Cover Image

HARDBACK
$59.95



View/Hide Left Panel

3
The Framework

Under the provisions of the Dietary Supplement Health and Education Act (DSHEA), dietary supplements are to be considered as foods and assumed safe unless the Food and Drug Administration (FDA) has evidence that the supplement or one of its ingredients presents “a significant or unreasonable risk of illness or injury” when used as directed on the label or under normal conditions of use. Since the FDA is not authorized to require or impose premarket safety evaluations for dietary supplement ingredients marketed for use in the United States before October 15, 1994, FDA itself must monitor safety data and gather and assess existing information on safety to determine if a significant or unreasonable risk is present.

Thus the purpose of the Framework1 described in this chapter is to provide a process for FDA to translate the results of their scientific review into a decision regarding regulatory action needed to protect the health of the public.

CONSIDERATIONS IN DESIGNING THE FRAMEWORK

The Framework consists of two components: (1) a process for prioritizing, evaluating, and describing available information to establish risk of

1  

A “framework” for safety evaluation of dietary supplement ingredients is characterized by the application of generally accepted scientific principles relating to adverse effects in order to make decisions of whether significant or unreasonable risk exists, thus overturning the a priori assumption of safety.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety 3 The Framework Under the provisions of the Dietary Supplement Health and Education Act (DSHEA), dietary supplements are to be considered as foods and assumed safe unless the Food and Drug Administration (FDA) has evidence that the supplement or one of its ingredients presents “a significant or unreasonable risk of illness or injury” when used as directed on the label or under normal conditions of use. Since the FDA is not authorized to require or impose premarket safety evaluations for dietary supplement ingredients marketed for use in the United States before October 15, 1994, FDA itself must monitor safety data and gather and assess existing information on safety to determine if a significant or unreasonable risk is present. Thus the purpose of the Framework1 described in this chapter is to provide a process for FDA to translate the results of their scientific review into a decision regarding regulatory action needed to protect the health of the public. CONSIDERATIONS IN DESIGNING THE FRAMEWORK The Framework consists of two components: (1) a process for prioritizing, evaluating, and describing available information to establish risk of 1   A “framework” for safety evaluation of dietary supplement ingredients is characterized by the application of generally accepted scientific principles relating to adverse effects in order to make decisions of whether significant or unreasonable risk exists, thus overturning the a priori assumption of safety.

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety harm, and (2) a set of science-based principles that serve as guidelines for evaluating risk to human health. For the Framework to be useful, FDA must have adequate resources for implementation. To be credible, it must be scientifically based and include guidelines for obtaining and integrating the totality of the information from many areas of science. The Framework should allow FDA to react to information, as well as to proactively gather information. It needs to be efficient and provide a system for updating information as new information becomes available. In providing a scientific infrastructure for the evaluation of the safety of dietary supplement ingredients, the framework must facilitate decision-making regarding a dietary supplement’s potential to cause harm when uncertainty exists. Adequate staff with appropriate expertise must be available within FDA to administer the process and evaluate the information. The Framework described here characterizes the nature of the scientific evidence that FDA is likely to encounter and describes a process for organizing this evidence to assess where a dietary supplement ingredient lies on a spectrum of concern2 (see Figure 3-1). As the level of concern increases, so does the potential for a “significant or unreasonable risk,” the standard warranting regulation under the Food, Drug, and Cosmetic Act (FDCA), as amended by DSHEA. I. THE PROCESS The process comprises three major components: Signal detection, Initial review of the signal, and Integrative evaluation. Signal Detection According to the DSHEA, it is assumed that dietary supplements are generally safe; given the large number of dietary supplement ingredients, it is unlikely that FDA will have the resources or the need to evaluate each ingredient uniformly to determine if it presents an unreasonable risk of illness or injury. Thus, at least initially, it is assumed that some “signal” will indicate that an ingredient’s safety may need to be reviewed. When a signal is detected, it is up to FDA to decide the next step once the credibility of the 2   The use of the term “concern” denotes a need for further investigation and inquiry by FDA based on a relative level of interest arising from initial information.

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety FIGURE 3-1 Diagram of the three components of the Safety Framework: signal detection, identification of level of concern in an initial review, and integrative evaluation, as well as how these components feed into FDA’s decision to take action.

