. I would not get concurrence in this room on a definition of safety. Safe is a subjective term.
At the same time, zero risk is unachievable. There is no zero risk. We face the challenge that we have prostulitized, if you will, that zero risk is achievable when, in fact, it is not. So the concept of safety and the reasonable or acceptable or tolerable risk are as much tied to a number of very human concerns like the outrage factor, the fear of the unknown, the question of whether or not a consumer has a choice.
Now, it also cannot -- risk cannot totally be defined as a mathematical entity. And I just want to share one small anecdote that impressed me with this. I was in a meeting in Paris, an international meeting in which we were talking about food safety at the farm level. And there were several presentations from the U.S. about farm-to-table risk assessment and the impact of the farm and the discussion of the potential carry-over risks of microbiological contamination.
During the discussion period, a Frenchman stood up and very impassionately said, "You Americans have forgotten what the jois de vie is all about. I want to eat my raw cheese. I want to drink unpasteurized milk. And I am willing to take the additional risk so that I may enjoy life."
We need to be very careful to recognize, in fact, that different cultures and different backgrounds and different personalities define safety differently. The challenge then of the risk manager, not to discriminate.
Point number nine, risk analysis is a dynamic process. It is not static. It is forever changing with new data. In fact, the document that was presented to you two, three, what, less than a week ago has already changed because this meeting will stimulate new data. And new ideas are coming to the people participating in this meeting. It is not a static situation.
If the risk management is successful, then -- in dealing with one of the contributors to risk, then something else will become more important.
Now, point number ten, the key to credible and effective risk analysis is trust, T-R-U-S-T, trust. And trust is built over time. And trust depends on openness and involvement. And trust is built first and foremost on the ability to listen.
It is interesting -- it can be very interesting that -- it is interesting to note that a good risk analysis in and of itself reduces risk. It reduces risk because of the increased education. In fact, some of the greatest risk management successes I experienced were situations in which we had sufficiently analyzed risks and involved stakeholders, that the stakeholders took actions without the necessitating for regulation.
Regulatory action is a very expensive, a very slow and a not very effective means for managing risk. We do not have a compliance force large enough in the United States to ensure 100 percent compliance with any regulation. So if one can achieve buy-in and trust and participation, then one may often accomplish greater risk management, in fact, simply in the process of doing risk analysis.
Point number 11, risk analysis presents -- risk analysis in the microbial field presents some new challenges. One cannot automatically take our toxicologic risk analysis and other risk analysis models and simply transpose those onto microbiological risk assessment.
I think as Dick Whiting pointed out very nicely, this microbiological risk assessment is a brand new and interesting area. It is also an area in which we are going to have to struggle with challenges that what is the risk manager's role in factoring in the -- for instance, the temperature abuse of the consumer. So what does -- where does the government's responsibility and industry's responsibility and the consumer's responsibility begin A very difficult question.
Well, lastly -- or just before I reiterate in some of my points, I would like to make one other. I would like to make a plea. And this is a plea for a unified approach. Interestingly enough, I believe that down deep, we all share the same goal. We are all consumers. I don't believe that there is industry out there or businessmen out there that consciously want to produce a product that harms human health.
I would also like to extend this plea in terms of the public health community. I am a veterinarian. I firmly believe that everything I do as a veterinarian I do because of public health. And there are great opportunities I believe for increased collaboration. There is no place in risk analysis for differentiating between good guys and bad guys, for incorporating finger-pointing and for demeaning our colleagues. That isn't going to help us achieve credible risk analyses.
All right. Let me reinforce then the points. Point number one, the risk manager must balance science and politics. Point number two, risk analysis is a tool. It supports rational decision-making in the face of uncertainty. Point number three, risk assessment and risk analysis are a process. The process is more important than the final output, probabilities or number.
Point number four, risk communication is far more than simply sharing the results. It means and requires the involvement of stakeholders in the entire process. Point number five, a systems approach, a pathways analysis supports the risk manager and risk management decisions.
Point number six, evaluation of one risk cannot be accomplished in an absolute and total vacuum. We must look to bring about or incorporate more holistic approaches. Point number seven, effective risk management must consider economics, cost effectiveness and practicality.
Point number eight, reasonable and acceptable risks are fluid concepts. Reasonable and acceptable risks are fluid concepts. Point number nine, risk analysis in and of itself is a dynamic process, not static. Therefore, the analysis itself will be continually changing.
Point number ten, the key to credible and effective risk analysis is trust. And last point, number 11, microbial risk analysis presents us with some new challenges.
I would like then to finish by sharing two things that I learned in my time as a risk manager that continue to be reinforced. Number one, there are some questions that the American public feel are too important to be left to scientists. We should all be humbled by that occasion. There are some questions that the American public feel are too important to be left to scientists.
