U. S. Food and Drug Administration
Center for Food Safety and Applied Nutrition
Office of Special Nutritionals
May 11, 1999

Good morning.

[GRAPHIC 1] In the next few minutes I would like to lay out a framework for today's discussions on the topic of implementing the Food and Drug Administration Modernization Act (or FDAMA) provisions -- provisions that provide for health claims and nutrient content claims on food labels. I will outline the tentative approach used by FDA in reviewing the first notification for health claims under the FDAMA, and at the same time highlight the various issues and questions that have surfaced in response to the approach.

[GRAPHIC 2] The FDAMA provisions for health claims and nutrient content claims have been variously described as an "alternative" for the existing petition process, or sometimes as a "fast track" for providing for new claims.

[GRAPHIC 3] Perhaps the situation can most easily be visualized as shown, although this is greatly simplified. For the existing petition process established under NLEA, the petitioner provides the scientific literature that is believed to establish the basis for the diet/health relationship, and a review is conducted -- headed by FDA but often in conjunction with others in the PHS -- to determine if the science supports the existence of the relationship. This review is subject to notice-and-comment rulemaking. Under the FDAMA provisions, an authoritative statement issued by a scientific body serves as the basis for the existence of the diet/health relationship (or for the nutrient content level) and in turn may be used to support a claim on the food label.

[GRAPHIC 4] For today's discussions, I believe it is possible to group the kinds of issues that need to be addressed into four broad categories: Are there general characteristics of authoritative statements? What is FDA's role and how should the agency interact with the scientific bodies? What should notifications for claims based on authoritative statements include? And what should implementing regulations address?

[GRAPHIC 5] As you have just heard from Mr. Lake, claims may be made when a scientific body has published an authoritative statement. The legislation specifies that a statement must be currently in effect. It also indicates that the statement in question must be about the relationship between a nutrient and a disease for a health claim, -- or, if for a nutrient content claim, it must identify the nutrient level. The legislation also indicates that the statement must be published, and that it cannot be a statement of an employee in his or her individual capacity.

[GRAPHIC 6] The initial question to be addressed, and one for which there has been some comment, is: Are all statements published by scientific bodies authoritative? The legislation refers to authoritative statements, not just to statements -- so is it possible that some published statements are authoritative and others are not?

FDA has interpreted the legislation to indicate that not all published statements from scientific bodies are authoritative. Others do not agree. This issue clearly is the first "fork in the road" for our discussions today -- and it is a point for which FDA requests comments and input.

[GRAPHIC 7] If, as FDA assumed, not all statements from scientific bodies are authoritative, the next set of issues focuses on: What distinguishes an authoritative statement? And then how is a statement of a scientific body determined to be authoritative?

[GRAPHIC 8] The statute identifies at least four characteristics of authoritative statements. But it does not seem that even with an in-depth consideration of these characteristics they can be used to distinguish an authoritative statement from the many other statements that might be published by a scientific body and are in currently in effect. So, what is to be done?

Before we discuss authoritative statements per se, let's backtrack to look at the notification review process, and to consider the process sequentially. And please note, for the remainder of my discussion I will be referring primarily to the situation for health claims. FDAMA does make provisions for nutrient content claims, and in most cases the issues are very much the same. I will highlight situations in which they are not.

[GRAPHIC 9] As Mr. Lake mentioned, the legislation requires that the notification be submitted to FDA 120 days prior to using the claim. The notification must identify the statement and provide the specific wording of the claim. During the 120-day time-period, FDA is expected to review the notification and, if appropriate, FDA can prohibit or modify the claim. Otherwise -- in the absence of FDA action -- the claim is authorized by statute.

[GRAPHIC 10] The first step is for FDA to conduct what might be called a routine check-in. The agency would look to see if all the pieces are provided within the notification, is a statement identified, is the source correctly cited and is the source a scientific body, is there a balanced presentation of scientific literature as required by FDAMA, etc.

But after this initial check-in, what happens? What are the steps to be taken and what is the framework for the overall approach?

[GRAPHIC 11] Clearly, from the agency's perspective, the scientific body is the best arbiter of whether its statement is authoritative. That is, a scientific body is in the best position to decide when one of its statements is authoritative under the FDAMA provisions.

[GRAPHIC 12] If FDA has a notification in hand, it needs to know if the particular statement in question is authoritative. What should the agency do? How should FDA get this input from the scientific body?

[GRAPHIC 13] To create a channel of communication between FDA and the scientific bodies relative to the FDAMA provisions, Secretary Shalala contacted federal scientific bodies and requested that they designate a person, with sufficient authority to speak for the scientific body, to serve as a member of an authoritative statement liaison group. This group came into existence and eventually, through this group, scientific bodies provided input as to whether statements attributed to them in the first notification were authoritative.

