![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
2
*
Agricultural Research Service, U.S. Department of Agriculture, Washington DC 20250,
Office of Dietary Supplements and the
Office of Prevention Research and International Programs, National Institutes of Health, U.S. Department of Health and Human Services, Bethesda, MD 20892
2To whom correspondence should be addressed. E-mail: piccianm{at}od.nih.gov
 |
ABSTRACT |
|---|
 TOP ABSTRACT FOOD DATABASESDIETARY SUPPLEMENT DATABASESDATABASES USED FOR ASSESSMENTS...CONTAMINANTSSUMMARY OF DISCUSSION GROUP...SUMMARY OF DISCUSSION GROUP...LITERATURE CITED |
|---|
8000 foods and 115 components; tables for compounds of special interest are also available. Nutrient databases need constant revision because of a constantly changing food supply. The completeness of analytical data varies from nutrient to nutrient. The National Center for Health Statistics developed and maintains a database of dietary supplements based on label information. To date, no verification of ingredients has been undertaken. The development of a dietary supplement database containing analytical values would require extensive resources but would be valuable. Databases for vitamin and mineral supplements are compatible with food databases. Databases for botanicals and other supplements include nonnutrient constituents that may not be documented in food composition databases. Gaps in food and dietary supplement composition data exist because of limited resources, changing availability of foods and products and the advent of new compounds of health interest. More data are needed on nutrients and other bioactive constituents in foods and dietary supplements. Analytical methods do not exist for all ingredients or active constituents in foods and dietary supplements. Research needs for further development of meaningful food and dietary supplement databases are similar.
KEY WORDS: • database • food composition • dietary supplement composition • chemical analysis • label database
Food composition databases are required for the analysis of data on intakes and therefore are vital in making intake estimates. The importance of accurate, reliable and up-to-date food composition databases has been acknowledged for some time and an infrastructure has evolved for keeping established databases updated in a timely fashion. The challenges facing those responsible for maintaining and updating these food composition databases include 1) a constantly changing food supply, 2) the advent of new compounds of health interest and 3) limited resources. Recent survey data have been interpreted to suggest that, for many Americans, dietary supplement use contributes a significant amount to nutrient consumption. An obstacle to further elucidation of the contribution of dietary supplements is that databases for them have only recently been created and at this time are still evolving. In addition, there is a growing demand for databases for other bioactive substances (e.g., many naturally occurring substances in foods, as well as intentional food additives, pesticides and contaminants).
Before the passage of the Dietary Supplement Health and Education Act (DSHEA4: PL103-407) in 1994 (1
), dietary supplements were traditionally defined according to common usage as vitamin or mineral supplements that are usually ingested in nonfood form to augment nutrient intake of individuals. With the advent of DSHEA, that definition now refers to a dietary supplement as "a product (other than tobacco) that is intended to supplement the diet that bears or contains one or more of the following dietary ingredients: a vitamin; a mineral; an herb or other botanical; an amino acid; a dietary substance for use by humans to supplement the diet by increasing the total daily intake; or a concentrate, metabolite, constituent, extract or combinations of these ingredients. This product is intended for ingestion in pill, capsule, tablet or liquid form; is not represented for use as a conventional food or as the sole item of a meal or diet; and is labeled as a ‘dietary supplement.’ " For the purpose of this discussion, a nutrient is defined as a substance (protein, carbohydrate, lipids, vitamin, mineral) that been established through rigorous research and accepted by the community to be necessary for growth, normal functioning and maintenance of life.
The effort to provide meaningful guidance on issues related to diet and health consists of a research enterprise focused on metabolism, epidemiology and evaluations of efficacy and safety of prevention strategies for and dietary treatment of disease. More specifically, considerable research effort is being devoted nationwide to the investigation of the relationship of the intake of dietary nutrients and nonnutritive substances, including phytochemicals to health promotion as well as the pathogenesis of many chronic degenerative diseases. The data generated from these efforts are used to develop dietary guidance for healthy individuals and the planning and implementation of national nutrition policy. At the federal level, major efforts in dietary guidance and disease prevention include the Dietary Guidelines for Americans (2
), the National Cholesterol Education Program (3
) and the 5-A-Day Program (4
). In addition, legislative actions such as the Nutrition Labeling and Education Act (5
) and other efforts have brought information on the composition of foods into every American household. Integral to the success of all of these efforts is the availability of reliable information on the composition of foods as well as other important sources of nutrients and dietary supplements including nutrients, botanicals and other potentially bioactive constituents that are naturally occurring in plants (phytochemicals) or animals (zoochemicals).
The challenge for public health policy and the nutrition research enterprise of maintaining accurate, reliable data on the nutrients and bioactive components available from both foods and dietary supplements is compounded by the dynamic nature of the U.S. food supply and a rapidly changing dietary supplement market. Continuous monitoring of these sources of nutrients and other bioactive components is essential to determining temporal patterns of intake and exposure at the individual and population levels and to tracking changes in product reformulation. Currently, data on food composition with respect to some essential nutrients are adequate, but gaps exist for other nutrients and dietary supplements that could have public health importance.
This paper is intended to address the current state of the art on the availability of composition data for foods and dietary supplements. The coverage includes descriptions of the adequacy of databases that are currently available for use in the compilation of dietary data in the What We Eat in America–National Health and Nutrition Examination Survey (NHANES) (the "integrated survey"). Gaps, needs and recommendations for the integrated survey in the future are outlined. Priorities and strategies for database completion are discussed. Technological and monetary considerations for developing a dietary supplement database similar to that available for foods are also addressed. An additional area discussed is the process for utilization of currently available databases for assessment of potential risks associated with exposure to other substances (e.g., pesticides, heavy metals and other possible environmental contaminants), many with potential deleterious effects in the food supply.
