Animal & Veterinary
Assessment of Risk: Water
Steve Shaub, Ph.D.
DR. SHAUB: Thank you, Dr. Sterner. Well, probably a lot of you are not aware of the fact that EPA Office of Water actually is considered one of those food agencies. So we really do have a link to the food. In the President's Food Safety Initiative, we were one of the members of the governmental groups that was identified to help protect that nation's food supplies. Next viewgraph, please.
Because we had a couple of questions that really needed to be answered for the panel I guess today, I kind of modified my slides a little bit to talk a little bit about some of the needs we have for the panel discussion. Within the EPA, we are actually now required by regulation to use a risk-based approach to how we actually develop our regulations to protect the general population.
And I would emphasize that generally all of our regs. are for the general population. But within the new criteria, we do have to evaluate and consider the risk of children and other sensitive populations.
One of the things that we kind of -- the approach we are using right now is the fact that if we do have a sensitive population that would have a significant or severe or fatal outcome from some chemical or microbial in water, then we will actually provide special guidance which would be presented to the people with this problem or their clinicians or whatever so that these people would be protected.
One of the best examples we have the EPA and CDC have put out guidance to the people who are affected or impacted by AIDS so that we actually have a boiled water guidance document out to them so that they can reduce their risk of cryptospridiosis which often can have a fatal or very severe outcome. Next viewgraph.
Just a couple of examples. In terms of how we are using risk as far as our development of our regulations, first of all, under the Safe Drinking Water Act which was re-authorized in 1996, again, we are trying to protect the general population.
And one of the things that is probably peculiar is the fact that we do have a risk number. Actually, we target one in 10,000 yearly risks to the general population as being appropriate for drinking water safety. And this is designated specifically against enteric diseases.
The approach that we have used is to establish a worst case organism. And this is possibly open to some suspect I guess in terms of our selection. But what we have done is established these worst case organisms based upon their probable occurrence in water, their potential to cause a disease, and their likelihood of getting through a water treatment system and actually then causing an exposure.
Classically and even currently, we are still working with two general worst case organisms. We currently require a three-log reduction of Giardia from water and a four-log reduction of enteroviruses -- excuse me -- in the treatment process to reach this risk target level of one in 10,000 yearly risks.
In the future, in fact, what we are working right now is in a meeting earlier this week with our EPA's Federal Advisory Committee to look at enhanced surface water treatment rules which will begin to initiate within the next couple of years. And we are changing from Giardia to Cryptosporidium as the worst case target because we know that the significance of this as far as getting through treatment is much greater than Giardia.
What we are trying to do is look at whether or not we need to target the removal requirements on the water shed concentration approach. In other words, do we really stage or increase our level of treatment based upon the likely occurrence of this organism in various types of water scenarios on a water shed basis.
So you may have some systems that may only have to remove two orders of magnitude of Cryptosporidium based upon a very low occurrence of the water. Others, you may have a very significant occurrence concentration which you may have to remove four or five logs of Cryptosporidium.
It is a very big concern of the industry and the water treatment industry because the potential cost associated with a five-log removal are very great. If everybody had to do that, basically, the additional cost to the water industry would be in the billions of dollars to implement those kinds of protection criteria.
One of the things that is unique about our -- having surface water treatment rules is that we don't have a maximum contaminant level like we do for most of the chemicals. And the main reason for that is that we can't really measure accurately the microorganisms that we are concerned about.
In other words, the enteroviruses and Giardia or Cryptosporidium, we just don't have adequate methods. So we have to use a treatment rule. So we do designate that a particular system has to have in place a capability to remove these levels of organisms which we think may occur in the source water.
One of the things also which is the fact that the states actually do the monitoring of the compliance of this. And then they report to the Federal Government or the EPA as to how well their various utilities are performing.
One of the things which is also associated with the development of the enhanced surface water treatment rule is the fact that when we are developing this, we actually have a risk-risk kind of a trade off which we are looking at.
So our risk assessment approach is more convoluted because when we protect against Cryptosporidium, we are also going to have to make sure that we are not in that process of treating introducing large amounts of disinfectants or disinfectant by-products that could be toxic to our consuming public.
So basically it is a balancing act. We want to make sure that we have a process that is going to get rid of the organisms, at the same time not to give a toxic load of disinfection by-products which are potentially carcinogenic to the population. Next, please.
As an example under the Clean Water Act, this is the other side of the EPA's water story. This is basically making sure that waters are swimmable, fishable and drinkable. And just as an example for how we are using the risk approach there, for recreational water criteria, we do have the risk-based approach. And this is against acute gastrointestinal disease.
And, basically, what has occurred in this is a number of indicators were actually tested during the late '70s and early '80s against various types of disease out-points, and particularly the acute gastrointestinal disease in actual epidemiology studies which they showed the relationship of the indicator organism levels versus the particular level of disease outbreaks which are actually occurring.