The four agencies listed on this slide were tentatively identified as the federal scientific bodies. The National Academy of Sciences (or NAS) is not a federal agency, but the legislation does specifically identify NAS as a scientific body for the purposes of FDAMA. Because of legal concerns, it was not clear how FDA should interact with NAS. And so, in addition to being interested in input as to how the agency can best work with the federal scientific bodies, we also need input as to how to facilitate similar communications with NAS.

[GRAPHIC 14] In considering how to conduct activities under this general check-in step, FDA believed there would be some benefit in tentatively identifying what could be considered general characteristics of an authoritative statement. FDA viewed such a listing as helpful to notifiers in distinguishing authoritative statements from the body of statements issued by scientific bodies. The agency also believed that a tentative listing, if developed, needed to be done in cooperation with the scientific bodies.

[GRAPHIC 15] In the spring of 1998, discussions with the federal scientific bodies, through the Liaison Group, were initiated. There was a dialogue about how the channel of communication could function, and in particular about the nature of authoritative statements as well as about what might be general characteristics of such statements.

You will hear from these members of the Liaison Group later today, but I believe that it is fair to describe these discussions as open, with a great deal of give and take. The interest -- and again you will be able to listen to representatives from the scientific bodies this afternoon -- was in establishing general descriptive characteristics with the expectation that the scientific bodies would make their own interpretation of the characteristics, as well as how the characteristics would be applicable within their agencies.

NAS, as I have mentioned, NAS has not been a member of the liaison group, but FDA representatives did meet with NAS and provided them with an overview of the approach being taken within the Liaison Group and the points being discussed.

This effort to pull together general characteristics of authoritative statements has been described by others outside the process as helpful and appropriate, and by some as inappropriate, misguided and even heavy-handed on the part of FDA. Today we are asking for your input as to their utility and appropriateness.

[GRAPHIC 16] The approach used was to turn to the legislative history of the FDAMA provisions to understand the nature of an authoritative statement. In this way, three characteristics were added to the four specified in the legislation: the statement has undergone a deliberative review conducted by the body, it is not about a preliminary or inconclusive relationship, and it reflects the official policy of the body.

[GRAPHIC 17] Concerning deliberative review, both the House and the Senate reports highlighted the characteristic of what they described as a deliberative review. So, FDA and the Liaison Group added the characteristic of deliberative review to the list of general characteristics. Neither FDA nor the Liaison Group attempted to define what is a deliberative review, its definition being left to each of the scientific bodies.

[GRAPHIC 18 ] Second, it was also noted that the House Report indicated that an authoritative statement would have a "presumption of validity" and that more "scientifically sound" nutrition information would be provided to consumers through claims based on such statements. This discussion suggested that authoritative statements would not be statements that were about preliminary or inconclusive relationships.

[GRAPHIC 19] Third, language in the statute indicates that authoritative statements are published statements of, for example, NIH or CDC and not of their subdivisions, nor of individual employees in their own capacity. In addition, the legislative history indicates that authoritative statements should reflect consensus within the federal scientific body. On this basis, the characteristic of representing a body's official position about the relationship was included. No definition for "official policy" was attempted, given the recognition that each scientific body had to determine its own official policy and what it would mean for there to be consensus within it. As I believe you will hear from some of the scientific bodies today, more extensive consideration on their part concerning the desirability of characterizing an authoritative statement as official policy is undergoing second thoughts, and of course is open for comment.

[GRAPHIC 20] These three added characteristics have been the subject of a great deal of discussion. Some people disagree that it is appropriate to link the use of "validity" and "scientifically sound" in the legislative history to the concept of not preliminary or inconclusive. Others are concerned that reference to "deliberative review" and "not preliminary or inconclusive" was in effect forcing the standard of significant scientific agreement (SSA for short) to be the defining characteristic of an authoritative statement. [GRAPHIC 21] FDAMA makes reference to the SSA standard, but not in the context of discussing an authoritative statement. As provided for by FDAMA, the issue of SSA comes further along in the process, after statements are determined to be authoritative. In reviewing the first notification that FDA received, the agency did not use the SSA standard as part of its review because the review did not get that far in the process.

However, there are certainly questions about the use of the SSA standard as part of the FDAMA provisions, but they need to be addressed at a different point in the process. So, we will come back to the issue of SSA in just a moment.

For now, we are asking input on these three added characteristics. They have not been defined within FDA's tentative approach, only identified. As general characteristics, are they useful for helping to sort authoritative statements from other statements published by a scientific body? Are they appropriate?