For the purposes of this paper, the universe of components of total intake consists of foods, beverages and the full range of dietary supplements as defined in DSHEA. Constituent databases have been divided into nutrients and other bioactive constituents, with the recognition that the latter group of substances consists of many bioactive constituents of possible positive or adverse health significance (e.g., conjugated linoleic acid, pesticides and heavy metals).
The paper is divided into two parts. The first is a historical overview of the available databases for food, dietary supplements and the process for utilizing the available database systems for risk assessment related to exposure to potentially harmful substances in the food supply. The second part summarizes the deliberations and recommendations of discussion groups assigned to address specific questions regarding the current status of the U.S. database system both in terms of content and process for updating and improving these databases in a manner that is responsive to the public health needs of the U.S. population.
|
FOOD DATABASES |
|---|
|
TOPABSTRACTFOOD DATABASESDIETARY SUPPLEMENT DATABASESDATABASES USED FOR ASSESSMENTS...CONTAMINANTSSUMMARY OF DISCUSSION GROUP...SUMMARY OF DISCUSSION GROUP...LITERATURE CITED |
|---|
8000 foods and 115 components. These databases are the foundation for almost all commercially available databases in the United States and serve as models for national and international database development (6
). Data for the NDBS are compiled from published and unpublished sources. Published sources include the scientific and technical literature, unpublished data obtained from the food industry and other government agencies and contracted research specified by ARS. Values in the NDBS are also derived from the results of laboratory analyses or calculated with the use of appropriate factors or recipes, as indicated by the source code in the nutrient data file.
The primary product of NDBS and related research is the Nutrient Database for Standard Reference (SR), the major source of food composition data in the United States. This database provides the foundation for most food composition databases in the public and private sectors. As food composition data are updated, new versions of the databases are released. Since 1992, updated SR data have been published electronically on NDL’s Web site (7
). The current Release 15 (SR15) contains data on 6222 food items and up to 113 food components; SR16 is scheduled for release in mid-2003 and SR17 is planned for release in 2004.
Other NDL activities have included the development of the Primary Nutrient Dataset, a subset of SR, consisting of 3000 foods and recipes that was used as the nutrient database for USDA’s Nationwide Food Consumption Surveys. Data for 52 components, including protein, total fat and individual fatty acids, total dietary fiber and carbohydrate (by difference) as well as minerals and vitamins, were quantified in several previous national surveys. NDL also maintains many composition tables that contain data for specific components or constituents of a limited number of foods and dietary components. Currently available examples of such tables are those containing data on individual carotenoids, isoflavones, trans fatty acids and vitamin K. Additional tables are being developed for individual flavonoids and choline. As data for these constituents are generated and joined with SR, the tables will be updated or discontinued. NDL also regularly publishes scientific papers describing the results of food composition research.
The SR has continued and will continue to serve as the primary source of food composition data for the integrated survey. NDL is expected to add components as they become available to expand the number of constituent dietary components. Anticipated additional components include individual carotenoids, total sugars, vitamin K and 
-tocopherol. Several traditional components (e.g., vitamin A, folic acid) will be modified to reflect methodological advances and changes in the mode of expression to be consistent with units specified by expert committees of the Institute of Medicine of the National Academies of Science in developing the Dietary Reference Intakes (DRI) (8
–11
).
Coordination of food database evaluation and revision
The National Food and Nutrient Analysis Program (NFNAP) is a collaboration of the National Institutes of Health (NIH) of the Department of Health and Human Services (DHHS) and USDA. NFNAP, under the direction of NDL, provides the framework for a comprehensive revision of the food composition data for foods that are consumed frequently and in large amounts and critical nutrients of public health importance and the framework for expansion of the food composition database to include new components of health interest. In addition, the structure of the NFNAP program allows continuous monitoring of the composition of those foods of interest. This integrated research program uses the results of national surveys of food intakes, recent advances in sampling statistics, data evaluation methodology, analytical chemistry and information on product alterations in the national food supply to determine issues of high priority. Data yielded by each phase of the program become an integral part of NDBS. When appropriate, these data are released to the scientific community, the food industry and consumers via the Internet and other data products.
The primary goals of NFNAP are the following:
1.Identification of key foods and critical nutrients for sampling and analysis.
National data on public health and research priorities are combined with data on food consumption and production patterns and used to identify major contributors to nutrient intake in the United States. About 1000 foods and ingredients account collectively for 80% of the intake for most nutrients. For any individual nutritional component, 5–200 foods may account for 80% of the population’s intake. Data from these foods are used as the basis for calculations of constituents in another 3000 recipes.
2.Evaluation of existing data for scientific quality. Using algorithms developed to review and categorize the quality of much of the existing analytical data, NDBS data are classified as being of fully acceptable analytical quality; unacceptable because of lack of documentation for important quality control categories; or nonrepresentative, needing replacement. Data quality indicators help prioritize analyses of specific foods and nutrients.
3.Design and implementation of a nationally based sampling plan. NFNAP uses a probability-based multistage design that minimizes bias and unpredictable accrual of data and fosters confidence that data are representative of the national food supply. Samples are collected in 12–24 locations nationwide. More samples are analyzed for the foods that are higher ranking nutrient contributors.
4.Analysis of sampled foods under USDA-supervised laboratory contracts. Food composition data are obtained through chemical assay using valid analytical methodology. Rigorous quality control programs using standard reference materials and thorough documentation procedures are used to maximize reliability and accuracy of data.