So, basically, this, you know, gives us our risk-based approach. And what we have come up with is the fact that we do allow 19 acute gastrointestinal illnesses per 1,000 swimmers per exposure a day for green waters and eight for fresh waters. So we really do have here, again, a -- we really do allow a particular exposure level and a particular illness level that can be associated with that particular activity.
The reason we don't have as stringent requirements is the fact this is a voluntary activity. People don't have to go out and swim in our nation's waters. Obviously, we don't like to see this kind of a scenario going on. But at least the public historically has accepted this as being appropriate for this particular kind of level and they accept this amount of illness.
One of the things that is of concern to us right now and we are trying to work on this is the fact that the current criteria are not protective against upper respiratory tract, skin, eye, ear, nose, throat, severe gastrointestinal diseases. They are only really known from a risk basis to be protective against acute gastrointestinal disease.
The way this approach works is that we do establish the criteria. And then the states adopt these. And they are actually the ones that are responsible for monitoring and ensuring that their beaches are safe. Next, please.
Okay. Turning to our current approach, as Dick Whiting mentioned to you, risk assessment really is a new science for microbials. And we have been working on this for a number of years. It's rather a slow pace, but we are starting to generate more speed now, especially now that we do have to have risk-based regulations.
Through a co-op with the International Life Sciences Institute, we have been developing a framework for how we should be dealing with pathogens in various types of water media. And actually, if you want to get the full detail because I am not going to be able to really get into it in much detail today, if you look at least a reasonable summary of where we are, look in Risk Analysis Sub-volume 16.
And one of the things this is -- it does fully consider the unique aspects of microbial pathogen exposures and human health effects. We recognize -- at least we think that the National Academy of Science-NRC model for chemical risk assessment really isn't appropriate, I guess in conflict with -- I guess we feel that really we need to address some of the more unique aspects of microbials and the host populations and the overall association of health effects and pathogen exposure.
And one of the things that did come out of this is that we pretty much followed the framework for the EPA's ecological risk assessment process which has actually gone through the EPA's risk assessment forum now and actually is considered a full-blown risk assessment protocol. Next slide.
Just to show you the general approach that is being used for the framework, it is really no different than anything else that you have probably seen as far as doing the risk assessment, as far as the general approach. We have the problem formulation which the concepts, focus and the breadth of magnitude and the target end points are developed.
Then we go through the analysis phase which actually is characterizing the exposure. In other words, where is the organisms out there and then what are their health effects. And going through risk characterization after that.
One of the things you will notice, we have these arrows. We think this is a highly iterative process. We think that all the way along through the risk assessment, that they really need to look back and see how you are addressing the problems, whether or not you are getting plausible answers and whether or not they are reasonable, going to other ways you develop risk assessments for other types of regulatory procedures.
One of the things, EPA actually has a formalized approach now to how we are doing risk characterization. It is in the draft right now, but we expect this to be completed probably early next year. And this lays out all of the criteria of what needs to be done when you are doing a risk characterization.
The -- to get down into the assessment end of things which is basically where I am going to focus the rest of the day, you have the characterization and exposure in which you are characterizing a pathogen, what makes it a significant concern from the standpoint of what kind of infection is it likely to cause and how is it going to be out there in the environment as to getting out into the exposure scenario and looking at the human exposure to that pathogen, and then coming up with the exposure profile where you have all the uncertainties, assumptions and various models and things like that which are used to actually establish that final analysis of the total exposure and the characterization of the human health effects and the host characterization, looking at the dose response analysis and the health effect.
Again, one of our weakest points we have in micro. right now is the fact that we don't have a lot of good dose response data to complete our risk assessment. Anyway, coming up with the host-pathogen profile again, all your assumptions, uncertainties, models, things like that which are utilized to then feed both of these into the risk characterization. Next, please.
Ilsie was kind enough to prepare -- I don't know if we can get it all in there now, yes, as you can plainly see. I just wanted to bring up -- and I realize this is too busy and too small to see. But one of the things -- actually, this is the water risk assessment framework which we have now versus the ecological framework versus the old NASA chemical risk assessment approach, CODEX approach and then -- I'm not sure what this one is. Maybe somebody else here probably knows.
But anyway, as you can see, if you look at all the various phases, I mean, really they are all pretty similar. I mean, there are little nuances in terms of how they are implied. But really, the end product really pretty much is almost always the same. Next, please.
Well, anyway, I am not going to have really a chance to really go through these in any depth. But for the analysis phase, I might just -- what I am trying to do is be consistent with where CODEX is going as far as their classical definition. So I will just -- I won't have time really to go through it anymore.
But just as far as pathogen characterization to evaluate the characteristics of the pathogen, or in our case, surrogates, we oftentimes don't deal in water with the direct pathogen. We are typically using surrogates such as E. coli or something like that to really determine the effect of the ability for the transmission to have caused disease in the host and some of the criteria and things which are incorporated into that. Next, please.
And just continuing on with other things that are part of that exposure scenario. One of the things which we are really concerned about is the strain differences, especially with Cryptosporidium right now. We are -- obviously, there has been three studies now done on Cryptosporidium and oral dose response. That is EPA sponsored.
