ATDEPARTMENT OF HEALTH AND HUMAN SERVICES

 

FOOD AND DRUG ADMINISTRATION

 

CENTER FOR FOOD SAFETY AND APPLIED NUTRITION

 

 

 

 

 

 

 

 

 

 

 

FOOD ADVISORY COMMITTEE

 

METHYLMERCURY

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Wednesday July 24, 2002

 

8:30 a.m.

 

 

 

 

Sheraton College Park Hotel

4095 Powder Mill Road

Beltsville, Maryland

 

PARTICIPANTS

 

   Dr. Sanford Miller, Chairman

 

   Ms. Catherine DeRoever, Executive Secretary

 

   MEMBERS:

 

   Alex D.W. Acholonu, Ph.D.

   Francis Fredrick Busta, Ph.D.

   Annette Dickinson, Ph.D.

   Johanna Dwyer, Ph.D.

   Lawrence J. Fischer, Ph.D.

   Marion H. Fuller, D.V.M.

   Joseph H. Hotchkiss, Ph.D.

   Lawrence N. Kuzminski, Ph.D.

   Ken Lee, Ph.D.

   Thomas J. Montville, Ph.D.

   Robert M. Russell, M.D.

   Mr. Brandon Scholz

   Michael W. Shannon, M.D.

 

   TEMPORARY VOTING MEMBERS:

 

   H. Vas Aposhian, Ph.D.

   Sarah L. Friedman, M.D.

   Ms. Jean M. Halloran

   Margaret McBride, M.D.

   Richard E. Nordgren, M.D.

   Clifford Scherer, Ph.D.

 

   GUEST SPEAKERS:

 

   Dr. Elizabeth Southerland

   Dr. Henry Anderson

   Dr. John Middaugh

   Dr. Charles Lockwood

   Dr. Diane Zuckerman

   Mr. Richard Wiles

   Mr. Michael Bender

   Mr. Bob Collette


C O N T E N T S

 

Call to Order, Dr. Sanford Miller    4

 

Environmental Protection Agency,

   Dr. Elizabeth Southerland    4

 

Questions of Clarification    21

 

Alaska Advisory, Dr. John Middaugh    53

 

Questions of Clarification    69

 

American College of Obstetrics and Gynecology,

   Dr. Charles Lockwood    81

 

Questions of Clarification    85

 

National Center for Policy Research for Women and

   Families, Diana Zuckerman    103

 

Questions of Clarification    116

 

Environmental Working Group

 

   Mr. Richard Wiles    127

   Ms. Jane Houlihan    135

 

Questions of Clarification    163

 

Wisconsin Advisory, Dr. Henry Anderson    180

 

Questions of Clarification    201

 

The Mercury Policy Project, Mr. Michael Bender    211

 

Questions of Clarification     226

 

National Fisheries Institute, Mr. Bob Collette    236

 

Questions of Clarification     252


P R O C E E D I N G S

    DR. MILLER:  Let me welcome you to the second day.  We have a lot of speakers and we need to make certain that everybody has enough time for their presentation.  I will remind you when you have five minutes left for your presentation, and then when the time is up I will remind you of that and, if necessary, we will wrestle for the microphone.

    We are destroying more trees again this morning; the pile of papers that you have been given is increasing logorithmically.  In view of the discussion we had yesterday about what was a tile fish and what was a mackerel, and so on, we have some pictures of both fish so we know what we are talking about.  We don't have quite enough copies for everybody and you will have to share them, but I think that will answer the question, yes, Virginia, there is a tile fish.

    Our first speaker this morning is from the EPA, Dr. Elizabeth Southerland.  Dr. Southerland?

Environmental Protection Agency

    [Slide]

    DR. SOUTHERLAND:  In the Office of Water at EPA we have a program that gives technical assistance to state and tribal health departments to help them put together fish consumption advisories, if they choose to do so.

    [Slide]

    Our program provides technical assistance, and we do a number of things.  First of all, we have a national guidance that talks about what types of species you should sample; what analytical methods are available to give low detection limits in those species that are sampled.  We have a risk assessment document that talks about once you get those concentrations in the fish, how would you calculate the number of meals that you want to recommend a person make of those species.  Then, we have a risk management guidance and we also have a risk communication guidance, and I will talk about that a little bit later.

    We also have a national database.  Since 1993 the states have voluntarily been giving us each year data of their fish consumption advisory.  So, if you go on our web site you can see all the fish consumption advisories that consist of state health protection throughout the United States and that, again, has been occurring since 1993.

    We also have national conferences and workshops.  Every year the states meet with us in a fish forum.  Generally we do this with the American Fishery Society.  This year it will be in October, in Burlington, Vermont when the states meet with us.  We will be talking this year about emerging contaminants, some new pollutants that have started to show up in fish, and we will also be talking, of course, as always about the benefits of eating fish.

    We also have grants for sampling and analysis.  In some cases a state will have a suspicion that there is some contamination in their fish.  They just need a little bit of seed money from us to actually go out and measure to see if, in fact, there is a contamination problem.  Whenever we can, we try to provide grants to states for them to do that.

    We also do special studies.  We have been working on a random stratified sampling of all the lakes in the country.  It is a big four-year study, multi-million dollar, in which we are randomly sampling lakes of all different sizes around the country and measuring them for over a hundred different contaminants.  It is not just the old banned chemicals that we have already been concerned about, but it is a lot of new chemicals, new pesticides that are currently being used and we want to check and see if there are some emerging pollutants that currently we are not sampling for on a regular basis.  So, that is one of the special studies we are doing.

    Finally, a number of times states, particularly in inter-state waters where there is a disagreement between states sharing a body of water over what kind of fish consumption advice they want to handle, they will often call us and we will provide technical assistance to the involved states.

    [Slide]

    I have done this just for lake acres.  A similar pattern, however, would be shown for rivers.  Across the X axis is the number of lake acres under advisory, and this is in millions of acres.  Then, along the Y axis we have the five most frequently detected pollutants.  Again, if you look at our web site you will see that state health departments have set fish consumption advisories for 39 different pollutants.  These, however, are the five most frequent.  Dioxin doesn't show on the graph.  There are 75 fish consumption advisories in the country for dioxin.  It is just that in terms of millions of lake acres it doesn't quite show on the scale.

    We have been measuring these advisories since 1993.  The states have been giving us this data.  So, you will see that the real trend in terms of increase in advisories is in mercury.  Again, we don't think that that is an indication that there is some new contamination source of mercury, it is that the states have more and more over the years begun to monitor their fish for mercury concentrations.  So, that is why you will see that that is now the most prevalent cause of fish consumption advisories put out by the state health departments, and it has been increasing, much more so than the other pollutants which have pretty much stayed about the same.

    [Slide]

    This is a really busy slide and I ask you to please look at the handouts.  We have them at the front table, if you haven't already picked one up.  This map shows several things and I will try to walk you through it and then we can talk more about it in the questions and answers, if you want.

    The states that are in white, they are ones that do not have any fish consumption advisories at all.  It is mostly in the West and Alaska and Hawaii.  The ones that are in pink, varying shades of pink, or striped pink, or red are states that do have fish consumption advisories.  They may have statewide advisories that say all the freshwater fish in this country are under some kind of mercury advisory, or they may point out individual water bodies for fish consumption advisories.  It is not necessarily all species of fish; it may be selected species of fish.  It may not be all waters; it may be certain waters.  So, it varies state by state but we are trying to show this to you on a national basis.

    The reason you have that blue line around the southeastern Atlantic coast and all of the Gulf of Mexico is that the states that border that coastline all have coastline advisories for mercury for a number of species of fish.  So, that is why you will see the blue there.

    The other thing of interest here is that there is a little square in 11 of these states, and many of them are on the northeastern side so you can't even see it in the little tiny state itself but it will be written next to the state name.  But it is a total of 11 states which have given their public advice on commercial species as well as on recreational species.  The reason they did this is because they realize people do not only eat fish that they catch themselves, but they also go to the store and to the restaurant and they eat commercial fish too.  So, these 11 states are giving advice to their public on eating commercial species of fish as well as recreational species.  Nine of the 11 states that issue this commercial advice include tuna in their recommendations because so much of the diet that many people eat of fish are tuna fish sandwiches.

    We used these states as an example in trying to do our national recreational fish advisory.  So, when I talk about that at the end of my presentation you will see that we included the FDA advice in with our recreational fish advisory to try to keep down confusion where we are seeing some difference for recreational fish than FDA is saying for commercial fish.  So, we use the example of these 11 states as a model to follow, and I will discuss that at the end.

    [Slide]

    I know we went over this ad infinitum yesterday, but the reason I am repeating it now is as a reminder of what we use.  EPA uses the RfD that the National Academy of Sciences recommended in July of 2000, and that is the 0.1 ug/kg body weight per day.  ATSDR has a different number and FDA has yet again a different number.  The reason I am going back through this is for the next slide.

    [Slide]

    If you look at the RfD that the states use for setting their mercury fish consumption advisories, it varies by the state.  Again, if you look at the top graph, I think it adds up to 38 or 39 states that give advice that they call out just for adults.  That means they are not differentiating women from men; they are just saying adults.  Of those, 25 of the states that give advice for adults will use the 0.3, which is the ATSDR number.  Seven of them use the National Academy of Sciences EPA number of 0.2 and then there are smaller numbers using RfDs on either side of the scale.

    If you drop down to the second graph, this is the number of states that are giving specific advice for women and children.  These states generally, 23 of them, use the 0.1 or the EPA RfD.  There are four of them that use the ATSDR number and then a few that use other numbers on here.

    So, when you drop down to women and children, it is generally the 0.1 that is being used to give their fish consumption advice.  It turns out that there is a total of 16 states who give both kinds of advice and use different RfDs depending on what public they are trying to look at.  So, we call that a two-tiered advisory.  What those 16 states do, they use the 0.3 RfD to give advice for women who do not wish to have children or are too old to and for men.  That would be the adults.  Then, for women of childbearing age and children they use the 0.1.

    So, that is generally what we would call two-tiered advice.  All the other states that give advice use a consistent RfD for adults versus women and children.  There is a lot of information there and, again, it is state by state so we can talk about that in questions and answers also.

    [Slide]

    When I talked about the national guidance that EPA puts out, again, we update this guidance about every two years just to keep current with the science.  Again, the sampling information will talk about what species and what kind of analytical methods you should use to get as low as possible detection limits.  Our risk assessment guidance is the type of equation that we would follow and that we use for our national recreational advisory.  It is basically set up to calculate allowable meals or recommended meals based on the fish concentration that you have.

    Again, we have risk management and risk communication guidance. The risk communication guidance that we have has some really innovative things that several of the states have come up with on getting the word out very effectively, particularly to sensitive subpopulations who may not be English speaking, who may be low income people, who are not going to be able to use the same kinds of communication techniques that we do where you are looking in a big, giant booklet with 700 pages of recommended fish consumption advice.  So, we have some really neat things in volume four that the states have come up with on how to communicate effectively.

    [Slide]

    Of course, we use our own equation to come up with our national recreational fish consumption advisory.  This is the one that we use.  Of course, we use the NAS recommended RfD of 0.1 ug/kg body weight per day.  We assume the body weight of 65 kg, looking at that as a woman's body weight.  We looked at a meal size as 8 oz uncooked.  Then, the concentrations of mercury that we looked at before we put together our advisory were all the ones that had been submitted to us by the states for our national listing of fish and wildlife advisories.  That is what that NLFWA stands for.  I am going to show you a chart that gives you all that concentration information in just a second.

    Basically, what we do is we take the maximum daily fish consumption rate, equal to the RfD times the body weight, divided by the concentration in the fish, and what I will show you next is a slide that shows the concentration of the fish and what that equates to in terms of a daily fish concentration rate.

    [Slide]

    When we look at this table, and that is using that calculation, if your fish concentration is 0.1 ug/kg, then you could eat nine meals per month.  If it is 0.2, you could eat 4.5 meals per month, and it goes on down.  This is generally how states will set up their fish consumption advisory.  They generally do not say don't eat any fish at all; the consumption rate is zero.  Instead, what they do is, based on the data they have about the concentration of their fish, they recommend a consumption rate because they want people to eat fish; they want to keep the fishery open as much as possible.  So, they will always produce their advisory in the form of what they recommend for fish consumption of specified species that they have concentration data on.

    [Slide]

    If you look at this, the 4.5 meals per month comes right in at about 0.2 and you will see how we came up with our concentration rate to fit our national recreational advisory for the country.  What we have here is a whole huge set of data that the states have given us of mercury concentrations they have monitored in their fish over the years.

    I am going to get to the statistics now.  We had 66,000 samples from 8,000 stations, and that was provided by 44 states to make up this analysis that we did for the national recreational advisory.  Alaska and five other western states did not provide any data at all.  So, these 44 states that we have are missing information from Alaska or for those five western states.

    However, if you look at the N over here, the sample size number, generally we have at least a hundred or more for each species, and in some cases thousands of samples.  We have the mean mercury concentrations if you go along the bottom axis, the X axis.  If you come up at the one meal per week number, it would be 0.16 ppm.  As you can see, that is protective for most of the species.  It looks like it is over-protective for those species at the bottom of the graph where the line is to the right of those concentration numbers.  However, if you look at those species, and we also did this analysis, they are generally ones that also have high PCB constituents.

    So here we go again, this is a mercury advisory, however, at the same time we don't want to focus only on mercury and then recommend that people eat lots of fish that may be high in PCBs.  So, what we decided to do, and this was a judgment call, is to call it at one meal per week because that would be protective for most species for mercury and it would not be, we felt, too over-protective for the species that were kind of low in mercury because in the 44 states that gave us data on those fish, they were high in PCBs.  So, again, a judgment call--we came up with one meal per week.

    [Slide]

    Here is the test of what we said.  First of all, we directed this not just at women of childbearing age but also nursing mothers and also young children.  We recommended one meal per week for untested waters.  This would be waters in which no one has any idea what the fish concentrations are.  From our national database of mean mercury concentrations we wanted to give some kind of rule of thumb to the public when there was no testing information available; no advice available from their state health department or tribal health department.  So, in that case, for that untested water we are recommending one meal per week.

    Because of the confusion over FDA also giving advice at the same time, and we did coordinate very closely on the release of our information with FDA's, we wanted to make sure and recognize this.  So, what we said is for commercial fish we are recommending that you follow the FDA advice, and then we go on and say for the highlighted area here, in yellow, therefore, if in a given week you eat 12 oz of cooked fish from a store or restaurant--the commercial fish covered by the FDA advice--then do not eat fish caught by your family or friends that week.

    So, again, we tried to follow the model of those 11 states that currently give commercial and recreational advice.  We recognize that people will eat a mix of fish generally in their diet.  So, if you are eating the full amount that FDA recommends for commercial fish, we are recommending that that week you do not eat your own fresh caught fish.

    DR. MILLER:  Dr. Southerland, five minutes.

    DR. SOUTHERLAND:  Thank you, I am almost done.

    [Slide]

    We also went on to say EPA recommends that women who are or could become pregnant, nursing mothers and young children follow the FDA advice for coastal and ocean fish caught by family and friends.  Again, our advisory was for lakes and rivers.  Generally, the commercial fish cover the marine types of fish.  So, we are also recommending that they follow FDA advice if they are catching their own ocean and coastal fish.  Then we go on to talk about other sources of protein.

    [Slide]

    I just wanted to throw this in because we were excited to have the NHANES data also.  We did a little different analysis than what was presented yesterday, and I am sure all of us will be doing different analyses over time.  But when we took the NHANES data, of course, we wanted to look at the split out if you followed our advice, if you had just one fish meal per week.

    So, what we did is look at the total mercury blood levels, along the X axis, for those people who ate more than one per week or those people who ate less than one per week.  What we saw for those people who eat one or more meal per week is the blood levels at or above 5.8 ug/L, which corresponds to the NAS EPA RfD of 0.1, and 15.3 percent of them were at or above the 5.8 ug/L.  For those who ate less than one meal per week, it was 1.9 percent at or above 5.8.

    So, when we look at this data, and everybody has their own interpretation as we heard yesterday, we feel like we have come up with a good recommendation for people to keep those blood levels at a reasonable level of one meal per week for waters that are untested and you have no idea what the fish contamination is like in those waters.

    [Slide]

    We have done a lot of outreach on our advisory.  Our advisory, remember, includes this connection to the FDA advice.  We have worked with ATSDR to distribute this to pediatricians and obstetricians throughout the country.  The 12 states that have statewide advisories, and that includes Alaska who says eat as much as you want, we have not sent this information to because we did not want to confuse the public.  Those states that have their own statewide advice are free to give their own advice, but to those states that do not cover all their waters and have a lot of untested waters, we have sent this information out.

    We have advisory brochures now translated into seven different languages.  Those are more generic advisories, not just for mercury, on how to trim and clean the fish to minimize your exposure to contaminants.  Then, we have participated in many medical conferences.  We even go to the midwives conferences, as well as pediatricians and obstetricians to make available our information.  That is it.

Questions of Clarification

    DR. MILLER:  Thank you.  Questions or comments?  Dr. Russell?

    DR. RUSSELL:  Thank you, that was very clear.  There is some confusion about serving size and how 8 oz was picked.  Could you clarify that for us?  There was some data shown yesterday where the serving size that a woman actually eats is more like 2.6 oz.

    DR. SOUTHERLAND:  Yes, the 8 oz came from what generally the state health departments had asked us to use as a recommended meal size.  Jeff, do you have any more detail?  We do all of our advice in conjunction with the states because ours is not a regulatory program; it is only a technical assistance program.  The states have generally used 8 oz uncooked, which works out to 6 oz cooked for an adult.

    DR. RUSSELL:  They may use it, but I am wondering does anybody know the origin of that.

    DR. HOTCHKISS:  Joe Hotchkiss.  I am just curious, to your knowledge, do all states or localities that have advisories say something to the effect that if you are pregnant or could become pregnant--

    DR. SOUTHERLAND:  Of childbearing age, yes, generally.  Jeff, is there any exception to that?

    DR. BIGEL:  [Not at microphone; inaudible]

    DR. SOUTHERLAND:  His question was is there any exception to states?  Do some states not say women of childbearing age?  Do they say only pregnant women or something more restrictive?

    DR. BIGEL:  [Not at microphone; inaudible]

    DR. SOUTHERLAND:  But generally when they say women, they say women and children together.

    DR. HOTCHKISS:  Thank you.

    DR. APOSHIAN:  I have two questions, one very short one.  Does the EPA have data going back, say, 25 years on fish from a given lake and whether the amount of mercury contamination is increasing?