5.Dissemination of data.
New data are incorporated into NDBS using rigorous statistical techniques. The resulting databases are subsequently released on the USDA Web site and include the 6000-food searchable database, nutrient-specific databases and portable document format (PDF) versions of many special-interest publications (6
).
Food databases summary
Food composition databases need to be constantly revised because the national food supply changes rapidly. Revisions are needed when constituents of raw foods or ingredients of prepared foods are altered (by breeding, changes in processing, new fortification rulings, etc.) and when new foods appear or particular constituents in foods become of interest because of potentially positive or negative health effects. There is also a need to constantly generate analytical data for foods and products identified through NFNAP. The development of enhanced or new methods is often necessary to identify and quantify new constituents and to support laboratory, clinical and epidemiologic research.
The completeness of analytical data varies from nutrient to nutrient. Analytical data for many of the vitamins of current public health interest such as folate, vitamin A (individual carotenoids), vitamin D and vitamin E are limited for many foods because of inadequate methodologies and availability of funding to support research for the development of adequate methods. Furthermore, many gaps exist for data on carbohydrates; to address this, total sugars, starches and dietary fibers are now being analyzed at the USDA for foods that are important carbohydrate contributors. However, analyses of resistant starch, amylose and amylopectin are not yet available.
|
DIETARY SUPPLEMENT DATABASES |
|---|
|
TOPABSTRACTFOOD DATABASESDIETARY SUPPLEMENT DATABASESDATABASES USED FOR ASSESSMENTS...CONTAMINANTSSUMMARY OF DISCUSSION GROUP...SUMMARY OF DISCUSSION GROUP...LITERATURE CITED |
|---|
Available dietary supplement databases for use in national survey data collection efforts are based on values declared on product labels rather than direct analysis as is done with foods. The National Center for Health Statistics (NCHS) of the Centers for Disease Control and Prevention (CDC), DHHS, created a database for dietary supplements in 1999 with additional support from the NIH Office of Dietary Supplements in 2001. The database contains information taken from the labels of dietary supplements (nutrients, botanicals and others) and calcium-containing antacids reportedly consumed or ingested by participants in previous NHANES household interviews. In addition to the overarching goal of determining the contribution of dietary supplements to total nutrient intake and ultimately to some measurable health outcomes, this database has two purposes: to link dietary supplement ingredient information to NHANES participants who report taking that dietary supplement and to serve as the basis for a list that is put on the household interviewers’ computers so that they can select dietary supplements for which information is already known and avoid recording the information repeatedly.
The NCHS database contains the following information on nearly 4000 products: name of the product; ingredients and amounts, if specified; serving size and form (e.g., 1 tablet, 2 teaspoons, 3 drops); name of manufacturer, distributor and/or retailer; and suggested dosage. Other information also entered includes date that the product was entered into the database; source of the product information (e.g., label, distributor, Internet, catalog); type of formulation based on the label name (infant/pediatric, prenatal, mature, standard); classification of each ingredient (vitamin, mineral, amino acid, botanical, other); and dietary supplement category based on the product’s name rather than ingredients, which may be different (single vitamin, multiple vitamin, single mineral, multiple mineral, amino acid, single botanical, multiple botanical, antacid, other). This information is taken from dietary supplement labels provided by the manufacturer or distributor. When labels are not available, the information may come from other sources such as company catalogs, the Internet, directly from the company, or the Physicians’ Desk Reference (12
). Health claims, certifications and descriptors that are not part of the name are not recorded.
Limitations of label-based databases for dietary supplements
Although label-based databases are a necessary first step, they are not wholly satisfactory. It is often difficult to obtain the necessary or appropriate dietary supplement labels. Manufacturer or retailer addresses on labels contain only cities and states, and despite searches on the Internet and in phone books, addresses often cannot be found. Some manufacturers and retailers choose not to respond to requests. Labels change as formulations change. Labels taken from the Internet may be old versions or may be inaccurate, and assessing the frequency of inaccuracy is not an easy process. Product information from the Physicians’ Desk Reference does not include an image of the product and the name used may not match the one on the label.
Manufacturers can and do change product labels and formulations at their own discretion. Consequently, a change in label, which may result in a name being recorded differently, may or may not correspond to a change in formulation. Similarly, for private label brands, label changes are also often at the discretion of the retailer or distributor rather than the manufacturer. Manufacturer representatives do not always inform database staff about product reformulations. Thus, it is difficult to track apparent name changes that do not indicate product changes and actual changes in formulations, which are important for ensuring that the correct product and formulation are recorded.
The NCHS database and others such as those at developed at the Pennsylvania State University and University of California, San Diego, are currently not available publicly. A main obstacle preventing their release is that in all cases the databases are constantly changing.
Dietary supplement databases summary
A need exists for a dietary supplement database that is underpinned with analytical data in a manner analogous to food databases. Analytical data would provide important dimensions to a supplement database, such as improved accuracy and precision and a measure of analyte stability that used available and appropriate internal and external standards. The major obstacles to establishing an analytically substantiated dietary supplement database are limited financial resources to support such an undertaking, the lack of established methods for some ingredients and the need to establish sampling frames that reflect the variability of various constituents.
Universal testing of dietary supplements is a huge task and seems unlikely to occur in the near future. Targeted testing of major name brand and private label brand nutrients would be a first step toward verification of dietary supplement ingredients. No broad-based verification of dietary supplement ingredients currently exists. ConsumerLabs.com (White Plains, NY) does test some product types and reports on whether they meet label standards and on the presence of contaminants. However, the names of products that do not meet standards are not reported and the number of products tested is limited.