We have almost a 50-fold difference in the human dose response associated with that. So -- and those are just the animal strains. Those aren't even the human strains. Nobody has done the dose response for those.
Moving on to the pathogen and hazard occurrence, this is the frequency of the appearance of a pathogen or its relationship to the surrogates in the media of concern. Some of the things real quickly that I think are really important to us is that there is a very dynamic situation in most water supplies.
It is not a constant. You have very, very large orders of magnitude, shifts in what is present in the water supply which impacts on your treatment efficacy and things like that. So it is very important from the water standpoint.
One of the other things which, of course, with water is important is the fact that microbes and certain types at least of bacteria especially amplify in water. Others die off. There is persistence of some based upon various types of water characteristics, things of that nature.
In the exposure analysis, it is to characterize the source and temporal nature of the human exposure to water-borne pathogens. Obviously, we have got recreational drinking. We've got sewage, sludge, waste waters, re-use of water, things of that nature. We assume 100 mls for swimming-associated exposures. And now we have come up with a new exposure level for drinking water which is 1.2 liters rather than the old two.
Some of the other things. I won't go through that. Everybody -- it is just common to every risk assessment.
As far as characterizing the human health effects, we need to evaluate the ability of the pathogen, again, or the indicator relationship to cause an adverse health effect under the prescribed set of conditions we are dealing with and just some of the tools which we have available to identify those approaches.
The host characterization it to evaluate the characteristics of the potentially exposed population that influences susceptibility to a pathogen. And, again, some of the tools and things which need to be considered. Next, please.
Again, characteristics that influence those effects. Obviously, all the things that humans do to cause them to be exposed and the various things which influence their ability to become infected.
The health effects, the clinical manifestations of disease associated with specific pathogens, we have to consider both the acute gastrointestinal disease, chronic disease, and diseases that might impact on other organs of the body, especially through various types of sequelae. Next, please.
Dose response analysis, to characterize the relationship between pathogen dose, infectivity and the manifestation and the magnitude of the health effects in that population. We have the various tools, epidemiology studies, feeding studies and animal studies. We have some real concerns for most animal studies whether or not they can really prescribe the human condition.
I know I am involved in a work group of FDA to look at this. I think there is a lot of problems with using animal models. Next, please. That was -- oh, okay. Gee.
DR. STERNER: My apologies. I forget to start the timer. But it was about four minutes that elapsed. Are there questions for Dr. Shaub
MR. : Yes, I've got one, Keith.
DR. STERNER: Thanks.
MR. : I wonder if you have any occasion to apply any standard for introduction of a pathogen into water by any kind of industrial or community activity. I am just thinking here of an analogy in the drug situation where the issue is really kind of creating a different type of pathogen by some activity. I am just wondering if there is an analogy in the water area.
DR. SHAUB: Well, certainly, we are very concerned about biotechnology, industrial things. We are very concerned about emerging pathogens from whether they are coming from, you know, other countries or whether or not they are coming from our own modification of our procedures in terms of bioengineering, modification of genetics, things like that.
We, I think with CDC, are keeping a vigilance for these kinds of things. And certainly, we have what we call the contaminant candidate list which we have identified nine pathogens which we think have emerged or re-emerged which either because of their health effects or the fact that they are now being found in the United States in water supplies and that they have the potential to defeat our current water treatment distribution systems.
We are looking at those. So if we do, we are going through the process with each one, a risk-based process to look at the potential to be a problem. If they are a problem, we will actually establish new regulations based upon their likely concern on a national basis. Is that kind of where you were going
MR. : Just one follow-up, Keith, or -- is -- how about for current pathogens Let's say, for example, in industry something is going to get introduced into water, treated water into the water supply. Do you sort of exercise a log reduction standard or something of that sort for viruses or bacteria that are currently existing
DR. SHAUB: Well, yes, I -- what we try to do basically is if we know what the general source water occurrence is, basically our whole scheme is to define the treatment requirements that would reduce that down to a level where we would have no more than that one in 10,000 yearly risk of infection.
So, basically, the treatment level is going to be geared to the source water concentration levels. In other words, we have a 103 level of source water. And then maybe we only need to remove maybe two orders of magnitude of that to maybe be protected. If you have a 105 level of material in source water, then you would have to boost your treatment up two orders of magnitude to give that same level of protection.
DR. STERNER: In the years that I've had as a speaker, my worst nightmare is to wake up far past my appointed time to go ahead and speak. And I am at a point as a moderator of embarrassment in that we do not have Mike Bolger having shown up or nor do I have a bio. Mike, you wouldn't happen to be in the audience, would you Seeing no favorable response, we will move to the next speaker. And I guess that unfortunately will -- well, fortunately will keep us well on schedule here and on task.
Addressing pathogens on meat will be Kenneth Petersen. He is a Senior Epidemiologist with the Food Safety and Inspection Service, FSIS. And he will present the USDA activities regarding risk assessment. Kenneth.