    DR. SOUTHERLAND:  Actually, the oldest data would be from the '80's.  Jeff, is there any analysis that shows that it is increasing or staying the same?

    DR. BIGEL:  I am not familiar with that analysis.  [Not at microphone; inaudible]

    PARTICIPANT:  I am not aware of data directly in fish, but there are studies that are done [not at microphone; inaudible] and that shows low levels of industrial leaching [inaudible] and then sort of leveling off [inaudible] but there has been about 80 percent reduction in use of mercury in the economy.  So, we expect to see some reduction [inaudible].

    DR. SOUTHERLAND:  In our data set, which is just late '80's and '90's, we are not seeing a change but that is not really long enough to see much of a change.

    DR. APOSHIAN:  The other question I have, perhaps you or someone might comment about it, I heard this morning that--and I want to be certain that it is correct and that is why I am asking for comment--that in those fish that have a high amount of mercury, as I understand it, and it is very important that it is clarified, those fish have a lower amount of essential fatty acids; that there is an inverse proportion.  Now, we have Dr. Mahaffey here who is an icon as far as this sort of thing is concerned.  She was very much involved in NHANES.

    DR. SOUTHERLAND:  I am going to have to call on Kate for the fatty acid issue.

    MS. DEROEVER:  Excuse me, could the speakers in the back please come to the microphone and introduce themselves so we have it on the record?

    DR. MAHAFFEY:  I am Kate Mahaffey, from U.S. EPA.  One of the things I have done recently is to sit down and look at some of the data on concentration of mercury in fish and the concentration of a couple of the essential fatty acids in the fish because the essential fatty acids are frequently cited as one of the major reasons for achieving benefits of fish consumption.  There are data that show that certain fatty acid in fish are important for the neurological development of the central nervous system of the fetus.  The abbreviation is DHA, and it is docosahexaenoic and it goes on from there.

    I looked at the species of fish that are the highest in mercury concentrations, the ones where the advisories are, which include tile fish, shark, king mackerel, swordfish.  They are not particularly high in this DHA fatty acid that is essential for neurological development in the fetus.  What the essential fatty acids seem most closely tied to is the percent fat in the fish which, of course, makes sense.  So, you are not really having a tradeoff between how much of the essential fatty acid you get.  In other words, it is not a one to one correlation.  You can select fish that are relatively high in essential fatty acids, things such as salmon, things such as some of the mackerels, and are comparatively low in mercury.

    On the other hand, if you look at swordfish, tile fish, king mackerel and shark, they are relatively more lean fish and are comparatively low in essential fatty acids.  So, it is not as though you give up the nutritional value.  You simply have to exercise wise choices in the kinds of fish you select.

    I have not done this for another fatty acid that has an interesting acronym EPA, not us, because the nutritional content of fish is cited as a benefit in terms of coronary heart disease, yet, there are some interesting data out of Europe, specifically out of Finland and also a multicenter trial, longitudinal cohort studies, and for some reason they measured mercury.  I honestly have no idea why they measured mercury in people's hair or people's nails, but what they found is that when the person's mercury exposure had been relatively higher, and in the Finnish cohort the demarcation was two or more parts per million in hair, the risk of coronary heart disease and the risk of deposition of fats in the carotid artery, and they imaged the carotids, was higher.  So, again, what is going on at least in this Finnish study is that the higher exposures to mercury seem to attenuate some of the benefits of fatty acids and the nutritional benefits of fish.

    Again, it is one of these deals where, depending on the fish that is chosen, you get more or less benefit of the diet.  People have said the Finnish study stands alone; we shouldn't ignore all the other studies that show benefits.  The Finnish study appears not to stand alone because I am now told that there are reports coming out from a multicenter trial in Europe that are showing a parallel kind of finding.

    You can't ignore decades of advice on nutrition, but I think we can expand that advice in a way that gives people the benefits of fish without necessarily the higher exposures to mercury.  So, for both the essential fatty acid that is important for CNS development in the fetus and also the risks of coronary heart disease it appears that you do not have to give up the nutritional benefits of fish in a tradeoff for mercury because it is not any sort of one to one correlation.  It has as much to do with how fat the fish are in terms of the percent lipid in body composition as anything else.

    DR. MILLER:  Dr. Lee?

    DR. LEE:  Basically, I have the same question about mercury over time in fish.  So, just to clarify that, one comment about the total mercury--I take it total mercury is not methylmercury that we are talking about over time.

    DR. SOUTHERLAND:  Oh, no, it is methylmercury that is measured in the fish.

    DR. LEE:  But in environmental exposure that increased and has now plateau'd.

    DR. SOUTHERLAND:  I have to refer to Arnie on the sediment core data.  Arnie?

    DR. KUZMACK:  This is Arnold Kuzmack, EPA.  That is total deposition of mercury of all sorts, mostly ionic mercury that is deposited.  That is the source of the mercury that gets methylated in the aquatic environment and accumulates in the fish.  Most of the methylation, of course, occurs in the top layer of the sediment at the bottom.

    DR. LEE:  And you are saying that is in the lake beds?  That kind of work has also been done on Antarctic cores?

    DR. KUZMACK:  What kind of cores?

    DR. LEE:  In the ice in Antarctica.

    DR. KUZMACK:  Yes, there is some work on ice cores as well which I think shows a similar pattern.  Lake core is done in various locations and typically shows peak levels, say, two to five times the preindustrial levels and what that is sort of depends on where you are located.  If you are near industrial areas it is likely to be a higher ratio.

    DR. LEE:  Thank you.

    DR. MILLER:  Dr. Fischer?

    DR. FISCHER:  I would like to ask whether the EPA recommends to the states the use of the EPA RfD for everyone--women, children and other adults, or do they recommend a different set of restrictions for those two groups, adults versus women and children?

    DR. SOUTHERLAND:  Right now we just have the 0.1 on the books for developmental effects.  Because it is listed for developmental effects we are recommending it for women and children, but we have not taken a position on this two-tiered approach that some of the states have done.  We do not have an IRIS value right now.  It varies for the general population.

    DR. FISCHER:  Why is that, that you don't do that?

    DR. SOUTHERLAND:  Kate, I would have to ask you, the IRIS program is our official?

    DR. MAHAFFEY:  Right.  The way EPA's reference does this work is that they are set for the most sensitive subpopulation.  It is not the most sensitive member of the subpopulation but the most sensitive subpopulation.  Since most of what we deal with does not know boundaries in the sense that if you deal with an air contaminant or water contaminant you can't very well separate out the exposures for men, for women or for children, the underlying philosophy has been that you work to protect the most sensitive subpopulation.  While this approach of a two-tiered advisory may be effective when you are dealing with a limited distribution or self- or family-caught fish, it is harder to enforce who eats the fish if it is winding up in commercial sales.

    DR. FISCHER:  You know, it seems to me very hard for you to convince people that fish is good for them and that there is benefit of eating fish if you are regulating in that way.  In other words, why are you restricting the consumption of fish to men of my age when, in fact, we aren't the most susceptible?

    DR. MAHAFFEY:  Well, until we know more about the cardiovascular risk, I don't know that I am ready to go there.

    DR. FISCHER:  I mean, I can't believe that you don't believe the immense literature out there on the benefits of fish consumption.

    DR. MAHAFFEY:  If you look closely at the studies, they are not unequivocally supportive of the benefits.  Some of those studies are mixed, and it is quite possible to choose kinds of fish that result in low mercury.  As you can see with the NHANES data that we showed, while the percent of people who eat fish one or more times a week is a lot higher, in the 6 and above blood level, 85 percent of those people were able to select fish that are comparatively low in mercury and have blood values under the reference dose.  So, it is more to do with the kind of fish selected than simply fish consumption.  It is as much which fish rather than fish or not.

    DR. FISCHER:  I understand that, but you can see the number of states who disagree with you here.  Look at the number of states who have the two-tiered approach and are using it.

    DR. SOUTHERLAND:  There are 16 states that have the two-tiered approach.  It is a little confusing from the way I had to present the graph but 16 states do both.  The other states, and there is a total of 43 that give advice on mercury, will use a consistent RfD.  So, this is a new thing that states are starting to work on, the two-tiered approach.

    I am going to have to make a point of clarification here, though I don't want to interrupt the conversation between you and Kate, but we have no regulatory authority.  EPA has no regulatory authority.  All we do is talk to the states about advice, and the states have no regulatory authority.  When they give advice on recommended meals, they are not seizing those fish out of people's hands and taking them to jail, or anything.  They are just giving advice; we are just giving advice.  There is no regulatory authority in EPA.  FDA is the only agency that has that.

    DR. MILLER:  Dr. Russell?

    DR. RUSSELL:  There is a lot of interest now in DHA and EPA with regard to membrane stabilizing effects and sudden death, prevention of sudden death.  Is there any data that high levels of mercury cancel out that effect?  Because this is totally different from coronary-artery disease.

    DR. MAHAFFEY:  I am assuming you are addressing that to me.  As I recall the Finnish study, there was a greater incidence of mortality in the people that had the higher mercury levels.  As you say, it gets into the underlying mechanism of oxidation and heavy metals such as mercury to promote oxidation.

    DR. RUSSELL:  Well, these fatty acids specifically stabilize membranes.  It is a physical-chemical thing.

    DR. MAHAFFEY:  Right, but it is a highly unsaturated fatty acid.

    DR. RUSSELL:  Yes, but it may not just be from oxidation.

    DR. MAHAFFEY:  Exactly.  Again, I am recalling this data from memory.  I would be pleased to follow-up and provide the paper to you.

    DR. RUSSELL:  Thank you.

    DR. MILLER:  Dr. Aposhian?

    DR. APOSHIAN:  I think it might also be a good idea, since we have Kate here because I have learned some things this morning that are very educational to me anyway, as you know there have been questions about the 60,000 children at risk that the National Academy of Sciences report pointed out.  Yesterday some people questioned this as being too high.  I believe Kate has some data that she might want to share with us, suggesting that the 60,000 National Academy of Science figure is too low.

    DR. MAHAFFEY:  Again, this is from memory.  I would be glad to go back and get the specific numbers, but as I am recalling this, there are about four million births in the U.S. per year.  If you take the NHANES value, the 99 alone showed about 10 percent of women with blood mercury of 5.8 and greater, the combined 99 2000, the number turned out to be, I think, about 7.8.  So, somewhere circa 8 percent.  If you take 8 percent of one million, you are coming out with about, I would think, 320,000.  If you take the 10 percent, it is 400,000 newborns a year.  If you apply the NHANES data and the number of births, that would be the estimate of the number of infants born each year where you would expect to see their initial blood mercury higher than the value that EPA believes to be safe.  So, we don't think the 60,000 is too high.  If anything, the data suggest it is comparatively low.  I am sorry, 10 percent of 4 million would be 400,000; 8 percent would be about 320,000.

    DR. DICKINSON:  [Not at microphone; inaudible]

    MS. DEROEVER:  Dr. Dickinson, would you please use the microphone?

    DR. MAHAFFEY:  I see what you are doing, 60 million women of childbearing age approximately, the data show about 9 percent in that age group in a given year are pregnant.  There is another number, I think it is 6.5 pregnancies per 1000 women.  You know, we went through the math.

    DR. FISCHER:  I would like to ask you to give us the calculation that you are speaking of--

    DR. MAHAFFEY:  Sure, that is fine.

    DR. FISCHER:  --so that we have it to look at.

    DR. MAHAFFEY:  No problem.

    DR. MILLER:  Dr. Scherer?

    DR. SCHERER:  Yes, Cliff Scherer.  I wanted to ask a question about the extent to which we have any information about the effectiveness of advisories.  Do you know about to what extent states or how people are following advisories?

    DR. SOUTHERLAND:  There have been some individual studies of that.  Actually, we have been working with Cornell University and some others to do effectiveness measures.  That risk commercial guidance document that we have actually has the best literature review of that type of effectiveness data.  What they have generally concluded is if you are talking about educated, middle class people who are doing it for recreational purposes, those books that the states will give you when you buy your fishing license are fine.  But when you are talking about people that have English as a second language or who are not buying fishing licenses, then that is obviously totally ineffectual.  What they have found is that things like posters--certain types of cultures react very well actually to comic book style posters.  They also have posting in different languages that can be effective, and also big press events.  Each year when a state updates their advisory, if they do a lot of press work.  ORSANCO has been doing a lot of the effectiveness studies because they do the fish advisory publicity for all of the Ohio river basin.  So, anyway, we have some good information on what works with what populations.

    DR. SCHERER:  Do we know anything about the percentage of people that are paying attention to those kinds of messages?

    DR. SOUTHERLAND:  I don't know if we have any percentage information.  I know in the Great Lakes, for example, they have done a number of studies and in the Great Lakes, because it is such a high group that are eating fish, it is very well publicized.  Also, in the Ohio River Basin I think they are getting fairly high effectiveness levels.  In other parts of the country, not at all because, again, it is the level of publicity and the type of publicity.

    That is why when we have sent our posters out, we send them to the pediatrician and obstetrician offices for them to post right there.  The women and children health clinics also like to get that kind of poster effect as opposed to a 700-page textbook.

    DR. MAHAFFEY:  Michael Bolger told me that the calculation, I guess, shows around 350,000 to 400,000 is in your books.

    DR. BOLGER:  If you look at your figure, you will see it gives you the number of women where it says 7 percent, but it is actually more like 8 percent based on data from Susan Schober.  So, the number of women on annual basis who are pregnant is about 276,000 women.  That is on an annual basis, the number of women who are pregnant who exceed the reference dose is about 276,000.

    DR. DICKINSON:  This is Annette Dickinson.  We had some discussion yesterday about the fact that that includes the ten-fold safety factor.

    DR. BOLGER:  That is correct.

    DR. DICKINSON:  So, given the fact that they exceed the RfD with a ten-fold safety factor and the effects on which the whole bench mark dose is based are effects in the most sensitive population, what is the implication of that, that they exceed the RfD?  Does it just mean their safety factor is less, or does it mean they are actually at risk?

    DR. BOLGER:  This is what I am going to talk about tomorrow.  What implications you draw really are some of the things I am going to try to address tomorrow so I would hate to get ahead of myself today.  But it is a very good question.

    DR. DICKINSON:  But we are thinking about it today.

    DR. BOLGER:  I understand.  The margin of safety issue is what you are getting at.

    DR. MILLER:  Ms. Halloran?

    MS. HALLORAN:  To go back to the point raised just a minute ago about consumer awareness, the Northeast States for Coordinated Areas Management, which is an association of air pollution agencies in the east, reported in May, 1999 in a survey that they apparently did that of about 75 percent of their respondents who eat fish on a regular basis, about half said they knew about advisories issued by states or FDA and one-third said they knew what they meant.

    DR. MILLER:  It is actually the question that I had, what data was there that looked at understandability?  I mean, reading that advisory, the combined FDA-EPA advisory I found it totally confusing.  Now, I may not be as clever about this as people who live near lakes, but I found that totally confusing.  There are three messages, all in one document.  I was just curious as to what kind of research was done.

    DR. SOUTHERLAND:  We have not done any consumer research on our advisory because it is so new, but we do know that the 11 states that are currently giving both commercial and recreational fish advice have dealt with this issue before of how to do the tradeoffs between what you get from a store or restaurant and what you get caught by yourself.

    So, we just followed their model in putting out our advice because, otherwise, it looked like the two were totally unrelated.  We were saying one meal per week; FDA was saying 12 oz.  They are entirely different species involved.

    DR. MILLER:  Mr. Scholz?

    MR. SCHOLZ:  Brandon Scholz.  I wanted to follow-up on a point that you made on EPA's outreach.  You said it appears that most of your outreach is to healthcare professionals who deal with pregnant women.

    DR. SOUTHERLAND:  And fishery groups, like the American Fishery Society.

    MR. SCHOLZ:  Do you do any other outreach to retail or to restaurants?  Any other distribution of your materials?

    DR. SOUTHERLAND:  No, ours is strictly directed at recreational fishing groups, like Bass Masters and American Fishery Society, as well as health groups.  The lead-off statement for us is fish caught by your family and friends.  So, we do not cross over into the commercial fish advice at all.  If anyone ever asks about commercial fishery we cite the FDA advice.  Again, EPA has no regulatory authority for fish consumption advisories.  We are strictly for technical assistance to the state health departments.

    DR. MILLER:  Some of you may have wondered why I have allowed this to continue on, not exactly my usual style for these kinds of things.  Not only is it an important issue, but our next speaker is not here.

    [Laughter]

    I just don't want you to get the wrong idea.  Johanna?

    DR. DWYER:  I wanted to agree with Dr. Miller that I found the advice a little confusing, and I kept coming back to the thought that for fish there are no recreational uses if you are fishing and you are the fish.

    [Laughter]

    I am also taken by some of the problems we hear on the dietary guidelines committee on the alcohol recommendations.  You know, if you think about it, most of the problems with alcohol seem to come from the ethanol.  Simplifying the advice to something where you can focus on that with a very, very simple message I think has gotten through to a lot of people, whereas 25 or 30 years ago it didn't.

    I guess what I am struggling with, and perhaps Dr. Miller is as well, for those of us who have to give a five-second sound byte in a clinic to a patient, I really need something that is a sentence or maybe two sentences.

    DR. SOUTHERLAND:  I think the way we have tried to hone in on it, and I hate to say it because, again, the focus groups have struggled with this, but it is the species type.  We deal with freshwater fish, fish that you would catch from a river or a lake.  The marine species are generally the predominant commercial species.  Now, what is the public's understanding when they see a species name as to whether it is freshwater or saltwater?  That is where the confusion is coming in I believe.

    EPA's original recommendation for FDA was to try to do lists of species and, apparently, the focus groups just found that too confusing.  Because if we had done a list of species, then we could have had unified advice from both EPA and FDA.  I believe we could have worked that out but it was just too much detail.

    MS. HALLORAN:  I hope I understand this correctly, the whole origin of the problem of EPA and FDA having to give different advice doesn't really come from a separate evaluation of the safety of freshwater fish and ocean fish.  Ocean fish is not safer than freshwater fish, if I am correct.  It is that FDA has made basically different judgments in the risk analysis.  Is that correct?

    DR. SOUTHERLAND:  And it is also the concentration of the species.  I gave you the data that we have, and we have, again, 66,000 samples from 44 states and we are looking at our concentrations.  I believe FDA is a little bit inhibited as they don't have as up to date data, and I don't know that they have as much data on their marine and coastal species as we do for our lake and river species.  So, they were also looking at the concentrations that they had in their database.