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety signal is evaluated and to determine the possibility that the ingredient caused the adverse effect noted (using the guiding principles outlined later in this chapter and discussed in detail in the chapters that follow). What Constitutes a Signal? FDA is likely to receive or become aware of a variety of signals suggesting potential risks to human health with the use of a dietary supplement ingredient. Signals may come to FDA’s attention and thus be “detected” through notice of regulatory action taken by other countries regarding a specific dietary supplement, through routine monitoring of medical and scientific literature, directly through it’s own Special Nutrition/Adverse Event Monitoring System, or through consultation with experts. FDA may also become aware of signals indirectly through reports in the media, through new data from animal experiments suggesting a specific risk, or through information provided by consumer protection advocacy groups. Signals can thus come from many sources and originate from many different types of scientific data. Given the significant number of dietary supplement ingredients, FDA’s attention should focus on signals that indicate a serious3 health problem may result from ingestion of a dietary supplement ingredient. Quality of the Signal In this first component little is done to evaluate the quality of the data because the focus is simply on signal detection. While some signals may result from concerns expressed by other expert bodies, such as those described in Chapter 2, or by case reports of adverse effects, the quality of the signal is not reviewed until the second component of the process (initial review of available information). The quality of the information behind the signals detected will be highly variable and in some cases may provide only weak evidence or be of little use or credence. Nonetheless, detecting these signals requires the attention of qualified professional staff at FDA and will result in a reaction by FDA (even if the reaction is only to consider the signal of little importance, as described in the next component). 3   “Serious” is defined as any experience resulting in any of the following outcomes: death, a life-threatening adverse experience, inpatient hospitalization or prolongation of existing hospitalization, a persistent or significant disability/incapacity, or a congenital anomaly/birth defect. Important medical events that may not result in death, be life-threatening, or require hospitalization may be considered serious when, based upon appropriate medical judgment, they may jeopardize the individual and may require medical or surgical intervention to prevent one of the outcomes previously listed (in accordance with 21 C.F.R. § 600.80 [2002] and 21 C.F.R. § 314.80 [2002]).

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety Proactive Initiation of Review In contrast to reaction based on detecting a signal, FDA may decide to proactively initiate a review of a dietary supplement ingredient due to high prevalence of use in the general population, high prevalence of use by a particularly vulnerable population, or other factors. More than likely, however, a signal indicating possible concerns will be the instigating factor in further review of a substance. Prioritization for Review One of the requirements of the study was to develop a framework that would include criteria for how the review of safety of dietary supplements and ingredients should be prioritized. It was suggested that a scheme to initially identify dietary supplement ingredients considered of higher priority for subsequent review based on only one criterion, such as end-organ toxicity in animal studies or the structure of one or more known compounds present in the ingredient be devised and applied to all dietary supplement ingredients. However, given the wide variety of dietary supplement ingredients available, the multiple forms of a specific ingredient that are sold, the voluntary and thus varying nature of the data available on an ingredient, and the wide variety of adverse effects that are possible for dietary supplements and the dependence of such effects on exposure levels, such a scheme is not feasible nor scientifically defensible. This is not to say that the availability of data from only one category is not enough to determine a higher level of concern. As emphasized in the following chapters on the various types of data, any one category of information can raise concern to a level that requires action by FDA. A hierarchy of adverse effects that warrant greater concern than others based on collective judgment is provided in Chapters 4 through 7; however, a formulaic or algorithmic approach that considers all the important variables—such as the dose at which such effects may occur, the relevance of the information, the information available suggesting the ingredient may be safe—is not useful given the multidimensional matrix that would be needed. The signal detection step, followed by an initial review of the information, should serve to identify those dietary supplement ingredients that are in need of further review and evaluation via an integrated evaluation. Initial Review of Available Information The second component of the Framework is to conduct an initial review of available information. First, the nature of the information generating the signal is examined to determine the appropriate level of concern

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety regarding a risk to human health. This component is not envisioned as a detailed analysis of data, but rather as an assessment of the concern level warranted by the nature of the evidence (e.g., quality of the report, applicability to humans, route of exposure) and whether the information raises questions that require further examination. Second, some effort can be made to gather easily available data to place the detected signal in context; such additional information may come from many sources, including other categories of data. Thus this initial review of the signal information need not be limited to reviewing only the information associated with the signal. For example, if the signal is a case report suggesting a possible problem in an elderly woman and clinical trials of the ingredient exist, these should be considered during the initial review. Level of Concern The outcome of the initial review is a determination of the initial overall level of concern to decide if an integrative evaluation is needed. Higher concerns warrant an integrative evaluation; lower concerns do not.4 A decision about an ingredient with a moderate concern level should be made after a review of other information to see if other signals are apparent; for example, if the initial signal is animal data that warrant moderate concern, a cursory literature search on the substance or a review of FDA’s adverse event monitoring system could be conducted to determine if other data about the ingredient raise concerns as well, leading to the need for further evaluation. Assuming that sufficient evidence may not be available from just one type or category of data to cause a higher level of concern, it is important for FDA to consider data from other categories to determine if a higher level of concern may exist. Decisions Possible Based on Initial Review When the initial review of the nature of the evidence available indicates a higher level of concern, FDA would then initiate an integrative evaluation process or possibly decide to take immediate action, if the concern is serious enough and the data are strong. If the level of concern is categorized as relatively low, FDA would continue to monitor signals and incorporate the 4   “Higher” level of concern is relative to the level of concern warranted by other evidence. The terms “higher” and “lower” are used to indicate that the level of concern is relative, rather than categorical.

