Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
A Origin anc! Framework of the Development of Dietary Reference Intakes This report is one of a series of publications resulting from the comprehensive effort being undertaken by the Food and Nutrition Boarcl's (FNB) Stancling Committee on the Scientific Evaluation of Dietary Reference Intakes (DRI Committee) and its panels and sub- committees. ORIGIN This initiative began in June 1993, when FNB organized a sympo- sium and public hearing entitled, "Should the Recommended Dietary Allowances Be Revised?" Shortly thereafter, to continue its collabo- ration with the larger nutrition community on the future of the Recommencleci Dietary Allowances (RDAs), FNB took two major steps: (1) it prepared, published, and disseminated the concept paper, "How Should the Recommencleci Dietary Allowances Be Revised?" (IOM, 1994), which invited comments regarding the pro- poseci concept, and (2) it held several symposia at nutrition-focuseci professional meetings to discuss FNB's tentative plans and to receive responses to the initial concept paper. Many aspects of the con- ceptual framework of the DRIs came from the United Kingclom's report, Dietary Reference Values for Food Energy and Nutrients for the United Kingdom (COMA, 1991~. 163
164 DIETARY REFERENCE INTAKES The five general conclusions presented in FNB's 1994 concept paper were: 1. Sufficient new information has accumulated to support a reassess- ment of the RDAs. 2. Where sufficient ciata for efficacy and safety exist, reduction in the risk of chronic degenerative disease is a concept that should be inclucleci in the formulation of future recommendations. 3. Upper levels of intake should be established where ciata exist regarding risk of toxicity. 4. Components of food that may benefit health, although not meeting the traditional concept of a nutrient, should be reviewoci, and if acloquate ciata exist, reference intakes should be established. 5. Serious consideration must be given to developing a new for- mat for presenting future recommendations. Subsequent to the symposium and the release of the concept paper, FNB held workshops at which invited experts discussed many issues related to the development of nutrient-baseci reference values. (FNB and DRI Committee members have continued to provide updates and engage in discussions at professional meetings.) In aciclition, FNB gave attention to the international uses of the earlier RDAs and the expectation that the scientific review of nutrient requirements should be similar for comparable populations. Concurrently, Health Canada and Canaclian scientists were review- ing the neeci for revision of the Recommended Nutrient Intakes (R:NIs) (Health Canada, l990b). Consensus following a symposium for Canadian scientists, cosponsored by the Canadian National Insti- tute of Nutrition and Health Canada in April 1995, was that the Canaclian government should pursue the extent to which involve- ment with the developing FNB process would benefit both Canada and the United States in leacling toward harmonization. Baseci on extensive input and deliberations, FNB initiated action to provide a framework for the development and possible inter- national harmonization of nutrient-baseci recommendations that would serve, where warranted, for all of North America. To this end, in December 1995, FNB began a close collaboration with the government of Canada and took action to establish the DRI Com- mittee. It is hoped that representatives from Mexico will join in future deliberations.
APPENDIX A THE CHARGE TO THE COMMITTEE 165 In 1995, the DRI Committee was appointed to oversee and con- cluct this project. It cleviseci a plan involving the work of seven or more expert nutrient group panels and two overarching subcom- mittees (Figure A-1) . The Subcommittee on Interpretation and Uses of Dietary Refer- ence Intakes (Uses Subcommittee), composed of experts in nutri- tion, dietetics, statistics, nutritional epidemiology, public health, economics, and consumer perspectives, was to (1) review the scientific literature regarding the uses of clietary reference stanciarcis and their applications, (2) provide guidance for the appropriate application of DRIs for specific purposes and identify inappropriate applica- tions, (3) provide guidance for adjustments to be macle for poten- tial errors in clietary intake ciata and the assumptions regarding intake and requirement distributions, (4) provide specific guidance for use of DRI values for incliviclual nutrients, and (~) identify research neecleci to improve the statistical underpinnings regarding quantitative applications of the DRIs for assessing and planning cliets for inclivicluals and for groups. Standing Committee on the Scientific Evaluation of Dietary Reference Intakes Panels Upper Reference Levels Subcommittee Ca, Vitamin D, Phosphorus, Mg, F Folate, B12, Other ~ Vitamins, Choline . Vitamins C and E, Se, ,8-Carotene and Other Carotenoids Vitamins A and K, As, B. Cl, Cu. Fe, 12, | Mn, Mo, Ni, Si, V, Zn l Energy, Carbohydrate, Sugars, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids l Sodium, Chloride, Potassium, Sulfate, Water ------------ I r , Other Food Components Alcohol ~ 1 FIGURE A-1 Dietary Reference Intakes project structure.
