DEPARTMENT OF HEALTH AND HUMAN SERVICES

FOOD AND DRUG ADMINISTRATION

CENTER FOR FOOD SAFETY AND APPLIED NUTRITION

NUTRITION SUBCOMMITTEE MEETING

TOTAL FAT AND TRANS FAT

Wednesday, April 28, 2004

8:10 a.m.

Loews L'Enfant Plaza Hotel

480 L'Enfant Plaza

Monet Room

Washington, D.C.

PARTICIPANTS

Norman I. Krinsky, Ph.D., Chair

Jeanne Latham, Executive Secretary

MEMBERS

Susan S. Baker, M.D., Ph.D.

R. Jean Hine, Ph.D. (Consumer Rep)

Guy Johnson, Ph.D. (Industry Rep)

Michael J. McGinnis, M.D., MPP

Susan T. Mayne, Ph.D.

Suzanne Pelican, M.S., R.D.

Barbara M. Shannon, Ph.D.

TEMPORARY VOTING MEMBERS

Alice Lichtenstein, Ph.D.

Eric Rimm, Ph.D.

C O N T E N T S

Page

Welcoming Remarks:

Dr. Robert E. Brackett 4

Call to Order and Opening Remarks:

Norman I. Krinsky, Ph.D. 7

Subcommittee Discussion 11

P R O C E E D I N G S

MS. WILKENING: I guess we will get things started this morning. I would like to introduce you to Dr. Bob Brackett who is the Director of our Center for Food Safety and Applied Nutrition who is here to say a few remarks to you.

Welcoming Remarks

DR. BRACKETT: Good morning to all of you.

One of the things I wanted to do in coming here and giving these remarks is actually welcome you to the Advisory Committee.

For those of you who have either followed the FDA Advisory Committees, and in particular with CFSAN's Advisory Committee, you don't realize how valuable you are to the Agency, as well as to public health and some of the scientific background that you bring to these.

I participated on the Food Advisory Committee before I came to FDA, and have been involved with several others within FDA, as well as a joint committee that we have with USDA, and I think two things that I hope that you will find, one of which is that you will get a sense of accomplishment from what you are doing, because some of the issues that you have to deal with are very difficult, as I understand, yesterday, it sometimes stimulates some lively debate, and that is one of the two things that we do want, but in the bottom, when we are all done, I think what we really want is some of the best scientific advice that we can get on many difficult issues.

Fats and trans fats is one of the things where I think there is the most debate. I came from a meeting yesterday where involving both there is academic people there, as well as food industry executives and some government people, and I think about 80 percent of the day or the discussions, and it was sort of free on a lot of different issues, centered on qualified health claims and trans fats.

So, it is of interest and it is something that we very much want to have the best scientific knowledge on, so that we can make decisions that can benefit the public, as well as give the regulated industry guidance on how they can label foods, how they can structure the claims that they wish to make, so it is quite important.

One of the things that I think that you will also find is that you will end up making some--if you haven't already--different types of associations and connections with others on the committee and also people that make public remarks during the committee that I think will be of benefit to you in your professional life. At least that was what I had found.

I think it a very enriching experience to do this, and if you haven't served on these advisory committees before, I think that you find it valuable to yourself.

So, I welcome you again and I really would offer that anytime that you have questions or concerns, please feel free to contact any of the people at the Center that are dealing with the committees, and me personally.

We do want this to be the very best that it can be.

With that, I will let you get on with your deliberations, because I know you have some interesting things to cover.

Thanks.

Call to Order

DR. KRINSKY: Thank you very much, Bob.

We will now call the meeting to order.

I would like to take this opportunity to introduce Dr. Michael McGinnis who unfortunately was not able to be with us yesterday, but who will be with us today. So, welcome. You missed out on some interesting discussions, and actually, of the three questions, we actually voted on one.

That reminds me, I was actually reminded by the Executive Secretary that this is transcribed, and not videotaped, so that, for the record, the vote on Question 3 was 8 yes, no no's, and no abstentions. That is being double-checked by Dr. Rimm. So, we have confirmation that that was the actual vote.

For future votes, I am sure the Executive Secretary will again remind me to put it in the oral recording.

I think that we can return to Question 1. The question has been rephrase.

DR. JOHNSON: Mr. Chairman.

DR. KRINSKY: Yes.

DR. JOHNSON: Could I say something before we get started officially? I spent a long time sort of thinking about what a headline for this meeting and the Washington Post might say, and the headline would be something like FDA Nutrition Committee concludes that trans fats are worse than saturated fats, and I guess I felt like that headline, taken out of context, would not really be reflective of our discussion, so what I wanted to do was just read a statement from myself into the record to make sure that it reflected that. I wrote it down, so I would get it right.

The statement is I believe that although the answer to Question 3 is technically yes, this black and white answer does not reflect the considerable uncertainty discussed by the Committee as to whether this difference is significant from a public health perspective at the level of trans fatty acid intake typical in the United States.

I just wanted to be on record to make that point.

Thank you.

DR. KRINSKY: That is fine, Guy, and I appreciate that comment, and I think one of the problems with questions that end up with yes or no answers is that they lose all of the shading that comes in the discussion, and we end up with a black and white question when, in fact, we are dealing with all of the shades between black and white, and even a vote of 5 to 3 would not necessarily get into the shading, so that, you know, we have the questions.

The Committee will vote on them, and that is what is recorded without the benefit of the discussion. But the FDA is here and is listening to our discussion, and I hope that they are appreciating the shades that we are presenting on these questions, the shading that we are presenting.

DR. JOHNSON: I guess my concern was that they had said that if they don't adhere to an advisory committee ruling, then, it is very uncomfortable for them, and I wasn't sure if by having a black and white vote, we put them in a position of kind of all or none, so I just wanted to officially acknowledge that we are in a gray area here.

DR. KRINSKY: Well, let me just point out that the decisions, the vote that we take does not go directly to the FDA, it goes to the parent Food Advisory Committee, and I only have limited experience, but the recommendations of subcommittees have not engendered long discussions by the Food Advisory Committee.

There may be a few questions that come up for clarification on the issues, but they don't then initiate a rehashing of what we have done, and then after that, then, the FDA takes that information and they do with it as they please. Is that--yes, so hopefully, they will pay attention to what we are saying. I am looking around the trying to make the FDA feel guilty.

[Laughter.]

DR. KRINSKY: All right.

Subcommittee Discussion

DR. KRINSKY: Let's then return, unless there are any other comments, to Question 1.

We have rephrased Question 1 so that we can answer it as opposed to the initial phrasing. So, the question, it has been suggested that the question now read as follows:

Does the current scientific evidence suggest a relationship between total fat intake and risk of coronary heart disease?

Now, I might point out that this question has been modified by the FDA for our benefit, and I would be certainly willing to entertain any suggestions on the part of the committee if they wanted to suggest a further revision of that question.

DR. RIMM: I presume that this question follows the first statement in Question 1. So, the first statement still stands.

DR. KRINSKY: That is correct. I didn't read that first question because I am under the assumption, and it's a correct one, that we can all read that first--

DR. MAYNE: I am sorry, Norman, I didn't have the full question. It is coronary heart disease risk, is risk what should be added?

DR. KRINSKY: The relation between total fat intake and risk of coronary heart disease.

DR. MAYNE: Risk of coronary heart disease, okay.

MS. PELICAN: I am going to ask guidance from other people who are on the committee, who are, you know, much more familiar with the literature than I, but I guess I am thinking that is--I mean it is good that the question has been rephrased, but I find that it is a little harder to answer in my mind with some of the statements that I think were in Chapter 11 from the IOM report.

With regard to I think, I just had made a note, page 22 of Chapter 11 talking about practically, it is very hard to avoid high saturated fat intake when fat is greater than 35 percent of total energy.

They way it is phrased now, it sounds like there is no level, I mean that there is no relationship, and I would say could we talk about that as a group and get some input from people who know the literature better than I.

DR. KRINSKY: Let me just point out that we are not even addressing the saturated fat issue in this question.

MS. PELICAN: Right, but if we think about the relationship between total fat and coronary heart disease, that statement that once you reach--there is, in my mind, a relationship between total fat at those high levels, that statement that came from the IOM report, I am just asking for clarification and thoughts from other people on the committee.

DR. LICHTENSTEIN: It is certainly possible to have a diet that is low in saturated fat and is high in total fat, and I think we have to view the question within the context of what the current dietary patterns are in the United States.

So, from that perspective, it appears that the question is clear and covers all the issues.

DR. RIMM: I mean I think if you look at the epidemiological evidence, there is the two, large-scale clinical trials that I mentioned yesterday, that randomized people to high fat diets--or that randomized people to low fat diets, and did not find any difference between the high fat and low fat, and that is only within a certain range. They obviously can't randomize people to diets that are 80 percent of calories from fat, so if you look at 40 percent of calories versus 30 percent, you can construct a diet that is 40 percent of calories from fat that is not high in saturated fat.

Then, there is the three or four prospective studies that just took free-living individuals, had them report what they eat, and within the range of those diets, which is usually somewhere in the range of 25 to 40 percent, you don't see any association between fat and coronary heart disease.

So, what we have to do, I think, if we send the message that the evidence doesn't suggest that there is any relationship between fat and coronary heart disease, then, I am sure that Guy will be talking to people already who are making, you know, canola oil and other oils, making food with other oils that are high in polyunsaturated fats or monounsaturated fats, so very easily, industry can change to fit the science, which is what has been done in the past, sometimes maybe we were wrong.

I think there is a lot better evidence now to suggest that it is not the fat, it's the composition of the diet.

DR. SHANNON: I can't help but be worried about how this will play, as Guy has brought up, because here, education is going to become very important. I think in the public mind, the difference between total fat and saturated fat may not be that clear, and I can just see some people thinking oh, now, fat is not related to heart disease.

I am not arguing against the statement, but I am concerned about how it will play in the public.

DR. JOHNSON: Could I just emphasize the context of the question is within health claims, and it is not the intent of anybody, I don't think, to broadcast this message. It is just whether health claims that have to do with cardiovascular disease would require foods to be low in fat. That is the issue here, and there are lots of controls on those health claims with respect to saturated fat and other things in the regs.

I don't think the idea here is to wave a red flag and say, hey, you don't have to worry about total fat anymore. It's within the context of a health claim.

DR. LICHTENSTEIN: I think in a sense what it does is it remediates a situation that does not allow the best communication of health information, and that is, if the total fat criteria precludes you from labeling something like canola oil or an unsaturated oil, or again, a fish that is high in omega-3 fatty acids, with the health claim, because of the total fat, and we know that the total fat is not related to the risk of developing cardiovascular disease, so it is not for the whole diet, although it for the most part is fairly consistent with the whole diet, but this is sort of individual foods, and I think we are in not in an unusual situation where some of the criteria for whole diets have actually been transposed to criteria for individual foods, and that has I think further complicated the situation.

DR. SHANNON: I think that this conversation is very good because I think what I am trying to say is I would like to hear this conversation here in this group, making it clear what we are talking about specifically, so the record will show that we are not misconstruing it.

MS. MAYNE: I mean one possibility might be to modify the question to include something about independent of saturated fat intake, because then I think it would be much clearer what we are discussing.