    DR. MILLER:  It seems to me, in response to that, listening to this discussion, it is not only that issue.  That is one issue.  The other issue is the fact that this increases the number of species that people have to worry about, and that is the problem with people saying they get confused when they see these lists even if they are identical.  Indeed, if FDA lowers its action level, if you will, its advisory level, that species list would increase even more.  The question from a procedural point of view is how do you give this advice to people in a way that they will use it and can use it?  As far as I can hear, that problem has not been resolved.

    DR. HOTCHKISS:  I want to make sure of your last statement.  FDA is using an ADI of 0.4; you are using an RfD of 0.1.  That is a four-fold difference.  If you run through the calculations, that seems to me to be a major difference in the two agencies' recommendations.  Granted, you have different databases and so forth, but the major difference is simply that either acceptable daily intake or reference dose, or virtually safe dose or whatever you want to call it, is a four-fold difference between the agencies.  Is that correct?

    DR. SOUTHERLAND:  Yes, that is correct, but they also did look at their concentration ranges for saltwater fish compared to the concentration ranges for what we have in freshwater.  At least in the lower 48 states our freshwater fish can be much more contaminated in some cases, particularly certain water bodies that have higher levels of mercury sometimes.  Again, we were working with a richer database and also concentration information.  But, yes, you are right, it is a combination of concentration and RfD difference.

    DR. MILLER:  Johanna?

    DR. DWYER:  I was very much surprised and pleased by the enormous number of chemical analyses that you have done on the mercury concentration of selected fish.  I think it was in our handout.  Does that go in any databases that are available on computer programs?

    DR. SOUTHERLAND:  Yes, those are all the data that have been submitted to us by the states that do this monitoring and it is all on our web site.  We have it all.

    DR. DWYER:  No, I meant databases like the kinds of things people put on a laptop computer and dieticians use for example.

    DR. SOUTHERLAND:  We haven't provided to those groups but they are readily available on our web site.  Again, it is all voluntary.  The states don't have to give us this information at all, but since 1993 most of the states have been giving us tons of information, not only information on their advisory but on the fish concentrations that they have monitored.

    DR. DWYER:  Is it collected in a random--how is it done?  How do you collect the data on these fish, and does it go into the standard reference database of the food composition for U.S.A.?

    DR. SOUTHERLAND:  To my knowledge, it only comes to us on a voluntary basis.  Each state has their own monitoring plan.  Some have a rotating basin approach and for each year they try to go to another basin and they measure fish tissue.  Others have a regular, you know, station that they monitor each year.  It varies by the state and we take all the data they give us because it has all been through state QA/QC and their laboratories, and they are using it to make their fish consumption advisory decisions so it is good enough for us.

    DR. MILLER:  Mr. Scholz?

    MR. SCHOLZ:  I would like to ask one more question.  You had mentioned that there is a Cornell study in place, or it has been done, gauging the effectiveness of the advisories?

    DR. SOUTHERLAND:  Cornell has an investigator up there that we have worked with that has done these effectiveness studies.  She has worked in the Great Lakes area and she has also worked in the Ohio River Basin.

    DR. MAHAFFEY:  There is also additional work in the State of Maine on interpretation of the advisories and risk commercial evaluation process understanding.  So, it is not as though this is a totally untapped area.

    MR. SCHOLZ:  No, I understand that.  I was just curious, is that a study that we can get?  I mean, is that available to us?  My question was is it ongoing or is it done?

    DR. SOUTHERLAND:  She has some that are ongoing; she has some that are under way.  So, I can get those to you, sure.

    MR. SCHOLZ:  It would be interesting because generally coming from the Great Lakes and the Wisconsin area, you know, there is the annual story in the newspaper at the beginning of the year that says, in short, all fish have mercury; don't eat it.  So, I would be curious what the study says because it doesn't seem that the way the press reports it is fair because it doesn't necessarily differentiate which fish, what level, for whom and, unfortunately, it is not a good way to come up with what we are trying to do.

    DR. SOUTHERLAND:  I know that certainly in the Ohio River Basin where she did her study there is a big drop-off at the time of the press release.  She did the analysis and she actually did a time series thing right around the press release she did questionnaires, telephone surveys and, of course, fish advisories.  Then she did it several months later and then several months after that and, of course, it really drops off over time.  That is why I think people are looking for things that are more permanent, like posters, signs or something either in the health clinics or actually at the point of fishing as something more substantive than just a big press release.  Again, you know, our risk commercial document, and I can get you that also, has tons of references of the studies that have been done to look at effectiveness.

    DR. MILLER:  Dr. Busta?

    DR. BUSTA:  I would like to get back to this collection of data from the states.  Do you have a distribution as to which states do the most analyses, and are they the states with the greatest pollution?  Maybe the states with greater pollution, are they sampling more and are they sampling mainly the polluted areas?  Have you any kind of information like that?

    DR. SOUTHERLAND:  Well, actually it started like that in 1993 when we began our program and I could say definitely yes, everything that was in the database was from suspected problem areas.  What has happened though, over time as people have become more and more concerned about mercury, and that is why they are starting to see these 12 statewide mercury advisories, as they went to more and more sites and checked to see if they had mercury concerns, they then said, you know, let's try pristine areas.

    This is what they did in Maine, for example, and that is when they found that even their pristine areas that did not have a discharge or a point source discharger, they found levels of concern for the mercury.  That is why you see more and more states having these statewide mercury advisories because they just said what are we doing here?  I mean, there is no sense waiting when we even have pristine areas because of the atmospheric deposition contributions of mercury.  We are just going to go ahead and have some general recommendations statewide, and then we will continue to try to go water body by water body and confirm or deny that assumption.  But right now that is what the statewides are based on, a general understanding that no matter where they looked they had some species that had concentrations of concern.  It is an ubiquitous problem, it truly is.

    DR. FISCHER:  I would like to see if we could get some information on the overlap between regulated commercial fish and sport fish--

    DR. SOUTHERLAND:  I am sorry, I lost you.

    DR. FISCHER:  What I am trying to see is what is the overlap between sport caught fish, which EPA regulates, and--

    DR. SOUTHERLAND:  We don't regulate; we give advice through the states.

    DR. FISCHER:  Excuse me, I made a big mistake there.  I am sorry, I should have learned that by now.  And, fish that the FDA takes care of, commercial versus sport caught.  In the Great Lakes basin you can buy whitefish and walleyed and a lot of what we would call sport caught fish, lake trout.  So, here are sport fish that are commercial fish.  I just wonder how much of the total consumption is confused in this way.

    DR. SOUTHERLAND:  There is definitely an overlap and the 11 states that have that little box in there that says they are giving commercial advice as well as recreational, they are the ones that definitely came up with this, and you will see there is a lot in the Great Lakes area because they said what difference does it make if they go down and catch this fish themselves or if they go to the nearby fish market where the fish that somebody else caught was put and they purchased it?  There didn't seem to be any reason for there to be a separation.  That is why the state health departments have decided to give advice across the board.  Like I said, 9 of the 11 states that do this include tuna in their recommendations because they have found so many people were eating tuna fish sandwiches and then adding onto that their recreational or other commercial fish.  So, they felt that they had to include the tuna fish in their advice too.

    DR. FISCHER:  Michigan hasn't dealt with this problem at all, and they give no advice on purchased sport caught fish.  I can't remember seeing other Great Lake states give it either.  I know that creates confusion in the public's mind as to what they should do.

    DR. SOUTHERLAND:  Several Great Lake states--Henry Anderson was going to speak next and I know he gives that type of advice.  Pam Schubat, in Minnesota, does.  We have a number of states that give combined advice for the Great Lakes states.

    DR. MAHAFFEY:  Just one comment, mercury, while it is a local problem in that there can be local discharge and local deposition, is also a national problem.  Part of the mercury that goes into the environment enters a high atmospheric level pool of mercury that then can be deposited in precipitation.  For example, it is the United Nations environmental program that this fall will do an assessment of mercury.  The European Union has adopted U.S. EPA's reference dose for mercury and, again, is doing broad work on fish in Europe because of concern for mercury.  There is not a clear separation between "locally caught fish" and fish that wind up in the commercial market.  They come out of the same water.

    DR. MILLER:  I think I am going to bring this discussion to a close.  I think the point has been made.  I think what is abundantly clear to me is that we are not really very close to a really effective communication system to get a relatively simple message that enables people to make appropriate decisions themselves.  I doubt that we will come up with anything better in our time but we ought to be thinking about how to approach this particular problem.

    I also think that we have spent quite some time this morning emphasizing the importance of the issue.  I am certain that this committee has a general recognition of the significance of the program and the importance of doing something about the problem.  I think the debate that we are having over which number to use is important in implementation of any plan, but I don't think it reflects the fact that the members of the committee in any way take this thing as anything but quite seriously.

    So, at this point I am going to call this section closed.  Thank you for this discussion, and we will move on to our next speaker.  I assume Dr. Henry Anderson is not here.  So, the next speaker is Dr. John Middaugh, from the Alaska Department of Health, talking about the Alaska advisory.

Alaska Advisory

    DR. MIDDAUGH:  Chairman Miller, and members of the Food Advisory Committee, I am John Middaugh, State Epidemiologist with the Alaska Division of Public Health.  I am here today to provide information to the committee on behalf of the State of Alaska as a public health physician with responsibility for protecting the health of the citizens of Alaska.  Thank you for providing the opportunity to bring to your attention Alaska's experience with fish advisories for methylmercury.

    I am pleased that the Food and Drug Administration is having this meeting.  I believe that the committee has an important opportunity to clarify roles in attaining shared national goals to protect the environment and to protect the public health.  To do so, it is essential to sort out federal agency authority and responsibility and to respect the balance between federal and state authority and responsibility.

    We also have before us new technology that can help us further scientific understanding.  Finally, we can ensure that we behave ethically, adhering to fundamental principles of "do no harm" and weighing benefits and risks.

    Alaska's experience with national fish advisories has uncovered several major areas of concern and revealed potential unintended adverse consequences.  Current national recommendations, a "one size-fits-all" approach, do not make sense in Alaska.  National recommendations for fish consumption are not consistent with available evidence and are not consistent with Alaska recommendations.

    Considerable scientific controversy exists over the risks of low-dose methylmercury exposure.  Data linking low-level methylmercury exposure to adverse health outcomes are weak.  Adverse neurodevelopmental outcomes documented are subclinical, detectable only by sophisticated tests of unknown long-term significance.  Results may be limited by potential confounding.  Leading studies have not found similar results, and ongoing studies hold the promise of providing important information in the near future.

    Advisories based upon risk assessment without consideration of well-established public health benefits of fish consumption have great potential to harm public health if reductions in fish consumption occur.  We have special concerns over the impact of fish advisories for Alaska natives and rural resident subsistence consumers who have few alternatives to fish.  The public health harm caused by fish advisories has been well documented, especially in Canada.  Fortunately, data from Alaska provide evidence that most, if not all, Alaskan exposures to methylmercury are below those of current concern, even applying conservative models.

    Finally, extensive international scientific investigation of Arctic contaminants under the Arctic Monitoring and Assessment Program of the International Arctic Science Council during the past eight years has led to the consensus Arctic recommendations that the health benefits of Arctic subsistence foods outweigh potential risks, and that local public health authorities need to take into account local information to craft dietary guidelines.

    We have a substantial body of scientific information on mercury in Alaska.  I would like to provide a brief summary of some of the most germane studies that provide evidence that determined Alaska's current dietary recommendations.

    I have provided a detailed copy of handouts because there is a lot of data and I know that I can't present it in 20 minutes.

    [Slide]

    First, we analyzed ancient human hair from mummies from the Aleutian Islands in Alaska.  The mummies were taken from islands out in the Aleutian chain during the 1920's when archeologists collected human remains and artifacts from burial sites.

    [Slide]

    This is an adult wrapped in reed tissue, and then underneath that wrapped in seal skins.

    [Slide]

    This was an infant in a basket.

    Our goal was to try to establish if methylmercury was present long prior to the industrial revolution and, therefore, represented naturally occurring exposure.  After receiving permission from the Aleut Corporation and the Museum of Aleutians, we collected hair samples from four infants and four adults that radiocarbon dating established to be approximately 550 years old, dating to about 1450 A.D.

    [Slide]

    The average level of methylmercury mercury in adults was 1.2 ppm in hair, and in infants was 1.44 ppm, with a range of 7 ppb to 4.61 ppm.

    [Slide]

    Segmental hair analysis showed patterns of higher and low methylmercury in centimeter segments, compatible with seasonal and event-specific changes in mercury exposure through a subsistence fish and marine mammal diet.  These results are consistent with a few other similar studies of mercury in ancient human remains, supporting the hypothesis that humans have always been exposed to naturally occurring mercury through fish and marine mammals in the diet.

    Unlike many areas in the continental United States, there are no local industrial sources of mercury in Alaska.  Extensive environmental sampling during the past ten years has documented that Alaska is one of the most pristine areas in the Arctic.

    I want to go over the next set of slides very quickly.  They show some of the wealth of sampling data of fish species.  These results are for 1993 from U.S. FDA.  In red, you can see for salmon many of the results are undetectable.

    [Slide]

    This is from the ATSDR criteria document.  Again, salmon is 0.035 ppm.

    [Slide]

    Again, our Department of Environmental Conservation lab, shows 1999 results and again in red are highlighted the salmon results.  They are almost all about 0.025 or non-detectable.

    [Slide]

    Researchers at our University of Alaska, measuring king salmon methylmercury levels are in this column, here, and all the results are very low.

    [Slide]

    Those are the same results in a graphic form.  All of these are in your packet so you can study them later and figure out which of the species you want to eat, but these are the same results from the University of Alaska.

    [Slide]

    Arctic grayling, a form of trout, are all levels that are very, very low.

    [Slide]

    Northern pike are one of the freshwater species with the highest levels in Alaska, but most are also below one part per million.

    [Slide]

    Alaska freshwater fish, we have turbot, sheep fish, dolly varden trout, sucker fish, rainbow trout, whitefish.

    [Slide]

    Our Department of Environmental Conservation for southeast Alaska shows all the species are very low except for salmon shark.

    [Slide]

    Out in the Aleutians there is a small number of samples but, again, Dover sole, ocean perch and yellow fin sole all have very low levels.

    [Slide]

    Cook Inlet, which is near Anchorage--all very low levels.  I would like to skip through this and then I will go on, but in the packet you can see that there is an extensive amount of sampling and of most importance is that the levels of methylmercury in all species of salmon are among the lowest of all species of fish, ranging from non-detectable to about 0.05 to 0.08.

    Dietary surveys in Alaska document a wide-ranging exposure to multiple fish species in marine mammals.  This overhead presents regional compositions of subsistence harvest by our rural residents in different parts of the State.  There are considerable variations by region most notable in amounts of fish and marine mammals.  Up in the north slope there is a lot of marine mammal that is whale, seal and walrus.  In some of the other areas of the State there is mostly fish and of the fish, mostly salmon.

    [Slide]

    As you can see, fish comprises about 60 percent of Alaskan's rural subsistence harvest.

    [Slide]

    Mean salmon consumption in Alaska far exceeds the current EPA and FDA consumption advisory amounts, and you can see in these dotted lines are the FDA and EPA recommended consumption advisory levels, and these are mean harvest data for fish for different communities in Alaska.

    [Slide]

    The economic and nutritional values of subsistence foods in Alaska are huge.  For the percent of population's required protein overall in Alaska subsistence harvest comprises about 65 percent or protein, 9 percent of total calories.  The estimated economic value in Alaska of subsistence harvest is 267 million dollars.

    [Slide]

    Currently, several major dietary surveys are under way in Alaska, including ones conducted by the Alaskan Native Tribal Health Consortium with EPA support, the Aleutian/Pribilof Island Association supported by the NIEHSS, the Alaska Native Health Board, supported by ATSDR, and the University of Alaska, supported NIH.

    In addition to these traditional sources of information, we also have new data on actual human exposure levels.  Dr. James Berner, Alaska Native Tribal Health Consortium, is the principle investigator of an Alaska native maternal-infant cord blood contaminants study.  This grassroots project was requested by local Alaska native communities.  Funding of this effort is from EPA and the National Center for Environmental Health of CDC.  In addition to actual measurement of heavy metals, persistent organic pollutants and radionuclides, long-term neurodevelopmental follow-up of the children is planned.

    Dr. Berner has given permission for me to share with you the initial results from this study.  For 52 mothers who delivered babies from the Bethel area of Alaska, the median blood mercury level was 4.65 ppb with a maximum level of 21 ppb, and for 29 mothers from the Barrow area the median blood mercury level was 1.1 ppb with a maximum level of 4.5 ppb.  Additional data include hair mercury results for 14 mothers with a median level of 0.94 ppm in Bethel and 0.48 ppm in Barrow, and a maximum level of 1.9 ppm.

    Recognizing that these two populations have high subsistence intakes, levels show no cause for concern.  The State recently established a statewide maternal hair biomonitoring program to provide, at no cost, measurement of mercury in the hair of all women who are pregnant.  We just had the first results that were called in last night of the first 12 women participating.  The hair mercury levels ranged from 0.03 ppm to 1.2 ppm, with a median of 0.26 ppm.

    [Slide]

    Balancing benefits and risks is essential in crafting public health recommendations.  Some of the benefits of subsistence lifestyle are nutrition, taste, sociocultural values, fitness, cost, children's education and ecological knowledge.  Some of the risks are accidents associated with hunting and fishing, and health risks include botulism, trichinosis and paralytic shellfish poisoning, for example.  Then, there are the risks of not eating the traditional foods, obesity, diabetes and heart disease.

    [Slide]

    There are also many sociocultural benefits of traditional foods, what food is to a culture.  The Alaskan native people have indicated that the issue of contaminants is the most important one facing them as a community.  They have identified values of the subsistence reliance on traditional food and fish as physical fitness, recreation, the healthy foods, being in tune with nature, sharing, that it saves money and the value to their culture.  Also, pride and confidence.  For their children, their education, the natural environment, survival skills, food preparation techniques, practicing patience and respect.

    [Slide]

    There are well-known public health benefits from fish consumption.  Fish provide high nutritional value, vitamins A, E and C, protein, energy, omega-3 fatty acids, monolipids, iron and zinc.  Omega-3 fatty acids have proven benefits in preventing complications from diabetes, preventing coronary heart disease and atherosclerosis and preventing complications from arthritis, to name just a few.  There are also major economic, cultural, spiritual and social benefits from subsistence practices.

    Experience in Alaska has documented adverse effects on public health and communities from fish advisories with subsequent abandonment of traditional diets.  Alaska natives are experiencing a major increase in the prevalence of diabetes.  Heart disease rates are increasing, and recent studies have documented vitamin A and D deficiencies.