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety information obtained into a monitoring database for future use if new data regarding the ingredient become available. Maintaining a database of specific issues to monitor would allow FDA staff familiar with the criteria outlined in Chapters 4 through 8 to systematically look for information that may address the data gaps. Similarly, data collected should be saved in case a decision is made to move to an integrative evaluation. If a decision is made to conduct an integrative evaluation, but a monograph is not subsequently prepared, then information and a summary of the thought processes involved in the integrative evaluation should be noted and filed for future consideration. Making data gaps and unanswered questions available to other interested parties such as the National Toxicology Program of the National Institute of Environmental Health Sciences (NIEHS) or the Office of Dietary Supplements, both part of the National Institutes of Health (NIH), Department of Health and Human Services, would allow them to incorporate these data needs on specific dietary supplement ingredients into their programs of work. In summary, once the initial level of concern based on the initial review of the signal is determined, FDA might decide that continued routine monitoring is needed, or it could decide to proceed with an integrative evaluation. This depends on the level of concern raised by the signal: ingredients provoking higher concern should proceed to the integrative evaluation; ingredients resulting in lower levels of concern would generally not proceed; and ingredients with moderate concerns might proceed after considering additional information not necessarily related to the initial signal, such as prevalence of use or concern related to a specific vulnerable population group. Since it is assumed by the DSHEA that dietary supplements are safe, there should be relatively few dietary supplement ingredients that will be categorized as of higher concern after the initial review and thus warrant further examination. This allows FDA to focus its efforts on dietary supplement ingredients that are strong candidates for regulation. Integrative Evaluation The third component of the Framework is to conduct an integrative evaluation for those dietary supplement ingredients that are deemed to warrant further investigation, based on the preliminary data reviewed in the second component and the resulting relative placement on the spectra of concern continuums. There are four aspects to the Integrative Evaluation component (see Figure 3-1): in-depth literature search and review, drafting a safety monograph based on this information, integrating the available data into an analysis to complete the monograph, and possibly referring the draft monograph and accompanying information to an expert committee for additional input prior to determining whether to take regulatory action.

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety Reviewing the Literature A critical review of the literature is a three-part process. First, multiple databases are searched for information on the dietary supplement ingredient and other substances with similar taxonomical, structural, or functional properties. Such searches are broad-based, and include information on safety and biological activity of the ingredients, including human data, animal data, and in vitro data. Second, each primary research paper is reviewed for internal consistency; for example, are proper methodologies used? Do data fit the conclusions? Are the associations real? Is appropriate information included? Are there chance, bias, confounding variables, a lack of coherence, or other significant internal issues or limitations that should be taken into account? Third, the external consistency of the research papers must be judged as a group. Are the studies coherent as a whole? Is there strength in the associations, general agreement, etc.? Studies can then be sorted into those that suggest that there is little risk of illness or injury when consuming the supplement ingredient, those that indicate a relevant concern for risk of illness or injury, and those that have equivocal results. Each should then be examined for flaws and strengths in accordance with the principles and concepts discussed in the subsequent chapters on each general category of data (Chapters 4 through 7). Focused Versus Broad-Based Evaluation. An integrative evaluation might be reactive to the signal and focused in nature, in that it is conducted to examine a specific moderate- or high-level concern about an ingredient, or it might be more proactive and broad-based, in that it is looking for any risk associated with use of the dietary supplement ingredient. As described above in the description of the signal detection component, a proactive integrative evaluation might be initiated simply because a large percentage of the population is using the ingredient, rather than as a reaction to a particular safety concern. The amount of information gathered depends on the nature of the harmful effect that is the focus of concern. If a focused evaluation is conducted, it is assumed that less information will be reviewed. However, the relative importance of an individual study is established in conjunction with an evaluation of other relevant literature. Clearly, data or information outside the primary safety concern may include information that has a direct bearing on the overall evaluation of the safety concern identified in the signal component. Thus a comprehensive review can provide information that may raise concerns in other areas not relevant to the focus, but which should not be ignored in a safety monograph.