So, it would be something like does the current scientific evidence suggest a relationship between total fat intake and risk of CHD, independent of saturated fat intake.

MS. PELICAN: That would certainly help address some of my concerns. I don't know, thinking about what Barbara is suggesting, too, I realize that on our written page here, you know, we have got the opening phrase, but could it say, "In relation to the health claims" in terms of, you know, FDA and health claims on labels, actually, to add that preface, because I think, maybe back to what Guy is saying, thinking about how a single question is interpreted or pulled out. I wouldn't mind having that qualification, too.

I was actually also going to suggest a small modification of including the caveat of saturated fat, and that probably take care of the concerns and also the potential interpretation of the recommendation.

Procedurally, can we do that, FDA?

MS. LATHAM: The committee can come up with a recommendation, yes, based on this modified question, that's fine.

DR. LICHTENSTEIN: And then we can vote on the modified question?

MS. LATHAM: Yes.

DR. RIMM: I don't know if we have enough evidence to say it's independent of saturated fat. That suggests that total fat is not associated with coronary heart disease when saturated fat is 30 percent of calories or when saturated fat is 10 percent of calories if we are saying it's independent of saturated fat.

I don't know if we can say the trials and the observational data test all levels of saturated fat. I mean we already have the health claim set in place for saturated fat, that you can't have a food that is high in saturated fat and say this is a whatever low fat food.

I think it is conservatively written right now, and I think it is appropriate because this matches the epidemiological evidence as we have it. I don't know, maybe I am not articulating the issue carefully enough, but I mean would we say it's independent of monos, independent of polys? I don't think so, I don't think we have enough evidence.

The evidence is three or four observational studies, two clinical trials, and that is not enough to say that at all levels of saturated fat, total fat is not going to be associated with heart disease.

DR. LICHTENSTEIN: My interpretation is that it would be just to tie together the two criteria formally, because as you said, that there is a separate one for saturated fat, so in this case it would just be to make sure that they would be linked.

DR. KRINSKY: As I understand it, the guiding rule now is that there can be claims for low fat, is that correct? And that there are exemptions, and those exemptions are based on the type of fat that is in the product, so that if you have a good fat, then, you have an exemption and you can get away from the low fat position.

Alice.

DR. LICHTENSTEIN: Well, that concerns me because the bar is then higher for those foods and frequently those are the foods that we would want to encourage people to consume, so that the playing field is not really level.

I think that is probably why this question has come up, because to go through the whole exemption process is quite cumbersome and it may actually deter certain foods or certain companies or purveyors from actually asking for the health claim, petitioning for the health claim.

DR. RIMM: The exemption I think we heard was for walnuts, was the most obvious case where the saturated fat was low and the other fats were good. You mentioned whole grains, which I think doesn't apply to this.

DR. LICHTENSTEIN: And soy beans and margarine and then the whole grains, so currently, there are four.

DR. RIMM: So, it is almost like this Question 1 refers to foods that are low in saturated fat, so it is not independent of saturated fat, it is only just as long as saturated fat is low.

DR. KRINSKY: But it doesn't state low in saturated fat.

DR. RIMM: No, it doesn't that, but that is sort of implied, because there is already an existing health--an existing food claim for foods that are low in saturated fat, so walnuts, if we said that here is Question 1, this is the eligibility criteria, fat is not associated with heart disease just as long as saturated fat is low, then, it would automatically sort of bring in walnuts, bring in fish, bring in everything for which we think there is no problem consuming high fat foods.

DR. JOHNSON: Assuming they are low in saturated fat. I mean the key is that all of the cardiovascular health claims that have been authorized so far require that the foods be low in saturated fat, and nobody is arguing with that.

It is just that they have gone through the process of having to exempt foods from being low in total fat so many times, it seems unnecessary to do that if the food is already low in saturated fat, and then there is all kinds of studies to show that it does good things to biomarkers, and so forth. I think that is what the point of this is.

DR. KOCH: Also, from an educational perspective, to tie the total fat and saturated fat together would probably be beneficial.

DR. KRINSKY: Well, Susan tried to do that in her amendment to the question. Does that actually accomplish what we want to accomplish that we just had in this discussion?

DR. MAYNE: One of my comments now, again one of the disconnects I am sensing is we are talking about a dietary pattern, but here we are considering the use for labeling in foods, and I am wondering if it might be useful to think about saying something that foods low in saturated fat, that contain high fat, are associated with coronary heart disease risk or something like that, because this is going to be applied to individual food items, and not a dietary pattern.

DR. KRINSKY: If that is the case, I wonder if it might not be useful to, in fact, insert the health claim preface in the first sentence at the beginning of this particular question, that is, for purposes of the health claim, does the current scientific evidence suggest total fat intake suggest a relationship between total fat intake and risk of coronary heart disease independent of saturated fat intake.

DR. LICHTENSTEIN: I think it would be helpful if the term "in an individual food" was inserted because that is really what this pertains to.

DR. KRINSKY: If we each insert two or three words into this question, it will become the mother of all questions, but this is what is known as a committee creating--crafting is not the word--creating a question that it wants to answer.

But give me your version again, Alice.

DR. LICHTENSTEIN: I have lost track of the whole sentence, but I was suggesting that the term "individual food" be in there, because there has been a lot of I think misunderstandings as far as whether some of these health claims are referring to individual foods or dietary patterns, and they can't refer to dietary patterns because you don't know what else is being consumed with these individual foods, so that we should just be real clear as to what it is actually referring to.

DR. JOHNSON: Actually, many of these health claims do say that the food must be consumed in the context of a diet low in saturated fat and sometimes cholesterol.

DR. KRINSKY: Michael.

DR. McGINNIS: I am reluctant to say anything since I missed some of the discussion and I am not voting today, but just two quick comments. I am presuming that since the reference is to the evidence base, our evidence is primarily on dietary patterns, and not on individual foods, so I am not sure that we don't get into some trouble if we try to introduce the notion of individual foods.

Secondly, I do think the educational value, if not the scientific accuracy of the qualifier for saturated fats, is an important one to include.

DR. KRINSKY: But the exemptions are for specific foods, not for diet, so though that should be put in the context of a diet, that is to say you don't want an exemption for walnuts while you are eating a diet that is supplementing the walnuts with a high fat, high saturated fat diet.

DR. LICHTENSTEIN: Not necessarily high fat, but not high saturated fat. Maybe what we should do is step back and decide which critical points we want to include in the question, and then it might be easier to craft a sort of trim question.

I mean the issues we want to get across the concept that the total fat content of the food is not related to cardiovascular disease risk, that the saturated fat content of the food is related to cardiovascular disease risk, and that this is for an individual food.

DR. KRINSKY: So, I will give you another version.

For the purposes of establishing a health claim of individual foods, does the current scientific evidence suggest a relationship between total fat intake and risk of coronary heart disease independent of saturated fat intake?

I will repeat that. The only problem I have with that, and it's a combination of everything that we have brought up so far, is that that last clause "independent of saturated fat intake," I think should sort of be earlier.

DR. McGINNIS: Grammatically, I would agree if you have total fat intake, independent of saturated fat comma.

DR. KRINSKY: I will reread this. Thank you, Michael. It is getting to be like a Rube Goldberg diagram on my sheet here.

For the purposes of establishing a health claim of individual foods, does the current scientific evidence suggest a relationship between total fat intake, independent of saturated fat intake, and risk of coronary heart disease?

Is something still missing? I wish we had a screen where we could put it up.

DR. LICHTENSTEIN: Can we say within the context of a diet that is low in saturated fat?

DR. RIMM: I think we want to educate people, and you want the message you get across, I think saying "independent" means very little to people who are trying to understand it. I think you have to say in the context of a low saturated fat diet, because that is where it has been tested.

DR. KRINSKY: Is there a computer up there? Can the computer be hooked up? Okay.

[Break.\

DR. KRINSKY: The Committee is back in order.

Now, we will proceed to some disorder. Alice left. Alice will be back.

Now, we have the following points, whether it is in appropriate grammatical form or not, we have the following points. We are talking about establishing a health claim. We are talking about individual foods. We are talking about the relationship between total fat intake in the context of a low saturated fat content, and risk of coronary heart disease.

Now, I understand that there is a suggestion for modification on my right.

DR. MAYNE: I am still vacillating between two different options, but, first, the health claim should be four individual foods instead of "of individual foods," is that right? Yes, okay.

Where I got stuck was the phrase there, "in the context of a low saturated fat content," I am not sure what that means. Does that mean foods, does that mean patterns, does that mean the overall diet?

One option is to simply clarify that. The other thing Eric and I were discussing was would it make sense to modify it something like, "For the purposes of establishing health claim for individual foods low in saturated fat," does it make sense, and then that would kind of get rid of the ambiguous clarification there.

But then it leaves then we are still dealing with total fat intake and CHD risk, so I am not sure which is the preferred option, but that is the alternative is to put the low in saturated fat intake with the individual foods at the beginning of the sentence. So, those would be the two choices.

DR. KRINSKY: So far the claims have been for foods that are low in saturated fat, so this, in fact, would be supporting the prior claims that have been accepted.

Guy.

DR. JOHNSON: I don't want to complicate things, but I have just talked to the FDA folks, and what they are interested in, I just think we are making this a little more complicated than it needs to be, what they are interested in is should there be an across-the-board requirement that foods that make cardiovascular health claims be low in fat.

They have already got saturated fat covered, they have got cholesterol covered. They are not looking for is there any condition where total fat could be associated with heart disease, they just want to know can they get out of having to exempt foods that are not low in fat from making a health claim if they are low in saturated fat and low in cholesterol.

I mean that is the question that they want, and, you know, as we talked about yesterday, the Heart Association is there, ATP3 is there, I think we are there. It is just we have kind of got ourselves into an exercise here of trying to worksmith something.

So, I just raise that for what it's worth.

DR. KRINSKY: Then, let me hear your suggestion for a simpler statement.

DR. JOHNSON: I should have figured that out before I opened my mouth, shouldn't I?

DR. KRINSKY: Yes.

DR. JOHNSON: Okay. One eligibility criteria that FDA has applied to most health claim regulations pertaining to heart disease risk is that foods bearing these claims be low in total fat.

Should this criterion be continued for future health claims?

DR. KRINSKY: Independent of the type of fat?

DR. JOHNSON: Those are already covered. There is already criteria for saturated fat and cholesterol. So, the question is should that criterion be an automatic part of health claims, yes or no.

DR. RIMM: I wholeheartedly agree with Guy on this one. I mean I think yesterday when we discussed it, we discussed it for 10 minutes and said this is a no-brainer and we moved on to Question 2, because the way it was stated was obvious, that there is no relationship between total fat and coronary heart disease, and as Guy said, all the other claims are covered by other things.

So, I think--now, I am going back on what we were saying--I don't think we need to educate people within this statement. I think the FDA is already educating people with the saturated fat and cholesterol, and we should just go back to the simple way that you reworded it, maybe even take out "For the purposes of a health claim," and just go back to the fact does the current evidence suggest that there is a relationship between total fat and risk of coronary heart disease.

We will answer is and let the FDA interpret it from there, because I think they just want the answer does fat cause heart disease.