    [Slide]

    For example, the prevalence of diabetes among Alaska natives has increased substantially in the past two decades.  We are concerned that Alaska natives may be on the threshold of a major epidemic, similar to those of the Pimas.  In addition, Arctic residents are faced with serious problems of alcohol use, lack of physical exercise and subsequent increases in obesity.

    The subsistence lifestyle and diet are of great importance to the self-definition, self-determination, cultural and socioeconomic and overall health and well being of indigenous peoples.  Alaska natives have voiced their fears and concerns about the safety of traditional foods.  However, native elders have also expressed concerns that the fear associated with the contaminants may cause greater harm than the actual presence of the contaminants themselves, and that health warnings regarding food consumption should only be made when there is strong evidence that the risks outweigh the benefits.

    There is a compelling need to incorporate benefits and risks in dietary recommendations.  The precautionary rule seems most appropriate for taking actions to reduce industrial and other man-made pollutants.  Alaskans have great concerns over the long-range atmospheric transportation of pollutants into the Arctic.  The State of Alaska supports the POPs treaty and efforts to reduce anthropogenic pollutants, and use of the EPA RfD in reducing mercury emissions.

    But in creating public health recommendations for fish consumption, it is essential to weigh benefits and risks.  Relying on the EPA RfD led to the question what is the public health risk of the uncertainty factor?  The Belmont Report provides the foundation for U.S. policies for the protection of human subjects.

    The report outlines basic ethical principles, the principles of respect for persons, beneficence and justice.  The Belmont Report formulated to general rules to complementary expressions of beneficent actions.  One, do not harm and, two, maximize possible benefits and minimize possible harms.  Justice has the sense of fairness in distribution or what is deserved.  An injustice occurs when some benefit to which a person is entitled is denied without good reason, or when some burden is imposed unduly.  The application of the general principles leads to considerations of risk/benefit assessment.

    In this context, the State supports the FDA's leadership in providing general public health-based dietary guidelines, including the flexibility to weigh benefits and risks and providing for substantial involvement of state and local public health agencies in applying local evidence in developing dietary guidelines.

    In response to the national fish advisories of January 2001, the Alaska Division of Public Health engaged in extensive consultations with Alaska stakeholders.  After reviewing all of the available evidence, the Division of Public Health issued consensus recommendations for fish consumption in Alaska.  The most important difference from national advisories is the following:

    The Alaska Division of Public Health continues to strongly recommend that all Alaskans, including pregnant women, women who are breast-feeding, women of childbearing age, and young children continue unrestricted consumption of fish from Alaskan waters.

    The State does not support national advisory recommendations to restrict fish consumption to 12 oz per week, nor the national advisory recommendations for pregnant women to restrict fish consumption to one meal per month.

    [Slide]

    The State, in consultation with the Food and Drug Administration, developed the following language that was included in the amended FDA advisory:  Some kinds of fish that are known to have much lower than average levels of methylmercury can be safely eaten more frequently and in larger amounts.  Contact your federal, state or local health or food safety authority for specific consumption recommendations about fish caught or sold in your local area.

    [Slide]

    The following agencies and organizations endorsed and contributed to the development of these recommendations:  The Alaska Department of Environmental Conservation, the Alaska Department of Health and Social Services, the Alaska Native Health Board, the Alaska Native Science Commission, the Alaska Native Tribal Health Consortium, the Aleutian/Pribilof Islands Association, the Institute for Circumpolar Health Studies of the University of Alaska Anchorage, the North Slope Borough, the University of Alaska Fairbanks and the Yukon Kuskokwim Health Corporation.

    [Slide]

    As part of the State's development of its dietary recommendations, the State also made a commitment to supporting increased monitoring of mercury levels in fish, supporting the ongoing research being conducted by Dr. Jim Berner in his maternal-infant contaminants study, and developing and implementing a statewide maternal hair mercury biomonitoring program.  These efforts are all underway.

    In conclusion, the State supports increased human exposure assessments such as the recent mercury studies of the NHANES by the National Center for Environmental Health of CDC; increasing human exposure assessments as proposed by the National Center for Environmental Health through expanding the national NHANES assessments to the state level; increased biomonitoring of fish species; increased consideration of benefits as well as risks; targeting fish advisories based on levels of mercury in key species and actual human exposure data; and increased and sustained efforts to reduce global anthropogenic emissions of mercury.  Thank you.

Questions of Clarification

    DR. MILLER:  Thank you.  Questions or comments?  Dr. Hotchkiss?

    DR. HOTCHKISS:  I think we would all agree that Alaskan fish are lower in methylmercury than fish available from most other parts of the country, which tells me if you don't throw mercury around in your environment you have fish with less mercury.  But my question to you is if the fish consumed in Alaska had methylmercury levels that were more consistent with the rest of the U.S., or at least the higher portions of the U.S., would your position be the same on this issue?

    DR. MIDDAUGH:  Well, I think that the methylmercury exposure is determined by fish, and I certainly think that there is an absolute need to have fish advisories especially for local contaminated areas.  I believe that there is a great opportunity with new technology to make accurate measurement of human levels of mercury, at very little cost, to combine the risk assessment methodologies with actual exposure levels which are showing us, at least in Alaska, that the levels that we actually measure are far below those that we would have predicted to have found having only used data from fish species and these dietary projections of assumed amount of exposure.

    I also believe that the data are probably quite skewed, and our experience is that when we monitor fish we find very few that have higher levels and almost all the other fish have very low levels.  I think that may be one explanation why we are seeing a disconnect between some of the predictions of exposure levels and some of the exposure levels when we actually go out and measure.

    The CDC and the Pugh Commission and the Trust for American's Health have all recommended increased biomonitoring to actually measure what the exposures are that are occurring among the U.S. population for these contaminants, and I think it is critical that we expand that knowledge database before we potentially warn people to avoid consuming a particular fish product that has huge documented public health benefits.

    DR. HOTCHKISS:  I agree, but I am a little confused by your answer.  Do we agree that in general fish caught in either marine or freshwater environments in Alaska are lower in methylmercury content than, let's say, fish caught in the region of the Gulf part of the U.S.?

    DR. MIDDAUGH:  I can for sure say that the fish in Alaska have very low methylmercury levels.  I believe that they are lower than in many other parts of the United States but I am not an expert on the levels in a lot of the rest of the country.

    DR. HOTCHKISS:  Let's assume for the sake of discussion that they are.  I think the data would bear out quite strongly that there is a difference from the Great Lakes, for example.  I am just curious about your position.  Is your position based on these low levels in Alaska, or is your position in general that we don't know enough about methylmercury nor the levels across the U.S. to make recommendations?

    DR. MIDDAUGH:  Well, there are certainly two parts to my answer.  The first one is that we were very pleased in Alaska, when we found all this data and did our measurements, to find that we have very low levels.  That made it much easier for us to develop our Alaska recommendations.  But they fly in the face of the national recommendations so we are confronted by having "Brain Food" on the computer that rural Alaska residents download which tell them not to eat their fish, and then what do they eat?  We are part of the United States.  So, that is a problem.

    Then, we also are concerned about the absence of benefit/risk evaluation in the national program.  So, I think that the easy part for Alaska is that our levels are lower so it made it easier for us to come to consensus recommendations for people in our State.  But we are also very hesitant about the reliance on the EPA RfD for crafting dietary recommendations for the American people.

    DR. HOTCHKISS:  Thank you.

    DR. MILLER:  Ms. Halloran?

    MS. HALLORAN:  Could you say more about your hair monitoring program, how big it is and how much does it cost?

    DR. MIDDAUGH:  We have just started this program, and the reason was to actually be able to try to provide ongoing surveillance evidence to document the validity of our dietary recommendations.  There are around 12,000 live births in the State of Alaska.  We just started the program in June.  It is available to all women, free of charge to their provider to send their hair in to the state lab where it will be measured and the results will be reported back to the provider.

    We believe that by doing so, one, we will be able to follow trends over time.  Two, if we find any evidence of unexpectedly elevated levels in any geographic area, any village, any sub-targeted component of the State, then we can go out and do more detailed evaluations and investigations to try to determine why the hair levels are higher and, if necessary, we can always develop targeted advisories.

    MS. HALLORAN:  How much is this costing?

    DR. MIDDAUGH:  We are using Frontier Geosciences.  Dr. Nicholas Bloom's lab is a consultant to our State Health Department and has been for some time, but we are setting up to run mercury in our new state laboratory.  So, we believe over the next six months we will be able to offer the mercury testing in our own state lab, not have to use a contractor.  That will lower our cost, and we are estimating that as we gear up with the volume we anticipate we may get by with less than $50 a test for all associated costs.

    DR. MILLER:  Dr. Shannon?

    DR. SHANNON:  I have a question and a comment.  My question relates to your discussion of increasing rates of diabetes and obesity.  In fact, you showed a slide of the increasing rate of diabetes.  It wasn't clear to me if you were relating that to changes in fish consumption or mercury exposure and exactly what point you were trying to make there.

    DR. MIDDAUGH:  Yes, very much so, there is a rich experience, well documented in Canada from 20 years ago when fish advisories were given in Quebec to the Ashkenazi, followed by complete abandonment of their subsistence intake of traditional foods.  That led to tremendous community problems, and we are seeing tremendous problems in Alaska not just from mercury but also from other contaminants brought to Alaska from long-range atmospheric transportation, and then these tremendous warnings against the dangers of the contaminants, and people are abandoning the use of their traditional foods, including fish consumption.  In the larger context of Alaska, the values of both nutrition and the potentials for severe public health problems have been played out and are a great concern.

    DR. SHANNON:  I am still not sure I understood.  You have clear evidence that the increased rate of diabetes and obesity is associated with decreased consumption of fish in Alaska?

    DR. MIDDAUGH:  We have anecdotal evidence to support that association.  It is clearly not the only factor going on.  There is a tremendous increase in life expectancy so all the chronic diseases are increasing.  But we also did extensive autopsy studies of Alaskan natives, looking at the association of omega-3 fatty acids and atherosclerosis, in collaboration with University of Louisiana and Dr. Jack Strong's group.  Fifteen years ago we showed that the amount of atherosclerosis among Alaskan natives was about half of that of non-natives and there was a strong association between the amount of atherosclerosis and omega-3 fatty acids measured in the coronary arteries and also in perirenal fat.  Now heart disease is increasing and we have evidence from dietary surveys that the amount of consumption of subsistence foods is declining.

    DR. SHANNON:  You didn't present any of those data.  Can you say a little bit more about that?  How much is fish consumption declining in Alaska?  How much has it declined over the last 10 to 15 years?  I don't remember you saying anything.

    DR. MIDDAUGH:  We don't have good quantitative data.  What we have are subsistence harvest records.  Harvest records are not dietary consumption records, but we have numerous anecdotal reports from the physicians around Alaska, from our nurses in the villages and, of course, the whole issue of contaminants has been one of headlines in the papers routinely for the last eight to ten years.

    DR. SHANNON:  I have to push you on this.  I think any good scientist would want more than anecdotal data that you think there have been important declines in consumption.  Do those data not exist, or why hasn't that been investigated?

    DR. MIDDAUGH:  I mentioned that there are four major dietary survey projects that have been developed in the last several years, funded by ATSDR, EPA, NIH and CDC, and we are hoping that that information will provide some of the quantitative evidence that we can use to compare dietary consumption practices today versus the data that was accumulated 15 or 20 years ago by nutritionists with the Indian Health Service.

    DR. SHANNON:  Maybe I misinterpreted your introductory comment, but if I understood you to say that you thought that low level of mercury exposure may not have an effect, and if it does, it is subclinical and, therefore, probably unimportant and it isn't clearly irreversible, my comment would be that I don't think that the weight of scientific evidence that we have now would support those statements.  Maybe that is not what you said.

    DR. MIDDAUGH:  Well, I would say that in Alaska, for example, we have the highest rate of fetal alcohol syndrome in the world and we have tremendous problems with infant mortality and nutrition and recent studies have documented borderline vitamin A and vitamin D deficiency among newborns and infants.  We have very serious, well-documented, very significant public health problems and we believe that abandonment especially of fish consumption but also other traditional food consumption would cause a great public health tragedy among Alaskans.

    DR. SHANNON:  Right, but we are here to talk about mercury and the issue is whether mercury exposure is important or not and it just seemed like you minimized it.  Again, my only comment is that I don't think that the weight of scientific evidence would support that.

    DR. MIDDAUGH:  I would only respond to that by saying that we are very concerned about mercury exposure in Alaska.  That is why we have developed some of these studies.  That is why we launched the statewide monitoring program because our available evidence is suggesting that, at least in Alaska, our exposures are very low, far lower than we would have predicted based on consumption and fish species mercury level monitory data, and it is something that we intend to pursue with great vigor in the future.

    DR. MILLER:  Dr. Dwyer?

    DR. DWYER:  Thank you for a very interesting presentation, and I wanted to congratulate you for this monitoring program that you are putting in place.

    The question I have is how will that work?  It really isn't necessarily the people in Fairbanks and Anchorage that you are interested in.  It is the people in the little, tiny towns that are very isolated and hard to get to.  I know a former colleague and friend, Betsy Nobman, has done some of the studies that Dr. Shannon was asking about and I know how difficult it is to get food consumption data in some of those places.  I wondered how you are going to get the samples of the hair and so forth in these remote villages that are very inaccessible in many cases.

    DR. MIDDAUGH:  Yes, actually Betsy Nobman is continuing to work on these dietary surveys that I have mentioned.  Dr. Berner's infant cord blood study is based on the rural hospital hubs to which pregnant women come in to deliver, and the Alaska Tribal Health Consortium in Anchorage has about 100 births per month, again, in many cases high risk pregnancies that are flown in from the villages.  But the hair program is a component of Dr. Berner's study so it is located in those rural hospital areas.  Then, should we find any evidence of elevated mercury, we can go out to the village and offer mercury testing not just to pregnant women but to women of childbearing age.

    So, initially we focused on pregnant women, one, because they are of greatest concern for their own actual exposures and the fetal exposure, but also in terms of just the amount of money we have available to support the program.  But the full intention is that should we find any elevated levels or higher levels than expected, then we can go back out to that area, the geographic area or to those villages, and do much more extensive testing and then liken that to detailed dietary exposure to determine if there is some unusual exposure or some unexpected elevated levels of mercury which we were unaware of, or for which a local advisory might be appropriate.

    DR. MILLER:  Other comments?

    [No response]

    Thank you.  I have just been informed that Dr. Anderson will be arriving here at about one o'clock.  The thunderstorms last night have prevented him from landing at Reagan.  In order to accommodate this, I would like to change the agenda just slightly.  I wonder, Dr. Lockwood, if you would mind making your presentation now, before the break?  I understand you will just need about five or ten minutes.

American College of Obstetrics and Gynecology

    DR. LOCKWOOD:  Thank you.  I am Charles Lockwood.  I am an obstetrician and I was the former chair of the American College of Obstetricians and Gynecologists, Committee on Obstetrical Practice that acts sort of like a clearinghouse for information that is then disseminated to the 40,000 or so fellows in the College.  Virtually all obstetricians and gynecologists in the United States are fellows in the college so it is relatively unique as a combination trade and educating society in medicine.

    The mission, of course, of the American College of Obstetricians and Gynecologists is to improve the health care of women and their fetuses, and we are somewhat confused and very, very anxious to be able to communicate with our fellows and, through our fellows, our patients precisely what message should be given to them regarding the safety of their food supply, and specifically the amount of fish that they can take in that is safe.

    The College is concerned particularly because we have been urging patients to eat fish for the past ten years because there is evidence that it may improve health outcomes, reduce the risk of preeclampsia, perhaps affect premature labor and so forth.  Although none of the literature is particularly robust, certainly the bulk of it suggests that fish intake may be beneficial.

    We are also very concerned that there may be, as has been implied already today and I suspect discussed yesterday, significant variability in the content of mercury in fish among different regions.  It is very nice that Alaska has such a low supply of mercury, and that is very good for the pregnant women in Alaska but we are very concerned about the other regions of the country that might have substantially higher contents of mercury and applying a uniform standard.  This is sort of the federalist response to the Alaska statement, we are concerned that applying a single standard may, in fact, underestimate the risk to the newborn.

    Although I certainly am very sensitive to the notion that different regions need to educate their people, our problem is that we can only communicate with our 40,000 obstetricians and that message has to be short, sweet, clear and concise.  It needs to be able to then be translated and given to their patients.  If we rely on the states to do that, since there is substantial variability in their monitoring and also in their communication of the levels that might be elevated for mercury in different water sources, that is going to lead to complete chaos amongst obstetricians and gynecologists.

    So, we do applaud the EPA and the FDA for giving us some guidelines and a fairly simple message to convey.  However, having said that, we are concerned that, in fact, we don't know enough about the neurodevelopmental effects of mercury.  The literature has been, at best, unconvincing.  We would like to urge the NIH and other federal agencies to support research to establish in a much more precise and rigorous way what mercury does to the developing infant's brain.  We would like fairly exhaustive studies done in primates and in vitro studies to assess the effects of mercury on nerve development, and so forth.  In addition, we would like far more detailed epidemiological studies, coupled with child development studies, to be able to get some sense of whether there is at least a crude correlation between fetal in utero exposure to mercury and subsequent neurodevelopment and its effects.  So, research I guess is the message there.

    We would also like studies to establish whether or not mercury is teratogenic.  It is pretty clear that in high concentrations it can induce fetopathy but it is not so clear that it causes birth defects and we would like to know that.

    Finally, I guess we would urge that if you are planning to change the RfD that is used by the FDA that you do err on the side of being conservative, err on the side of, for example, accepting the Institute of Medicine's recommendations because we don't want to discover 15 years from now that, in fact, the level was too high and we have had an effect on the development of the next generation of our citizens.  Thank you.

Questions of Clarification

    DR. MILLER:  Thank you.  Any comments or questions?  Yes, Dr. Nordgren?

    DR. NORDGREN:  Dr. Lockwood, your organization is in the front lines on this issue, and I think it sounds like your organization has very carefully studied the issue that we are all facing here.  My question to you, and the reason I am asking this question is, dealing with fetal alcohol syndrome in the past, I think many obstetricians were way behind the eight ball as far as recommendations in their practice as to good scientific knowledge.  My question is do you think within your organization this information is filtering down?  That is a hard question to answer.

    DR. LOCKWOOD:  It is a hard question to answer.  The life of an obstetrician is difficult.  We have really added substantially to the burden of information that we have to convey to our patients.  Some of that is actually regulated by states and some of it has, in a sense, been de facto regulation by the College endorsing certain programs and policies, for example, cystic fibrosis screening and universal HIV screening, and so forth so that a substantial amount of time is spent by the average obstetrician counseling patients.