The Office of Dietary Supplements (13
) has initiated an analytical methods program, with the goal of developing rapid and reliable procedures for identifying active or marker compounds in dietary supplements including botanicals. A component of the program is the development of standard materials for reference during analysis. When these methods and standards become available and are disseminated in the near future, the process of developing an analytically substantiated database for dietary supplements will be greatly facilitated.
If a comprehensive database were to be established, NFNAP, as implemented by USDA for foods, would be a suitable model for the systematic sampling and analysis of dietary supplements. The five key areas of information integral to the quality assurance of analytical data in NFNAP—identification and quality assessment of a sampling plan, sample handling, number of samples, method of analysis and analytical quality control—would be applicable to the dietary supplements. In general, the USDA data quality evaluation systems used to generate confidence codes for the composition of individual foods could also be applied to the evaluation of data in the dietary supplements database.
|
DATABASES USED FOR ASSESSMENTS OF OTHER CONSTITUENTS |
|---|
|
TOPABSTRACTFOOD DATABASESDIETARY SUPPLEMENT DATABASESDATABASES USED FOR ASSESSMENTS...CONTAMINANTSSUMMARY OF DISCUSSION GROUP...SUMMARY OF DISCUSSION GROUP...LITERATURE CITED |
|---|
|
CONTAMINANTS |
|---|
|
TOPABSTRACTFOOD DATABASESDIETARY SUPPLEMENT DATABASESDATABASES USED FOR ASSESSMENTS...CONTAMINANTSSUMMARY OF DISCUSSION GROUP...SUMMARY OF DISCUSSION GROUP...LITERATURE CITED |
|---|
Although national nutrition surveys are primarily used to assess the adequacy of food consumption or the intake of a specific nutrient, the U.S. Environmental Protection Agency (EPA) uses the data to assess risk from dietary exposure to pesticide residues present in or on foods, beverages or water consumed. The authority of EPA to regulate pesticides is derived from the Federal Insecticide, Fungicide, and Rodenticide Act of 1946 and the Food Quality and Protection Act of 1996 (14
,15
).
EPA regulates pesticides to ensure that their use does not pose unreasonable risks to human health or the environment and that exposure to pesticide residues in food is safe. These determinations rely on the process of risk assessment. In assessing risk, EPA’s methodology includes all sources of exposure (e.g., food; drinking water; incidental exposure in and around the home, school) and the inherent toxicity of the pesticide. The EPA Web site includes a comprehensive discussion and listing of EPA, USDA and U.S. Food and Drug Administration guidance and policy documents containing detailed methods and data for assessing exposure to pesticides from the foods that we eat (16
).
A database was developed in July 2000 through a joint effort of the EPA Office of Pesticide Programs and the ARS Food Survey Research Group for estimating human exposures to pesticide residues through food and beverage consumption. The Continuing Survey of Food Intakes by Individuals (CSFII) bases this database—the Food Commodity Intake Database (FCID)—on food and beverage consumption data from the USDA 1994, 1996 and 1998 CSFII surveys. FCID contains a translation file that assigns each food ingredient reported consumed to an EPA raw agricultural commodities classification and subsequently expresses the contaminant or substance as grams per 100 g of commodities consumed. The food items correspond to established USDA food codes, and the recipes identifying each food code’s ingredients are the representative recipes developed for additional food consumption surveys.
The translation to agricultural commodities is necessary because pesticides are approved for use on a commodity basis, not on a food basis. A food may contain numerous agricultural commodities. An example of this translation process is the conversion of the food "apple pie" into its agricultural commodity components: wheat flour, the oil constituents of shortening, peeled apples, beet and sugarcane sugar and cinnamon. FCID contains consumption data for 2 nonconsecutive days for 21,700 individuals, including 11,800 children from birth to age 19 y. The data are expressed in terms of agricultural food commodities for the 5831 different foods and beverages that the surveyed Americans reported eating. FCID is available through the National Technical Information Service (17
).
Once the human dietary exposure to pesticides is determined from the types and amounts of agricultural commodities consumed plus known amounts of pesticide residues on all of these commodities, the potential toxicological effects of the pesticides on human health can be factored in and a dietary risk assessment for a pesticide can be completed.
Lead and heavy metals
EPA uses methodology incorporating food intake, contaminant residue and toxicology data in estimating the levels of heavy metal contaminants in food. For example, the estimation of mercury exposure in the United States is based on fish consumption. In the Mercury Study Report to Congress, the EPA Office of Air Quality Planning and Standards and the Office of Research and Development estimated the dietary intake of fish and then mercury from fish intake for the general U.S. population (18
). Methylmercury is a developmental toxin that accumulates in the muscle tissue of both marine and freshwater fish. It may produce adverse effects on the nervous system of the developing fetus or growing child.
Unlike consumption of other dietary components, such as bread or starch, which are almost uniformly consumed, consumption of fish is highly variable across the U.S. population. An estimation of general fish consumption patterns can be gleaned from data from the integrated survey and, hence, can serve as the basis for estimating general levels of potential mercury exposure from fish intake. The national survey data can also provide additional information on fish intake in specific subpopulations (e.g., children and women of child-bearing age) as well as estimates of quantity and species of fish consumed. With this knowledge of fish intake over time, differences in species consumed and source data on the fish (local, imported or from aquaculture or fish farms), dietary risk assessments of mercury intake from fish can be evaluated.