DR. LICHTENSTEIN: I would just like some clarification from the FDA as to whether, by doing that, which I think is perfectly appropriate, whether that is going to complicate the educational message, and then that's it, whether that opens them up to some misinterpretation, potentially misinterpretation or exploitation of the specific statement, because otherwise, scientifically, I think we all agreed yesterday it was accurate.

DR. KRINSKY: Can we get some clarification, please, from somebody at the FDA whether the simpler question, and I can assure we can make this question more complicated, we are not through yet, whether the simpler question would be adequate for the FDA for their purposes of educating the public.

MS. WILKENING: FDA will have the benefit of the entire transcript, so we will put this in context when we use it in rulemaking.

DR. KRINSKY: Thank you.

DR. SHANNON: Could you read the question as you posed it, because I found that very helpful.

DR. JOHNSON: I didn't write it down, but I will make it up again. One eligibility criterion that FDA has applied to most health claim regulations pertaining to heart disease risk is that foods bearing these claims must be low in total fat. Should this criterion be continued for future health claims?

DR. LICHTENSTEIN: I withdraw my concern about the educational implications or interpretation, because I was just reminded that we are just advising, we are just answering a scientific question, we are not crafting any kind of health claim or statement or anything like that.

DR. SHANNON: But I still like Guy's, the way he stated that question. To me, it is a much clearer thing.

DR. KRINSKY: Let me go back to the beginning of the morning, which seems like hours ago, and that is the modification that the FDA originally suggested for this question.

Does the current scientific evidence suggest a relationship between total fat intake and risk of coronary heart disease?

That was the modification that we started out with.

DR. RIMM: May I suggest, Mr. Chairman, that we take a vote?

DR. LICHTENSTEIN: I will second that.

DR. KRINSKY: I question has been called which eliminates all further discussion according to page 82 or Roberts Rules.

So, I would like to ask the voting members to raise their hands if they agree with the question: Does the current scientific evidence suggest a relationship between total fat intake and risk of coronary heart disease?

All in favor?

DR. SHANNON: What are we voting on? Are we voting on that's the question, or are we voting on the answer to the question?

DR. KRINSKY: We are voting on the answer to the question, and the question calls for a yes or no answer.

Let's have a show of hands of those that agree that the current scientific evidence suggests a relationship between total fat intake and risk of coronary heart disease.

All in favor?

All opposed?

All abstaining?

DR. KRINSKY: There were six people that voted no, two people abstained.

Now that we have finished that simple question, we have done two out of three.

Amy, thank you very much. You can turn that off because we don't want to look at that anymore, although I hope that the FDA has appreciated what we have tried to do in that rather complicated question. We are just trying to repeat what the FDA already has in place.

MS. PELICAN: I just wanted to add. Maybe it is in the same spirit of Guy, who I think has really helped us make sure that we are looking beyond just an immediate question even though our obvious focus is what FDA wants us to address.

But I wonder if I might just add just a comment because I am thinking about sort of with all the consumer and industry and media focus on fat these days, if not FDA, but in terms of proceedings and what might get out about what this committee has agreed to, a statement such as, in the light of widespread consumer, industry and media interest in fat in relation to weight management, I would like to make a clarification that dropping the requirement for health claims for foods to be low in total fat is based on evidence for coronary heart disease only, and should not be inferred to mean that total fat is not an important consideration in other areas of intake, for example, energy density and energy balance.

DR. KRINSKY: Any discussion?

DR. McGINNIS: I would encourage a comment of that sort. I think it would be very helpful to clarify the record.

DR. KRINSKY: Susan.

DR. BAKER: Yes, I actually wanted to make a comment for a couple of moments here. I hope the committee will acknowledge that we are making recommendations based on data that has been obtained in adults. There is precious little, if any, data available in children.

I would also like the committee to acknowledge that what children eat is very important, not only to their health at that moment, but also to their long-term health. We have very little data on that.

We have some information that suggests that depending on how much of a particular fatty acid a child has early in life, maybe their blood pressure is a little different at other ages. We know for specific minerals the cognitive function, anxiety, and mental disease correlates with minerals in the first year of life.

I also hope that we will acknowledge that, at least for children under two years of age, that food labeling has no acknowledgment of the amount of fat that is contained in that food for specific reasons.

So, I hope that we can recognize that much of what we are doing here pertains to adults, and while at some time we may find that it is very important for children, right now we simply don't have that information available.

DR. KRINSKY: I am glad you said that, Susan, and that will be obviously in the minutes of this meeting for the benefit of the FDA. I am not sure what type of action we as a committee can take on this, but it is in the record, and I think that is the important thing.

DR. BAKER: Yes, I didn't expect us to take any action, I just wanted us to acknowledge that our work is very important for adults. We don't have any idea how relevant or important it is for children, maybe more so, maybe not relevant. But thank you.

DR. KRINSKY: So, let me go back to Susie's comment. Could you do me a favor and reread that?

MS. PELICAN: In light of the widespread consumer, industry, and media interest in fat in relation to weight management, I would like to make a clarification. Dropping the requirement for health claims for foods to be low in total fat is based on evidence for coronary heart disease only, and should not be inferred to mean that total fat is not an important consideration in other areas of intake, for example, energy density and energy balance.

DR. KRINSKY: That is absolutely true, and the question is I am not quite sure how we do this apart from you are also putting that in the record of the minutes of this particular committee.

I am not sure that we are in a position to make any further recommendation than the coronary heart disease issue, which we have been asked to vote on. We obviously were not asked to vote on total fat in obesity, which would also be a no-brainer, and we could do in a day or two with the group's ability to discuss these issues.

MS. PELICAN: No, it's just a clarification, thanks.

DR. LICHTENSTEIN: I think that is a totally different discussion, and I think we haven't reviewed the literature and we shouldn't even get into it at this point.

DR. KRINSKY: I don't want to get into it at this point, and I withdraw the comment from the record, and I will personally see to it that it is blotted out of the final record. When it is transcribed, there will be a transcription error.

DR. McGINNIS: I think what she was asking for was a statement similar to Guy's, a personal comment to enter into the record.

DR. KRINSKY: Let's return to where we were yesterday, and that deals with Question 2. I will read it.

The Dietary Guidelines Committee may suggest that less than 1 percent of energy should be obtained from trans fatty acids (2 grams per day for a 2,000 kcal diet).

Does the scientific evidence support this level?

We spent a good deal of time yesterday in trying to answer this question and trying to see whether, in fact, there was scientific evidence for supporting any level, let alone this level.

As I recall the discussion, there did not seem to be very strong scientific evidence for supporting a level of this nature, and I would be interested in comment from the committee.

DR. RIMM: I think, I don't know if it was Guy yesterday or Robert Earl when he spoke, someone had mentioned that, if used, this will likely be used as a target if we look at the label, that the 1 percent trans would be put on there as a target, much the way we have a target for saturated fat of 20 grams, a total fat of 65 grams, sodium of 2,400 milligrams, total carbohydrates 30 grams, and fiber 25 grams for someone who has 2,000 calories, if you want to refer to your label that we got yesterday.

So, again, as I said yesterday, I don't think that levels of that amount for any of those macronutrients or for sodium have a necessary threshold where there is benefit or risk. I think it is really put there as a target, and, you know, Dr. Enig would like to have a target of zero percent, based on her talk yesterday, and I think we have the realization that there are trans in dairy and beef and therefore a target of zero percent would not really be rational, but I think a target of 1 percent would not be that far off if we think of removing all trans fats from hydrogenated foods if we remove those from the diet.

DR. LICHTENSTEIN: I think we could perhaps deal with this in two separate ways, and, first, answer the question regarding the scientific evidence and then perhaps propose guidance to the FDA that in the absence of hard scientific evidence for a specific number, that a target be established consistent with the targets for saturated fat and cholesterol.

DR. MAYNE: My only comment is that my understanding is that those targets were set at a time when intake was higher than that, so, for example, the 30 percent fat was when U.S. intake was 37 percent of calories from fat, so setting a lower target was desirable, but from our discussions yesterday, it sounds like the intake of trans fatty acids in the U.S. is really very close to what this is right now.

So, if we are trying to set a target, I mean we heard different, that it is not clear yet where the target level is or what the U.S. intake is. We heard some discussion yesterday that it is not entirely clear what the intake levels are, but if we are at 2 percent of energy or 1.5 percent of energy, I think we should definitely make the target lower because consistent with the IOM report, as low as possible.

So, how certain are we of what the current intake levels are? I mean we heard different figures yesterday and that there is a lot of ambiguity in the data.

DR. LICHTENSTEIN: Yes, I don't think we have any idea what the actual intake is, but it probably is above 1 percent, and, yes, any target that you would want to set, you would want something below what the current intake levels are.

MS. PELICAN: If you would bear with me, I just took some time to write some things down last night, so I am going to be reading this.

I want to say how much I valued everyone's input through all the discussion, but particularly this one, to thank Eric and Alice for being really truly our expert consultants.

I am not comfortable voting to establish a D.V. for trans fat, and it is not that I don't want to have one. As an educator, I would welcome a D.V. that is both defensible and workable, but I don't see that those two bases are fully covered.

First, in terms of being defensible, it doesn't seem like we have enough of a basis to distinguish the desirability of 1 percent from a half a percent to 1.5 percent.

I know several comments have been drawn parallels with the saturated fat guideline of 10 percent or less, but this brings me to my second dimension or area of hesitation, and that is the workability.

I was struck by a couple of points that Norman and Susan made. Norman noted that a product with a half a gram of trans fat would display, if we used that 1 percent, 25 percent in the D.V. column.

Susan noted a problem with measurement, that being the difference in 0.1 grams from 0.4 to 0.5, would change a label value of from trans not being listed or from zero to listed as 25 percent.

With saturated fat, we often deal with amounts that are ranging from like 1 to 20 or more grams, and those are amounts where the 0.5 gram increments at those levels are small relative to the total, whereas, a 0.5 gram increment with the trans represents a huge proportion of that total.

So, just educationally, I think this could be really confusing. My hope would be that we could get more information on what an upper level might be, better trans fat composition data and more research, and on that note with more research, I would like to refer to the ruminations of two other important groups, the Institute of Medicine and the Dietary Guidelines Committee.

It wasn't until last night when I was reading one of the Federal Registers that we just got yesterday from April 19th, I hadn't realized that in addition to the 2002 IOM report in which that report, the 2002 did not, you know, give any kind of a DRI value for trans.

The 2003 report suggested an approach--and I didn't realize that there had been a 2003 report--suggested an approach to estimate minimum trans fat intakes with a nutritionally adequate North American diet, and use that value to establish a D.V. for trans fat. But again they, themselves, did not establish a D.V., they are looking to others to do that.

With regard to the Dietary Guidelines Committee, I find it very helpful to know that they are considering a limit of 2 grams or 1 percent of energy intake. I would encourage FDA to use whatever quantitative value that the Dietary Guidelines Committee sets as a basis for doing consumer research, both FDA-conducted research, as well as extramural research by others, to assess consumer understanding and use of trans fat labeling with a D.V. equal to the dietary guidelines value.