[GRAPHIC 22] Continuing now with questions about process, given the seven general characteristics and based on what FDA experienced with the first notification, the agency considered whether there was utility in and whether it was appropriate for FDA to conduct a kind of screening or initial sort relative to a statement identified by a notifier as authoritative.

Here some examples from our experience with the first notification:

[GRAPHIC 23] One statement identified in the first notification as an authoritative statement in support of a health claim about chromium was: "Scientists must often draw inferences about the relationships between dietary factors and disease from animal studies or human metabolic and population studies that approach issues indirectly." This statement is found in the Surgeon General's Report. It does not identify a substance nor refer to chromium, nor is there reference to a specific disease. FDA therefore believed it could determine that the statement was not authoritative.

[GRAPHIC 24] A second example raised the issue of preliminary and inconclusive. "If the findings hold up in further research, eating more ..... " and the statement goes on to discuss beta-carotene from fruits and vegetables. Is the statement about a preliminary or inconclusive relationship? In this case, FDA believed that the language of the statement suggested a preliminary and inconclusive relationship, and FDA, in turn, asked the scientific body to whom the statement was attributed to determine whether the statement was authoritative. Could FDA have determined whether the statement was authoritative?

[GRAPHIC 25] Similarly, another statement submitted in the first notification was from a review article in a professional journal by a scientific body that read: "Antioxidant micronutrients, especially carotenes, vitamin C, and Vitamin E, appear to play many important roles in protecting the body against cancer. They block the formation of chemical carcinogens in the stomach, protect DNA and lipid membranes from oxidative damage, and enhance immune function." In reviewing the source material, FDA noted that the next sentence in the article stated: "Nevertheless, many important questions need to be answered before either micronutrient supplements or food fortification can be recommended as a cancer prevention strategy to the general population." This raises questions about the use of context.

[GRAPHIC 26] And finally, from our experience, the agency does believe that as a general matter it could evaluate a statement such as this one, again from the first notification: "High dietary carotene and possibly vitamins C and E and folate are associated with reduced risk for cervical cancer." The statement is about a diet/disease relationship and is not qualified, and FDA is not easily in the position of judging the basis upon which this statement rests. It would seem to require input from the scientific body.

The agency did review the statements included in the first notification and did discuss in the preamble to the interim final rules how the statements matched or didn't match the various characteristics that had been tentatively identified. [GRAPHIC 27] However, because it is acknowledged that the determination of whether a statement is authoritative rests with the scientific bodies, is a preliminary screening appropriate and useful? If so, how can it be defined or specified so to ensure its acceptability? For instance, can guidelines be developed? Can there be clarification as to the meaning and application of the so-called "context" which surrounds the statement within a paragraph, page or chapter?

[GRAPHIC 28] You will hear from the scientific bodies later today concerning their view on moving through these various steps, and what will be their approach as individual agencies and as members of the Liaison Group. An important question that FDA would like answered is: How can the agency best establish the needed working relationship with the scientific bodies given that ultimately the regulatory responsibility of addressing inappropriate claims rests with FDA? Given that in general the determination of whether a statement is authoritative rests within a scientific body, should each statement be forwarded to the body for review? Are there any other options? If a scientific body declines to review a statement, what are FDA's options at that point? What if a scientific body declined to review individual statements, but issued a policy guideline instead? How should FDA use such a document?

[GRAPHIC 29] As we move along the schematic, other factors come into play and questions crop up. When a decision passes the step at the point where a scientific body determines whether the statement is authoritative or not, the process returns to FDA. FDAMA has given FDA several bases on which to prohibit a claim. For instance, FDA may prohibit a claim when the statement upon which the claim is based is determined not to be authoritative. And in most of the nine interim final rules published in response to the first notification, this was the basis for the agency's acting to prohibit use of the claim; prohibition occurred because the statement was not considered to be authoritative.

But if the statement is authoritative, the process continues along toward other considerations relative to the claim. And at this point, after the statement is determined to be authoritative, it is possible that the issue of SSA could arise.

[GRAPHIC 30] The legislation, while not requiring FDA to determine SSA affirmatively, does provide that FDA may prohibit a claim if there is not SSA. So, it seems that at some point after the statement is determined to be authoritative, the standard of SSA could come into play. But what is the practical application of this provision, what is FDA to do? How can FDA consider SSA during the initial review process? For example, should FDA request comment from all the scientific bodies? Are there other approaches that FDA could use or should be taken into account? Clearly the issue of SSA is an important one and we specifically are requesting input to help us clarify the best approach. Again, we never reached this point with the first set of claims.