In general such risk estimates are based on data from both national and local food intake surveys and the use of techniques such as long-term dietary histories, short-term dietary recall methods, questionnaires to identify typical food intake and food frequency, identification aids showing pictures of selected species of fish and survey data from recreational anglers as well as Native Americans. The accuracy and utility of such dietary assessments for estimation heavy metal exposure are inherently linked to the accuracy of the dietary intake and food frequency data collected and to sample sizes adequate to support statistical confidence. As with all national surveillance efforts to link exposure to health outcomes, biomarkers from NHANES and other surveys are needed to further enhance the ability to make these types of diet–health connections.
|
SUMMARY OF DISCUSSION GROUP DELIBERATIONS AND RECOMMENDATIONS |
|---|
|
TOPABSTRACTFOOD DATABASESDIETARY SUPPLEMENT DATABASESDATABASES USED FOR ASSESSMENTS...CONTAMINANTSSUMMARY OF DISCUSSION GROUP...SUMMARY OF DISCUSSION GROUP...LITERATURE CITED |
|---|
) and consisted of members from the range of stakeholder groups attending the workshop. The discussion group membership consisted of representatives from across the spectrum of federal agencies, academic programs and organizations involved in the national nutrition monitoring enterprise. Discussion points
The discussion groups agreed that the development and maintenance of accurate, reliable and current databases is informed by the efforts of nutrition and environmental science researchers and their efforts to continuously improve their scientific understanding of the role of specific constituents in food and dietary supplements on health. The development of databases for both foods and dietary supplements is supported by many of the same steps including development of more precise definitions of individual bioactive components, active ingredients and marker compounds; improvement of the specificity of analytical methods; performance of appropriate analyses on representative samples; and collation of this information into meaningful databases.
The discussion groups urged that those responsible for database development continue to monitor changes in the food and dietary supplement marketplaces and obtain analytical information on new products as they become available. Although the essential features of food and dietary supplement databases are similar, at present they are at different levels of development.
Improving the data and databases
The discussion groups identified a number of strengths of available food composition data and databases for use in the integrated survey to assess dietary intake patterns of Americans (Table 1
). Similarly, the discussion groups listed problems, gaps and constraints of the databases currently used in the integrated survey (Table 2
).
|
|
Since the passage of the DSHEA, a plethora of dietary supplements has appeared in the marketplace. It has been estimated that 40% of U.S. residents consumed dietary supplements of various sorts (20
). Although sales data suggest that vitamin and mineral supplements constitute the bulk of purchases, a multitude of botanical supplements and various combinations are also used by consumers (21
). Dietary supplements of all varieties are marketed both as supplements with a single active ingredient and in various combinations of vitamins and minerals, botanicals and either animal- or botanical-derived substances with many active ingredients, some of which may not have been identified.
As noted previously, in the integrated survey intakes of dietary supplements from a nationally representative sample are collected via dietary supplement recall over the previous month (22
). The NCHS database of products that respondents reported consuming is compiled from information declared on labels. The discussion groups concluded that this database and its utility for the generation of accurate dietary supplement intake data would greatly benefit from a system of analytical verification.
The discussion groups viewed the ultimate goal of obtaining dietary supplement intake information in the integrated survey to be a refinement of the ability to capture and disseminate information on the exact products and amounts taken by survey participants for determination, to the extent possible, of any interactions between those behaviors and specific health outcomes. The discussion groups recognized, however, that given the explosion of the number of dietary supplements in recent years as well as their changing names and formulations, these goals may not be achievable at present. The groups emphasized the need for the testing and verification of product ingredients to determine the level of detail that is reasonable and the precision that is possible in the tabulation of intakes. Documentation of all dietary supplements in detail is useful for public health purposes but it requires a large time and funding commitment at the data collection, database and data release levels. Therefore, it may be necessary to set priorities for the categories of information that have the greatest potential for utility.
The discussion groups placed a high priority on the determination of the contribution of nutrients derived from dietary supplements to overall nutrient intakes because of the need to calculate total dietary intake of nutrients, particularly those implicated in disease prevention. The discussion groups acknowledged that to gain a total appreciation of all the sources of nutrient intake it would be necessary to eventually include data on all commercially available products currently marketed that might contain specific nutrients. For example, nutrient content of commercial water and beverage products and medications will need to be considered, particularly as the use and market for these products continue to expand.
The discussion groups made specific reference to the importance of the form of the nutrient and bioactive substances, particularly in the context of issues related to bioavailability. For example, form specificity is presently of interest for provitamin A activity (amount from ß-carotene, etc.) and iron present as heme or nonheme iron. However, in some circumstances it may not be necessary to know the form of many nutrients because bioavailability information is lacking. Additionally, the discussion groups stressed that even in such cases, the brand name of the product may be important because of differences among manufacturers. Total nutrient data are most commonly tabulated in generic terms—that is, as the nutrient itself, not its molecular form. When this is the case, collection of generic information is probably adequate. In the 2002 NHANES, generic rather than brand name information was collected for 17 commonly consumed single- or double-ingredient vitamins and minerals (i.e., nutrient and amount).
The discussion groups were aware that the need for brand-specific data for basic multiple vitamin and mineral preparations is currently being assessed. The number of commonly used brand name products is limited, and a few private label manufacturers supply a high percentage of store brands with formulations generally similar to those of major brands. However, an increasing number of specialty multiple-component preparations that emphasize conditions such as stress; products that are gender specific; and products with the addition of botanicals, energy enhancers and so forth are now being marketed. It is becoming difficult to distinguish between standard and specialty products, and product names are not always a good guide to ingredients. It may be possible to try to collect generic information on standard multivitamins and multivitamin–mineral combinations as well as those for children and seniors. The discussion groups concluded that the tolerable degree of error in generic data collection that would be acceptable to the user community needs to be determined. In addition, a need exists to evaluate the degree to which values on labels and contents in products agree for individual components of dietary supplements.