In conclusion, I will quote a friend of mine who says one of the good things about being a nutritionist, that we will never starve because we get to eat crow every few years.

I think it is better to promise less detail now and add more later when we have it than to specify it now and change the specifics later and have something be very confusing. I don't think it serves consumers or industry or FDA well.

Thank you.

DR. KRINSKY: Thank you very much, Susie, that was an excellent statement. Again, this is a statement that I am sure will be appreciated very--it is appreciated by me--and I think by the other members of the Committee, and I hope it will be appreciated by the FDA, as well.

So, we are in a position where we are virtually asked to vote on whether we want to establish a D.V. or something comparable to a D.V. by establishing a value that we are asked is there scientific evidence to support a value, a level of trans fatty acids, and that is troubling to me.

Susan.

DR. MAYNE: Again, clarifying the question, I mean it doesn't explicitly say should we establish a D.V. It is kind of implicit in the question, but to me, they are quite different.

Again, coming back to my comment I made yesterday, I just think that if you do have a D.V. there, there is no way, there is so much uncertainty in the amount of trans fatty acids in the individual food items, that the D.V. is going to have plus or minus 25 percent error for each individual food item.

So, I think it makes no sense to have a D.V. of 2 grams per day when we have plus or minus 0.5 grams measurement problems with the content in an individual food item. You can't operationalize it.

DR. KRINSKY: Eric.

DR. RIMM: Again, I think that maybe we are thinking that the FDA is not going to think about this a little bit more. It is very possible they can change the definition of how to measure or how to list trans fat or how to do the D.V., and I am sure that industry can measure trans fat down to less than 500 milligrams, plus or minus 500 milligrams, and we measure micrograms of folate, so I am sure it is very possible to measure grams of trans fat down to lower levels, so it is very possible that it won't have to be listed on a per half-gram or a per 25 percent D.V. increment.

So, I think if we vote on the question and then put it to the FDA saying now you need to be more creative about how you are going to list on the label, because there is evidence to suggest that there may be risk at 2 grams, so maybe we should maybe we should make the increment smaller, between zero and 2.

DR. LICHTENSTEIN: I would just like to say something about the measurement, that we have to keep in mind measuring trans fatty acids is not like measuring something like folate, that you have got many, many isomers in there. When you think about it, the difference in a cis versus a trans isomer, especially if you have something like 18 carbons, and you have 3 double bonds, they have 1 in the trans position, it may be in different positions, and with each one of those measurements, you have error, and then you are accumulating and adding up the error, so that is just a technical point as far as measuring the accuracy of measuring.

The other thing, I just want to say one thing about the Dietary Guidelines Committee and recommending a D.V. or any number. I am not sure, with all due respect to that committee, that they have any more resources or access to information than the IOM Committee or this committee with respect to actually coming up with a number.

So, whether they come up with a number or not, I am not sure that that is particularly applicable to this situation.

DR. RIMM: The only thing is I think it would be great if the USDA and the FDA worked together. I think it would be very nice if both groups came to one point, so there wouldn't be confusion by consumers and by industry, because in the past, there have been differences in how D.V.s and RDAs were set.

So, it would be nice if they did come up with a similar number.

DR. KRINSKY: This is an apolitical committee.

DR. RIMM: I am a very apolitical person.

DR. HINE: I would support what Eric just said and I would also like to add something about folic acid and measurement error that might indeed--

DR. LICHTENSTEIN: A bad example, I am sorry.

DR. HINE: Okay--be applicable here. I believe that two people from the FDA measured the actual content of folic acid in breakfast cereals versus what was on the label, and found in some breakfast cereals that there was up to 3-fold higher level that was actually present in the food.

I think this may indeed be of concern in this area, too, and would ask the technical people to comment on this. I think of doughnuts, for example.

DR. JOHNSON: Let me talk a couple of things. One, this measurement thing, I agree that in a purified system with sophisticated analytical techniques, you can get very precise in determining trans fatty acids, but we are dealing with a commercialized food product, and as you mentioned yesterday, it could be 10,000 gallon batches of something based on biological materials.

Think about a meat-containing product, for example, that would have varying levels of naturally occurring trans fatty acids. To that you may add exogenous sources of trans fatty acids. There are formulation variability, there is homegenicity in food products, even formulated products, that is a real issue.

Then, there is analytical variability, and all of this makes it very difficult to be precise down to 100 micrograms in a food product. It just doesn't work from a practical standpoint, that is just the way it is.

Then, I can't help--forgive me for replowing this ground--but the studies that we have on the effect of trans fatty acids in the diet are conducted at levels beyond--most of them are conducted at levels beyond typical intakes, and there is a paucity of data points in the area around 1 percent of the diet.

Granted, there are epi studies, but I know there are potentially other factors involved there, so, you know, if I am FDA and I create a D.V. and put it on a label for trans fatty acids, and they say where are the studies that show that it makes a difference from 3 to 2 to 1 percent, and they say, well, gee, we don't have much, I think that is not a very defensible position particularly when we are talking about a nutrition facts panel. It really should have sound science behind it.

I will get off my soapbox again.

DR. McGINNIS: Just a question, Mr. Chairman, about the discussion around the wording of the question. Did you discuss yesterday, as you apparently did in Questions 1 and 3, the rephrasing of this as a binary response yes/no, is that what you are intending to do?

DR. KRINSKY: Yes. Just for the purposes of simplifying the responses, that if, in fact, we can get yes/no answers, it will make it much better than an answer of yes, but I don't think this and I do think that.

Susie.

MS. PELICAN: Just to respond back to Alice, which I think she was commenting just with this idea of the Dietary Guidelines Committee, I guess my thought was not that I wouldn't think that they have access to different scientific data than we do, but just to say in some ways their charge is different than ours.

I had made a couple of notes during one of the presentations yesterday that I think this question FDA is asking with the idea that could a D.V. be set. So, I guess that was my thought and what I shared was just that I just am not comfortable answering yes to this with the idea that FDA is looking to then develop a D.V. with their best interpretation, but at something around this level.

I guess my point was to encourage FDA to take whatever the Dietary Guidelines Committee sets, not to put that on the label, but to use that in research.

So, I cleaned out one of my nut labels yesterday. I had chips yesterday, today I have my mixed nuts, and just looking and having it blank for trans or just as we do for polyunsaturates and monounsaturates, or as I shared yesterday, with that asterisk with the disclaimer, but to do that consumer research and see how do people use and understand that information if there were to be a D.V. at 2 grams or 1 percent energy level.

DR. KRINSKY: But if, in fact, a D.V. is set at 2 grams per day as one possibility, and your label says zero, your label could contain 0.4 grams, that is to say, 20 percent of the D.V., and that would just vanish because nobody would know about that.

MS. PELICAN: That's right, and that is my concern, and that is why I am saying I would not do that.

DR. RIMM: Well, based on what you heard yesterday, Suzanne, do you think that a target should be set? Do you think that we should have something on the label that gives a target for people to shoot for to stay below, so if they see the trans fat and the 1 gram, they have some context?

MS. PELICAN: I guess that, you know, on the one label that I got where it said, "Intake of trans fat is encouraged to be as low as possible, which I think is taken right from the IOM report, I mean that is one option, or just on this label, there is nothing next to polyunsaturates and monounsaturates.

That is where my point was, that I would rather see less. To me, less is better right now than to try to put something out with these percentages that go from zero to 25 percent in a 0.1 gram increment. So, I would let FDA with their best judgment--

DR. RIMM: My concern would be that if a label like the one you had, or if you have ice cream and it says 9 grams of saturated fat and 2 grams of trans fat, has an asterisk at the bottom that says, "Keep trans as low as possible," someone will look at it and go oh, I only have 2 grams of trans, I have 9 grams of sat, I am keeping my trans as low as possible, but they are at what we would consider eating too much trans.

So, I would be concerned about not having some target in some context rather than just sort of a vague statement saying keep it as low as possible.

I mean I do agree there should be something on the label that is a little bit more than just grams, but I am not sure just having a vague statement is sufficient.

MS. PELICAN: I have a response, but I think Barbara wanted to respond first.

DR. SHANNON: All I wanted to say was if we had the same data, by all means, having a target there is most helpful to the public, but having targets change is extremely confusing to the public, and right now I am very uncomfortable with whether we have the data for a target.

MS. PELICAN: I was just going to add that that is where the consumer research, if we could put something on now, and I realize I say that and for industry to do this, there is a lot of expense to end up changing labels down the road, but I think for the sake of minimizing confusion and promoting understanding, I think educators would have a better base for then, if the label said either there was nothing under trans or there were the two asterisks that said minimize intake as much as possible.

Educators can take that dietary guidelines, if it turns out to be 2 grams, we can develop materials around that, and we can do research to say do people understand that, but I think to go from nothing to that 2 grams is really risking a lot both in terms of scientific, that sort of defensibility, as well as workability.

DR. JOHNSON: Let me talk about education again. We had talked about putting the E back in NLEA, and I think this is an area where we are really going to need to do that.

I can tell you from experience that consumers do not understand the daily values thoroughly, and I don't think that there has been sufficient education for them to do that.

I used to talk with the 800 number people at Pillsbury on a regular basis, and many consumers don't understand percents. They don't look at a label and go oh, let's see, that means I need to eat 10 servings of this in order to--they just don't make that connection.

In fact, when NLEA first came out, if you look at the label, under total fat, the D.V. is 65 grams. We had a lot of people that said how come you have got 65 grams of fat in all of your products.

I mean we are sophisticated and we understand these things, but you are asking the label to do a lot, so I think that particularly if a D.V. for trans means that you are either free or an excellent source, with the difference of a half a gram, you are really going to fuel that confusion, and I am not sure you want to go there. So, let's educate consumers.

Then, the other thing is, as was mentioned, the IOM Applications Report suggested that a D.V. for trans fatty acids be developed on the basis of current intakes, and so forth, and whatever data is around.

We are chasing a moving target right now, and this is a very complicated sort of process that needs to be done thoughtfully, I believe, and to pull an arbitrary number out of the air without really getting into the intake data and really figuring out what the ramifications are, I think would be premature and counterproductive.

So, I would like to reinforce the thought of let some people thoughtfully look at the best data available and figure out how to implement that recommendation, and I don't think it is going to happen in the next six weeks.

DR. BAKER: I just want to echo what Suzanne said. I am a clinician for a good deal of my time, and my practice sees about 5,000 patients a year. My practice is very typical for pediatric practices across the country. About 30 percent of our patients are Medicaid.

All of our educational materials are read by a reader, so that they are not above a sixth grade level, and even that sometimes stretches people's ability to understand what we are trying to communicate.

When I or my dieticians try to work with label reading, it is very difficult for people. The daily value is absolutely a concept that doesn't make it, and particularly for children, it is irrelevant.

I really want to support what Suzanne said in terms of giving less information rather than more. The numbers are very difficult. When people have to do a percent of a percent or something like that, they just don't have the math skills.

I think we may do a disservice by trying to put in more numbers and more percents than we really have hard data for, and then if we change it, as we have done with total fats and bounced all over the place, people have no confidence in us.

So, I really hope that maybe we can take what you have suggested under serious advisement, and that is not perhaps support a particular number at this time.