    So, if you are attempting to pay your $115,000 a year malpractice premium, as you would in Long Island for example, and you are attempting to support your staff and so forth, and increasingly large staff since more and more counseling is done by nurses, etc., and you are facing a 50 percent reduction in the average reimbursement of your services by managed care organizations, which we have over the past ten years, and, therefore, you are forced to see many more patients in a shorter period of time, it becomes increasingly more difficult to add to the burden of information in the time that is available.

    This is not to say that we don't have an obligation to do the very best we can with every patient, but from a practical perspective, there is only so much we can do; there is only so much information we can convey and at some point this all becomes crippling.  You know, I live in a lovely Ivy League world with plenty of time to spend with every patient and, yet, even for me it is becoming incredibly difficult to give all the information that we have.  We have to talk about exercise.  We have to talk about vaccinations.  We have to talk about infectious disease exposure.  We have to talk about screening for aneuploidy or Down's syndrome.  Now we have to talk about cystic fibrosis screening; a variety of other ethnic specific genetic disorders, and on, and on, and on.

    So, to be able to convey the complexity of these issues to the average pregnant woman is impossible.  It can't be done.  We can provide educational resources.  We can provide information on web sites.  We can have patient handouts, and we are more than happy to do all that, but we do need a very simple, clear message to our patients and I think that the College was happy to pass on the information that the EPA and the FDA produced but we understand that, in fact, that may be a fairly--depending on your political perspective--conservative or liberal recommendation.  So, we will follow whatever the recommendations are, but our bias would be that those recommendations be as conservative as possible from the perspective of the developing fetus.  I hope that answers your question.

    DR. MILLER:  Dr. Lee?

    DR. LEE:  I was just following up on a casual mention you just made.  Is there a single source or clearinghouse of information for pregnant women to go to for recommendations on prenatal care and diet?  Do you point them to a particular site, or do you have just a whole array of brochures?

    DR. LOCKWOOD:  The bible for provision of prenatal care, if you will, is contained in "Guidelines for Prenatal Care," the fifth edition of which is about to come out.  Having helped write that, I am embarrassed to say that I don't believe we incorporated any recommendations about mercury.  There may be time, and I will work on that right away to be able to do that.  It would be nice, of course, if you changed your recommendations you would do that in the next 25 minutes so I would be able to do that--

    [Laughter]

    The second line of information, and the one that is much more time sensitive is the Committee opinion which is rendered by the Committee on Obstetrical Practice and that is distributed via publication in the Journal of Obstetrics and Gynecology to virtually all of the fellows of the College.  Rarely will we actually do direct mailing.  For example, if tomorrow the New England Journal of Medicine pointed out that exposures of mercury are substantially higher than the current recommendations would allow caused neurodevelopmental abnormalities, we would send a Committee opinion and literally mail it to every single member of the College with new recommendations.  So, we can effect rapid change and those opinions are looked at very carefully because they are construed by our trial lawyer colleagues--I hope none of them are in the room--as the standard of care in the United States.

    DR. LEE:  That is a very good mechanism for communicating directly with your members, but is there any attempt to communicate directly with your patients?

    DR. LOCKWOOD:  We do assume that physicians speak to their patients, particularly if the topic is critically important.  But, in addition, there are handouts that are available that are published by the College that most obstetricians and gynecologies have in their office.  So, if there is a new recommendation that is going to be made, we could certainly incorporate that into the general educational efforts the College makes directly to patients.

    DR. MILLER:  Dr. Dwyer?

    DR. DWYER:  I just wondered, given the current state of information we have available, what do you intend to put in the ACOG handbook?  What is the one-liner that you want your ACOG members to have?

    DR. LOCKWOOD:  We would put in the "Guidelines for Prenatal Care" what we already put into our general publication.  There is yet a third layer of communication and that is a newsletter that goes out to all of our fellows, and we have already put in that newsletter the recommendations of the EPA and the FDA regarding the amount of fish to eat and the various other aspects of whether, you know, it was commercially obtained.

    DR. DWYER:  But do you really expect an obstetrician to have enough time to go through all of that?  I mean, what would be the one-liner?

    DR. LOCKWOOD:  Yes, these are one-liners.  Most obstetricians do read that.  Whether they incorporate it into their practice is up to them.  We don't have any way of monitoring at this point what obstetricians actually tell their patients in their offices.

    DR. MILLER:  Dr. Acholonu?

    DR. ACHOLONU:  Thank you.  I have a very similar question to Dr. Dwyer's.  You made a statement, and please correct me if I have put it down wrongly, that we don't know enough about the neurodevelopmental effects of mercury.  I have been sitting here, listening to summaries and reading about the Faroe Islands Study and the Seychelles study.  What is the ACOG opinion of those studies?

    DR. LOCKWOOD:  We certainly understand, based on the exposures that occurred in the Minamata Bay, what substantial exposure would do to the developing fetus, and that the manifestations of cerebral palsy and major retardation and the other various neurodevelopmental abnormalities that occurred in response to that exposure are pretty straightforward.  I don't think there is much debate about that.  The thresholds that have been calculated on the basis of those exposures I think are robust and no one is going to dispute them.

    Reviewing the data from the Faroe Islands and the Seychelles Islands, as well as the New Zealand study leaves us a little bit more lost.  It is unclear whether or not you can set a specific exposure level that would be safe and, conversely, one that would represent the lower limit of absolute risk.  It is unclear, sort of like the fetal alcohol story, whether there is an absolute discrete threshold below which there is no risk to neural development, or whether this is a continuous exposure.

    I think we would like to see literally a dose response study and, if you will, a time course study to inform the recommendations that you all are going to be making.  But I don't see at this point, based on the information that I have been given, that you can say that a certain level is safe or that a certain level is absolutely unsafe.  In general, what we have ended up doing with fetal alcohol is to say that we are not sure that any level of alcohol intake by a pregnant woman is safe.  Most obstetricians will say it is okay to have a glass of wine once a week or so and it probably is okay but, in fact, since we never could establish an absolute limit to alcohol exposure that could be deemed safe we general proscribe the use of alcohol in pregnancy.

    I suppose at this point, if we are left with increasingly concerning information about the lack of a lower limit of mercury exposure, pregnant women will stop eating fish, but there are a lot of health benefits to eating fish and it is a relatively cheap source of protein.  There may be some additional benefits of reducing oxidative stress that might induce preeclampsia or per-term delivery; may affect fetal growth restriction by impairing placentation.  So, there are a lot of reasons to think that fish might be useful for pregnant women to take in but apparently, if we want to be absolutely safe, we have to tell them that they have to go to Alaska and eat fish with very low levels of mercury.

    DR. ACHOLONU:  I think you have answered most of my questions, but did I understand you to advocate that NIH and other groups should give research grants to your organization or to other people do to research on the effect of mercury on the developing fetus?  Is it to your organization?

    DR. LOCKWOOD:  Not through our organization, we don't conduct original research.  It would be up to the appropriate academic medical center to do that.  But, yes, I think we need to have more research on the topic.

    DR. ACHOLONU:  Do you have any more areas where you want the research to be concentrated because a lot of work has been done on the effect of mercury on children and prenatal effects?  What area of research do you want this thing to be concentrated on?

    DR. LOCKWOOD:  We would like a body of research that informs our counseling about fetal alcohol syndrome, which is a substantial amount of research, done in animal models and in humans, with long-term follow-ups, and an enormous effort made to understand precisely what the effects of fetal alcohol exposure were.  I don't think there is anywhere near the quantity or quality of research that was done in that area in this area.

    Now, in a world of limited resources that may not be a reasonable expectation, particularly if the stock market goes down again today.  But, in fact, I and the rest of our Committee and the rest of the College's leadership is not convinced that the information that is available is adequate.

    DR. MILLER:  Ms. Halloran?

    MS. HALLORAN:  Some women in the United States don't get prenatal care from a physician.  Do you have any advice on how to get this message to those women?

    DR. LOCKWOOD:  Well, there are several reasons women don't get prenatal care.  One rare reason might be physical access.  That might be a bigger issue in Alaska.  But, generally speaking, people who don't seek prenatal care are less likely to be responsive to messages about the health of their fetus or behaviors that should be avoided that might impair the health of their fetus.  So, I would wonder right off the bat whether whatever I recommend or am about to recommend would be cost effective in that particular group of patients.  That is a real dilemma with fetal alcohol syndrome, drug abuse and a whole variety of other exposures.

    But it is a little hard for me to imagine that that population can be easily accessed and that that information could be easily conveyed or be readily accepted.  There are issues with language barriers, and there are issues with poverty and crime, and many other things.

    The March of Dimes has been very, very effective I think in spreading the message that folic acid supplementation is something that should be begun by women of all childbearing ages.  There is evidence that the amount of folic acid used in women who are anticipating being pregnant has increased substantially, and I think if a solid recommendation comes out of this body and it is based on good scientific evidence and everybody can get their hands around it, and feels comfortable with it, the March of Dimes would be an excellent organization to help disseminate that information, provided, of course, that they also were comfortable with the content.  But they seem to do a much better job at reaching everybody than we do as physicians.

    DR. MILLER:  Dr. Nordgren?

    DR. NORDGREN:  I would just like to follow-up on Dr. Lee's question.  I don't think you really answered his question.  You didn't say your organization has a consumer web site like, for instance, the Academy of Neurology that has a web site so that people who want information about neurologic disorders can go to.  Your organization doesn't have such a thing?

    DR. LOCKWOOD:  We do.  We do have a web site that consumers can access and there is patient information there.

    DR. MILLER:  Dr. Friedman?

    DR. FRIEDMAN:  I just want to make a little comment about the research that you are advocating.  I am from the National Institute of Child Health and Human Development, and I will take the message back, but most of the research that NIH supports is investigator initiated research.  So, I think the word needs to go out to researchers in your community to ask those questions and through submitting publications.

    DR. LOCKWOOD:  An RFA would help.

    DR. FRIEDMAN:  Okay, I will pass on the word.

    DR. MILLER:  Dr. Fischer?

    DR. FISCHER:  Dr. Lockwood, I would like to ask whether you think that the fellows in the College would support and participate in a national monitoring program for mercury exposure.  To do that would require a lot of effort from a lot of people, I suppose, including those in your organization.  Do you think that that would be possible?

    DR. LOCKWOOD:  I think not only would it be possible, but there would be great enthusiasm for it.  I was very intrigued by what Alaska is doing.  I think that sort of cuts to the chase and gives us a lot of valuable information.  Speaking for my organization, I don't know what my bosses are going to say but I would say that that would be something we would enthusiastically embrace.  You know, if we had envelopes and we could snip some hair and it was relatively a straightforward process, I think we would be happy to contribute to that.

    DR. KUZMINSKI:  I have one more question.

    DR. MILLER:  Go ahead.

    DR. KUZMINSKI:  I would just like to return, Dr. Lockwood, to the question asked earlier.  Perhaps you could give us an example from another area of a message that works for your fellows in the College.  You mentioned that you put in the EPA and the FDA recommendations on fish consumption.  We have had discussion here that one person reading all of those messages together might get confused.  So, as a guidance to this committee, can you give us the one-liner that has been referred to from another area of advice that gynecologists or obstetricians give to their patients?

    DR. LOCKWOOD:  As an example, I think folic acid is a good one.  We recommend that all pregnant women take a milligram of folic acid.  We recommend that, in fact, you begin the consumption of folic acid really during your childbearing years and certainly if you anticipate being pregnant in the near future since folic acid reduces the risk of neural tube defects only if the exposure occurred prior to the development of neurulation in the embryo.  That is a message that seems to be working.  The prevalence of neural tube defects is dropping.  Enrichment of the food supply may have helped as well, but it is clear that most pregnant women currently begin folic acid supplementation prior to conception.  So, there is an intervention that works.  It is cost effective.  It has reduced a really terrible birth defect, and it is done in a way that doesn't invoke abortion or other things that society doesn't like to discuss.

    DR. MILLER:  Thank you very much, Dr. Lockwood.  We are going to have a break now.  Would you please return in 20 minutes?  That makes it 11:05.  Thank you.

    [Brief recess]

    DR. MILLER:  Before we go to the next speaker, an issue has come up.  Over the last two days we have been receiving enormous amounts of information with sets of numbers derived from different sources, and so on, and Dr. Kuzminski has a recommendation which I think we would follow in order to clarify some of this.

    DR. KUZMINSKI:  Thanks for bringing it up, Dr. Miller.  I am just reflecting a little session three or four of us had after Dr. Miller adjourned us yesterday afternoon, but I know I certainly, personally, would find it helpful as part of the total information flow if I had in front of me--and perhaps the rest of the committee would find it helpful too--something that we could get from the FDA on one piece of paper, the various levels--we hear 0.1, 0.3, 0.4 from the various sources and agencies, sources for those numbers--how that translates into blood level mercury; how that translates to hair level mercury; how that translates to consumption of perhaps high level fish, low level fish, medium level fish.  Again, just as a piece of information, one part of the jigsaw puzzle as I mentioned yesterday, of the total information flow to help us have deliberations.

    As I keep going back to the five questions that the agency has posed to us, the charge and the questions, I think that one-page summary would be helpful.  I have been making notes.  I am not sure that they are correct and I just want to avoid any chance of error.

    DR. MILLER:  What I am going to do is to ask the secretariat to get together with our FDA colleagues and provide us with as much of that information as they can, and get this distributed to the committee before tomorrow.  Yes, Robert?

    DR. RUSSELL:  Included in that, there was a translation being made between parts per million of fish and the 0.1, 0.3, 0.4.  I would like to see that as part of the column as well.  By the way, it was never explained exactly how that is transformed, parts per million into 0.1 ug/kg/day.

    DR. MILLER:  I think it was based on consumption.

    DR. RUSSELL:  Assuming 8 oz, a half a pound of fish per serving?

    DR. MILLER:  Well, that ought to be specified.

    DR. RUSSELL:  Yes, that is what I would like to see, if we could.

    DR. BOLGER:  Since I am the one who is going to have to do this, I want to make sure I understand what you want.  I mean, I understand how this gets terribly confusing because you have these different terms that are used and it is unclear what they really mean, but they are really different terms for the same thing.  So, I am trying to get a handle on what it is you want.  I mean, trying to translate an ADI, MRL, RfD, TDI, whatever you want to call it, those are all different terms for the same thing.  They are safe levels of exposure.

    The FDA's ADI is really not relevant for this consideration because the ADI was based on the adult endpoint; it was not based on the fetal endpoint and was never part of the consideration of this advisory because the advisory is focused on fetal sensitivity.  So, that is really not germane.  The TDI WHO is also based on the adult endpoint; it is not based on fetal.  So, that too is not germane to the advisory.

    Translating a dose level, a safe dose level to a concentration in fish is all predicated on what residue level you use in the fish and what level of consumption you use to derive an estimated exposure.

    DR. MILLER:  I think that this will take a real long time and should be helpful, not confusing.  Can I suggest that you and Larry Kuzminski get together for a couple of minutes and see if you can clarify what it is.

    DR. BOLGER:  Right, so I have a clear idea of what it is they want.

    DR. KUZMINSKI:  I would be glad to do that, but I would invite anyone else on the committee to join us.

    DR. MILLER:  I don't want you to run out now.

    DR. BOLGER:  Well, can I leave?

    [Laughter]

    DR. MILLER:  You not only can leave; you have to leave.  Our next speaker is Dr. Diana Zuckerman, of the NCPR, to talk about their recommendations.

National Center for Policy Research for Women and Families

    DR. ZUCKERMAN:  Thank you very much.  I am Dr. Diana Zuckerman.  I am president of the National Center for Policy Research for Women and Families.  I am really delighted to be here and I thank you for inviting me to speak.

    CPR for Women and Families is a nonpartisan, nonprofit organization that reviews scientific and medical research, and explains the implications of that research for public policy and for the health and well being of women, children and families.  Our mission is to ensure that research information is made available and understandable for policy makers and the public, to support policies that benefit public health, and to help ensure that consumers can make educated choices.

    Just as an aside, my own training is in epidemiology and psychology, and my policy perspective comes from working in the House of Representatives, the Senate, briefly the White House and, most recently, working for and with many nonprofit organizations that are very focused on consumer issues.  This is a combination that doesn't come in too handy very often, but I think might actually be the right combination for what I am going to talk about today.

    Our Center is very concerned about methylmercury exposure, especially for children and pregnant women.  We believe that the Food and Drug Administration's current efforts at protecting the American public from the health risks of methylmercury are not adequate to protect the public or to educate them so that they can protect themselves.

    We are concerned because the FDA does not adequately monitor methylmercury levels in commercial fish supplies.  These levels may change over time but the FDA does not collect data to determine if that is true so we think surveillance is really essential and needs to be improved.

    We are concerned because the current FDA advisory is incomplete.  The advisory should be revised to include information about tuna.  Although the levels of methylmercury in tuna, and especially canned tuna, are lower than in other fish that are included in the current advisory, the amount of tuna consumed is typically so much higher that a public health perspective requires that the FDA widely disseminate risk information about fresh tuna and canned tuna.

    Our third main point is that we are concerned because FDA's dissemination of information about methylmercury exposure has not reached most consumers.  Even health-conscious consumers are unaware of the overall risks of methylmercury in fish and they don't know which fish pose the greatest problems.

    More than a year and a half ago I attended a small meeting with Joe Levitt and other consumer groups to talk about FDA's plans regarding methylmercury advisory.  It was really an excellent meeting and the major focus of the discussion, as I recall it, was whether the FDA should include information about canned tuna in their advisory.  It seemed to be already assumed that fresh tuna would be included.

    Most of the consumer groups strongly urged that canned tuna be included in the advisory and we spent a lot of time talking about how to make that information available to consumers; what do you need to do to make sure that consumers know not just about tuna but about all the fish involved in the advisory.  We talked about the need to have labels on food that is sold in supermarkets and fish markets, and also about the need for information on restaurant menus.

    So, I was certainly very disappointed and actually extremely surprised when the advisory came out and it didn't mention tuna at all, and very disappointed at the lack of dissemination of information once the advisory came out.

    It seemed to us that FDA was making little or no effort to inform consumers of these risks at the time when it would do the most good, which is when they are buying fish when they are in the market, in the supermarket or ordering it in a restaurant.

    Based on the National Academy of Sciences report, we are convinced that the FDA should be doing more to protect vulnerable populations, as you know, pregnant women, women who might become pregnant, nursing mothers and young children.  In our experience, the National Academy of Sciences is actually pretty cautious.  So, when they suggest that 60,000 newborns each year might be at risk for neurological problems due to methylmercury we take that estimate very seriously even though we understand very clearly that it is just an estimate.