The discussion groups were aware that degree of error presently occurring for content of dietary supplements, even for multivitamin–mineral supplements in list databases, is unknown. The error, even with these products, may be considerable. For example, in 1994, the Food and Drug Administration published comparison data of declared vs. determined amounts for 300 food products (23
). For some 66 foods analyzed for thiamin, riboflavin and niacin, 38% were found to be outside acceptable limits of analytical measurement.
Use of dietary supplements with ingredients other than vitamins and minerals was also a significant point of interest for the discussion groups. However, the degree of detail needed during collection of intakes of all specific nonnutrient-containing dietary supplement products is unknown. The extent to which brand-specific information will be used in analyses when data are released is not clear. Little is known about the health-related effects of constituents other than vitamins and minerals, their active ingredients and form or even the amount actually in the dietary supplement. Fewer people take these products than take vitamin–mineral supplements. Therefore, the discussion groups were willing to accept that recording generic information may be sufficient for some purposes. For example, it may be acceptable simply to record that an individual took a garlic, gingko or valerian supplement. Without analytical verification, as is often the case, the amount consumed of these products may be unknown because the dose of the active ingredients (if known) may vary greatly from product to product.
It was understood that the NHANES sample size for individuals using these products is likely to be quite small, although the discussion groups were confident that these data would nevertheless be useful for describing supplement use behaviors. With additional knowledge of these substances and with testing of products for content, data from the integrated survey may be used to investigate the use of products that meet standards and their health effects. However, the constraints of a small sample size will continue to limit meaningful evaluation in NHANES, and other surveys must also be contemplated.
Barriers to improving the integrated survey databases
Among the barriers identified by the discussion groups to improving nutrient databases for foods were limited funding to perform representative sampling and analysis, paucity of validated constituent-specific methods of analysis, gaps in knowledge about stability of constituents of interest and lack of access to proprietary information from the manufacturers regarding proportions of ingredients. The discussion groups noted that the same barriers apply to the development of reliable databases for dietary supplements. In addition, little analytical data are available in the literature. Also, a barrier specific to the analysis of botanicals is the lack of knowledge of the active components and the lack of methods for quantifying these components. The discussion groups strongly urged that resolution of these issues should be a primary focus of a federal committee charged with the development of the database for dietary supplements.
Improvements indicated for food composition databases
For many years NDL provided values for 53 components to the Food Survey Research Group team that conducted the CSFII. Recognizing that for the integrated survey NDL will add data for additional components for which DRI have been defined, the discussion groups offered the following additional suggestions:
-tocopherol need to be reviewed because this is the parent compound against which intake data must be compared. In general, data are not available for the specific forms of tocopherols in vitamin E databases, but the DRI committee recommended that only specific stereoisomers of -tocopherol should be considered for vitamin E activity (10
). 40 g/kg, whereas selenium in wheat from Kansas is only 0.5 g/kg (24
). Scientists at the USDA laboratories in Grand Forks, ND, found that selenium in bread loaves with the same label, sold together, may vary by a factor of 10 (J. Finley, unpublished research, 2001).
In addition to this list of suggestions, the discussion groups commented on several other key aspects of the current available databases and the plans for their use. The discussion groups paid particular attention to completeness of the database for use in the integrated survey. The subset of 3000 of the 6000 foods developed by the NDL for use in the integrated survey database is fairly complete for the foods commonly eaten by Americans. The data set used for the analysis of the 1994–1996 and 1998 CSFII surveys provided mean values for 52 nutrients, including proximate components, vitamins, minerals and individual fatty acids. The USDA Food Surveys Research Group used this data set to calculate the nutritional content of 7000 foods reported by survey participants, which allowed the calculation of the nutrient contents of unique recipes reported by participants. The same procedures will be followed in the integrated survey but the database used will be enhanced and include more nutrients.
The discussion groups concluded that one critical gap is the lack of current, unbiased and representative data on new food products; reformulated products; and new forms of products. These include fortified cereals, beverages, meal replacements and special diet preparations. Recently, new formulations for conventional products have proliferated in the marketplace to meet consumer demands for healthful products that are convenient to prepare and consume. Also, many traditional products have been fortified with additional nutrients in various amounts. Many brands of new meal replacement products complement are now for sale, and because these represent a major portion of food intake, it is important that their food composition be known.
The discussion groups offered the following strategies for filling current gaps in the food composition databases:
The discussion groups made specific note of the importance of use of bioinformatics as an essential key for incorporating existing food composition analytical values into the databases. NDL and other organizations that compile and disseminate food composition data have developed standardized algorithms to estimate and calculate food composition values. Algorithms include widely accepted protein and energy factors as well as statistical procedures for reviewing and aggregating data and methods for imputation and recipe calculation. NDL developed an expert system-based approach for collecting and evaluating the quality of existing or published analytical values for components in foods. After the evaluation of data, quality indicators are assigned to the foods on a component-by-component basis to permit the data user to judge the appropriateness of the estimate. NDL is developing a multinutrient system to be incorporated into NDBS. This system will be used to assess the quality of data for various components by evaluating the quality of the plan for collecting samples, methods for handling the samples, validity of the analytical method, nature of the quality control and numbers of samples analyzed.