DR. RIMM: Let me say also I agree with what has been said. I think there is no question that education has to be an important component of what comes out of this, because, you know, I think part of the issue is related to not just fat, but anything in the diet, there is not good enough education at all levels for adults and for children.

It took I think 20 years to get sort of the basic food pyramids out, teaching to children, and now they are teaching them new food pyramids, so I think there has to be a lot of time and concerted effort spent on education.

Along those lines, I think that for trans fat, you know, what we waiting for? Is there really going to be 10 or 15 more years, until Alice does three more trials, until there is one or two more? There is not going to be huge long-term clinical trials on trans fat, which is what we like to see, what we saw for total fat, but there is not going to be huge trials in that for trans fat.

So, I agree that maybe we can't come up with a daily value for trans because of the complication of the small amount of grams it takes, but giving a message that there should be a target of 2 grams and putting trans fat as 2 grams in the bottom part, and just leaving the daily value blank, would be a message to individuals, it would be something educators could look it and say, look, you have 1 gram of trans fat, you shouldn't have more than 2 grams in your diet, and forget the percentages.

I think that would give a strong message that there is concern that having foods that are too high in trans fat, above 2 grams per day is a bad idea.

I think, you know, there is four observational studies. There is one study done where you measured trans in adipose, there is all the metabolic studies that we discussed yesterday, and they all sort of point to the same relationship, that there is a linear association between trans and risk of coronary heart disease.

In addition to that, something we didn't even discuss yesterday is that when you take an oil like soybean oil or canola oil, which has alpha-linolenic acid in it, 18.3, which is important for a child's development, and important for reducing sudden death, that is the first oil to hydrogenate.

So, in addition to looking at the impact of trans and lipids, what we haven't even talked about is other health effects from the oils that we are getting rid of because we are hydrogenating in oil.

So, I do hope, it would be nice to send some message about trans, and that is something that we could start educating people now, instead of this committee meeting 10 or 15 years from now and saying, look, we have, you know, six pieces of evidence, but we still don't have enough evidence, let's wait for another 10 or 15 years.

So, the IOM spent a year and a half reviewing all the available evidence, and there has been nothing really new in the last, you know, year since we have been--maybe a little bit new--since we met last.

What we came up with is that really there is no reason to have trans in the diet, so going up to 1 percent, a lot of people have beef and dairy, you know, sufficient amounts, and you can still be below 1 percent.

So, I think we should send the strong message because I think the scientific evidence does support a level for trans for which there could be harm.

DR. LICHTENSTEIN: I would like to bring up a somewhat different issue. First of all, I think we need to come to some resolution on Question 2, but then I was sort of thinking about the example that Eric used earlier as far as ice cream that may have 8 grams of saturated fat and 2 grams of trans, that there should be some way of communicating that information, and it goes back to something that I mentioned yesterday, and that is combining sats and trans.

We have just voted that trans and sats are not the same, however, they both increase risk of cardiovascular disease. It would be unfortunate to lose the trans information if there are inadequate data to set a D.V., so another alternative would be to represent trans and saturated within the same target of less than 10 percent and express that on the label.

DR. JOHNSON: I just wanted to go back to something that Eric said, and it is my contention that we are already educating people about trans fatty acids. It is going to be a mandatory part of the label starting in 2006. Those numbers are already starting to show up, and I can tell you that consumers are engaged about trans fatty acids, and they will be hearing more and more about it, and whether there is a little 2 in the footnote of the label, I don't think there is strong evidence that that is going to be a motivator to people at all.

I think they have already gotten the message, the industry is responding, and there has been an awful lot of progress being made already, and I would hate to complicate the message for consumers by getting more detailed than we really need to be.

DR. KRINSKY: I am looking at the label again, and the trans fat line is the line that is going to be implemented in 2006, but there is no percentage of daily value that is going to be there, so you are just going to have this line sitting, hanging sort of free.

Now, are we expecting too much of the consumer to see this line and know what to do with it? That is an issue that I think we should be concerned with, because if a consumer sees that the trans fat line is going to have 1 gram, and the saturated 8 grams--

DR. RIMM: Great food.

DR. KRINSKY: Well, great food or not great food, how are they going to interpret that unless additional information is given? For example, we do have that saturated fat should be less than 20, whatever the figure it, it is 20 grams on our sheet, so there is some indication there for the consumer of what the goal should be.

I guess the basic issue is should there be a goal for the trans fat. Is there scientific evidence for a goal for the trans fat?

Jean.

DR. HINE: I would like to support what Eric originally said, and that you just commented on. I think people look at the labels and if there is no qualifier for trans fat, they think the number looks low, so this is a fine food.

So, there should be some type of qualifier that indicates at least our concern we don't know anything about the duration of exposure, especially for children, nor do we know very much about the duration of exposure for adults, and in what matrix they get their exposure.

DR. KRINSKY: Barbara.

DR. SHANNON: I am just asking a question, and I think it has been answered. Do we have the option of leaving the line hanging there, but putting "less than" statement at the bottom, in other words, not putting a percentage, but still having the statement of less than whatever, is that option available?

DR. KRINSKY: Not to this committee. We are not the ones that will be writing or have any real responsibility for writing the label, however, we can make suggestions, and then that can be acted on or ignored depending on whoever is listening to the suggestion.

Alice.

DR. LICHTENSTEIN: It would be helpful to me if perhaps we could address the question and then address the follow-up, because we keep going back and forth between the question and then what other options there are, so I am going to call the question.

DR. KRINSKY: As a member, you are certainly free to call the question, but as the Chair, not hearing a second to the call, it would indicate that the Committee is still not prepared to vote on the question.

If anybody were prepared to vote on the question, they would have seconded, but it is now no longer possible to second it, because the Chair has interrupted with his comments.

However, I am not getting a sense of a lot of input from the Committee, and there does not seem to be a single point of view. I think one of the problems is the word "scientific" in the question.

I think that some of the members of the Committee may feel that this is appropriate, but may be a little bit concerned about the strength of the scientific evidence, and as all of us know, everybody has their own definition of strength of scientific evidence.

I don't want to review that, thank you, I don't want to put that on the table because we won't get out of here at 4 o'clock this afternoon if we open up that can of worms.

The other thing, Alice, is that once we vote on the question, we will have completed our responsibilities.

DR. LICHTENSTEIN: Then, I withdraw it because I thought then we could go on and discuss some of the alternatives that have come up and make some specific recommendations to the FDA. I didn't realize that would terminate the meeting.

DR. KRINSKY: Well, I would be perfectly happy to hear alternatives and suggestions and follow-ups once this question is answered regardless of how it is answered. We could answer it yes, but, and we can answer it no, but, and I think the buts may be as useful, if not more useful, to the FDA than, in fact, the simple answer to the question.

Eric.

DR. RIMM: So, let's go back and read the question. So, the first part of the question says the Dietary Guidelines Committee is already thinking about this.

DR. KRINSKY: "May suggest."

DR. RIMM: May suggest this, so the Dietary Guidelines Committee presumably has read the IOM report very carefully.

DR. KRINSKY: I would hope so.

DR. RIMM: And may suggest that a trans fat target of 2 grams be set. Then, the next one says, "Does the scientific evidence support this level?"

It doesn't say does the scientific evidence, you know, absolutely prove this level, it just says does the scientific evidence support this level.

I think we are not making a definitive thing saying does it absolutely cause coronary heart disease, it says does it support this level.

DR. BAKER: May I ask Guy or anyone else here who has more experience than I do, if the only trans fatty acids that were available in your diet were those that occur in naturally--and I use the word advisedly--occur in food products, such as those from ruminant animals or dairy products, how much trans fatty acid would a person actually take in?

You are a hunter/gatherer, and you have your cow and you milk that cow, and you kill a wildebeest every now and then, how much?

DR. KRINSKY: Are you representing that particular individual?

DR. RIMM: I am from Wisconsin, and we do a lot of huntering and gathering there. Those are not normal diets, though.

I think someone from the FDA pointed us to the page yesterday where we had the gram amounts of trans you get from a serving of beef and dairy, and they were quite low, below 1 gram per day.

I will say that in the Nurses Health Study, when we look we look at grams of trans from ruminants versus grams of trans from hydrogenated oils, it was only the trans from hydrogenated oils which related to coronary heart disease, it wasn't the specific trans that you can identify from ruminant animals.

So, I think there actually are probably biological differences because you get like 100 different trans from hydrogenating oils, and in the ruminant animals you get one or two different types of trans.

DR. BAKER: Thank you very much. That is very helpful. The question was if we are going to try and set a lower limit or we are discussing some sort of number, it would be really important to know what kind of number would not be bad, what kind of number we might actually want to have in people's diets, particularly if those are the naturally occurring ones.

MS. PELICAN: Just to follow up, the Federal Register, that part of our background papers, 41-470, also lists animal products that give a separate total, if that might be helpful here, and I think what Susan just suggested is, in fact, what the IOM, the 2003 report did encourage, what is a nutritionally adequate diet and how much trans would we get a from a nutritionally adequate diet.

But I find this table on 41-470 of the Federal Register helpful.

DR. RIMM: It is also on page 8-45 in Section 5 of our booklets where it has servings of meat, range from 0.01 to 0.21 grams per serving, butter is 0.4 grams per serving, milk is 0.22 grams per serving.

DR. HINE: But what is notably missing from there, rather vaguely, the fourth line up from the bottom of the paragraph, "Therefore, foods that are contributors of trans fatty acids, fried foods, doughnuts, and french fries," and a lot of people eat a lot of those things, there is no estimate or range for the content there.

DR. RIMM: Usually, french fries are in the range of 4 to 6 grams per serving.

DR. HINE: Four to 6 grams per serving?

DR. RIMM: Grams, correct, and I believe doughnuts are in the range of 3. We have done some analysis, Frank Sachs has done some analysis in our lab of a few chain french fries, and the grams per day were in the range of 4 to 6 per serving.

DR. LICHTENSTEIN: But those analyses were done at least five years ago.

DR. RIMM: Correct.

DR. LICHTENSTEIN: So, we really would need data to say if it is actually that high at this point, because--

DR. RIMM: We were told at McDonalds--sorry--that the large M company were not able to change their hydrogenation methods, which is why they are not taking trans out, so it is true, we have not done the analysis in the last five years, but it is well above what you get in meat and butter.

DR. KRINSKY: But the point, Eric, is epidemiologically, the natural trans fatty acids are not associated with coronary heart disease risk.

DR. RIMM: To the best of our ability. I mean from the trials that Alice has done, those were not necessarily done on meat products, they are done on hydrogenated oils, which is what we are talking about, the whole list of different trans, not the trans from ruminant animals.

DR. BAKER: I can't help but make a comment, and please forgive me because I live with this day to day, what the heck am I going to tell my poor patients? I means we have trans fatty acids here, some are good, some are bad, the label says trans.

I am very confused, and I can't imagine that I am going to serve people very well, and I just point this out as something, if we are looking to label, to help people make good decisions or to make recommendations, so somebody can write a label to help people make good decisions, how are we serving that goal?