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety Relevancy of Data. Gathering data and reviewing it for relevance provides the scientific base upon which FDA can substantiate its conclusions. Data that are not relevant to safety or to the concern in a focused integrated evaluation need not be incorporated into the report; however, acknowledging that it was considered and deemed irrelevant will be helpful if the information has been characterized by others as substantiating safety. Drafting a Safety Monograph In most cases, the integrative evaluation will be documented in a monograph5 that summarizes the categories of data available and their use in drawing conclusions about the potential risk associated with use of the ingredient. Evidence obtained either from only one category of data or from integrating all the categories that results in an increased level of concern should result in a higher priority for development of a safety monograph. A monograph need not be developed for every dietary supplement ingredient, as it is assumed that only those ingredients with moderate or higher concern levels following the initial review will be subject to an integrative evaluation (essential nutrients represent a special case; see Annex 3-1). The monograph may not need to cover every concern, in which case a focused integrative evaluation and resulting monograph would be completed. In a few cases, where available information obtained in the initial review results in a highly significant level of concern, it may be necessary to undertake regulatory action prior to or without developing a monograph. The development of a monograph may be resource and time intensive, especially when initiated proactively and thus with a much broader focus. However, the development of a monograph provides a method to document in a systematic format the evidence on which FDA can base a regulatory decision. The science-based guiding principles described in the following sections of this chapter, and explained in detail in the following chapters, should be used to reach a decision regarding whether there is an unreasonable risk of illness or injury. The general types of information to be collected and used in the integrative evaluation and thus collated in a monograph are listed in Box 3-1 and include a description of the ingredient (e.g., constituents, different types of preparations, typical intake amount and duration, historical use) and available information about toxicities and safety (human data, animal data, data describing risks associated with related substances, and in vitro data). In most cases, this information will be gathered from the medical and 5   A monograph is defined as a “learned treatise on a small area of knowledge” (MerriamWebster, 2001).

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety BOX 3-1 Safety Monograph DescriptIon of the Ingredient Constituents as appropriate: chemical classification and structure; for a botanical ingredient, genus, species, part of plant; for an animal ingredient, genus, species, part of body Description of individual components, alternative forms or secondary constituents Descriptions of different dietary supplement preparations, how they may differ in constituents, and amounts ingested in ordinary use Safety Information (in narrative form, referencing data tables) Human safety data (serious and nonserious adverse effects): historical use, if applicable; adverse effects from clinical trials, clinical case reports, and adverse event reports; interactions (i.e., drug interactions); consequences of unusually large intake and/or chronic cumulative use Animal studies In vitro studies Biological activity of related substances (structurally, taxonomically, and functionally related) Other Relevant Information (if known and applicable) Sources of ingredient Conditions of use suggested or recommended in labeling or other marketing material Cautions about use from historical use, labeling, or other marketing material Usage patterns (prevalence of use in the general population, use by vulnerable groups) Information on regulatory actions, including those of regulatory agencies in other countries Available information on physiological and biochemical aspects (bioavailability, distribution, metabolism) Summary and Conclusions Summary Conclusions about the safety of the ingredient, based on the strength of the scientific evidence Unresolved issues and uncertainties in the available data Data gaps and future research recommended Literature Search Strategy Literature Cited Tables of Key Data Evaluated

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety scientific literature. However, additional information may be obtained by requesting information from clinical investigators who have published reports about the particular ingredient, as well as by requesting information from industry (e.g., distributors and manufacturers) and other stakeholders. The collected information should be collated into a draft safety monograph. The monograph should be prepared using a standard format to summarize all the data collected on the ingredient (see Annex 3-2 to this chapter for a more detailed discussion of monograph preparation). Integrating the Data to Determine Risk The data evaluation component of the integrative evaluation should be conducted on the initial assumption that consumption of the supplement should not present an unreasonable risk of illness or injury, as is assumed in DSHEA. To overturn this assumption, the end result of the review should demonstrate that there is an unreasonable risk of illness or injury to the consumer. When evidence on a dietary supplement ingredient presents a moderate or higher level of concern relative to this risk, data from other categories should be considered to evaluate biological plausibility and consistency. Integration within and across the other categories of data will help determine if an unreasonable risk exists by looking at the overall picture. Such an analysis can be represented by creating a causal model diagram—a tool to organize the data to visualize how the different types of available data link together to establish risk (described in Chapter 10). For example, in reviewing the potential for concern in the use of saw palmetto for the prototype monograph described in this report, data from all categories were integrated to make a conclusion about risk (see Chapter 11 and Appendix H). The principles described for considering the various types of data and modifying factors (Chapters 4 through 9), as well as the principles described for how to integrate among and within categories of data (Chapter 10), are applied in the integrative evaluation. They should be followed in assessing and weighing the different types of evidence that enter into the decision. They are summarized in the conclusions in the safety monograph. The conclusions should describe: The relevance of the evidence; How the dose, manner of use, and product affect conclusions about risk; The seriousness of the potential harm suggested by the evidence; and The quality and strength of the evidence.