The discussion groups strongly endorsed the value of statistical modeling to provide information on when variations in nutrient content may significantly affect human nutrition. That information can be used to determine where more in-depth analysis is necessary in surveys or in the food composition databases. For example, if tomatoes have widely different lycopene levels according to variety and preparation method and dietary assessment of lycopene is of interest, various entries may be needed to account for key sources of variability. If the mean value is adequate, one query may be enough. There are currently sufficiently detailed (i.e., for model development and validation) data on 20 foods.
The discussion groups had specific comments with regard to the accuracy and precision of food composition database, noting in particular that accuracy depends on how the food composition data are to be used. In general, the integrated survey database is used to characterize the intake of various population subgroups rather than the intake of individuals. However, many researchers and other users of the database require enough precision so that over the long term the database also is accurate for individuals. Based on these considerations, the discussion groups urged that the databases must therefore provide high quality data, especially for frequently consumed foods, because these foods have the most influence on intake.
Data are needed for generating a list of nutritional components that are of public health significance at present and are relevant to the biochemical or biological measurements included for the participants. In addition, because a 3- to 5-y lead time exists for the addition of new components to the databases, a forward-looking vision is needed to identify new biologically relevant components that may be needed in future monitoring and research.
Improvements indicated for dietary supplement databases
There was a consensus among the discussion groups that dietary supplement databases are incomplete at present. The discussion groups listed the following priorities for addressing this inadequacy:
The discussion groups identified the following gaps in dietary supplement databases:
The discussion groups offered several strategies for filling the gaps:
|
SUMMARY OF DISCUSSION GROUP CONCLUSIONS AND RECOMMENDATIONS |
|---|
|
TOPABSTRACTFOOD DATABASESDIETARY SUPPLEMENT DATABASESDATABASES USED FOR ASSESSMENTS...CONTAMINANTSSUMMARY OF DISCUSSION GROUP...SUMMARY OF DISCUSSION GROUP...LITERATURE CITED |
|---|
. The participants recognized a need for comprehensive database systems for foods and dietary supplements, each built on a foundation of analytical chemistry data that can be supported and harmonized to provide a complete estimate of dietary intake of nutrients and other constituents. The currently available food composition database has a long history but needs continuous and expanded support. Other constituents to be added to it include bioactive compounds of potential beneficial public health significance as well as compounds with known negative effects. The dietary supplement database should be analytically verified and expanded to include not only nutrients but also other active ingredients of botanicals and other complex formulations.
|
Food database
Maintenance of an accurate and comprehensive food database requires continuous monitoring of the food supply, defining the sources and magnitude of the variability of food composition, increasing knowledge of the various forms of specific nutrients and expanding collaboration with the food industry to obtain analytical data.
Continuous and thorough monitoring of the food supply is essential to building and maintaining a current database because the food supply changes frequently. Up-to-date, specific and unbiased data on nutritional components will improve research on the relationship of foods and their constituents to chronic degenerative diseases such as cancer, cardiovascular disease and diabetes. To this end it is also important to define the sources and magnitude of the variability of the composition of foods. To date it has not been possible to quantify the variability of nutrients in foods because of the lack of sufficient nationally representative data. The variability in nutrient content for the foods consumed by a single individual cannot be determined, but it is possible to quantify the variability in nutrient content for foods that have been sampled and analyzed to gain some estimate of this variability on a population basis. It is also important to examine both the variability from analytical measurement errors and from the natural variability in a food.
More information on the various forms of some key nutrients and bioactive components in foods is also needed. If more specific information on certain categories of foods and nutritional components was obtained in the survey and database values were available for them, the survey information would be more useful to researchers investigating how specific nutritional and bioactive components function in health maintenance and disease prevention.
In general, the food industry is cooperative in providing data for some nutrients, especially those listed on the Nutrition Facts panel of labels. Some industry sectors are more cooperative than others. Expanded cooperation with all sectors to acquire needed data will be of benefit in enhancing the completeness of the databases.
Dietary supplement database
There is a need for a dietary supplement database that is analytically tested and verified. The database that currently exists is based on label data and is a reasonable starting point. Initial efforts should be directed toward assessing the validity of the label data using a comparison with analytical data for a limited number of highly consumed dietary supplements. The findings from this comparison should be used as a basis for establishing a comprehensive dietary supplement database.
To maximize the effectiveness of a dietary supplement database, it will also be necessary to identify an agency to advocate for and champion the project, determine where it will be permanently housed and delineate some of its basic characteristics. Workshop participants suggested that the NIH Office of Dietary Supplements, by virtue of its congressional mandate, was an appropriate and logical choice to champion the database. Further, they recommended that the appropriate home for the database would be a federal agency that has nonregulatory responsibilities, such as the USDA and DHHS’s NIH and NCHS.
Criteria recommended for establishing and maintaining the dietary supplement database included management by a permanent staff and a database design that would permit seamless integration into the food composition database. Workshop participants also recommended that the information in the database should be made publicly available to all stakeholders (i.e., researchers, industry and government). The budget estimated to create, update and maintain a dietary supplement database was $2.5 million per year.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
3 P. Peter Basiotis, Mary M. Bender, Bernadette K. Bindewald, Alicia L. Carriquiry, Anne K. Courtney, Nancy T. Crane, Kevin W. Dodd, Katie Egan, Kathleen C. Ellwood, Susan E. Gebhardt, Joanne F. Guthrie, James M. Harnly, Joanne M. Holden, Clifford Johnson, Susan M. Krebs-Smith, Paul M. Kuznesof, Carol E. Lang, Margaret McDowell, Alanna Moshfegh, Pamela R. Pehrsson, Kathy Radimer, Amy F. Subar, Christine A. Swanson and Wayne R. Wolf.