DR. LICHTENSTEIN: I think what we should do is tell people to minimize their intake of saturated and trans fatty acids. I think that is the best advice that you can give people, and they can do that by limiting their intake of animals fats from both meat and dairy, and limit their intakes of commercially fried and baked products, because that is a major source of hydrogenated fat in the diet.

DR. RIMM: Yes, even the ones that we have here, which show up as zero as trans because they are all less than 0.5, so I think if you want to give a specific message about trans, it would be to lower trans as much as possible in addition to what Alice said.

DR. BAKER: I just want to be a little careful about dairy particularly for children. It is the most bioavailable source of calcium that we have.

We know that we have a tremendous long-term problem with bone health, and if we are going to comment on dairy, perhaps we want to say low fat or something like that. We really are a having a problem getting adequate calories in, particularly with the numbers that we are being given currently.

DR. LICHTENSTEIN: Absolutely, if you were going to make specific recommendations. My comment was specifically for dairy fat, and also my understanding is these goals are for individuals above the age of two, and not below the age, but clearly, it would specifically be low fat and non-fat dairy products, which is part of the dietary guidelines in the IOM report.

DR. McGINNIS: I may be totally out of order here.

DR. KRINSKY: No one has been out of order in the past, and there is no reason why you should start being out of order.

DR. McGINNIS: But it seems to me that there is a strong consensus that the specification of a certain number that would represent a maximal daily intake is not possible given the current scientific evidence unless I have misheard the discussion.

Hence, the answer to the question, does the current scientific evidence suggest a specific level that would be an acceptable maximal daily intake of trans fatty acids in terms of risk of coronary heart disease is one that we ought to be able to answer.

DR. LICHTENSTEIN: I call the question.

DR. McGINNIS: Second.

DR. RIMM: I think the word is target, though, it is still a target that we want people to shoot for.

DR. KRINSKY: We have got the question called on this question.

DR. McGINNIS: I can't second, I can't vote. Sorry.

DR. KRINSKY: That's true. That is a good point, Michael, as to whether you can second, although you can't vote on the action, and I don't recall Roberts Rules of Order that much, but I don't believe that you can second.

DR. LICHTENSTEIN: If nobody wants to second it, then, it shouldn't be seconded.

DR. KRINSKY: To me, the problem is the question doesn't address what we have been discussing. That is the basic problem. We have been discussing a lot of issues with respect to trans fatty acids. We already have established that they are more adverse than saturated fatty acids as far as coronary heart disease is concerned.

We have been talking about the fact that their intake should be minimized with the exception of children. We do not want to deny them milk, et cetera.

But as I said, I believe yesterday, I have a lot of trouble talking about recommending a little amount or a large amount without any quantitation.

So, really the question addresses are we at the stage where there is sufficient evidence for quantitation, not that we are setting--well, yes, we would be setting this, unless I am wrong, as a daily value, less than 1 percent of energy, less than 2 grams per day.

Wouldn't that be interpreted by the FDA as our recommending this as a daily value? It may or may not be interpreted. I am looking for head nods on the table.

Do you want a five-minute break? Let's take a break for a few minutes, because this is how I am interpreting this. If we vote in the affirmative, we are essentially saying that there should be less than 1 percent of the energy value as trans fatty acids, and that to me is the same as recommending a daily value for trans fatty acids, which would, in fact, change this sheet completely because under those circumstances, one would now have room for percent, and one would have, at the bottom, less than 2 grams per day.

DR. RIMM: You don't know if they actually would make the percent, they may just put the 2 grams at the bottom.

DR. KRINSKY: That is true, but it seems to me the logical outcome would be that they might do something of that nature. Certainly, because of the inaccuracies of the measurement and the difficulty with the small amount where less than half a gram, which could be 25 percent, would not be recorded, it may not make sense to put the percentage down, but that is their call, not our call.

Let's just wait a wait a few minutes.

It is now 9:45. Checkout time is noon, although they will make an exception for you at 11:30, and we can take our break now, our 10:15 break now. We are going to break for 15 minutes.

[Break.]

DR. KRINSKY: I would like to reconvene the Committee and would ask for the members to resume their seats and pay attention to the Chair.

I want to get this correct, that according to the FDA, if we were to vote in the affirmative on this issue, it would not necessarily be establishing a D.V., it would simply be saying that we agree that less than 2 grams per day of trans fatty acids would be useful, valuable, healthy.

DR. LICHTENSTEIN: Then, it sounds to me like the question has to be rephrased because that is not how it is written.

DR. KRINSKY: Well, let me just say that it is referring to the preceding sentence, and I would be certainly willing to have it rewritten, but it is referring to the preceding sentence, the Dietary Guidelines may suggest that less than 2 grams per day for a 2,000 kcal diet should be obtained from trans fatty acids.

DR. LICHTENSTEIN: And then does the current scientific evidence suggest.

DR. KRINSKY: Support this level. Okay.

DR. LICHTENSTEIN: So, we have to vote on what is written, not what the interpretation is.

DR. KRINSKY: That's right--well, no. With respect to a D.V., we have no input on that. We are not establishing a D.V.

DR. LICHTENSTEIN: I am not suggesting a D.V., I am just saying that we need to vote on what is actually written on this page or it needs to be officially changed.

DR. KRINSKY: Did you have a comment, Michael?

DR. McGINNIS: Well, I did. I guess that it sort of picks up on this to some extent. As it is written, I would have taken the essence of the discussion to lead us to a negative answer to this question in terms of the conclusion about the evidence suggesting a specific point, so I do think the question would have to be reworded if we are seeking otherwise.

Let me just take 60 seconds to revisit what I had interpreted the essence of the discussion as indicating.

What I heard us trying to do is to accomplish certain specific things. One is to be faithful to the evidence. Second, given the fact that there is reference to the Dietary Guidelines Committee here to not get crosswise with the Dietary Guidelines Committee and meddle unnecessarily in their deliberations, and thirdly, to leave the FDA with sufficient flexibility in label design, in order to do that, my interpretation, which is not a proposal because I don't want to muddy the waters with a proposal, but my interpretation of what Question 2 in essence would say is that as a signal, that trans fatty acid intake should be reduced to the lowest practical level, the Dietary Guidelines Committee may suggest that intake be reduced to less than 1 percent of energy, 2 grams per day from scientific perspective, then the question goes on, does the current evidence suggest an acceptable maximal daily intake of trans fatty acid for reducing the risk of coronary heath disease.

That divorces the issue of the practical guidance of the Dietary Guidelines Committee from this committee's judgment on the adequacy of the scientific evidence.

Now, that may lead us down the wrong path. If there is a simpler wording change that can be effected in the nature of the question, then, that seems a perfectly reasonable approach.

DR. KRINSKY: Thank you.

Guy.

DR. JOHNSON: Well, I get the sense, I mean I agree that based on the question that is written, the sidebar conversations that have gone on here, and the general discussion leads to the irrefutable conclusion that there is no hard scientific evidence to support that, and at the risk of boring you again, the studies that were listed in Ascherio diagram show that based--that is 861--I am a person who has a fair amount of experience in petitioning FDA for health claims, and they have very little sense of humor in interpreting the strength of the data on this--but when you petition for a health claim, you have to have strong evidence to support your position.

Here are the data that are available to support a level of trans fatty acids. If you look at the trials in that study, there are three that were conducted at levels below 4 percent of the intake of trans fatty acids, arguably, two that are at the current estimated level of intake or less.

Both of those studies failed to find a significant difference between the LDL level of the control diet and that containing trans fatty acids.

We hop over to the HDL side, which is what led this committee to conclude that there is a difference between trans fatty acids and saturated fats at high levels, a caveat which did not make it into our answer.

There are six studies below 6 percent of calories of trans fatty acids. Five of those studies found no statistically significant change in HDL, and one did.

If I submitted those data to FDA, and said you need to come up with a D.V. for trans fatty acids, they would laugh. Those are the data that is there.

Now, we are basing an assumption that 1 percent is supported by hard data. It is an assumption, and I suppose we are at liberty to do that, but let's not kid ourselves into thinking that the data that is in the published literature supports that. It does not, and if we come to that conclusion, shame on us.

DR. KRINSKY: Well, that gets back to the issue that I raised earlier, Guy, and that is the definition of scientific evidence, and we all have our own definitions. It is quite clearly what yours is, and that is that there is not adequate scientific evidence to adopt this particular position.

DR. JOHNSON: That is clearly my position, and my contention is that my position is based on the evidence. I don't see any other way to interpret it. This is what the data show. This is peer-reviewed data that has been published in the literature, and I have been hesitant to say this, but the Ascherio diagram was not in peer-reviewed literature, it was published as sort of an editorial on the New England Journal, and my contention is that by looking at the ratio and forcing it through the origin, it's misleading and has been misinterpreted.

Now, I can't argue that at high levels of trans fatty acids, there is a statistically significant difference, I am there, but that is based on intakes that are beyond typical intakes in the U.S.

If you look at these studies, I don't see how you can come to any other conclusion, and I don't see how FDA could come to any other conclusion, and that is assuming that HDL is given the same sort of weight as a biomarker that LDL is, and if you look again at the Federal Register documents we have, there is a lot of discussion about how HDL is not risen to the same level as LDL in terms of a cardiovascular biomarker.

FDA used it in the cost analysis section, but they viewed the ratio of HDL to LDL with some skepticism, and again shows that the way FDA interprets data is very rigorous and rightly so, because no matter what they do, somebody beats on them, so they need to make sure that what they do is very evidence based.

By forcing them to ignore the data that are there because we think that there is a dose-response relationship, when a reasonable scientist can conclude just the opposite by looking at the data, I think is a very dangerous road to go down.

DR. KRINSKY: Just for my own personal clarification, does that mean that nobody beats on the Dietary Guidelines Committee when they come up with a recommendation? Does anybody know?

DR. LICHTENSTEIN: I think since none of us have been to the Dietary Guidelines Committee and we are not privy to what the conversations were, I really don't think that we can make any determination about what is going on in that committee.

Now, I certainly understand that there should be some consistency between the Dietary Guidelines Committee and the FDA, but at this point, I think it is totally premature to even address that issue.

DR. KRINSKY: Well, I know that when I first saw this question, I was troubled by the "may suggest" in the sentence, and found a little bit of difficulty in trying to interpret how we were supposed to address a "may suggest" issue.

DR. JOHNSON: The way I interpreted that was the Dietary Guidelines are thinking about it. If they do, what do you think?

DR. KRINSKY: But we are not asked if they do, we are asked does the scientific evidence support this particular value.

DR. JOHNSON: Based on the assumption that they took that position.

DR. KRINSKY: Well, but we can then move independently of their assumption, of their position, and we can address our own position, and that is to say do we feel that the normal U.S. diet should contain less than 2 grams per day of trans fatty acids, or does the scientific evidence indicate that the American diet should contain less than 2 grams of trans fatty acids per day.

DR. JOHNSON: And as will be no surprise to this committee, my point of view is a resounding no, as I said yesterday, and other people sitting around this table have also said.

DR. SHANNON: Would you repeat again how the FDA will interpret our answer if it is yes?

DR. KRINSKY: Well, as I understand it, and I sit to be corrected, and their understanding would be that we concur that the diet should contain less than 2 grams of trans fatty acids per day, that may or not be included in the nutrition facts label, that is going to be up to the FDA and up to rulemaking, and all of that nature, but it may, in fact, given them a leg up in terms of doing something like that.