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety plasma volume10 to estimate the maximum achievable concentration from a single dose. These assumptions may not be accurate, but they do provide a reasonable basis for making decisions. Bioavailability is further discussed in Chapter 5 as it relates to route of exposure when considering exposure of animals through non-oral routes. It is also discussed in Chapter 6, when looking at concentrations of substances that are similar to endogenous substances. In Chapter 10, the concept is discussed in terms of integrating data that may appear inconsistent. Proof of Harm To evaluate the safety of dietary supplements, it is necessary to determine if an unreasonable or significant risk exists—not to have complete evidence that a dietary supplement ingredient causes a serious adverse event—which is a lower standard than conclusive proof. The difference between proof of harm and risk of harm should be considered when judgment rather than strict interpretation of facts needs to be made. Absence of Evidence Absence of evidence of risk does not indicate that there is no risk. In some cases, some data will indicate a risk, while other data will not suggest the risk exists, producing what could be interpreted as an inconsistency. Even if a study showing lack of adverse effects is reported, if the study is not adequately designed to identify risk (e.g., not sufficiently powered, incompletely reported, does not include positive controls, or otherwise has inadequate mechanisms for detecting adverse events), it is not scientifically valid to use such information to mitigate suggested risk from other sources. This concept is discussed in Chapter 4, as it relates to comparing different types of human data. It is also discussed in Chapter 10, as it relates to comparing human and animal data. Considering Consistency and Biological Plausibility In many circumstances, data will need to be collated within the same category or across several categories to determine the appropriate overall level of concern. In integrating observations across categories of data, consistency and evidence of biological plausibility should raise the level of 10   Plasma volume is used when actual volume of distribution is unknown. If the substance is known to distribute across the cell membrane or into a different distribution space, division by body water volume or other volume would be appropriate.

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety concern. This weaving together of available information can be facilitated, and conceptually illustrated, by the use of causal evidence models. A causal evidence model (see Figure 3-2) provides a structure to help interpret available data from a number of sources in order to address a specific safety question, and is explained in detail in Chapter 10. The model describes the relationship among a dietary supplement ingredient, potential adverse health effects, and biological effects by depicting the relationship as linkages illustrated with arrows. The type of arrow illustrates the type of evidence: convincing data are depicted by solid arrows, and weak or less conclusive data are depicted by dashed arrows. The path between a dietary supplement ingredient and an adverse health effect illustrates the strength of their potential relationship. When the available information is integrated, multiple links between the dietary supplement ingredient and a given health outcome are illustrated by multiple arrows. Evidence from all types of study designs may form linkages to aid in determining the extent of association between dietary supplement exposure and adverse health effects or outcomes. Causal models are useful when a single type of evidence is weak or does not illustrate a relationship, but other related information is available, as may often be the case with dietary supplement ingredients. FIGURE 3-2 Diagram of causal model relating information to an adverse health effect.

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety UTILIZING THE FRAMEWORK The request from FDA to develop a framework for evaluating the safety of dietary supplement ingredients also included a request that prototype monographs for six dietary supplement ingredients be developed as examples of how the Framework should be applied. Chapter 11 provides case studies of how the available evidence for six dietary supplement ingredients could be evaluated using the spectra of concern discussed in this chapter and described in detail in the following chapters. Given the fact that these are prototype monographs, they should not be considered as representing findings related to these six dietary supplement ingredients. Rather, they are examples of how to approach reviewing and evaluating the various types of available information on dietary ingredients. Appendixes D through I contain summaries of the six prototype monographs. (The full prototype monographs are available on the web, at www.iom.edu/fnb.) Appendixes J and K contains examples of two focused prototype monographs to show how FDA could focus on determining a level of concern related to one specific adverse effect or outcome when identified. Conducting a broad-based comprehensive assessment would typically identify all data about the dietary supplement ingredient and would lead to a description and evaluation of other adverse effects—which would be a resource-intensive process. SUMMARY This chapter outlines a system for conducting a review of the safety of dietary supplement ingredients. Conducting the safety evaluation consists of three components: signal detection, an initial review of the available information, and, when needed, an integrative evaluation. Based on detection of a signal or a proactively initiated review of a dietary supplement ingredient, FDA evaluates the detected signal by conducting a brief initial review of readily available information to determine whether there is a need for a comprehensive review, termed an integrative evaluation. When an integrative evaluation is undertaken, FDA, or a contractor of FDA, prepares the initial draft monograph that is a collection and review of available safety information. In some cases, the integrative evaluation process, during which a draft monograph is developed, may provide sufficient evidence for FDA to decide on a course of action without use of an external advisory committee or public input. However, it is expected that when the data are not sufficiently definitive for FDA to make a decision about whether to take action, it will request the assistance of an advisory committee to review the information. The external advisory committee, if constituted, reviews the draft monograph, determines if additional information should be collected, and holds