    Given my training in epidemiology, I am really very interested in data and I strongly believe that we need better data, and it would certainly be preferable to be making these kinds of decisions based on better data.  The American public relies on the FDA to require or to collect data so that we will better understand the risks of exposure, in this case exposure to methylmercury in the fish that we eat or the fish that we want to eat.

    It seems to me that the epidemiological research suggests that methylmercury in fish can potentially pose very serious problems especially, of course, in the developing fetus.  But there are two ways to be exposed to this.  One would be to consume fish that are high in methylmercury.  That is what is in the advisory right now.  But what is less understood is what happens when women consume large amounts of fish that have moderate amounts of methylmercury, fish like tuna.

    My feeling is this is not rocket science.  We can get just as overweight from eating lots of light ice cream as we can from eating a smaller amount of Haagen-Dazs.  So, it is an issue of not just which fish but how much fish people are consuming and, obviously, tuna is a popular fish.

    We believe tuna should be included in the advisory because Americans eat a lot of tuna and American women eat a lot of tuna, both canned tuna and fresh tuna.  We looked at the government data that you all have in your book.  We looked at even the U.S. Tuna Foundation's estimates that the one percent of women that eat the most tuna eat almost 7 oz a week.  Then, there were other estimates that were for 8 oz or more.  Then we were wondering what about the top half of one percent, how much tuna are they consuming?  That is still a lot of women in this country that are being exposed to potentially harmful levels.

    Canned tuna is a convenience food because you can buy it now and eat it almost any time, and because almost anybody in the United States can afford it.  Just last week, the CVS stores offered cans of brand-name tuna for 44 cents a can.  There just aren't that many main courses in America that you can buy for that amount of money.  Most of them, including other favorites for children like hot dogs and bologna, are perceived to be rather unhealthy.  So, this makes canned tuna especially appealing to low income women, including pregnant women and the mothers of young children.

    If you believe, as we do, that an FDA advisory should reflect the science, then I think you will agree that a very popular fish such as tuna needs to be included in the advisory.  If scientists at the FDA believe that the level of methylmercury in canned tuna is not sufficient to warrant being included in the advisory, it seems to us they need to prove that.  They need to prove it by providing current data.  It is not enough to say that the evidence is unclear; they need to collect the evidence that will either support or refute those concerns.

    The current FDA advisory is entitled, quote, an important message for pregnant women and women of childbearing age who may become pregnant about the risks of mercury in fish.  At the bottom of the page it mentions, almost casually, that, quote, it is prudent for nursing mothers and young children not to eat these fish as well.  I think it is obvious that those warnings deserve more attention than they have in the current advisory.

    We all know that most foods have risks and benefits.  We know that cheese provides calcium, but it can also be high in fat.  Juices can provide vitamins, but pediatricians warn parents of young children to limit their juice consumption.  So, it makes sense to us to include both sides of the issues, the benefits and the risks, for fish as well in an advisory and in any kind of information that is being disseminated.  But we also share your concerns that it is difficult to do this succinctly and in ways that consumers can understand.

    So, we think that it is important what is in the advisory, but also how that information is disseminated to the general public.  Obviously, the advisory needs to reach more people.  For example, how many people know that the FDA believes that pregnant women and women who might become pregnant should limit their consumption of cooked, store-bought fish to an average of 12 oz a week?  I would have to say I have asked several health-conscious, fish-eating types of people and none of them were aware of it.  In fact, I asked a pediatrician who teaches at a major university and is active in the American Academy of Pediatrics, a man who does not eat red meat and considers himself very health-conscious, and happens to be married to me, and--

    [Laughter]

    --he barely knew what methylmercury is, and he certainly knew nothing about the 12 oz limit for pregnant women.

    What efforts has the FDA or the fish industry made to get information from the FDA advisory into women's magazines, parenting magazines, or other publications that are read by women of reproductive age?  Why not have PSAs on TV talking about this issue?  Because if they are on TV, they are going to reach a lot more women than anything else that the FDA could possibly do.

    Since the FDA advisory states that fish consumption should be limited to an average of 12 oz per week for pregnant women and women who might become pregnant, why not place that information directly on menus and all fish products that are sold, including canned tuna?  Well, I think we know why not.  This is a very controversial issue.  But we can't let public policy be dictated by concerns about controversy.

    In the absence of complete information about methylmercury contamination of fish, we believe that the FDA should warn vulnerable populations not to eat swordfish, shark, king mackerel, tile fish and fresh tuna since they have previously been shown to contain unsafe levels of methylmercury.

    We believe that the risks outweigh the potential benefits since at risk consumers can simply switch from eating one kind of fish to another kind of fish.  Fish are healthy but there are many different kinds of choices of fish.  We also believe that consumers should be advised to limit their consumption of canned tuna.

    We realize that companies are concerned that people will stop eating tuna, particularly canned tuna, if there was such an advisory.  I looked at the data from the focus groups that was on the CD disc that you all received, and obviously those focus groups were quite persuasive in saying we want simple information--is the fish good or bad?  We want to know which fish are good.  We want to know which are bad and we want to not eat the ones that are bad.

    But as somebody who has worked on research for a long time, and those of you who have know focus groups are not necessarily the best source of information about what people actually do in their real lives, particularly when a focus group is first given information that might be shocking or disturbing, they are likely to respond in a more radical way than they would as time goes on.

    Let's face it, millions of Americans still eat hot dogs despite warnings about nitrites, and remember when there was a time when it seemed like nobody was ever going to eat a hot dog again.  We eat processed foods despite warnings about salt content.  We still eat fresh fruits and vegetables, not as many as we should, despite warnings about pesticides.

    I think there would be an initial shock if there were labels on food and it would probably result in lowered consumption for a short period of time, but over time people learn--they will learn with fish, just as they have learned with other food products, that moderation is the key; that there are certain foods that might not be for pregnant women but that doesn't mean that other people can't eat them, and that there are certain foods where the levels are such that eating in moderation is important and they shouldn't overdo it.  We certainly managed to get that message out on a lot of other foods, whether it is ice cream or alcoholic beverages or cakes or cookies, otherwise our kids would be eating nothing but cakes and cookies I think.  So, there is no reason to think that in the same way that we teach people about moderation in all kinds of foods we can't teach them that about canned tuna.

    DR. MILLER:  Five minutes.

    DR. ZUCKERMAN:  Five minutes?  Fine.  When consumers purchase swordfish, shark, king mackerel, tile fish and fresh tuna that is either prepackaged or packaged at a fish counter, we believe the package should bear a label that tells pregnant women, women who might become pregnant, nursing mothers and young children not to eat it.  At restaurants, a similar warning should be on menus if those items are served.

    Labels are also needed for canned tuna, and we urge that the FDA advise vulnerable populations to consume canned tuna infrequently until a more comprehensive analysis of the methylmercury content of canned tuna can be performed that is more reassuring than the information we have now.  Obviously, we think that analysis should be done as soon as possible.

    Our recommendations are consistent with the precedents that have been set to provide clear warnings for pregnant women and children, even in the absence of data establishing a specific risk.  I think the most obvious example would be the warnings on alcoholic beverages.  There are warnings on every alcoholic beverage that pregnant women should not consume them, even though there really are not data that support the idea that no pregnant women should ever drink any alcoholic beverage.

    My final brief remark has to do with regulatory standards.  We urge the FDA to set a regulatory limit that is consistent with the EPA standard since that has been scientifically justifiable for the protection of public health according to the National Academy of Sciences.  We strongly believe the FDA should monitor the levels of methylmercury in shark, swordfish, king mackerel, tile fish and fresh and canned tuna, and if problems arise and with this monitoring they find that the levels are higher than expected, we believe that those fish should be removed from the market.

    Thanks very much for the opportunity to speak with you today, and I am happy to answer any questions.

Questions of Clarification

    DR. MILLER:  Thank you.  Comments or questions?  Dr. Dwyer?

    DR. DWYER:  I wonder if you know of any advisories in Scandinavia or other countries that have taken actions on this, and the effectiveness of those advisories.

    DR. ZUCKERMAN:  I am not familiar with that.  You know, I don't have that information.  We do know that there are all kinds of advisories but people get a lot of their information from the media, and I think it is very unfortunate that the FDA across the board does not make better use of the media in getting the word out on a wide variety of issues but, in this particular case, I think, you know, people do care about what they eat and I think that magazines and newspapers and TV would care about this.

    It is a little scary working with the media on issues like this because you are so afraid you are going to get these extreme headlines and these very extreme statements so you just have to keep at it until the message comes out that is accurate.  I am not saying that is easy, but I am saying it is possible, and I also have no doubt that the folks who sell these products will do their very best to promote them and that will counterbalance the warnings.  So, I think that we can get to a point where consumers are reasonably educated.  They are not going to perhaps understand all the nuances, but they will be reasonably educated and will be able to make reasonable choices and we will have provided information to enable them to do that.

    DR. APOSHIAN:  Ten years ago our laboratory did a study at the University of Arizona where we studied students as far as the urinary mercury was concerned for, I think, it was a 24-hour period.  When we got the results in we had three outliers, about ten percent at least, maybe a little more.  An outlier is always a problem.  You want to know why.

    So, when I talked with each of these outliers, and they happened to be women, young college women, I asked them what they had had for dinner or lunch, what they had eaten because part of receiving their honorarium was signing something that said they had not eaten seafood for a week.  That was spelled out, seafood.  Each of them said they had had a can of tuna fish for lunch.  And, I said, "how could you do that?  You accepted our honorarium; you signed an agreement."  And, they said, "we just didn't think of tuna as seafood."

    Now, in much of the literature that we have, they talk about swordfish and they talk about fish, and when I did my little survey on how much mercury was in tuna fish cans, I went to each can of tuna fish sold at the supermarkets in Tucson and nowhere does the word "fish" appear on a can of tuna.  My wife, when I talked this over with her, and I should even mention this, my wife, who is a 40-year collaborator of mine, said "everyone knows that tuna is fish."  But I wonder with the educational level in our country--

    [Laughter]

    --how many people really know that tuna is fish, and I would really like to suggest that it is very easy for a company to put the word "fish" on a can of tuna.  I think that there is a real need for education, as I am sure you all realize, for this whole problem of mercury in all kinds of fish.

    DR. MILLER:  I hope that that experience doesn't reflect the quality of your students--

    [Laughter]

    DR. APOSHIAN:  I must say that one of these young women was a graduate student, a year away from her Ph.D. and she is now a department head of the leading biotech company in the world, a very educated woman.

    DR. ZUCKERMAN:  I know you didn't exactly have this question but I do want to follow-up on the issue of outliers because I think it is really important.  I will admit to something that I probably never said to anyone in my life, and that is that when I started graduate school I only knew how to make four things pretty much, hamburgers, hot dogs, tuna noodle casserole and tuna salad.  So, there are people who eat a lot of certain foods, particularly students who are certainly of reproductive age.  So, there are outliers out there and we have to care about the outliers and if we don't have warnings that warn those outliers, there are people who can be harmed.

    DR. MILLER:  It just occurs to me, as we were talking before, I think the real problem is to translate knowledge into action, and that is where the hard thing comes because very often from a number of studies that have been done over the years, people, particularly students, can be extremely knowledgeable about the subject but their behaviors are unchanged.  I don't know the solution to that problem; I am not sure anybody does.

    DR. ZUCKERMAN:  Yes, I agree but I guess I don't think it is the role of the FDA or even the responsibility of the FDA to do that.  I think it is the role of the FDA and the responsibility of the FDA to make products safer by regulating and to get information so consumers can make the right choices.  I think the FDA should do everything it can to make that information available in a way that people can use, but ultimately it is our decision.  We make a lot of foolish choices in how we eat and how we live our lives, but it shouldn't be based on lack of information.

    DR. MILLER:  Other comments or questions?  Yes?

    DR. HOTCHKISS:  FDA has in its arsenal of public health tools a number of avenues that it can pursue, along with what at least are naturally occurring or adventitious toxicants--one could take aflatoxin, for example, that FDA has chosen to take regulatory action for above a certain limit--rather than the way it has chosen methylmercury which is more to try to advise consumers about the safety or lack of safety.  What would your organization's opinion be about what choice FDA has made in the case of methylmercury?  Would you rather see something more along other adventitious toxicants, lead for example, or do you think they have chosen the right path in terms of warnings to consumers?

    DR. ZUCKERMAN:  I am not sure I understand the question.  I will answer the best I can, and please follow-up if I miss the point here.  Obviously, we think that first of all you deal with advising consumers.  We also think that the regulation has been inadequate.  We think that there is more information needed and more action needed if the information provides clear concerns about toxicity levels.

    I guess the bottom line is it is not enough to have a two-page piece of paper that says this is what the FDA believes, and the FDA did have something in their magazine that has nice illustrations but, still, I think it is clearly at the college level, college reading level and, anyway, who reads that?  So, I think that the FDA has really not gotten the word out.  I mean, people do know a lot more about lead, for example, and it is much more rigorously regulated in a variety of ways but I think some of those are state and local decisions, I believe, not just FDA decisions.  I don't know if I am responding adequately.

    DR. HOTCHKISS:  I think you are but let me be more specific.  In the case of lead, for many years FDA had a very rigorous research program and in the days when there was lead used in manufacture of cans, rather than going out and telling select groups of consumers at large to limit their consumption of certain canned foods that they knew would be high in lead, they chose to regulate product and not try to have the consumer self-regulate.  They could do that, I presume, in the case of methylmercury and I was just wondering about your opinion about whether they should take that approach or the current approach.

    DR. ZUCKERMAN:  Sure, I agree.  I mean, the preference would be to regulate the product and make it safer.  I guess there are situations where a product can be safe for most people but not be safe for pregnant women or nursing women for example.  Certainly, FDA has gone in the direction of putting a lot of responsibility on the consumers.  Our organization feels they have gone too far in that direction, and I guess this is just one example of it where, yes, it is nice for consumers to have choices, but in this country people believe that the FDA makes sure that products are safe.  So, no matter what you tell them, there is this underlying belief I think that if this product is being sold in my supermarket it is safe, and if these pills are being sold in my drug store, they are safe but, unfortunately, that is not always the case because they can be safe for some people and not others.

    So, I agree that the FDA should be doing more to regulate the product, but if there are reasons why certain products can be healthy and good for many people but not for pregnant women, then I think it is okay to have warning labels but they have to be really clear ones so that pregnant women, nursing mothers and mothers of young children know that.

    DR. MILLER:  Dr. Friedman?

    DR. FRIEDMAN:  I want to ask you to help me resolve a cognitive dissonance.  Okay?  You were emphasizing the importance of more research surveillance data and for more information about the connection between exposure and fetal development and infant development.  So, you are saying there isn't enough knowledge out there about the ill effects.  On the other hand, you are speaking with great enthusiasm about enforcing regulations maybe or advisories that are much more stringent than the ones that are out there, and making it more known to greater segments of the population.  So, how does this fit together?

    DR. ZUCKERMAN:  That is a fair question.  Let me clarify, my concern about surveillance is actually not so much about lack of epidemiologic research.  I mean, there is epidemiologic research; it is progressing.  There is some, I think, very clear data that methylmercury exposure is dangerous to children, to fetal development under certain conditions.  What we don't know is exactly what conditions.  But I think it is very clear that it is dangerous.

    So, when I was talking about surveillance I was really talking about the levels in fish, current levels in fish.  I think that a lot of the data are based on older information and we need current information.  So, when I was talking about surveillance, that is actually what I meant.

    DR. FRIEDMAN:  So, you think there is enough in the mixed literature that exists now to warrant the recommendations that you made?

    DR. ZUCKERMAN:  Yes.  I mean, I would personally like better data.  I don't want to say that I think the data we have is adequate.  You know, I am a researcher by training and I always want more data.  But I think that there is enough.  I think the previous speaker said something about erring on the side of caution and I think there is every reason to err on the side of caution in a situation where there are other alternatives that pregnant women can eat.  You know, if this was the only fish available and fish was so great nutritionally for other reasons, that would be different.  But there are other fish available.

    DR. MILLER:  Any other comments or questions?

    [No response]

    Thank you.

    DR. ZUCKERMAN:  Thanks very much.  I have copies of my statement which I can leave.

    DR. MILLER:  We are now supposed to break for lunch.  We are scheduled to come back at 1:30.  I would ask you, if you would, to come back about one o'clock because we would like to get Dr. Anderson in this afternoon and we need to make up some time in that respect.  Thank you, all, very much.

    [Whereupon, at 11:45 a.m., the proceedings were adjourned for lunch, to reconvene at 1:00 p.m.]

A F T E R N O O N  P R O C E E D I N G S

    DR. MILLER:  We are ready to start the afternoon session.  I want to remind the speakers that it is important that they keep exactly to their time so everybody has a fair shake at making their presentations.  I will provide a five-minute warning before the end of their time.  The first speaker this afternoon is Mr. Richard Wiles and Ms. Jane Houlihan.  Ms. Houlihan instructed me to point out she doesn't have a doctorate and she didn't want me to continue calling her doctor.  I want you to understand it was her idea, not mine.

Environmental Working Group

    MR. WILES:  Thank you.  I am just going to read a little statement and then Jane will make the bulk of the technical presentation.  My goal is just to give you the goals of our presentation and our ongoing work on mercury.

    I am Richard Wiles.  I am senior vice president of the Environmental Working Group.  EWG is a nonprofit environmental research advocacy group, with offices in Washington, D.C. and Oakland, California.  We are entirely foundation funded.  We have no members and accept no money from industry or government.

    EWG has a long track record in working with the pesticide program at the U.S. EPA on issues directly relevant to the issue before the committee today.  We have developed and presented two probabilistic exposure models to the EPA scientific advisory panel, one for acute exposure to organic phosphate insecticides and another for chronic exposure to arsenic by arsenic-treated lumber.  Both of these models were embraced by the EPA and have formed the basis for significant changes in the way that EPA assesses exposure and risk to these types of substances.

    The acute exposure model for OP insecticides is particularly relevant to today's discussion.  This model has moved EPA away from regulating pesticides on the basis of average exposures or even fixed point statistical estimates such as the 98th percentile to risk assessment methods and regulatory policies for non-cancer health risks, such as those we are dealing with today, that are designed to identify and predict, in EPA's case, at least 99.9 percent of the most vulnerable population.

    We are deeply troubled by the FDA's antiquated exposure and risk assessment model for methylmercury, and the fact that these models and methods have produced a mercury health advisory for pregnant women that allows thousands of unborn children to be exposed to unsafe mercury levels each year.