In the absence of a yes vote on our part, they are not in a position to incorporate that particular number.

So, that is my understanding of the FDA's position.

DR. KRINSKY: I think that Eric was up first.

DR. RIMM: I just wanted to respond to Guy's comment or one of his comments. I think that it is too bad that Guy and I used to be friends before this meeting. I do disagree with him a little bit on several points, and I know that there have not been tons of trials in the low level of trans fatty acids, which is where the rub sort of is, even though the four or five observational studies do assess trans fatty acid at the lower levels.

Apparently, Guy does not think that is scientific evidence, but to the point that Ascherio's review was a review and published in 1999, and not a peer-reviewed, the paper that we did get sent to us via e-mail, that someone contributed, which was Mensink's review, Mensink, Zock, Kester, and Katan, who I think are some of the leading scientists in this area, did a meta-analysis of 60 control studies, and their conclusion, the first line of their conclusion is, "The replacement of trans fatty acids with unsaturated fatty acids from unhydrogenated oils is the single most effective measure for improving blood lipid levels."

So, that is the main point of this meta-analysis, and the meta-analysis reviews the eight metabolic studies, and they ranged--unfortunately, it doesn't have as nice a picture, so we can't see all the studies or we can't see anything that was new since the Ascherio review in 1999, but I think it does point to the fact that there is a dose-response between trans fatty acids and the total cholesterol-HDL ratio, which is what they have used as the outcome measure.

Unfortunately, there is not a PowerPoint presentation that I can make on this, but all of us have this paper from the AJCN 2003. I will get the exact reference. AJCN 2003, Volume 77, pages 1146 through 1155. I don't know if everybody has that or not, but it was something that I believe either Mary Heonig [ph] or someone else contributed to our meeting.

DR. KRINSKY: I believe this was sent to us by the FDA.

DR. RIMM: Oh, the FDA sent this to us as additional information, so this is peer-reviewed, and this is put together by some of the leading scientists in the world in this area. I don't think Guy, Alice, or I, you know, going through this, could come up with a better interpretation than they have done, because this is what they do.

In the eight studies that specifically examined trans monounsaturated fatty acids, the intake of trans ranged from zero percent to 10.9 percent of energy, and these values included the trans fatty acid-free control diets in these studies, and then it goes on to discuss them. Unfortunately, it doesn't have as pretty a picture, but I really think there are metabolic studies to support the fact that there is a dose-response between trans and lipids.

It goes along with the observational studies, and I don't want to rehash the data, you know, just like for saturated fat, there was not obvious level where we picked, and just like for sodium, there was no obvious level.

The USDA has picked a target of 2 grams, and I think the evidence supports that a low level of trans would be the healthiest.

DR. LICHTENSTEIN: I want to go back to what you were saying, Norman. I think there are two issues here. I think one is does the scientific evidence support a specific number, and as I have said before, I don't think it supports a specific number.

The other issue is should there be a target, and we have already voted that trans fatty acids have an adverse effect on cardiovascular outcome, and everyone has sort of re-reviewed that, therefore, should a target be established, and I think that that may be reasonable, but there are two different points.

DR. KRINSKY: The target would be based on scientific evidence?

DR. LICHTENSTEIN: I don't think that there are adequate scientific evidence to come up with a specific number. Could we try to come up with a target as was for saturate fat, which was really a target, as was pointed out earlier by Susan, the intake levels were higher than the target in the United States, it made sense to aim towards a lower number? We don't really know what the trans intake is in the United States, but I think we would all agree we would benefit if it was decreased.

Therefore, again, it is reasonable to set a target. In that way, I think that what we do can be defended, because we can't go beyond what the scientific data tells us, and it is not a black and white. If you consume below this number, you are not going to change your risk of cardiovascular disease, and if you consume above this amount, you are going to increase your risk of cardiovascular disease, it is a continuum.

DR. KRINSKY: Could you give me the target number you are thinking of?

DR. LICHTENSTEIN: I am not thinking of a target number. I think of things in small doses, so what I would like to do is deal with one issue, and then discuss the other issue, because right now we keep going back and forth, and we are mixing up the discussion of whether there is adequate scientific data to come up with a specific number and whether it's advisable to set up a target, one of the implications would be to the FDA.

DR. KRINSKY: I understand we are mixing things up, but I would not be adverse to, in fact, having a statement as you have made, and to hear what your target number would be independent of our voting on this particular question.

Then, we could move forward with that particular proposal and then return to the question, to Question 2. That is why I was asking you what the target number should be, and the reason I am asking you is that 24 hours ago, not quite, I was disturbed by talk about low and high. To me, that is not very scientific.

DR. LICHTENSTEIN: Well, I don't have a target number. I can certainly react to other people's target numbers, but I do have a position on the second question, and my position on the second question is totally independent on whether we can set a target number or what the target number would be.

DR. KRINSKY: Again, I want to point out that when we complete our action on the second question, that is all that the FDA has asked us to do, so I would like to sort of get any of our other input in before we vote on Question 2.

DR. LICHTENSTEIN: So, is what you are saying that once we vote on Question 2, we have to stop the deliberations?

DR. KRINSKY: We have no other responsibilities.

DR. LICHTENSTEIN: Is that true?

DR. KRINSKY: Oh, I have been told that we can come up with recommendations, but when we vote on Question 2, will you then give me a target number?

DR. LICHTENSTEIN: I don't have a target number. I can certainly support or not support, I didn't realize I was expected to come here with a target number.

DR. KRINSKY: You weren't, but you are the one that raised the target number issue, and if you are willing to support or not support target numbers, you must have something in mind, you know, if somebody says 5 grams a day, you might be willing or not willing to support that, if somebody says half a gram, et cetera.

So, you want us to come up with some values. Okay, fine.

DR. LICHTENSTEIN: I am just suggesting that that might be appropriate, but I am just trying to move the discussion forward, and I feel that sometimes we are spinning wheels, and I am sure that nobody wants to hear me repeat lots of stuff that I have already said, and it seems like we should be able to address at least the question that was given us, and then we can decide if we want to go further or not as a group.

DR. JOHNSON: Well, I will just reiterate my point that I don't think there is enough scientific evidence to support that, but let me say that whatever we do, please, please, please let us not lose sight of the fact that we are dealing with both trans fat and saturated fat. The practical realities of the matter is if you beat the industry into acting prematurely on trans fat, there is a possibility that you are going to increase saturated fat just by the realities of food chemistry and technology that is available.

So, we need to be cognizant of what we do here can have unanticipated consequences in terms of the total dietary content of fatty acids that could raise LDL.

Also, I can't resist--this is again based on my experience with health claims with FDA--a meta-analysis of studies does not constitute a dose-response based on the way FDA interprets health claims.

If there aren't studies double-blind, randomized, controlled studies in the dosage that you are interested in talking about, they don't anticipate a regression line through the studies based on a meta-analysis as dose-response.

DR. KRINSKY: Your position, Guy, by the way, is reinforcing the Mensink position because what their quotation is, that was read to us a moment ago, risk is reduced most effectively when trans fatty acids and unsaturated fatty acids are replaced with cis-unsaturated fatty acids, so that it is not simply the reduction of trans and saturated, it is the replacement with the cholesterol lowering fatty acids, and this, of course, the question doesn't address that at all.

You had a comment, Eric.

DR. RIMM: I guess I wasn't sure on what Guy's point was. There are all randomized, double-blind trials. It is just a summary of those by giving you a better dose-response. I don't think you could have one study that would have 15 levels of trans. This is a way to look at it across the different dose-response in trans.

DR. JOHNSON: Well, a true dose-response would be a single study that fed graded levels of trans, and the data in the meta-analysis, most of those studies again are conducted at levels higher than the current intake of trans.

So, you are extrapolating the data from an area above typical intakes in the U.S., down into an area that is atypical, and assuming that that is a linear relationship. That is an assumption, I am sorry, but it is.

DR. LICHTENSTEIN: That statement that was read for Mensink and somebody and Katan, sorry, that sort of got back to the total fat issue, which we have already addressed, that you substitute trans and sats for unsaturated, but we still have the option, if we could get beyond the question, to then perhaps propose a target for trans plus sats if that is what we want people to decrease, and we could discuss having separate ones, but there are multiple options.

MS. PELICAN: I appreciate FDA's clarification that this is not necessarily to be used to set a D.V., because that certainly was one of my concerns, so I appreciate that, but I am going to go back to something that Susan Mayne said yesterday, and this is without putting you on the spot, Susan, if you could just see if you are still of the same mind.

I think we were talking about linear and threshold effects, and you said we do lack data points, but you prefer seeing assuming a linear relationship, because we don't see a threshold, and I guess my question is if the idea in thinking back to the IOM, the 2002 report, if their recommendation was that we consume as little trans as possible, is there consistency with that question, does the scientific evidence support a level at 1 percent, 2 grams, 1 percent of energy, does that idea of linearity support that, that yes, this is low. If you could respond to that, that would be great.

DR. MAYNE: Just two comments. One is on the Ascherio study which looked at the clinical intervention studies, and just looking at the data points, I see no evidence that there is a exponential relationship, so therefore, with the lack of evidence that it is nonlinear, the more conservative approach is to assume that that is a linear relationship. So, I assume that people in risk management toxicology, if they saw that relationship, and they were using this for public policy, they would take the more conservative approach in the absence of evidence contradicting that.

The second one was the observational epi data, which was not included in the Ascherio study. That looked at the quintile distribution and did a statistical test to see whether there was evidence that the relationship was linear, so that is the kind of thing that we would look at, for example, in the Nurses Health Study and the Health Professionals Follow-up Study, to see if there is significance that there is a linear relationship, because sometimes we will see associations, let's say, at the highest quintile versus the lowest, and we run a statistical test. They will say, well, there is no evidence that that is a linear relationship.

So, those were the two things we were looking at. When I just eyeballed the estimates, the relative risks from the Nurses Health Study and the Health Professionals Follow-up Study, it looked pretty linear.

So, even though the second quintile was not significantly different than the first quintile, the relative risk estimates increased at each level of intake in several of those prospective epidemiological studies.

DR. KRINSKY: But isn't that the same as saying that if you do an experiment, and you are looking at the difference between 1 percent and 2 percent, you may not find a statistical difference, but if you were to do 1 percent, 2 percent, 4 percent, 8 percent, then, you might interpret the data as linear.

DR. MAYNE: That's correct.

DR. McGINNIS: Picking up on this notion of consistency and boldly going into the notion of targets that has been requested, I will actually offer a proposal for people to shoot down as a trial balloon.

Something on the order of although current scientific evidence does not indicate a specific acceptable intake for trans fatty acids, it is consistent with reducing trans fatty acid intake to a level of less than 1 percent of energy (2 grams per day for a 2,000 kilocalorie diet).

DR. KRINSKY: That, to me, is a very reasonable statement that we can discuss, but I am getting the feeling that, in fact, that statement should be discussed after we carry out our vote.

I know that the question has been called twice and has been rejected resoundingly by the Chair, and seeing no further comment, the Chair is prepared to call the question.