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety sessions for input from the public. It then recommends modifications to the draft monograph as appropriate and summarizes concerns based on the evidence. The completed monograph, with input from the external advisory committee, is then made public in an easily accessible format. In any scientific evaluation, at least four categories of information can be considered informative for evaluating the risk of ingesting chemicals, including dietary supplements: human data, animal data, in vitro data, and information about related substances. Evidence in any one of these four major categories can provide considerable guidance regarding the ingredient’s safety. The chapters that follow provide detailed information on the use of this information, and how to integrate the available data to determine the extent to which an unreasonable risk of illness or injury from ingestion of a dietary supplement exists. ANNEX 3-1 APPLYING THE SAFETY FRAMEWORK TO REQUIRED NUTRIENTS Essential nutrients (i.e., vitamins and mineral elements) are unique compared to many other categories of dietary supplements in that much more data regarding adverse effects of overconsumption in humans are available. Structures of the vitamins have been elucidated, relative activity of closely related compounds determined, and, at least at physiologic doses, biological activities for these compounds are reasonably well characterized. Most have been characterized in terms of potency in standards such as those produced by the U.S. Pharmacopoeia, as is required by law when monograph standards are available for them. Several nutrients are regulated as generally recognized as safe substances, approved food additives, and as over-the-counter and prescription drugs. It is for this reason that the essential nutrients are considered differently within the Framework than other types of dietary supplement ingredients. As summarized in Chapter 2, a system for reviewing data about the safety of vitamins and essential elements already exists. Since 1940, the Food and Nutrition Board of the National Academies has been commissioned by federal agencies to set Recommended Dietary Allowances (RDAs) for nutrients; more recently this was expanded to include other reference intake levels for nutrients, now collectively termed Dietary Reference Intakes (DRIs) (IOM, 1994, 1997). With the exception of only a few nutrients, scientific data on all vitamins and mineral elements, and some other nutrients, have been reviewed recently through the DRI process (IOM, 1997, 1998b, 2000, 2001, 2002, 2004). While the RDAs are designed to be recommendations for intake to ensure that the needs of almost all apparently healthy individuals in a

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety population group (such as women over 70 years of age or adolescent boys) are met in order to avoid nutrient deficiencies and to decrease risk of chronic disease, other reference values included as part of the new DRIs provide upper levels of intake for vitamins and mineral elements that, if consumed below the specified level on a continuing basis, should not cause specific identified adverse effects of overconsumption (IOM, 1998a). These upper levels of intake are called “tolerable upper intake levels,” or ULs. For the vitamins and the mineral elements that have been established as required by humans, a considerable amount of primary data relating to animal or experimental studies, human studies, and in vitro studies are available, and these data have been reviewed as part of the DRI consideration of the UL. The DRI review can thus provide substantial data and background information for an FDA evaluation; the process for incorporating this information is provided in the following sections. Signal Detection for Nutrients When nutrients present in dietary supplements are suggested for use at levels greater than established ULs, it is appropriate to be concerned, and thus this is an initial “signal” analogous to that described for other dietary supplement ingredients. However, the initial review in response to the signal should be focused on the DRI review of the serious adverse effects that were identified as potentially occurring at high intake levels, recognizing that there has been an uncertainty factor applied to ensure that few, if any, individuals will be adversely affected at the UL level. New information (post-DRI review) on adverse effects of consuming a nutrient also serves as an initial signal. In summary, it is appropriate to a priori consider any marketing of nutrients as dietary supplements above the UL to be potentially of some risk, but whether the risk is unreasonable will depend on the data available. Nonetheless, vitamins and mineral elements are not innocuous substances. Consumption at high levels of some nutrients is associated with illnesses and death as documented in the DRI reports (IOM, 1997, 1998b, 2000, 2001, 2002, 2004). Initial Review for a Nutrient Whenever the safety of a vitamin or mineral is considered, the first step should be to consult the results of the DRI UL process. It is possible that a UL was not established for a nutrient when the data were reviewed, but the review process and the data considered are included in the specific DRI report and serve as a credible, nonbiased review of the scientific data available at the time of review. If a UL was established for the vitamin or mineral under question, it is important to consider the following:

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety Was the information suggesting concern available before the publication of the most recent DRI review, and were these data considered in the evaluation of ULs? One limitation is that the reviews are conducted at specified intervals, so it is possible that safety issues relating to dietary supplements that are essential nutrients might be newly identified within the time period after the last DRI review and prior to an upcoming updated review. Is the ingredient in use the same formulation or is the dosage outside the ranges previously reviewed, and thus beyond the coverage of the most recent DRI review? Was the intake level under question addressed in the DRI report? There is increasing use of very high doses of nutrients in dietary supplements, so various biological activities, toxicities, and adverse effects may have been incompletely addressed (or not verified) in the most recent DRI review. These reviews necessarily must cover issues related to intakes over a wide range, looking for the lowest intake level at which adverse effects are noted, so that the issues at the highest levels of intake are not the sole (or primary) focus of the reviews. What was the critical endpoint or adverse effect used to set the UL? Was it a benign reversible adverse event or a serious and irreversible condition? Is concern about consumption above the UL directed primarily at specific populations, or does it apply widely? If a higher level of concern is directed primarily at specific populations, it may be appropriate for the integrative evaluation to focus on that concern as it relates to that population group. It is necessary to consider the basis for the UL, with less concern being warranted for non-life-threatening or self-limiting effects, as well as the recommended dose of the supplement to determine the level of concern. The conclusions of the DRI review should be given much greater weight than other data available at the time of the review but not considered in it. This allows an immediate determination of the level of concern and thus a fairly rapid determination of the need to go forward with an integrated evaluation, following similar procedures to that of other dietary supplement ingredients (as described in earlier sections of this chapter). Integrative Evaluation for a Nutrient It is quite possible that situations will arise where, due to information which becomes available after the DRI review on the severity of the adverse effect, or the vulnerability of a population at risk, there will be a higher level of concern and thus a need to go further than the DRI review to