166 DIETARY REFERENCE INTAKES This second report from the Uses Subcommittee examines the appropriate uses of each of the DRI values in planning nutrient intakes of groups and of inclivicluals; an earlier report presented information on the appropriate uses of specific DRI values in assess- ing cliets for groups and for inclivicluals (IOM, 2000a). Each report presents the statistical underpinnings for the various uses of the DRI values and also indicates when specific uses are inappropriate. This report reflects the work of the DRI Committee, the Uses Sub- committee, and the Subcommittee on Upper Reference Levels of Nutrients. ISSUES OF RELEVANCE FROM PAST DIETARY REFERENCE INTAKE REPORTS Methodo;togy to Develop Estimated Average Requirements and Recommended Dietary A;~;towances When Requirements for Nutrients Are Not Norma;l;ly Distributed For most of the nutrients for which Estimated Average Require- ments (EARs) have been established, the required assumption of distribution of requirements is that of symmetry about the mean. In the case of iron, a nutrient of concern in many subgroups in the population in the United States, Canada, and other areas, require- ments are known to follow a nonnormal distribution (IOM, 2001~. Thus, a different method was neecleci to determine the intake of iron at which half of the inclivicluals would be expected to be inacle- quate in the criterion used to establish acloquacy (the EAR), and also to construct an intake level at which only a small percentage of the population would be inacloquate (the RDA). Similar adjustments were macle for clietary protein (IOM, 2002a). If the requirement of a nutrient is not normally clistributeci but can be transformed to normality, its EAR and RDA can be estimated by transforming the ciata, calculating the both and 97.5th percen- tiles, and transforming these percentiles back into the original units. In this case, the difference between the EAR and the RDA cannot be used to obtain an estimate of the stanciarci deviation or the coefficient of variation because skewing is usually present. Where factorial modeling is used to estimate the distribution of requirements from the distributions of the incliviclual components of requirement, as was clone in the case of iron recommendations (IOM, 2001), it is necessary to add the individual distributions (con- volutions). This is easy to do given that the average requirement is simply the sum of the averages of the individual component distri-
APPENDIX A 167 buttons , and a s tan ciarci deviation of the combine ci clistribu tion can be estimated by stanciarci statistical techniques. The 97.5th percen- tile can then be estimated (for a further elaboration of this method, see Chapter 9 and Appendix I of Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium, and Zinc tIOM, 20014~. If normality cannot be assumed for all of the components of requirement, then Monte Carlo simulation is used for the summa- tion of the components. This approach models the distributions of the incliviclual components and randomly assigns values to a large simulated population. The total requirement is then calculated for each incliviclual and the meclian and the 97.5th percentile are calcu- lateci directly. As was the case for iron (IOM, 2001), the underlying joint distribution is approximated and a large number of inclivicluals (100,000) are randomly generated. Information about the clistribu- tion of values for the requirement components is mocleleci on the basis of known physiology. Monte Carlo approaches may be used in the simulation of the distribution of components, or where large ciata sets exist for similar populations (ciata sets such as growth rates in infants), estimates of relative variability may be transferred to the component in the simulated population (Gentle, 1998~. At each step the goal is to achieve distribution values for the component that not only reflect known physiology or known direct observations, but also can be transformed into a distribution that can be modeled and used in selecting random members to contribute to the final requirement distribution. When the final distribution representing the convolution of components has been cleriveci, the meclian and 97.5th percentiles of the distribution can be directly estimated. It is recognized that in its simplest form, the Monte Carlo approach ignores possible correlation among components. In the case of iron, however, expected correlation is built into the mocleling of require- ment where components are linked to a common variable (e.g., growth rate) so that not all sources of correlation are neglected. Life Stage Groups Nutrient intake recommendations are expressed for 22 life stage groups, as listed in Table A-1 and clescribeci in more detail else- where (IOM, 1997~. If ciata are too sparse to distinguish differences in requirements by life stage and gender, the analysis may be pre- sented for a larger grouping. Differences are indicated by gender when warranted by the ciata.