    But let me be clear.  We are not asking the FDA to adopt our model per se any more than they should adopt the models of the seafood industry per se.  What we are saying is that the FDA needs to conduct its own probabilistic risk assessment through a public and transparent process and issue a comprehensive list of fish that women should avoid during pregnancy and, equally important, and I want to emphasize this and I am going to emphasize it over and over again, a list of fish that are low in mercury and high in omega fatty acids that women should eat more of during pregnancy.

    The charge to the committee was to determine whether the agency's consumer public health advisory on methylmercury is adequate to protect the health of those who follow that advice.  That is a quote from the charge.  Our 2001 report, "Brain Food," addressed this question exactly and found, without question, that the current advice is not safe for those who follow it.  Indeed, we found that if FDA's advice were followed nationwide one out of every four pregnancies would be exposed to a maternal blood mercury level above the NRC recommended level for at least one month.  That is a quarter of all pregnancies.

    Let me make clear our goals.  We want the FDA to, one, adopt the NRC, NAS blood level for methylmercury.  We know that is beyond the charge of this committee today but we mention it because it is critical.

    We want FDA to conduct a one-time sample of the top 40 or 50 most consumed fish, particularly those where they have very limited data, so that the agency is operating from a position of knowledge when advising pregnant women on fish consumption.  This testing should include important seasonal sport fish that can be a significant source of mercury for thousands of pregnant women.

    They should conduct and make public a state-of-the-art Monte Carlo style exposure and risk assessment of fetal mercury exposure.  They should issue a mercury health advisory that protects 99 percent of the pregnant women at least for methylmercury while, at the same time, recommending fish and other foods that are low in mercury and high in omega-3's.  This second step is critical because it is the list of safe fish that makes possible a truly comprehensive list of fish that pregnant women should avoid.

    We are not asking for tolerance of methylmercury nor a lowering of the mercury action level.  The action level is not enforceable; it is not enforced; and it would do as close to nothing to protect pregnant women and their children from mercury as any action the government could take.

    Above all, we are asking the FDA to change the way that it looks at fetal risks for methylmercury exposure.  All the pregnancies at risk for methylmercury occur above the 98th percentile of exposure, as Jane will explain.  But this is a huge number of pregnancies, as many as 400,000 per year.  FDA has been overlooking these exposures and the seafood industry has been more than eager to support this policy.

    This attitude and approach is no longer acceptable but, more importantly, it is no longer necessary.  There are sufficient data on mercury levels in canned tuna, tuna steaks and several other fish to support their addition to the FDA methylmercury health advisory.  For scores of other important fish, the FDA only needs better data on fish contamination with mercury to issue a sound advisory.  The agency stopped monitoring seafood for mercury in 1999, as I presume you have heard, and we presume that everyone would agree that this program needs to be restarted and that the FDA needs to monitor the seafood supply for mercury.

    With better data on mercury contamination of fish, the FDA could monitor mercury exposure for nearly all important fish with reasonable confidence at the 99th percentile of exposure and above, enough confidence to support a strong protective mercury advisory for pregnant women and to produce a list of fish that are low in mercury and safe for consumption during pregnancy.

    A protective mercury health advisory for pregnant women is not about banning tuna consumption, although we feel quite strongly that pregnant women should avoid canned tuna.  And, it is not about denying women the benefits of omega-3 fatty acids.  It is about developing a health-based standard for mercury exposure that protects all the nation's unborn children from unsafe levels of mercury, while providing nutritional, low mercury alternatives for these same women.

    The seafood industry and their consultants would love this, an argument about the future of the fish industry and about how ill-informed consumer advocates want to deny women and children the proven health benefits of fish consumption.  Nothing could be further from the truth.  Women should eat more fish.  There is no doubt about that.  At the same time, women should be protected from mercury.  There is no doubt about that.  What the seafood industry would have you believe is that you can't have it both ways.  You either have no fish or suck it up and eat the mercury.  It is very effective PR and it is simply not true.

    In the midst of this onslaught, the FDA rightly sees itself with two obligations, to protect the fetus from methylmercury exposure and to provide those developing babies with the best nutrition possible.  The problem is that the agency has adopted essentially the industry view that these two goals are mutually exclusive.  Until today or this meeting, the agency had essentially given up trying to protect women from mercury in seafood.  They stopped monitoring seafood for mercury in 1998.  They have not adopted the NRC recommendations for mercury levels in blood, and they have not provided a sound rationale for not doing so, and they have not been able to provide the Congress, or anyone, with a physical, actual copy of a risk assessment underlying their health advisory.

    In the end, the agency limited its methylmercury health advisory to a short list of fish based on antiquated science and a misplaced fear that a longer list would deprive women of the nutritional benefits of fish.  Yet, at the same time, the agency has done nothing to help women identify the fish and other foods that are low in mercury and high in omega-3 fatty acids.  This is particularly ironic because it is precisely the creation of a good fish list that would provide the nutritional rationale for a comprehensive list of fish to avoid during pregnancy.

    When Jane and I met with Joe Levitt and his staff, he asked us what we wanted and I will tell you today what I told him then.  Don't laugh--we want a refrigerator magnet or a card a woman can put in her wallet with two lists of fish on it, one, a list of mercury-contaminated fish that women should avoid when pregnant and, two, a list of fish that are low in mercury and high in omega-3's that pregnant women should eat more of.  Very simple.  This is not a complicated problem.

    We then want the FDA to work with healthcare providers, not the seafood industry, to ensure that every pregnant woman in America gets these simple lists, just like lead.  No regulations.  No tolerances.  No bans on any fish.  Just the best information based on the best science available on how pregnant women can protect their babies from mercury and provide them with good nutrition at the same time.  Thank you.

    MS. HOULIHAN:  I have some copies of the presentation I am about to give.  Good afternoon.  As Richard mentioned, I want to share with you, maybe not so briefly, some of the analyses that we have done regarding methylmercury exposures in women of childbearing age.  I understand you had some presentations made yesterday that also looked at this issue so some of this should sound a little bit familiar.

    [Slide]

    First of all, I want to talk briefly about what led us to do this work.

    [Slide]

    As you have all heard by now, on January 12, 2001 FDA issued its first consumer advisory to pregnant women regarding methylmercury in seafood.  That advisory contained two main points.  First of all, pregnant women should completely avoid eating four fish, shark, swordfish, king mackerel and tile fish.  The second part of that recommendation is that pregnant women can safely eat up to 12 oz of any other kind of fish throughout pregnancy.

    We have done work on methylmercury in the past and we are familiar enough with the mercury data to know that there are some fairly high mercury fish that did not appear in FDA's advisory.

    [Slide]

    So, we asked ourselves the question at that time what would happen if a pregnant woman followed FDA's advice and actually did eat 12 oz, or about two servings of fish a week, except for shark, swordfish, king mackerel and tile fish.  The pertinent question is would her mercury exposures exceed the safe levels that have been derived from available epidemiology studies, levels recommended by the National Academy of Sciences and applied by the Environmental Protection Agency for example.

    [Slide]

    I am going to jump straight to the answer to that question before I give more details of the model.  The answer is, and I will explain the graph in a second, that the model shows that if pregnant women followed FDA's advice and ate 12 oz of supposedly safe fish a week, and ate that in normal national consumption patterns so they are far more likely to eat canned tuna, for instance, than any other kind of fish, more than a quarter of all pregnancies would be exposed to mercury at levels above the reference dose for at least a full month of pregnancy.

    The graph is a little complicated but, of course, the X axis is total time during pregnancy.  That is any three-month, any four-month, five-month chunk of time during pregnancy.  On the Y axis we have the percent of babies exposed in utero to methylmercury levels above the reference dose for that given time span or longer.

    For instance, if you look at one month on the X axis and go up through the graph, you are seeing about 28 percent of all pregnancies that would be exposed to methylmercury above the reference dose for at least one month of pregnancy.  You also see maybe midway through pregnancy, say, four and a half months, that you are almost at one in ten pregnancies being exposed above the reference dose for half their pregnancy if women actually ate 12 oz of fish and followed FDA's recommendation.

    I will get into the details of that model and how we produced that answer, and also in this presentation I will focus in on some other examples.

    [Slide]

    This is really the outline of my talk.  First I will talk about our modeling method and go into more detail on the results and how those were produced.  Second, I will describe some of the underlying biological parameters in the model.  That would be things like how a woman absorbs and excretes mercury.  Third, I will talk about fish consumption and now much fish do women really eat.  Of course, there aren't a lot of women eating 12 oz of fish a week but I will look at proportions of women who do eat a lot of fish, based on national databases.  Fourth, I will talk about methylmercury exposure, in particular focusing on canned tuna because it is the top seafood eaten in the U.S. and a lot of high seafood consumers focus pretty heavily on canned tuna.  Fifth, I will highlight some other examples from real-world data user examples that are found in government and market company databases of real women reporting large amounts of fish that were actually eaten during different periods in their life.  Last, I will just sum up what we would like to see come out of FDA on the subject.

    [Slide]

    Modeling approach, the underlying structural technique in this model is called a Monte Carlo method.  It is a probability method.  What it is really good at is allowing for an accounting of biologic variability.  Differing individual consumption patterns and ranges of mercury concentrations in seafood are all incorporated into the model.  So, compared to, say, looking at an average woman paired up with a high end consumption, this model lets you look at the full range of exposures across the population accounting for the fact that people are different; they eat in different ways; and they are eating seafood that varies widely in its concentrations of mercury.

    Second, the underlying mathematical representation in this model, is a non-steady state or transient one-compartment pharmacokinetic model, developed and verified--one place it is presented is Dr. Gary Ginsberg's work from the Connecticut Department of Public Health in peer reviewed literature.  It is a model that looks in particular at mercury concentrations in blood and how they vary with time as a woman eats seafood.

    How do we vary the way that women eat seafood?  We account for biologic variability in a couple of different ways.  We rely on data from CDC's National Health and Nutrition Examination Survey, or NHANES, and we also rely on studies of absorption and excretion of methylmercury that appear in the peer reviewed literature, which were summarized in a paper written by Alan Stern, of the New Jersey Department of Environmental Protection, that appeared in peer reviewed literature in 1997.

    [Slide]

    The model also accounts for the different ways that people eat seafood, different consumption rates.  The scenario I showed you first, which is everybody is eating 12 oz of fish a week, is a hypothetical scenario.  But today I will present data that is based on real consumption data from national databases that include low fish eaters and high fish eaters as well.  We used two different databases to do this.  One is CDC's newly released 30-day recall seafood consumption study that includes over 1000 women of childbearing age and what they ate in the past 30 days.  We use that in combination with the National Eating Trends database from a major market survey organization, called the NPD Group.  This database we use in particular because it is really powerful.  It contains data on almost 8000 women so it is many, many more women than CDC's recall file.

    [Slide]

    Lastly, mercury concentrations in seafood, the model accounts for the widely occurring concentrations of mercury in different fish species.  To account for that, we compiled mercury data from seven different government databases of mercury and fish tissue.  These came from FDA, NOAA and EPA, seven databases and 50,000 records altogether of numerous different fish species.

    [Slide]

    I will just talk briefly about how a Monte Carlo model works, for those of you who haven't used this technique before.  It is conceptually pretty simple.  You know, you start off any model with the exercise of what questions you want answered.  So, we are first of all answering the question what would happen to mercury exposures if pregnant women followed FDA's advice and ate 12 oz of fish a week throughout pregnancy, barring the four black-listed fish.  Would her mercury exposure exceed safe levels derived from available epidemiology studies?

    The first step in the model is that the model begins simulation.  It basically creates a woman and the model randomly assigns that woman a unique combination of body weight, blood volume, the fraction of mercury that will be absorbed through the gut and the fraction of that mercury that will be distributed to blood as opposed to what is stored in other tissues, like the brain, and an elimination rate constant which is a measure of how quickly a woman can excrete that mercury from her body.  So, that is basically the model, a person in all her associated biologic parameters that all come from measured data in the peer reviewed literature.

    [Slide]

    The second thing the model does, it assigns this woman an initial blood concentration.  In other words, women are eating seafood throughout their life and at some point, you know, a woman who decides to have a child gets pregnant and she has a starting concentration of methylmercury in her blood.  That is called her initial methylmercury concentration and that is where the model starts her off as she begins to eat seafood through pregnancy.

    The third thing the model does is it allows the woman to eat the prescribed number of fish servings through pregnancy, so in this case, if I am doing two servings, two 6 oz servings of fish a week, it will loop her through two meals per week in the model.

    [Slide]

    As it does that, it does the fourth thing, which is track the woman's blood mercury concentrations as she eats fish through pregnancy.  The program maintains statistics on her blood mercury distribution through time.  So, in other words, a woman eats the fish.  A fraction of the mercury in that fish is absorbed through the gut and is distributed to the blood.  Through the bloodstream, it is then distributed to the various tissues in the body and then is excreted through time, depending on that individual woman's excretion capacity.

    This is where the Monte Carlo part comes in.  It repeats that, in this case, 299,999 times for example.  So, in essence, it simulates 300,000 unique women and at the end of the model we can compute composite blood mercury statistics on that whole population of modeled women.

    So, you can see how this modeling technique is really powerful because you are essentially creating women in this model based on real biologic variability and some of them will be particularly sensitive to mercury, will absorb a higher quantity, excrete a lower capacity, and those women, in combination with high consumption rates, is where some of the biggest problems lie and the model takes all that into account.

    [Slide]

    When we did this exercise of two servings of fish a week, as I have already discussed, this is one of the basic answers that came out of the model, that women really cannot safely eat two servings of fish a week through pregnancy and FDA's advice.  If women actually followed it, you know, more than a quarter of all pregnant women would be in a zone that exceeds a safe level for at least a month of pregnancy.

    [Slide]

    Another thing that this model lets us do is segregate fish according to their relative safety.  For instance, I can construct fish scenarios where I say let's let a woman eat sea bass and nothing but sea bass and see what that does to the answers in the model, how women fare.  It turns out that sea bass, tuna steaks, halibut and white croaker are high enough in mercury, according to the measured data, that if women routinely ate those fish, or ate almost any of them, they would put themselves into a zone where their mercury levels might exceed safe levels.

    This chart is a little hard to read but it does segregate out by species which is one valuable thing we can see.  On the X axis is how many meals per month of that particular fish a woman would be eating, so a meal per month up to two meals per month basically on the bottom of the chart.  Then, the Y axis is the percent increase in the number of women whose blood mercury level exceeds the reference dose for more than a month of pregnancy.  So, this is fixed at one month.

    But you can see that sea bass, tuna steaks, halibut and white croaker have really high mercury concentrations relative to the fish that fall out at the bottom of the chart.  At the bottom of the chart we have things like salmon, wild Pacific salmon, farmed catfish, farmed trout, summer flounder.  They are all much lower in methylmercury concentrations and, therefore, are associated with much lower risk levels as women eat them through pregnancy.  So, the take-home message is that individual species vary pretty widely in their mercury concentrations.

    [Slide]

    I want to briefly go over some of the underlying biologic parameters in this model.  The biologic parameters that govern absorption and excretion of mercury, as I mentioned, were taken from the NHANES program and from combinations of studies that appear in the peer reviewed literature, largely summarized by Stern in a paper from 1997.

    Body weight of women is assigned at the beginning of the model.  We used data for 4935 women of childbearing age from CDC's NHANES survey.

    Blood volume is calculated from body weight.  Blood volume correlates really strongly with body weight and increases, of course, as body weight goes up.

    Elimination rate constant is a measure of how quickly a woman can excrete mercury from the body.  It follows a first order exponential decay pattern according to a number of studies in people, with a mean and standard deviation that I give here.

    Fraction of ingested dose that is absorbed--every woman has her own unique capacity to absorb mercury through the gut after she eats the fish.  The fraction of that absorbed dose is then distributed to the blood.  So, the mercury enters the blood and is distributed throughout the body, goes to various tissues where it is stored, brain and other tissues, but some of it remains in the blood and that is the concentration that we keep in the model.

    [Slide]

    This is just what the initial concentration distribution looks like.  This is data from NHANES arbitrarily separated out by age of women so you can actually see all the dots.  But there are 1645 women whose blood mercury was measured in CDC's latest NHANES data release.  One reason I post this is because it has become pretty apparent from work, in particular by Ben Raines, in Alabama, that when you focus studies on women who are frequent fish consumers, these women aren't necessarily represented in this data set.  Ben Raines tested mercury in 65 women and he found that 51 of those women exceeded the reference dose in his measurements.  So, some of them are a substantial amount higher than what is shown in this population of 1645 women.

    What we do know is that in that data set, although it is a large data set from CDC, we are not necessarily representing women who eat a lot of fish.  There are a lot of women missing from that.

    [Slide]

    Fish consumption--I just want to talk a little bit about the fish that women eat.  As I mentioned, we rely on two primary databases to look at consumption.  One is the 30-day seafood consumption recall file from CDC's NHANES program released a couple of weeks ago.  The other is the National Eating trends database that gives us data on approximately 8000 women.  I focus on that database here because it gives us so much more resolution at the tail where women are really eating a lot of fish.

    When we break that data down we find a lot of women in that tail and this is, in particular, focused on canned tuna.  So, this graph basically highlights how many women are eating a lot of canned tuna.  Let's just look at that center bar which is five or more servings of tuna per week.  This is women who eat five or more canned tuna servings per week.  When we extrapolate the data out to the population of women of childbearing age that corresponds to 52,000 women.  There are also women who even eat seven servings or more of tuna a week.  Three or more servings of tuna, 210,000 women of childbearing age, when you extrapolate this data, they are eating at least three servings of tuna a week.  If we focus on the 52,000 women that are eating five or more servings of tuna a week, I can safely say that all those women will be above the reference dose for methylmercury.

    [Slide]

    So, I did a simulation looking at real consumption patterns pared down to only canned tuna so this is nothing except canned tuna.  This is measured data from 8000 women.  They are true consumption patterns, extrapolated to represent all women of childbearing age.

    I want to talk for a minute about concentrations of mercury in canned tuna because that, of course, is an integral part of this model and the scenarios that I do focus on are canned tuna.  In our database we have 479 test results for canned tuna.  Most of these come from FDA.  FDA tested 219 samples in a special canned tuna survey in 1993.  We have 115 samples from FDA's seafood surveillance database before they stopped testing domestic seafood; 27 samples from the total diet study and we have supplemented that with relatively new data from the State of Florida with 118 samples that they submitted to FDA about the time that the advisory was issued.

    So, we have a lot of canned tuna data.  You can see that it varies widely in concentration, averaging about 0.2 but there are many individual cans of tuna that have concentrations far above that.  I understand Dr. Aposhian tested cans of tuna and had a result even exceeding what is on this chart.  So, there are a lot of cans of tuna that have very high levels.