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety determine if a significant or unreasonable risk of illness or injury exists when consuming a nutrient as a dietary supplement at its suggested level of intake. This is the integrative evaluation component. An integrative evaluation should use as its basis the DRI review and analysis, recognizing that for some nutrients (i.e., vitamin K, β-carotene, arsenic, chromium, silicon, thiamin, riboflavin, vitamin B12, pantothenic acid, biotin, potassium, and sulfate) the available data, while reviewed in the DRI series, were deemed insufficient to develop a UL based on requirements of the model of risk assessment used. This does not indicate that high intakes pose no risk of adverse effects (IOM, 1998a, 2000, 2001, 2004), but that a thorough review by an expert group could not identify dose- response evidence from chronic intakes that would provide a basis for establishing the level at which adverse effects might occur. If a UL was not established for the vitamin or mineral element under question, it is important to consider the following: Was the substance reviewed by the DRI process, even if a UL was not established? Is there new evidence suggesting risk that was not available at the time of the DRI review? Was significant concern about serious harm expressed in the DRI review, even if a DRI could not be established because of limits in the data or acuteness of the adverse effect (e.g., arsenic)? It is assumed that the integrative evaluation for a nutrient would contain the same general components as for other dietary supplement ingredients, with the exception that significant information is already captured by virtue of the DRI process and its reports. In situations where new data or information indicate that higher concern is warranted, the nutrient would enter the evaluation in the same manner as other supplement ingredients. Nutrients are usually well-characterized chemically and thus there is less concern about active ingredient identification and function. ANNEX 3-2 MONOGRAPH PREPARATION AND PUBLIC ACCESS Preparation of Draft Monographs In evaluating evidence indicating that an ingredient may present an unreasonable risk to human health, a comprehensive examination of the literature is required, recognizing that not all studies are relevant to ascertaining the safety of a dietary supplement ingredient. The initial step in

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety preparing a monograph is to gather as much information as possible from the published literature and other sources regarding the potential hazards of consuming the supplement. Multiple comprehensive databases, such as MEDLINE (NLM, 2003a), TOXLINE (NLM, 2003b), and EMBASE should be searched (Elsevier, 2003). In addition, NAPRALERT (Farnsworth, 2003) can be searched if the supplement is a natural product or botanical. To search for potential ingredient-drug or ingredient-ingredient interactions, the Metabolism and Transport Drug Interaction Database (UW, 2003) is a useful tool. Abstracts and titles should be reviewed for relevance to the adverse event or harmful effect of concern. Although review articles may be useful for the purpose of providing literature references and an overview of the data, review of the original articles from the peer-reviewed literature is essential to obviate any bias or unsubstantiated opinion of the authors of the review. Information in other non-peer-reviewed literature that raises concerns about adverse effects should not be ignored. In order to get as much information as possible, FDA should request the voluntary submission of safety data information from industry and other stakeholders. This request may be made through notice in the Federal Register and through the FDA website. FDA should also request information directly from manufacturers and distributors of the ingredient under consideration, if they are known. Collecting descriptive and safety information and organizing and summarizing the information into a draft safety monograph will require significant expertise and resources. Time and other resources required for completion of the draft monographs are likely to vary, depending on the amount and complexity of safety-related information available for the ingredient under consideration, as well as the focus of the monograph. FDA may choose to prepare monographs internally, or it may choose to contract the work out to organizations, individuals, or both. The extent of time and effort devoted to preparation of monographs on dietary supplement ingredients will depend on FDA’s prioritization of need. FDA could set priorities and develop a complete list of substances warranting monographs first. Alternatively, it could retain one or more individuals or groups to develop monographs and determine the need for individual monographs on an ongoing basis as priority setting proceeds or as new needs emerge. The former approach may be more cost effective to implement, given that the latter approach might not provide continuity in workload. However, preparing monographs for substances considered of high priority will require more resources up front if given to contractors and will be dependent on information available at the time high-priority substances are identified. As discussed previously, monographs, whether prepared internally or