DR. LICHTENSTEIN: Second.

DR. KRINSKY: No, no, I think you are out of order, Dr. Lichtenstein, but I appreciate your support of the Chair's position.

So, we have discussed this question enough. Any disagreement? No.

So, the question is: The Dietary Guidelines Committee may suggest that less than 1 percent of energy should be obtained from trans fatty acids (2 grams per day for a 2,000 kcal diet).

Does the scientific evidence support this level?

All those who want to answer yes, please raise your hand.

All those who want to answer no, please raise your hand.

The Chair votes.

The no's carry this question 5 to 3, so that we have responded to the question put to us by the FDA, but we are not through with our deliberations, and the FDA will allow us to continue? Okay.

So, let's go to your statement, Michael, and if you could you re-read it for the benefit of the Committee.

DR. McGINNIS: Although current scientific evidence does not indicate a specific acceptable intake for trans fatty acids, it is consistent--

DR. BAKER: Excuse me, I am a really fast typist, I don't claim to spell, but would you like to see this up there?

DR. McGINNIS: Sure.

DR. BAKER: Would that help you?

DR. KRINSKY: We would be indebted to you for a long period of time.

DR. BAKER: You could start reading it.

DR. McGINNIS: Okay. Although current scientific evidence does not indicate a specific acceptable daily intake for trans fatty acids, it is consistent with reducing trans fatty acid intake to a level of less than 1 percent of energy (2 grams per day for a 2,000 kilocalorie diet).

For the purpose of a vote, you would have to rephrase this as a question, but have at it.

DR. KRINSKY: Thank you very much, Michael.

Let's have some comments.

DR. LICHTENSTEIN: I have more just an alternative concept to consider, that the IOM report also recommended, that saturated fat be as low as possible, and thinking from both an educational perspective and a conservative perspective, could we not consider recommending that trans plus saturated fatty acids be less than 10 percent of energy?

DR. KRINSKY: Would you have any concern or are there circumstances where one can, in fact, have foods that would contain a relatively high trans to saturated fat content?

DR. LICHTENSTEIN: That could potentially happen, but on the other hand, it would address the issue of manufacturers taking trans out and potentially putting in saturated fatty acids, and we know that although, gram for gram, trans increase risk of cardiovascular disease more than saturated fatty acids, we also know that the proportion of trans and saturated fatty acids in the diet is quite different.

So, yes, I guess theoretically, that could happen, I think it is highly unlikely, and I think what we have heard is that using the Nutrients Facts Panel, as confusing as it is, I think the least information, the clearest information is probably the best from the perspective of public education and use.

DR. RIMM: I don't disagree at all. I mean I think that we really need to educate as best as possible with saturated fat and trans fat, however, I think there is two or three pages in the Federal Register that were sent to us saying why they didn't put those together, because the would have meant that by taking trans out, that suddenly there was a new D.V. for saturated by adding trans as saturated that changed what the D.V. mean for saturated, so I think they separated the quantitative amounts purposely because trans is different from sats, even though the educational message should be have as little as possible, I think that combining them into one percentage then makes the interpretation of saturated fat is more difficult. That is my reading of the Federal Register report.

DR. JOHNSON: Here we go again. Part of the issue there is stearic acid, and there seems to be pretty good agreement that stearic acid is neutral from a biomarker standpoint. If we are going to get into this, I think we need to recognize that fact.

I do agree that the message really should be broad enough to be actionable for consumers who are interested in managing their risk of heart disease, and the simpler the message, the better.

DR. RIMM: Which is why I think Michael's message is quite clear.

DR. KRINSKY: Except it doesn't indicate any disease relationship here, and do you think that we should reducing risk of coronary heart disease in the statement?

DR. RIMM: Well, I think that that statement may change, because I think there is a growing body of evidence that trans affects diabetes, that trans may impact obesity, that trans may have other effects where the evidence is not as strong, not as yet strongly documented by metabolic studies and by observational studies.

So, I think this makes it very clear because a 12-year-old picking up a package of Doritos will say, ah, I am not worried about heart disease, so I think if you just say limit it as much as possible, I think that would give a pretty clear message.

I don't know if we are in the business of making messages, but I think this is consistent with all of our discussions that we have had today and yesterday, saying that there is scientific evidence to support it, keeping it less than 1 percent would be best.

DR. KRINSKY: We have made a lot of messages in the last day and a half, which are not up on the screen, but they are on the tape of this meeting.

Alice.

DR. LICHTENSTEIN: Maybe what we could do is just encourage the FDA, if there is a major reassessment of the Nutrient Facts Panel, to ultimately consider the possibility of creating a category of the fatty acids that increase risk of cardiovascular disease or other chronic diseases, accepting that my suggestion of trans plus sats might not be a good idea, or might not be scientifically defensible because of the issue of stearic acid.

DR. KRINSKY: Maybe more importantly would be a comment about fatty acids that decrease the risk of coronary heart disease.

DR. LICHTENSTEIN: Potentially, but I think the major problem in the United States right now is excess caloric intake, so I think we need to be very careful if we get into that arena of recommending increased intake of anything, but at this point, I think we can feel pretty confident that decreased risk of trans fatty acids and certain saturated fatty acids is advantageous.

DR. KRINSKY: A point well taken.

DR. McGINNIS: For the purpose of consistency with the other two, if you want to rephrase this as a question, just transpose the "it" and the "is."

DR. KRINSKY: Is it consistent, and then end with a question mark.

DR. JOHNSON: I have got to make the observation that I think we are dealing largely with a matter of semantics here. Basically, we are saying that the data is consistent, but not supportive. It sounds kind of like a qualified health claim, I guess.

But I really encourage the Committee to make sure that--and to consider the possibility of unintended consequences here. What you are doing here is raising a big red flag about trans fatty acids in the current food supply, and this is going to have the effect of really drawing a lot of attention to trans fatty acids, and there is, I think, a reasonable likelihood, in fact, there is probably consumer data out there to verify that you are going to minimize the importance of saturated fats in the mind of the consumers.

You know, we are making this up as we go along, and not basing this on scientific evidence, and I ask the Committee to consider the fact that there could be unintended consequences particularly in the reality of the food science that dictates the way foods are reformulated to remove trans fatty acids.

DR. RIMM: I agree that if we take the trans out, something else will be there, because we are not going to reconstruct all foods, but I guess I am not quite sure what you are proposing then.

Do you think that we should just leave it alone and not make a comment about trans? We have already said trans is worse than saturated fat. Should we not make a message at all about the health impacts of hydrogenated oils?

I think that, you know, shame on industry if they want to take hydrogenation out and put something else in before testing it and doing studies themselves, and I think they should take responsibility. I mean I think there should be some onus on industry also, if you are going to create a food product, that you should also do some testing, and I don't think you should just put the onus on this committee. There has to be some responsibility on both sides.

DR. JOHNSON: Well, I would contend that industry is already responding heavily in the trans fatty acid area. Let me point out that many of the studies that we are talking about were funded by industry.

I was part of ILSI when they funded the Judd II study. That was like a million bucks. So, they have been very science driven in this whole thing, and I really believe that they are trying to respond to the demands in the marketplace, which is ultimately what industry does best, and they are doing a pretty good out of it.

They can't be held accountable for the level of technology that is available to make these changes in the food supply. They are doing the best they can. I haven't made this point up until now, but I was at Pillsbury when the whole issue of animal fat really took off.

When I started work at Pillsbury in 1989, all of our prepared dough products were made with beef tallow, and there was all kinds of criticism about that, oh, my God, animal fat, cholesterol.

Now, there weren't any trans fatty acids, and we responded to the demands of the marketplace and the public health community to get rid of animal fat, and the technology, the best technology available at the time that everybody agreed to was hydrogenated soybean oil.

It gave the functionality that consumers demand, it didn't have any cholesterol, and the industry was regarded as being really responsible for doing that.

Now, we have got new data which nobody anticipated, and the industry is getting beat up, and we are being forced to take another leap. I just caution that we look before leap as much as possible, and the industry is trying to do that with the constraints that they have.

I would like to see more of a partnership in addressing this issue and before we force the issue by saying 1 percent of energy from trans is the number. Consumers aren't going to make the distinction between whether the evidence is consistent or supporting it. That distinction is going to be lost.

I just think there is a danger in showcasing trans fatty acids particularly when although we agreed that trans fatty acids have a different effect on heart disease than saturated fats, it was with the big caveat that didn't get included in our vote that it is at high levels, whatever that is. There is no evidence, at least in my opinion, they are worse at levels that are currently consumed.

That is my last soapbox speech.

DR. SHANNON: I really appreciate Guy's soapbox, but I would like to come back to a few other statements.

First of all, in relation to what you read us would be FDA's interpretation, your last statement was in the absence of a yes, and we have just voted in the absence of a yes.

Then, in the absence of a yes vote, we voted no, there is not the evidence to support, scientific evidence.

So, if we stop now without any kind of follow-up statement, we have left it there, it's the absence of a yes, and that is where we have left it.

Then, I want to come back to this label. Well, first of all, you have said be careful about showcasing trans fats. They are already showcased, they are going to be showcased on the label, so they are showcased, and they are showcased right here on this label.

Let's say it's a label that says there is 6 grams of saturate fat--well, let's say it says 10 grams of saturated fat and 2 grams of trans, a lot of people are going to say woo, this is great, there is just 2 to 8 unless there is some other indication.

Now, what we are doing with this statement is just saying to FDA take that into consideration. Down the road, we are opening the door to continue this consideration of helping out with that 2 versus 10, which is going to be showcased on there.

So, that is just I feel some follow-up statement to our no vote is needed.

DR. KRINSKY: Thank you, Barbara.

Again, not hearing any further discussion on this, Michael suggested that we move on this question, on this statement and this question. What it does is acknowledge that there is not adequate current scientific evidence, but that we do encourage reducing trans fatty acid intake, and we have set a goal or recommended goal in this statement.

So, I would like to call this question, and it again calls for a yes or no answer.

Those of you that are in favor of transmitting this statement to the FDA as an addendum to our votes on the three questions that they supplied us, please raise your hand and indicate yes.

And those voting no?

And abstaining?

As I read this, we have 6 yes votes, no no votes, and 1 abstention. Therefore, this Committee is transmitting this statement to the FDA to use for whatever purposes they see fit, or if not used as they see fit.

With that, if I see no other red lights on, I am going to adjourn the first meeting of the Nutrition Subcommittee of the Food Advisory Committee of the FDA and express my appreciation to all of you for your participation, to Jeanne Latham for the wonderful executive secretary function in whispering in my ear occasionally, and to the FDA staff sitting off to our side, and particularly to the audience behind us. I do want to apologize for facing my back to the audience for this day and a half routine, but apparently those are the rules and regulations of the FDA, Article 6, Section 4, or whatever it is.

So, as far as the Committee is concerned, luncheon will be served, and the luncheon is available at 11:00 a.m. this morning for those of you that eat early, and you are free to lunch with us or not lunch with us.

For those of you that would like your notebooks FedEx'd back to your homes or offices, Amy will see to that.

If there are no other comments, thank you all very much and we adjourned.

[Whereupon, at 11:00 a.m., the meeting was adjourned.]