4 Abbreviations used: ARS, Agricultural Research Service; CDC, Centers for Disease Control and Prevention; CSFII, Continuing Survey of Food Intakes by Individuals; DHHS, U.S. Department of Health and Human Services; DRI, Dietary Reference Intake; DSHEA, Dietary Supplement Health and Education Act; EPA, U.S. Environmental Protection Agency; FCID, Food Commodity Intake Database; NCHS, National Center for Health Statistics; NDBS, Nutrient Databank System; NDL, Nutrient Data Laboratory; NFNAP, National Food and Nutrient Analysis Program; NIH, National Institutes of Health; NHANES, National Health and Nutrition Examination Survey; SR, Nutrient Database for Standard Reference; USDA, U.S. Department of Agriculture.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTFOOD DATABASESDIETARY SUPPLEMENT DATABASESDATABASES USED FOR ASSESSMENTS...CONTAMINANTSSUMMARY OF DISCUSSION GROUP...SUMMARY OF DISCUSSION GROUP...LITERATURE CITED |
|---|
1. U.S. Congress (1994) Public Law 103-417. Dietary Supplement Health and Education Act (DSHEA) 1994 U.S. Government Printing Office Washington, DC.
2. U.S. Department of Health and Human Services and U.S. Department of Agriculture (1995) Dietary Guidelines for Americans 1995 U.S. Government Printing Office Washington, DC.
3. U.S. Department of Health and Human Services (1990) National Cholesterol Education Program (NCEP) Report of the Expert Panel on Population Strategies for Blood Cholesterol Reduction. National Institutes of Health No. 90-3046 1990 National Institutes of Health Bethesda, MD.
4. Kurtzwell, P. (1997) Fruits and vegetables: eating your way to 5 a day. FDA Consumer March :17-23.
5. U.S. Congress (1990) Public Law 101-445. Nutrition Labeling and Education Act of 1990 1990 U.S. Government Printing Office Washington, DC.
6. Shils, M. E., Olson, J. A., Ross, A. C. & Shike, M. (1999) Modern Nutrition in Health and Disease 1999 Williams & Wilkins Baltimore, MD.
7. U.S. Department of Agriculture, Nutrient Data Laboratory () Available at http://www.nal.usda.gov/fnic/foodcomp/(accessed on May 27, 2002).
8. Institute of Medicine (1997) Dietary Reference Intakes for Calcium, Phosphorus, Magnesium, Vitamin D, and Fluoride 1997 National Academy Press Washington, DC.
9. Institute of Medicine (1998) Dietary Reference Intakes for Thiamin, Riboflavin, Niacin, Vitamin B6, Folate, Vitamin B12, Pantothenic Acid, Biotin, and Choline 1998 National Academy Press Washington, DC.
10. Institute of Medicine (2000) Dietary Reference Intakes for Vitamin C, Vitamin E, Selenium, and Carotenoids 2000 National Academy Press Washington, DC.
11. Institute of Medicine (2001) Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium, and Zinc 2001 National Academy Press Washington, DC.
12. Medical Economics Staff (2002) PDR Physicians’ Desk Reference for Nonprescription Drugs and Dietary Supplements 2002:2002 Thomson Medical Economics Montvale, NJ.
13. Office of Dietary Supplements () NIH Dietary Supplement Research Centers: Botanicals. Available at http://ods.od.nih.gov/grants/diet_suppl_centers.html (accessed November 7, 2002).
14. U.S. Congress (1947) Public Law 100-460, 100-464 to 100-526, and 100-532. Federal Insecticide, Fungicide, and Rodenticide Act of 1947 1947 U.S. Government Printing Office Washington, DC.
15. U.S. Congress (1996) Public Law 104-170. Food Quality Protection Act 1996 U.S. Government Printing Office Washington, DC.
16. U.S. Environmental Protection Agency () Information on Guidance, Policy Documents, and Databases. Available at http://www.epa.gov/epahome/topics.html (accessed on May 27, 2002).
17. National Technical Information Service () Food Commodity Intake Database. Order No. PB2000-500101LVK, CD-ROM. Available at http://www.ntis.gov (accessed October 13, 2002).
18. U.S. Environmental Protection Agency (1997) Mercury Study Report to Congress. EPA-425/R-97-009, December 1997. Available at http://www.epa.gov/oar/mercury.html (accessed on May 27, 2002).
19. Dwyer, D., Picciano, M. F. & Raiten, D. J. (2003) Introduction. Future directions for the integrated CSFII-NHANES: What We Eat in America–NHANES. J. Nutr. 133:576S-581S.
20. Ervin, R. B., Wright, J. D. & Kennedy-Stephenson, J. (1999) Use of dietary supplements in the United States 1984–94. Vital Health Stat 11:1-14.
21. U.S. Nutrition Industry: Top 70 Supplements 1997–2001 () Nutrition Business Journal. Available at http://www.nutritionbusiness.com (accessed on October 13, 2002).
22. Dwyer, J., Picciano, M. F. & Raiten, D. J., Members of the Steering Committee (2003) Estimation of usual intakes: What We Eat in America–NHANES. J. Nutr. 133:609S-623S.
23. Life Sciences Research Office (1994) Analytical Data and Label Review of 300 Food Products. Prepared under FDA Contract No. 223-92-2185 T.O.5 1994 Federation of American Societies for Experimental Biology Bethesda, MD.
24. Combs, G. F., Jr (2001) Selenium in global food systems. Br. J. Nutr. 85:517-547.[Medline]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