And number two, the joy and benefit of being a risk manager, here is the interest paradox. If, in fact, we can successfully prevent disease, then we will be criticized for wasting resources on a problem that doesn't exist. If, on the other hand, we do not prevent disease, we will be criticized for not having taken sufficient action.
So I stand before you ready to be criticized. Thank you very much.
DR. STERNER: You will notice that Dr. Hueston left time to answer questions. I told him to try and be controversial. I am not sure that I saw anything but heads nodding yes, yes, yes here. That's leading the cheer here I think. Anybody who has a question, please go to the microphone.
MR. : Thank you, Will. I was one of those nodding my heads through most of that. I do want to make a comment in defense of the risk assessors and the risk managers, at least that USDA which I am most familiar with. Not too many years ago, decisions were made, you know, behind the rooms, you know, behind the closed doors of the administrators and such. And then we came to the point where there were decisions made. And then when we got risk assessors, then we were given the challenge of providing assessment to support the decision.
I think at USDA, which I can speak most closely about, that has changed greatly in the last five years. And I think with the E. coli 0157 risk assessment that is being presented this afternoon downtown and such things, and the S. e. risk assessment, that it shows that we have had a change. And so the world has changed.
DR. HUESTON: I agree wholeheartedly. And I think the openness and sharing the openness is a very important point. And I think there has been progress. And I certainly don't want to demean that in any, shape or form. Yes, sir.
MR. : I would like to ask a question on risk communication. In Europe in the food area, we got a reaction in our consumers which is quite dramatically different from that in the U.S. with regard to, if you like, hormones in beef, antibiotics as growth promoters and recently and most dramatically, genetically modified food.
We got it wrong. Have you got it right How have we managed to communicate an element of hysteria rather than perhaps a rational thought
DR. HUESTON: Well, as you may or may not -- that is a superb question. I will try -- I will share some of my thoughts. As you may or may not know, I have served -- I just finished serving six years as a member of the Spongeiform Encephalopathy Advisory Committee in the U.K.
Some interesting differences -- there are lots of differences between Europe and the U.S. For all of those -- those of you who have had the pleasure of living in Europe, it is a considerably different environment. And I certainly found in the U.K. -- and that is the area in which I have the most experience -- a couple of interesting things.
One, risk analyses are done in the U.K. in the spirit of secrecy. And one has the official Secrets Act which one can hold up and say this is a secret. And you are legally precluded -- the newspapers, in fact, are legally precluded from publishing that secret. If they publish, they close the newspaper and haul the publisher off to jail.
Number two -- and we are struggling. You know, in six years of SEAC, you have watched -- it has been interesting in the last six years to watch the whole evolution. Public meetings -- I have never attended a public meeting like this, that involved SEAC.
I served for a short time on the Spongeiform Encephalopathy Advisory Group -- the Transmissible Spongeiform Encephalopathy Advisor Group for the United States. All the meetings were held in public. They were all open. Anyone that wants to attend sits in the back. Everyone has a chance to comment. All of the SEAC meetings are held in private.
There have over the years, therefore, built up some public feelings about the role of the government and about what is going on. And they are a little different than what happens in the United States. Now, that is one thing.
I think as well that there are cultural differences. I mean, certainly within the European community itself on some of the very issues you mentioned, huge cultural differences in terms of people's willingness to look to the future of GMOs.
So all I am saying in the end is I don't think there is a right answer. What may be the right answer for the United States today may not be the right answer for Nigeria.
If you have heard or read or followed the WTO discussions in Seattle, as well as some of the other discussions that have gone on, even CODEX meetings, a number of the developing countries have stood up and said, "Do not" -- "It is not appropriate for you developed countries to set a standard that determines food safety in our countries. We are still concerned about food security, the provision of an adequate food supply."
So I think it has to be looked at very carefully and in the context of each individual country.
MR. : If I could just comment, I think one overriding reason that -- from our side is that we don't have anything equivalent in terms of respect, authority of the FDA and the CVM. And I think rather late in the day, countries are now feverishly trying to establish food safety agencies along the lines of the FDA which will have that respect. I wait with interest and I rather doubt they will have it when it comes. Thank you.
DR. HUESTON: Good point. Interesting to watch.
DR. STERNER: Further questions for Dr. Hueston Thank you, Will.
DR. STERNER: Our final speaker this morning hails from CVM. Dr. Linda Tollefson is the Director of the Office of Surveillance and Compliance at the Center for Veterinary Medicine. Her D.V.M. degree is from the University of Illinois and her master's in public Health is from Johns Hopkins University.
Dr. Tollefson was one of the developers of the National Antimicrobial Resistance Monitoring System, known as NARMS. For those of you who are TLA challenged as I am, that is three-letter acronym challenged, the FFDCA stands for Federal Food, Drug and Cosmetic Act. I actually had to go ask Linda.