168 DIETARY REFERENCE INTAKES TABLE A-1 The 22 Life Stage Groups for Which Dietary Reference Intakes (DRIB) are Given Life Stage Groups Infants 0-6 mo 7-12 mo Children 1-3 y 4-8 y Males 9-13 y 14-18 `7 19-30 31-50 y 51-70 y > 70y Females 9-13 y 14-18 y 19-30 y 31-50 y 51-70 y > 70y Pregnancy < 18 y 19-30 y 31-50 y Lactation < 18 y 19-30 y 31-50 y NOTE: Differences in DRls are indicated by gender when warranted by the data. Reference Heights and Weights Used in Extrapolating Dietary Reference Intakes for Vitamins and E;tements The most up-to-date ciata providing heights and weights of incli- vicluals in the United States and Canada when the DRI process was initiated in 1995 were anthropometric ciata from the 1988-1994 Third National Health and Nutrition Examination Survey (NHANES III) in the United States, and older ciata from Canada. Reference values cleriveci from the NHANES III ciata and used in previous reports are given in Table A-2. The earlier values were obtained as follows: the meclian heights for the life stage and gentler groups through age 30 years were iclentifieci, and the meclian weights for these heights were baseci on reported meclian Body Mass Inclex (BMI) for the same inclivicluals. Since there is no evidence that weight should change as adults age if activity is maintained, the reference weights for adults age ci 19 through 30 years were applied to all adult age groups. The most recent nationally representative ciata available for Cana- . . dians (from the 1970-1972 Nutrition Canada Survey tDemirjian, 19804) were also reviewoci. In general, meclian heights of children from 1 year of age in the United States were greater by 3 to 8 cm (1 to 2.5 in) than those of children of the same age in Canada measured two clecacles earlier (Demirjian, 1980~. This difference could be partly explained by approximations necessary to compare the two data sets, but more likely by a continuation of the secular
APPENDIX A TABLE A-2 Reference Heights and Weights for Children and Adults in the United States Used in the Vitamin and Element Dietary Reference Intake Reportsa through 2001 169 Median Body Reference Reference Mass Index Height, Weightb Sex Age (kg/m2) cm (in) kg (lb) Male, female 2-6 mo 64 (25) 7 (16) 7-12 mo 72 (28) 9 (20) 1-3 y 91 (36) 13 (29) 4-8 y 15.8 118 (46) 22 (48) Male 9-13 y 18.5 147 (58) 40 (88) 14-18y 21.3 174 (68) 64 (142) 19-30 y 24.4 176 (69) 76 (166) Female 9-13 y 18.3 148 (58) 40 (88) 14-18 y 21.3 163 (64) 57 (125) 19-30 y 22.8 163 (64) 61 (133) a IOM (1997, 1998a, 2000b, 2001). Adapted from the Third National Health and Nutrition Examination Survey, 1988-1994. b Calculated from body mass index and height for ages 4 through 8 years and older. trenci of increased heights for age noted in the Nutrition Canada Survey when it compared ciata from that survey with an earlier (1953) national Canaclian survey (Pett and Ogilvie, 1956~. Similarly, median weights beyond age 1 year derived from the recent survey in the United States (NHANES III) were also greater than those obtained from the older Canaclian survey (Demirjian, 1980~. Differences were greatest cluring adolescence, ranging from 10 to 17 percent higher. The differences probably reflect the secular trenci of earlier onset of puberty (Herman-Giciclens et al., 1997), rather than differences in populations. Calculations of BMI for young adults (e.g., a meclian of 22.6 for Canaclian women com- pareci to 22.8 for U.S. women) resulted in similar values, thus incli- cating greater concordance between the two surveys by adulthood. The reference weights used in the previous DRI reports (IOM, 1997, 1998a, 2000a, 2000b, 2001) were thus based on the most recent ciata set available from either country, with recognition that earlier surveys in Canada inclicateci shorter stature and lower weights cluring adolescence than clici surveys in the United States.
170 DIETARY REFERENCE INTAKES New Reference Heights and Weights As cliscusseci earlier, when the DRI process was undertaken in 1994, the references heights and weights used were clevelopeci baseci on NHANES III ciata on BMI for children and young adults (IOM, 1997~. Given the increasing prevalence of overweight and obesity in both adults and children (HHS, 1996), use of such population ciata is of concern. However, recent ciata providing heights and ideal BMIs for adults (Kuczmarski et al., 2000) and new growth charts for infants and children have allowoci the development of new refer- ence heights and weights (Table A-3) that should more closely ap- proximate ideal weights baseci on low risk of chronic disease and acloquate growth for children. These new values were used in the DRI report published in 2002 (IOM, 2002a) and will be used in subsequent DRI reports until they neeci to be revised baseci on new ciata or because of a conceptual neeci. TABLE A-3 New Reference Heights and Weights for Children and Adults in the United States Previous Median New Body Mass New Median New Indexa Median Reference Reference (BMI) BMIb Heightb Weights Sex Age (kg/m2) (kg/m2) cm (in) kg (lb) Male, female 2-6 mo 62 (24) 6 (13) 7-12 mo 71 (28) 9 (20) 1-3 y 86 (34) 12 (27) 4-8 y 15.8 15.3 115 (45) 20 (44) Male 9-13 y 18.5 17.2 144 (57) 36 (79) 14-18 y 21.3 20.5 174 (68) 61 (134) 19-30 y 24.4 22.5 177 (70) 70 (154) Female 9-13 y 18.3 17.4 144 (57) 37 (81) 14-18 y 21.3 20.4 163 (64) 54 (119) 19-30 y 22.8 21.5 163 (64) 57 (126) a Taken from male and female median BMI and height-for-age data from the Third National Health and Nutrition Examination Survey (NHANES III), 1988-1994; used in earlier DRI reports (IOM 1997, 1998a, 2000b, 2001). b Taken from new data on male and female median BMI and height-for-age data from the Centers for Disease Control and Prevention/National Center for Health Statistics Growth Charts (CDC/NCHS, 2000; Kuczmarski et al., 2000). c Calculated from CDC/NCHS Growth Charts (CDC/NCHS, 2000; Kuczmarski et al., 2000), median BMI and median height for ages 4 through 19 years.