[GRAPHIC 31] Finally, we need to seriously consider this last "home stretch" of our procedural process. There are some issues that must be addressed, issues that impact on what should or could be required in a notification, and that bring up questions about the relevance of existing NLEA-based regulations for claims that would be provided for under FDAMA provisions.

[GRAPHIC 32] Clearly, a notification submission is required by FDAMA to include the identification of an authoritative statement, a stated claim, and a balanced presentation of the scientific literature. We would find it useful to have comments that address the balanced presentation -- what should it be, what should it include, and how should FDA evaluate or use it?

But as many of you may know, there are quite a number of specific requirements already in place for health claims and for nutrient content claims. For instance, Section 101.13 lays out the general principles for nutrient content claims and covers a range of topics from expressed/implied claims to location on food label panel and referral statements. Subpart D gives the specific quantitative criteria for the claims. How does FDAMA account for these existing provisions? Should they be incorporated into the implementation process for FDAMA claims?

Likewise, for health claims, there is Section 101.14 which deals with various definitional issues for such claims as well as disqualifying levels and substance eligibility criteria. There are provisions such as the so-called jelly bean rule. Again, are these applicable to claims made under the FDAMA provisions? In the case of health claims, specifically, what happens if the authoritative statement does not provide information on the effective level of the substance that is the subject of the relationship. Isn't it important to be able to specify how much of a substance is needed daily in order to have the effect? Is this necessarily information that would be found in an authoritative statement? How does the agency deal with providing for a claim when the effective level is not addressed? Is it important to be able to specify how much of the substance is needed daily because claims on foods with too little of the nutrient would be misleading to consumers?

Another related issue focuses on the need for an accepted analytical method to determine the presence of the substance in a food. Again, is it likely that an authoritative statement would provide such information? Is this the responsibility of the notifier?

[GRAPHIC 33] Finally, Section 101.14 (d) provides for the wording of health claims and focuses especially on the need for the claim to be an accurate reflection of the science as well as other provisions. Given the importance of claim wording -- both so that it can accurately reflect the science of the relationship and so that consumers can fully comprehend the claim within the context of the total daily diet -- how can or should such decisions be made?

In short, given this overview, what are the approaches to be used, what about the existing provisions -- How do they fit into the FDAMA provisions? What specifically should the notifiers be required to provide beyond the authoritative statement, the claim wording, and the balanced representation? How does FDA make decisions about appropriate wording of claims, especially in terms of its reflecting accurate the state of the science, when the science review is conducted outside FDA? Basically: What is needed and who does it?

[GRAPHIC 34] I will end our series of questions -- our framework for discussion -- by asking three broad and final questions that have great impact:

(1) As many of you may know, the FDAMA provisions for health claims only apply to conventional foods and do not provide for their use on dietary supplements (although this is not the case for nutrient content claims). FDA considered this omission, and also considered the interest in providing a level playing field for both supplements and conventional foods. In January, the agency proposed to expand the provisions to include dietary supplements. The comment period closed April 6 -- but we are open to discussion on this topic today.

(2) Regarding existing health claims, given that (a) there is a clause within Section 403 of the act added by FDAMA that refers to FDAMA claims as being those not already authorized by the Secretary, and (b) the Senate report mentions that once an authorized claim exists, claims cannot be made unless they are consistent with the existing provisions, how does this affect efforts to modify existing health claims through FDAMA notifications? FDA has taken the tentative approach that the FDAMA notification process cannot be used to modify existing health claims. We'd like your input on how these provisions are to be interpreted and used by FDA.

And (3), the legislation is silent as to whether notifications received by FDA for use of authoritative statements are to be made public so that the world at large knows that the statement is undergoing review and what the potential claim will be. Or, alternatively, whether the information in such a notification is to be considered confidential or proprietary. Because FDA has not yet made a determination about this issue, it has been cautious and did not immediately place the first notification received in a public docket. But there are some who believe such notifications should be made public when FDA receives them. The agency may need to make provisions for one approach or another, and we'd be interested in hearing your views on this.

We hope this framework is helpful to you in approaching our discussion for today. Admittedly what has been outlined is a large set of complex and interrelated questions, but they are also important questions. FDA is eager to hear discussion and to gather information about any and all of the concerns that surround the use of authoritative statements as the basis for health and nutrient content claims. Thank you and we look forward to the rest of the day's discussion.

^*Presented by Dr. Christine Lewis, Office of Special Nutritionals, Center for Food Safety and Applied Nutrition, Food and Drug Administration

Hypertext updated by kwg/ear/dms 2005-OCT-19