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety by a contract organization, should be evaluated to determine if the conclusions could be improved by input from additional expert judgment. The decision to undertake a monograph internally or by contract to an outside group will depend on FDA’s resources and internal expertise. The monographs developed should not be considered static documents. New information should be added as it becomes available, and an organized process for adding information should be developed. The process should also include periodic reviews of monographs to determine if additional external reviews are appropriate. Public Access to the Monograph After the advisory committee’s summary is shared with FDA, the revised monograph and the advisory committee’s summary should be posted on FDA’s website. One of the important components of DSHEA is that the public should be educated about dietary supplements. FDA thus has a responsibility to educate consumers about the safety of dietary supplement ingredients, and the public availability of the final monographs can be an important aspect of the educational process. The monographs can provide the public with a reputable summary of the available information and scientific uncertainties about the inherent safety of the supplement ingredient. Importantly, public access to information from an advisory committee will add to the quality and strength of the available scientific literature. ANNEX 3-3 THE USE OF AN EXTERNAL ADVISORY COMMITTEE The decision to refer a dietary supplement ingredient to an external advisory committee rests with FDA, which has the authority to refer to such a committee for any reason deemed necessary, as discussed in the text. The external advisory committee needs to include experts in critical key disciplines. FDA has significant experience in establishing advisory committees and already has rules regarding membership (e.g., conflict of interest) in place. Possible approaches that FDA may wish to consider include: A standing committee of about seven persons, with the option to add one or two scientists with special expertise, as needed, for the review of individual substances. A standing committee of five scientists representing core disciplines, and the addition of three or four special experts depending on the nature of the ingredient, the data to be evaluated, and whether a focused or broad-based evaluation was required. The presence of individuals with expertise in either the ingredient under review or the purported adverse effect of the

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety ingredient is critical to providing a well-documented review of the literature where the data are equivocal. To ensure that the critical evaluation of the information contained in the monograph and related information is as free of conflict of interest and as objective as possible, the external advisory committee should be composed of expert scientists who have appropriate training, education, and experience. Whether as a result of the appointment of a committee by FDA or by contract with a scientifically based, nonprofit organization, examples of expertise to be included are toxicology, preferably with expertise in safety evaluation; pharmacognosy; clinical pharmacology; nutritional science; epidemiology; biostatistics; clinical trials; medicinal chemistry and structure–activity relationships; pharmacokinetics; consumer behavior related to dietary supplement use; and public health, as well as ad hoc consultants with expertise in specific fields on an as-needed basis (e.g., specialists needed to evaluate particular ingredients, such as experts on oriental medicine, herbalists, veterinary toxicologists, or clinicians with relevant experience). Advisory committee members should be selected based on their disciplinary expertise rather than as representatives of stakeholder viewpoints, and they should not have a financial stake in the outcome of the process or otherwise have a real or perceived conflict of interest. After the external advisory committee is assembled, a draft monograph should be released, and the public should be provided with an opportunity to comment on the completeness of the data included, as well as on the strength and relevance to humans of the different types of evidence. Industry and other stakeholders should be given time during meetings of the external advisory committee to provide input into the process. The external advisory committee should provide advice on the further refinement of the draft monograph as it reviews all the data and summarizes its conclusions. REFERENCES Elsevier. 2003. EMBASE (Excerpta Medica Database). Online. Available at http://www.elsevier.nl/homepage/sah/spd/site/locate_embase.html. Accessed March 13, 2004. Farnsworth NR. 2003. NAPRALERT (Natural Products ALERT) Database. Chicago: University of Illinois at Chicago. Gillette JR. 1976. Application of pharmacokinetic principles in the extrapolation of animal data to humans. Clin Toxicol 9:709–722. IOM (Institute of Medicine). 1994. How Should the Recommended Dietary Allowances be Revised? Washington, DC: National Academy Press. IOM. 1997. Dietary Reference Intakes for Calcium, Phosphorus, Magnesium, Vitamin D, and Fluoride. Washington, DC: National Academy Press. IOM. 1998a. Dietary Reference Intakes: A Risk Assessment Model for Establishing Upper Intake Levels for Nutrients. Washington, DC: National Academy Press.

OCR for page 85
Dietary Supplements: A Framework for Evaluating Safety IOM. 1998b. Dietary Reference Intakes for Thiamin, Riboflavin, Niacin, Vitamin B6, Folate, Vitamin B12, Pantothenic Acid, Biotin, and Choline. Washington, DC: National Academy Press. IOM. 2000. Dietary Reference Intakes for Vitamin C, Vitamin E, Selenium, and Carotenoids. Washington, DC: National Academy Press. IOM. 2001. Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium, and Zinc. Washington, DC: National Academy Press. IOM. 2002. Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids. Washington, DC: National Academy Press. IOM. 2004. Dietary Reference Intakes for Water, Sodium, Potassium, Chloride, and Sulfate. Washington, DC: The National Academies Press. Merriam-Webster. 2001. Merriam-Webster’s Collegiate Dictionary. 10th ed. Springfield, MA: Merriam-Webster. NLM (National Library of Medicine). 2003a. PubMed. Online. Available at http://www.ncbi.nlm.nih.gov/entrez/query.fcgi. Accessed March 13, 2004. NLM 2003b. Toxline. Online. Available at http://toxnet.nlm.nih.gov/cgi-bin/sis/htmlgen?TOXLINE. Accessed March 13, 2004. UW (University of Washington). 2003. Metabolism and Transport Drug Interaction Database. Online. Available at http://depts.washington.edu/didbase. Accessed September 24, 2004.