    [Slide]

    I ran the model using real consumption patterns for women of childbearing age.  As it turns out, the consumption comes from 7319 women of childbearing age who eat canned tuna.  We find that even forgetting all other fish, canned tuna consumption alone drives high blood mercury because it is such a high proportion of seafood that is eaten in the United States.  There are about 40,000 pregnant women every year, and this is based on real consumption data so this is a picture of what is actually happening, an estimation of what is happening.  There are 40,000 women in the U.S. every year, pregnant women who exceed the reference dose for at least three months of pregnancy based on canned tuna consumption alone.  That is the bar that is associated with three months.  So, they are over the safe dose for at least three months of pregnancy.  That is 40,000 pregnancies.  That is one percent of all pregnancies in the U.S. just from canned tuna.

    If we think about women exceeding the safe dose for at least a month of pregnancy, we are up to about 90,000 pregnancies going over a safe dose just from canned tuna consumption.  So, you can see that the canned tuna consumption really drives a good percent of the risk in the U.S. in terms of women who are facing high methylmercury exposures.

    [Slide]

    I next focus in on a particular woman in a model, just a little window of what is happening in the model to the mercury in her blood.  Between days 12 and 22 of the model she had a number of eating occasions in which she ate canned tuna that had quite a bit of mercury in it.  You can see that her intake is exceeding her excretion capacity.  That is what happens in this model, blood levels can build up and then decline.  If you exceed excretion capacity, your mercury levels go higher and higher, and then maybe you eat some more mercury fish and levels decline but the model is accounting for all of that.  So, you can see her mercury level is building up through this ten-day period just because of the tuna fish that she is eating.  She is a frequent tuna consumer.  She is eating it almost every day but the same thing happens for women who are eating it less frequently.

    [Slide]

    The next thing I did was a basic reality check on the model to see if it was consistent with what CDC had measured.  The top blue line is the real world.  That is the distribution of blood mercury levels that CDC measured in women of childbearing age, and it represents--what?--11600 women.  That is the real world.  Blood methylmercury concentration is on the X axis.  The Y axis represents the percent of women who are at or below that methylmercury level.  For instance, about 20 percent of all women are at or below 2 ppb of methylmercury in blood.  That is how you would read that.

    Now, if I plot on top of that one of the answers from our model, the red line, that red line is the model results.  It is a snapshot of blood mercury levels at model day 100.  I could pick any day of the model.  This happens to be model day 100.  That is the distribution in all 300,00 women in the model at model day 100, what their blood mercury distributions looked like.

    As you would expect, you can see, number one, canned tuna consumption doesn't account for all the methylmercury in the population but it accounts for a substantial fraction.  It is consistent with consumption patterns in the U.S.

    The second thing you can see, and this is important, if you subtract one line from another, so if you take background exposures that are represented by CDC's data in that top line and you subtract out the canned tuna line and you are basically saying let's look at what happens if people don't eat canned tuna, you suddenly find the exposures are substantially lower and you are getting a much higher percentage of women who are in a safe zone throughout pregnancy.  So, the bottom line is canned tuna is a big driver of women exceeding a safe dose during pregnancy just because consumption is so very high, especially at the tail end of the data.

    [Slide]

    That is canned tuna.  That is real good consumption data.  I want to point out a couple of other examples of particular women where we know their consumption pattern in detail.  First of all, FDA conducted some focus groups in the fall of 2001 which are real interesting reading if you are so inclined, but there are a number of women in these focus groups who talked about how much fish they were eating or did eat during pregnancy.

    One woman said in these focus groups, "when I had my first son I had gestational diabetes and I was put on a very strict diet, and one of the things I could eat was a can of tuna and two tablespoonfuls of mayonnaise for lunch every day, five days a week."  She said, "so every day, five days a week for lunch, for seven weeks of my pregnancy, so for seven weeks I am eating more than they recommend."  So, she ate tuna every day for seven weeks, and that is perfectly consistent with some of the data that is in the measured databases.

    A second woman said, "I was doing it, eating canned tuna, because I was planning to get pregnant and as part of the Atkins diet you can have stuff like that, fish or chicken.  So, I ate a bag of salad, a can of tuna.  That was my lunch."  So, she was on a special high protein diet and she was eating a lot of canned tuna.

    Let's look at what the databases tell us about how consistent those anecdotes are with real-world data.  Well, in CDC's NHANES 30-day recall seafood consumption database, there is a 29-year old woman who reported eating 30 servings of tuna over 30 days.  She eats canned tuna every single day.  In the National Eating Trends database two women, ages 22 and 24, reported eating 15 servings of canned tuna over 14 days, about a can a day.  A 23-year old woman reported eating 14 servings of canned tuna over 14 days.  So, there are people for whom canned tuna is a staple.  They are eating it every day for lunch or dinner.

    [Slide]

    I took one of these examples and ran it through the model.  I looked at the focus group woman from Calverton, Maryland who said she ate a can of tuna every day for seven weeks of pregnancy.  She delivered early so that was the end of her pregnancy, but during that seven-week time, if you think of the Y axis as the percent chance that she would go over a safe dose for this given time span during her pregnancy, let's look at four weeks.  For that seven weeks of pregnancy there is about a 65, 70 percent chance that she will go over the safe dose for well over half of the rest of the duration of her pregnancy.  So, she has substantial chances that she is over the safe dose through almost all of the remaining period of her pregnancy.  That is just seven weeks.  So, for women who really are eating a lot of tuna throughout their entire pregnancy, they are getting into very high probability that they are exceeding the reference dose through almost all of their pregnancy.

    [Slide]

    The next thing I would point out is that this is women who happened to eat fish that Ben Raines tested from the Gulf of Mexico.  He tested a handful of fish.  One was an amber jack that was caught in the Gulf of Mexico and had a mercury concentration of 1 ppm.  The other was a yellow fin tuna that had a concentration of 0.83 ppm of methylmercury.  These are averages based on number of samples.  If a woman ate that fish for four servings of that fish at the beginning of her pregnancy so she is only eating four servings of fish only over four days--the time span during pregnancy is shown on the X axis and the percent chance that she would be exposed to methylmercury reference dose for that time span or longer is on the Y axis.  So, if we just look at, say, one month she is going over the reference dose for at least a month of pregnancy.

    We see for the yellow fin tuna there is a 60 percent chance she would be over the reference dose for at least a month.  For the amber jack, she would be up to an 80 percent chance that she would be over the reference dose for a full month of pregnancy.

    So, you can see these fish with high concentrations which are in the database--these fish are out there, they are being sold in the marketplace, and a very few servings of those fish will really drive a substantial percent chance that a woman would go over a safe limit for a long period of her pregnancy.

    [Slide]

    Last, our recommendations to FDA--because FDA has really such a limited consumer advisory to pregnant women, a lot of states are taking up the slack and trying to give women supplemental advice.  There are at least nine states that advise women to limit their tuna consumption, and I think Michael Bender will talk about this in more detail later.  Connecticut, for instance, you know, fresh tuna once per month, canned tuna one to two meals a week; Wisconsin, one meal per week of canned tuna, one meal per month of fresh tuna; Michigan, one meal per week of canned tuna.

    You know, FDA is advising women that they can safely eat up to 12 oz of fish a week that includes canned tuna and tuna steaks, and these states have said, wait a minute, that might not be safe.  So, they have tried to issue some supplemental advice to women to get the message out about mercury in fish that FDA does not have in their advisory.

    [Slide]

    This is a plot of omega-3 fatty acid content, a rough estimate of omega-3's in various kinds of fish versus the mercury concentrations in those fish.  The wide bars are the omega-3 fatty acid content.  The first point is that fish vary pretty widely in their omega-3 content.  This is an average level based on USDA's nutrient database for that group of fish that I show on the X axis.  Herring, sable fish, salmon, very high in omega-3's.  Scallop and clams are on the other end of the graph, generally a lot lower in omega-3's, and everything else is in between.

    On top of that you see the dark bars.  That is average methylmercury concentration in those fish.  It is interesting to note that there doesn't seem to be much correlation at all between omega-3 content and methylmercury concentration.  The good news in that is that they are not mutually exclusive and it is possible to come up with a list of high omega-3 fish that are low in methylmercury.  One really great example of that is wild Pacific salmon, which is the third fish over on the bottom.  It is very high, on average, in omega-3 fatty acids and very, very low in methylmercury.  So, that would be a great fish according to the data that we have in-house for pregnant women to eat during pregnancy.  They could safely eat it and still get the nutrients that they need.

    Let's also juxtapose a couple of examples of how many of these fish FDA has tested.  For herring, which has fabulously high omega-3's on average, FDA has eight samples.  Whitefish, high in omega-3's and you would want to be able to recommend that to pregnant women and FDA has tested two whitefish.  They have tested two bluefish; about ten of some of the species of sea bass, ten snapper, ten perch.  So, methylmercury sampling has not been sufficient, to say the least.

    [Slide]

    FDA's summary, their own summary that is on their web site--they released a summary of methylmercury concentrations in seafood after they released their consumer advisory, and it shows that their monitoring program has major testing gaps.  They focused their testing really heavily on seafood that women are advised not to eat.  For swordfish, for example, they have 598 samples listed; for shark, 394 samples.  I pointed out some of the other fish that have promising sources of omega-3's for women, and there are only two tests for whitefish, two for bluefish, eight for herring, ten for ocean perch, nine for orange ruffy.  In so many cases the mercury concentrations have not been adequately characterized that you could actually be certain that women are getting good advice if you recommended those fish for them to eat during pregnancy.

    [Slide]

    The bottom line, I just want to reiterate some of the things that Richard said.  This is what we would like to see FDA do.  We would like for FDA to adopt the NAS recommendations for a safe level of exposure.

    We would like FDA to conduct a one-time sampling program for the most consumed fish species.  For so many of these fish there is not enough data to fully characterize mercury concentrations.  We want FDA to do a major one-time sampling program of, say, the top 50 fish eaten in the U.S. to really get a grip on what is going on with the mercury.

    Third, we would like FDA to conduct and make public a risk assessment, to do a risk assessment.  We are recommending a probabilistic exposure risk assessment that would allow FDA to get really good resolution at the tail end of the data where women are consuming a lot of fish.

    Number four, we would also like FDA to issue a public health advisory that protects a very high percentage of women from mercury exposure that exceeds the NAS recommendations and that ensures women eat low mercury fish, high omega-3's.  As we saw in the canned tuna data, about one percent of women are going over a safe dose under model predictions for at least three months of pregnancy just from canned tuna consumption, and that really is what is happening based on all the data that we have put together.  Those exposures could be mitigated if FDA would give the public, particularly women, comprehensive advice on what fish to avoid during pregnancy and what fish they should eat during pregnancy that are high in omega-3's and low in mercury.

    That is all I have to say.  So, is it question time?

Questions of Clarification

    DR. MILLER:  Thank you.  Comments?  Dr. Nordgren?

    DR. NORDGREN:  I have two questions.  The first is how accurate do people estimate these 30-day recall things are for actual fish consumption?  The second question is, is there any evidence out there--if fish are like humans, there must be tremendous levels in variation in species so if you measure a shark in Cape Cod versus somewhere else, why aren't their levels fluctuating, and why only one sampling?

    MS. HOULIHAN:  Why not more sampling?  So, the first question was how accurate is the 30-day recall file.  That is a question I don't have an exact answer to, but I can tell you that it is the most comprehensive long-term survey done by the government on safe food consumption.  That is not the underlying data in the scenarios I presented today.  I used the National Eating Trends database, which records what you eat as you eat it for 8000 women.  It is a food diary that is kept for a two-week period.  It gave me so much better resolution at the tail end of the data that I used this for the canned tuna scenarios.  I think the 30-day recall file is generally consistent with the NPD data though there is a substantial fraction of women in that who are eating a lot of seafood, reporting eating a lot of seafood.

    The second question was why not test more.  We are sort of calling on FDA to do a big one-time sampling program because we think that would be a fabulous start.  And, that is the kind of underlying data that you would want FDA to have as they go through doing their own risk assessment and exposure analysis.  There are many species of fish that I was not able to include in my model because there is just not enough data.  So, for FDA really to take a good initial comprehensive look at this, they need to do a comprehensive sampling program.  I would advocate for continuing to test for mercury, but I think a big one-time testing program of a lot of samples for the top fish is a really good place to start.

    DR. MILLER:  Johanna?

    DR. DWYER:  Do you mean actually do the mercury analyses in these various types of fish, or do you mean collect data that already exist in the literature and put them into some uniform database?  Which is it?  Are you calling for the analytical work, or informatics work, or both?

    MS. HOULIHAN:  Both would be fabulous because there are a number of states who have tested some marine species that maybe FDA doesn't have data for.  But, you know, FDA should conduct their own comprehensive sampling program at the same time.  It makes a lot of sense for FDA to try to get together a comprehensive database of what has been done to date as well for methylmercury testing in seafood.

    DR. MILLER:  Dr. Russell?

    DR. RUSSELL:  Thanks for the very clear analysis that you have given.  Your analysis, of course, is based on the acceptance of the NAS guidelines, which was based primarily on the Faroe Islands study.  The other serious scientific attempt, it seems to me, to arrive at a safe level was the ATSDR study or analysis using as the lead study the Seychelles study, which came out with a level that was 0.3 instead of 0.1 as being a safe level.  Can you explain how you analyzed or what you thought of that ATSDR analysis and why you rejected it, or why it is being rejected in favor of the NAS study?  Is it because it is more conservative and you want to err on the safe side, or are there some other reasons, scientific reasons?

    MS. HOULIHAN:  You know, in the world of risk assessment or toxicity risk assessment in particular, it is pretty traditional when you are trying to protect public health to look at the study that showed measurable effects at the lowest levels and to use that as a basis for public health protection.  I think that makes a lot of sense.  Instead of choosing a study that showed a higher level, you want to do something that actually ensures a higher level of protection.

    No one is disputing, I think, that they are credible studies, the Seychelles, Faroe and all the other studies that have been conducted, but the idea behind risk assessment is to take the data as a whole, look at the most sensitive endpoint and apply that as the starting point for your reference dose.

    DR. RUSSELL:  Just to carry that one step further, I agree in general.  Another question that I would have for you is do you think that the Faroe Island study is more comparable to what happens in the U.S., or do you think the Seychelles study, with regard to dietary patterns, is more comparable?

    MS. HOULIHAN:  I think there are a lot of women in the U.S. who eat a lot of seafood.  I will say that.  I know that there were questions brought up during the National Academy study of possible confounding with PCBs, and those questions were addressed.  It is not PCBs, it is methylmercury that seems to be associated with the neurodevelopmental decrements.  There are other studies as well that show consumption patterns that closely resemble ours where we don't have those sort of episodic spikes where effects are seen in those consumption patterns as well.

    I would also say that when we are modeling methylmercury in blood we are not really reproducing the spikes.  In the model results that I showed, when you see a spike in the model, that is not the true spike.  I did neglect to point that out, that that is already the post-distributed mercury concentration.

    DR. MILLER:  Dr. Lee?

    DR. LEE:  Somewhat as a follow-up, I admire your conservative approach, erring on the safety of the women and the children, but both the Faroe and the Seychelles studies were done in a double-blind fashion so you remove any investigator bias.  But it seems to me that in this modeling study that you did you selected data based on a conservative outcome that you knew.  So, some of the model variables that you have picked, like the body weight and the blood volume and elimination, could have been influenced by the need to demonstrate the outcome.

    MS. HOULIHAN:  You are saying that I chose biologic parameters that would drive up mercury exposures?

    DR. LEE:  No, I am saying that you didn't choose your parameters in a double-blind or even in a single-blind fashion.  You were looking at data that would help demonstrate the need for action.

    MS. HOULIHAN:  I chose the underlying biologic parameters that are in the model based on what is in the peer reviewed literature and I didn't cut anything out.  Say, for instance, elimination rate constants and the volume that is absorbed through the gut, those were taken from a survey that Dr. Stern, from New Jersey, did where he considered all the available data from the peer reviewed literature and I don't think he purposely cut any data out of his analysis.

    DR. LEE:  Okay, that was a review that Dr. Stern did?

    MS. HOULIHAN:  Right, a paper he published in 1997.

    DR. LEE:  So, is that all the available information on that absorption?

    MS. HOULIHAN:  Well, that is an issue that FDA should take up as they build their own model because there have been data published subsequent to Dr. Stern's review that should also be included in those kinds of distributions.  That is an important question.

    DR. LEE:  Okay, thanks.

    DR. MILLER:  Dr. Fischer?

    DR. FISCHER:  I would I guess have a comment that your whole presentation seemed to indicate, at least to me, that you knew what level of mercury exposure was safe and what level was unsafe and that, of course, was the EPA RfD.  Except, I think we all, from  our discussions up to this point, have heard other people say that they weren't certain exactly what level is safe and what level is unsafe.

    The fact is that we really don't know where the safe level is at this point.  What we have to work with is a number that comes from a risk assessment procedure which can be used for regulation.  So, the terminology that is used, safe versus unsafe, when used in the way you are using it tells the public that they are being subjected to a lot of risk that they believe is actual.  But we know there is a lot of uncertainty.

    I just wish, for example, when you presented your data from your model that you would have shown us, instead of percent of women above a certain regulatory level, the model blood levels of methylmercury of these women, or hair levels so we could compare with actual blood and hair levels that have been seen, like the NHANES levels and others.  It would allow at least me, who looks at blood levels and hair levels, a better picture of actually how well your model is doing.

    MS. HOULIHAN:  Well, you will see in your slides that I did present a picture of the blood levels, the composite model results from all 300,000 women and compared them directly with CDC's background distribution.  So, you can look through your presentation slides and see exactly the distribution that the model predicted from real consumption patterns.

    I agree there will always be uncertainty in these kinds of problems for what level is toxic to a particular person, what is safe, what is not safe, is there a safe level of exposure but in the end, when you are doing a risk assessment and trying to protect public health you have to make decisions.  For instance, what percentage of the population are you trying to protect from what level of exposure, and you have to make those decisions in order to do the risk assessment and protect public health.  Whether that is a bright line, a fuzzy line, a grey line, you know, I don't know but right now there is not really a line; it is just don't eat sharks or king mackerel or tile fish, and that is not really doing it for keeping blood levels down to levels that the National Academy and the Environmental Protection Agency say are protective of public health.

    DR. FISCHER:  I understand perfectly the decision-making process.  It is communication that I am talking about.  Whether it is scientific or not scientific, it seems to me we should be honest and truthful about how we are presenting the information to each other and to the public.  It is a matter of semantics I guess.

    MR. WILES:  We would be glad to provide the committee with blood level distributions for any particular slide that was up here, if that is going to help clarify matters in any way.

    DR. FISCHER: