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.
Summary This report is the second of a series intencleci to provide guidance on the interpretation and uses of Dietary Reference Intakes (DRIs). The term Dietary Reference Intakes refers to a set of at least four nutrient-baseci reference values that can be used for assessing and planning cliets and for many other purposes. Specifically, this report provides guidance to nutrition and health professionals for applica- tions of the DRIs in dietary planning for individuals and groups, as well as providing the theoretical background and statistical justifica- tion for these applications. A previous report examined the use of the DRIs in dietary assess- ment (IOM, 2000a). Dietary assessment using the DRIB, whether for inclivicluals or groups, involves a comparison of usual nutrient intakes with nutrient requirements and examines the probability of inacloquate or excessive intake. Dietary planning, on the other hand, aims to optimize the preva- lence of cliets that are nutritionally acloquate without being exces- sive. Dietary planning may be clone at several different levels. It may refer to an incliviclual planning a meal and food purchases, a food service manager in an institution planning food acquisition and menus, or a government agency planning large nutrition-related or food assistance programs. For the purposes of this report, clietary planning applies to planning intakes, rather than the amount of food purchased or served. Throughout this report methods for planning nutrient intakes of inclivicluals and methods for planning nutrient intakes of groups are clistinguisheci, as these are two very different applications. 1
DIETARY REFERENCE INTAKES Some of the clietary planning activities most relevant to use of the DRIs include incliviclual clietary planning, clietary guidance, institu- tional food planning, military food and nutrition planning, plan- ning for food assistance programs, food labeling, food fortification, developing new or moclifieci food products, and assuring food safety. This document presents a framework for how the DRIs should be used and interpreted for these purposes. WHAT ARE DIETARY REFERENCE INTAKES? The Dietary Reference Intakes (DRIB) are a set of nutrient-baseci reference values that expand upon and replace the former Recom- mencleci Dietary Allowances (RDAs) in the United States and the Recommencleci Nutrient Intakes (RNIs) in Canada. The new DRIs clicker from the former RDAs and RNIs conceptually in that (1) where specific ciata on safety and efficacy exist, recluction in the risk of chronic degenerative disease is inclucleci in the formulation of the recommendation rather than just the absence of signs of deficiency, (2) the concepts of probability and risk explicitly underpin the determination of the DRIs and inform their application in assess- ment and planning, (3) upper levels of intake are established where . . . . .. . . ~ . . ... ^ - . . ,, ~ clata exist regarding risk ot adverse nealtn erects, and to) components of food that may not meet the traditional concept of a nutrient but are of possible benefit to health are reviewoci, and if sufficient ciata exist, reference intakes are established. A nutrient has either an Estimated Average Requirement (EAR) and an RDA, or an Adequate Intake (AI). When an EAR for the nutrient cannot be cletermineci (anci therefore, neither can the RDA), then an AI is established. In aciclition, many nutrients have a Tolerable Upper Intake Level (UL). A brief definition of each of the DRIs is presented in Box S-1. An important principle underlying the former RDAs and RNIs, as well as the new DRIB, is that these are standards for apparently healthy people not values that are meant to be applied to those with acute or chronic disease or for repletion of previously deficient individuals. The chosen criterion of nutritional acloquacy or adverse effect is different for each nutrient and is iclentifieci in the DRI nutrient reports (IOM, 1997, 1998a, 2000b, 2001, 2002a). Requirements are typically presented as a single number for various life stage and gender groups rather than as multiple endpoints except in the case of vitamin A. A more detailed discussion of the origin and frame- work of the DRIs is presented in Appendix A. Recommended in-
SUMMARY 3 takes for the nutrients examined to ciate are presented at the end of this book. Box S-2 provides a brief introduction to appropriate uses of the DRIs for planning. IMPLEMENTATION OF DIETARY PLANNING FOR INDIVIDUALS AND GROUPS Regardless of whether cliets are being planned for inclivicluals or for groups, the goal is to plan usual cliets that are nutritionally acle- quate, or stated another way, such that the probability of nutrient inacloquacy or excess is acceptably low. For inclivicluals, the goal of planning is to achieve usual intakes that are close to the Recom- mencleci Dietary Allowance or the Acloquate Intake (AI) . For groups, the goal of planning is to determine a usual intake distribution that results in a low prevalence of intakes that are inacloquate or at risk of being excessive. The Estimated Average Requirement, AI, and Tolerable Upper Intake Level are used in planning the cliets of groups. Figure S-1 schematically shows the various steps involved in imple- menting clietary plans for inclivicluals and groups. Details of each step are discussed below.
4 DIETARY REFERENCE INTAKES USING DIETARY REFERENCE INTAKES TO PLAN DIETS FOR INDIVIDUALS Planning diets for individuals involves two steps. First, appropriate nutrient goals should be set, and second, a dietary plan that the individual will consume must be developed. This is most frequently accomplished using food-based guidance systems.
r _ ~: z z z o z ~: [L Q ~0 C~ O C~ =2 .= C~ ~o .= C~ ~ _ C~- U) au au au au C~ U) _ ~C Q 11] O ~ o' _.~_ ._ U) ~ C~ au au C~ S Q ~0 U) au o l - U) O C~ au _ C~ U) ~ O .2 m~ ~ 'c' Q au au S au ~C Q ~0 g U) C~ au au _ . _ ° ~ a, c~ au S~ C~ t, _ C~- ~ _ U) .0 g .m -a, a3- au . _ ~ _ au au S U) U) au S ~ C~ a, ~o $ Q ~ C~ au U) au C~ - au ~ s U) C~ au ' a, ~ o QO = Q GO u' ~ au . ~ . _ C~ o o_ ~ S S CO U) Q Q C~ U) U~ N ~ U) .> ,~ o ~ ._ ~ a~ ~ ~ g O S a~ (d au ~0 ~C <0' -Z~ O co au.' S ~ C~ ~ C~ U) C~ o ->au ~Q U) o s~ a, o S ~0 ~ ~0 = ~0mo O,00 O a~ CO ~ U) o ~ C' ~ C' O ~- g ~ a, Cl, Q U) 1 U) .e au S o au ° Q ~ ~o U) U=) ~ ~ a, O.-° g Q cO a, u'= '= ~ ~ O au .= ~0 ._ $- ~0 o ._ U, .o .= VO - V' ~i
6 DIETARY REFERENCE INTAKES Setting the Goal The goal for incliviclual planning is to ensure that the cliet as eaten has an acceptably low risk of nutrient inacloquacy while simulta- neously minimizing the risk of nutrient excess for all nutrients for which Dietary Reference Intakes (DRIB) have been established. When planning for inclivicluals for nutrients such as vitamins, min- erals, and protein, a low risk of inacloquacy is planned by meeting the Recommencleci Dietary Allowance (RDA) or Acloquate Intake (AI), and a low risk of excess by remaining below the Tolerable Upper Intake Level (UL). There are neither adverse effects nor clocumenteci benefits associated with exceeding the recommencleci intake, provicleci intake remains below the UL. Planning is always for usualintake, clefineci as an incliviclual's intake over a long period of time. In some cases it may be appropriate to use a target other than the RDA for inclivicluals. The RDA provides assurance that the probability of inacloquacy floes not exceed 2 to 3 percent. However, nutrition- ists and other planners may clecicle to use a different definition of what is an acceptably low probability of nutrient inadequacy. If so, the rationale should be clearly stated. The EAR is not used as a goal in planning incliviclual cliets. By definition, a cliet planned to provide the EAR of a nutrient would have a 50 percent probability of not meeting an incliviclual's require- ment, and this is an unacceptable degree of risk for the incliviclual. The situation for energy is quite different. In this case, there are adverse effects to inclivicluals who consume intake above their requirements over time, weight gain will occur. This difference is reflected in the fact that there is no RDA for energy, as it would be inappropriate to recommenci an intake that exceecleci the require- ment of 97 to 98 percent of inclivicluals. The only DRI available for energy is the EER Estimated energy requirement), which reflects the estimated average energy expenditure associated with an incli- viclual's sex, age, height, weight, and physical activity level. As such, it exceeds the neecis of half the inclivicluals with specified character- istics, and is below the neecis of the other half. Although the EER may be used as an initial planning goal, body weight must be moni- toreci and intake acljusteci as appropriate. Finally, it is necessary to consider the recommencleci distribution of energy from the macronutrients fat, carbohydrate, and protein (IOM, 2002a). For example, for adults, their energy consumed from fat should be between 20 and 35 percent.
SUMMARY 7 Developing Dietary Plans Dietary plans will usually be clevelopeci using fooci-baseci clietary guidance. In the past, clietary reference stanciarcis (e.g., the former RDAs in the United States and Recommencleci Nutrient Intakes in Canacia) have been used to provide fooci-baseci clietary guidance in several ways. These include developing national food guides and clietary guidelines for healthy inclivicluals, providing consumer infor- mation on food and supplement labels, and serving as a reference stanciarci for nutrient content and health claims. When clietary ref- erence stanciarcis are revised, there will be unavoidable time lags until food guides and information on food and supplement labels are assessed and revised, if necessary, to reflect the new nutrient stanciarcis. When these gaps occur, cliets of inclivicluals must be planned using more cletaileci ciata on nutrient composition, such as those found in food composition databases. Information on food and supplement labels may be useful for estimating macronutrient contents (e.g., energy, fat, and fiber), but may be less useful in situations where the labeling reference stanciarcis do not reflect the current recommencleci intakes. Planners may wish to start with cur- rent food guides and then check to be certain that the resulting cliets meet the RDAs and AIs without exceeding the ULs. The Bottom Line: Planning Individual Diets The goal of planning cliets for inclivicluals is to have a low proba- bility of inacloquacy while minimizing potential risk of excess for each nutrient. In most cases, this is clone by meeting the RDA or AI while not exceeding the UL. This can be accomplished by using food guides such as Canada's Food Guide to Healthy Eating (Health Canada, 1991) or the U.S. Food Guide Pyramid (USDA, 1992), although supplemental information such as food composition data- bases should also be used in situations when these guides may not reflect the DRIB. Gaps or excesses identified can then be remedied by planning to alter the type or amount of foods in the various food groups, by using fortified foods, or by using supplements. USING DIETARY REFERENCE INTAKES TO PLAN DIETS FOR GROUPS Planning diets for groups is a multistep process. It involves identi- fying the specific nutritional goals, determining how best to achieve these goals, and, ultimately, assessing if these goals are achieved.
8 DIETARY REFERENCE INTAKES The goal of planning for groups is to determine a distribution of usual nutrient intakes that provides for a low prevalence of inacle- quate intakes and a low prevalence of intakes that may be at poten- tial risk of adverse effects clue to excessive intake. This proposed framework thus shifts the focus of planning away from using clietary recommendations in clecicling what to offer or serve to what is ulti- mately clesireci in terms of the distribution of usual nutrient intakes in the group. By focusing e~ctlic~tly on the distribution of usual nutrient intakes of a group, the framework for planning presented below is, in many respects, a new paradigm. The procedures used for planning intakes of groups differ clepencling on whether the group is relatively homogeneous (e.g., a single life stage and gentler group, such as women 31 to 50 years of age), or is composed of a number of subgroups that differ in nutri- ent and energy requirements. Planning for Homogeneous Groups The important steps in planning cliets for a homogeneous group include: · selecting the goals, including the acceptable prevalence of inacl- equacy and prevalence of intakes at risk of excessive intake, for each nutrient of interest; · estimating the target usual intake distributions for each nutri- ent; · planning a menu to achieve the target usual intake clistribu- tions; and · assessing the results of the planning. Selecting the Goals The first step in planning for groups is to select the goals: what will be considered an acceptable prevalence of inadequate intakes and what will be consiclerecT an acceptable prevalence of intakes at potential risk of adverse effects. These decisions need to be macle for each nutrient of interest that has an Estimated Average Require- ment (EAR) or Tolerable Upper Intake Level (UL). One approach is to aim for a prevalence of inadequacy of 2 to 3 percent and a prevalence of intakes at risk of adverse effects of 2 to 3 percent.
SUMMARY 9 However, higher or lower prevalences could be selected for either, and the selected prevalences may vary by nutrient. Goals may also be set for nutrients with an Acloquate Intake (AI). In these cases, the goal will usually be to achieve a meclian intake equal to the AI. For energy intake, the goal is to provide the mean estimated energy requirement (EER) for the group. In aciclition, planners will usually wish to specify goals related to macronutrient distribution, such as ensuring that the energy from fat is between 20 and 35 percent for adults. Estimating the Target Usua;t Nutrient Intake Distribution For nutrients with an EAR, the next step in planning group cliets is to determine the usual intake distribution that will meet these goals. This process neecis to be repeated for each nutrient of interest. A target usual nutrient intake distribution has an acceptably low preva- lence of inadequate or excessive intakes, as defined by the proportion of individuals in the group with usual intakes less than requirements or greater than the UL. In most cases, the prevalence of inadequate intakes is estimated as the proportion of the group below the EAR, and the preva- lence of excessive intakes is estimated as the proportion of the group above the UL. In order to select a target usual nutrient intake distribution, it is necessary to make some assumptions about usual intake clistribu- tions for the group of interest. In some cases, the planner may have information on the current intake distribution for the group, and can use this information to plan the new intake distribution. In other cases, it will be necessary to use intake distributions from similar groups (for example, using ciata from national nutrition sur- veys). In either case, the distribution of usual intakes is needled, with the effect of day-to-day variation removed. Because intake dis- tributions are seldom normal, it is usually not possible to determine the distribution from just the mean and stanciarcT deviation of intakes. Percentiles of intakes are almost always neeclecl. Next, the planner neecis to position the intake distribution so the nutrient intake goals are met. For example, if a planner clecicles that the prevalence of inacloquacy in the group should be set at 2 to 3 percent, then the usual nutrient intake distribution of the group should be positioned such that only 2 to 3 percent of inclivicluals in the group have usual intake less than the EAR. Using the EAR as a
10 DIETARY REFERENCE INTAKES cut-point for estimating the prevalence of inacloquate intakes builds directly on the approaches previously clescribeci for assessing intakes (IOM, 2000a). It is appropriate to use the EAR as a cut-point for estimating the prevalence of inacloquate intakes for all nutrients with an EAR, except iron. Because the iron requirements are not normally clis- tributeci, it is necessary to use published tables showing the clistribu- tion of iron requirements in order to estimate the prevalence of inadequate intakes (IOM, 2001~. Because the available intake distribution will not usually be cor- rectly positioned to meet the nutrient goals, the planner must move it up (or clown) by Hilling (or subtracting) a constant amount of the nutrient to each point on the distribution until the appropriate prevalences are obtained. When the distribution is correctly posi- tioneci, it becomes the target usual intake distribution. Assuming there are no changes in the shape of the distribution, the amount of the shift can be calculated as the aciclitional amount of the nutrient that must be consumed to recluce the proportion of the group that is below the EAR. For example, the EAR for zinc for girls 9 to 13 years old is 7 mg/day. Current data from the Third National Health and Nutrition Examination Survey show that about 10 percent of the girls have intakes below the EAR. If the goal is to plan intakes so that only 2 to 3 percent are below the EAR, intakes neeci to be increased. The amount of the increase can be calculated as the difference between the current intake at the 2nci to 3rci per- centile (which is 6.2 mg/ciay) and the clesireci intake at the 2nci to 3rd percentile (the EAR of 7 mg/day); the difference is thus 0.8 mg/ciay. That means that the distribution of intakes neecis to shift up by 0.8 mg/day in order to have only 2 to 3 percent of the girls with intakes below the EAR. The same procedure should be followoci to determine if the clis- tribution meets the goal of a low prevalence of potentially excessive intakes. For zinc, the UL for girls 9 to 13 years of age is 23 mg/day. The 99th percentile of their current usual intake distribution is 15.5 mg/day, so even if the distribution is shifted up by 0.8 mg/day, the 99th percentile (16.3 mg/ciay) is below the UL. The median of the target intake distribution is a useful summary measure, as it can be used as an initial tool in planning menus. Assuming that the shape of the intake distribution does not change as a result of planning, the meclian of the target intake distribution is calculated as the meclian of the current usual intake distribution, plus (or minus) the amount that the distribution neecis to shift to make it the target usual nutrient intake distribution. In the zinc
SUMMARY 11 example above, the distribution neecleci to shift by an aciclitional 0.8 mg/ciay. The meclian of the current zinc distribution for these girls is 9.4 mg/ciay, so the meclian of the target usual intake clistri- bution would be 9.4 + 0.8 = 10.2 mg/ciay. The meclian of a target intake distribution will usually exceed the Recommencleci Dietary Allowance (RDA) because the variance in usual intakes exceeds the variance in requirements. The RDA for zinc for girls is 8 mg/ciay, but the target meclian intake is 10.2 mg/ciay. Thus, selection of the RDA as the meclian of the target usual intake distribution is not recommencleci as it results in a percentage of inacloquacy greater than would likely be selected with more careful consideration. Planning a Menu to Achieve the Target Usua;t Intake Distributions After the planner has estimated a target usual intake distribution for each nutrient of interest, this information neecis to be opera- tionalizeci into a menu. Menu planning involves several steps: 1. Establishing an initial goal for the nutrient content of the menu that is baseci on the target usual nutrient intake distribution. 2. Determining what foocis to offer that will most likely result in a distribution of usual nutrient intake that approximates the target, and thus attains the clesireci prevalence of nutrient acloquacy. 3. Determining the quantities of foocis to purchase and serve. Step 1. Establish an initial goalfor the nutrient content of the menu. It might appear logical to use the meclian of the target usual intake distribution as a goal for the nutrient content of a menu. As clescribeci earlier, this would be projected to leaci to an intake clistri- bution with the clesireci prevalence of nutrient acloquacy, assuming that the shape of the distribution clici not change. However, in almost all group-feeding situations, nutrient intakes are less than the nutrient content of the foocis provided (i.e., food is not completely consumed). Furthermore, many planning applications involve offer- ing a variety of menu options from which the members of the group will select foods. For these reasons, the planner might aim for a menu that offers a variety of meals with a nutrient content range that includes, or even exceeds, the meclian of the target nutrient intake distribution. It is necessary to set initial planning goals for all nutrients of inter- est. For nutrients with an AI, it is not possible to estimate the preva-
12 DIETARY REFERENCE INTAKES fence of inacloquacy, and the goal should be to achieve meclian intakes at the AI. Thus, the AI can be used as a planning goal if the distribution of intakes for the group of interest is similar to the distribution of intakes that was used when setting the AI. For energy intake, either a mean EER or the mean of the current energy intake distribution should be cletermineci. An EER may be calculated for a reference person that is typical for the group of interest, or more accurately, by using an average EER for the members of the group. However, accurate estimates of heights, weights, and physical activi- ty levels are neecleci to estimate an energy requirement, and these are often not available. Thus, even though it is known that energy intakes are often unclerreporteci, the mean of the distribution of energy intakes may also be used as the target in the planning pro- cess. In either case, monitoring of body weight should occur. Step 2. Determine what foods to offer. After all the nutrient targets have been set, the planner must select foocis that will provide this average level of nutrient intake. To con- vert nutrient intake targets into food intakes, planners will usually rely on food guides such as the U.S. Food Guicle Pyramid (USDA, 1992), Canada 's Food Guide to Healthy Eating (Hearth Canada, 1991), published menus, and previously used menus to design a menu that is likely to result in the target level of acloquacy. This will typically be an iterative process, often assisted by nutrient calculation software that allows interactive changes to menus and then recalculates the nutrient levels at each step. In aciclition to achieving goals for preva- lence of inacloquate intakes and prevalence of potentially excessive intakes, goals for acceptable macronutrient distribution ranges (IOM, 2002a) will also neeci to be consiclereci. Step 3. Determine the quantities offoods to purchase, over, and serve. Designing menu offerings to meet an intake target is a difficult task. Because food selections and plate waste vary among groups, and among menus within groups, the appropriate procedures for deter- mining the foods to offer depend heavily on the particular plan- ning context. In aciclition, the amount to purchase to be able to offer or serve must take into account food waste clue to preparation losses.
SUMMARY A.s.sessing the Results of the Plan 13 The final step in planning intakes is to assess the results of the planning process. Such an assessment would follow the procedures for assessing group intakes (IOM, 2000a). There are several reasons why assessment is a crucial component of the framework for group planning. First, planners typically can control only what is offered to inclivici- uals in the group, not what they actually eat. Because the goal of planning is to achieve an acceptable group prevalence of inacle- quate nutrient intakes, it is clear that to judge the success of the planning activity, assessment of intakes must occur. Furthermore, the distribution of intakes that was chosen as the starting point for the planning activity often will not be taken from the group whose intakes are being planned. For example, it may be necessary to start with intake distributions from national surveys. Thus, the planner is making an assumption about the applicability of the distribution to the group of interest. In aciclition, a crucial assumption was macle when selecting the target meclian intake that shifting the distribution of intakes to a new position would not change the shape of the distribution. If the shape changes, then the estimated target meclian intake may be incorrect. The shape of the distribution is likely to clepenci on many factors, including food preferences, the types of foocis served, and the amount of food neecleci to meet each person's energy neecis. Thus, there are several reasons to believe the clistribution's shape may change if a different selection of foocis were served. This is another reason why assessment is a crucial component of good planning. Planning group diets is an iterative, ongoing effort in which plan- ners set goals for usual intake, plan menus to achieve these goals, provide these new menus, assess whether the planning was success- ful, and then modify their planning procedures accordingly. Planningfor Nonhomogeneous Groups If nutrient or energy requirements (or both) are not uniform across a group, the approach to planning can vary. In some cases it may be possible to target the most vulnerable subgroup (i.e., that with the highest nutrient requirements relative to energy neecis) for a specific intervention. In other cases it may not be possible or practical to target the vulnerable subgroup, and in these situations,
14 DIETARY REFERENCE INTAKES a nutrient density approach can be used. Even within a group with the same nutrient requirements, energy requirements may vary sub- stantially, and the nutrient density approach may also be applicable. Nutrient density is dined as the ratio of the content of a nutrient to the energy provided by the food item, diet, or food supply. It is expressed as the unit weight of the nutrient per 1, 000 kcal or per M] of energy. A simple nutrient density approach for heterogeneous groups is to determine the subgroup with the highest target meclian nutrient intake relative to their estimated average energy requirement. Energy re- quirements can be obtainecT by using the current average energy intake of the subgroup, or by calculating the average EER for the subgroup. For example, in a hypothetical group of men and women combined, assume that the vitamin C target meclian intake for the men is 138 mg/ciay, and the target meclian intake for the women is 116 mg/ciay. If the average EER for the men is 2,600 kcal/ciay, then their target meclian vitamin C intake, expressed as a density, is 138/2.6, or 52 mg/1,000 kcal. If the average EER for the women is 1,800 kcal/ciay, then their target meclian intake, expressed as a clen- sity, is 116/1.8, or 64 mg/1,000 kcal. Thus, the women require a higher vitamin C density in their cliets. In this simple approach, the planner would use the target meclian vitamin C density for the women in the menu planning process, and would assume that the men's intake would also be acloquate. However, the simple approach floes not consider the actual clistri- bution of nutrient densities within the group. A new method of planning for heterogeneous groups is proposed in this report. Its goal is to develop a target nutrient density distribution for each sub- group, and then choose the highest target median density from these distributions as the nutrient density to be used in planning. There are three steps to deriving a target usual nutrient density intake distribution: 1. Obtain the target distribution of usual nutrient intakes for each subgroup of interest. 2. Combine the target distribution of usual nutrient intakes with the usual energy intake distribution in each subgroup to obtain the target distribution of usual nutrient intakes expressed as densities. 3. Compare the estimated target meclian intake density for each discrete subgroup to identify the highest nutrient density and use this density to set planning targets for the whole group.
SUMMARY 15 This approach is theoretically more likely to provide an accurate estimate of the appropriate target meclian intakes for heterogeneous groups, although the practicality of its use in planning has not been tested. For either the simple approach or the target nutrient density clis- tribution approach, this selection process would then be repeated for each nutrient of interest for the group, and planning a menu to achieve these targets would proceed as clescribeci above. For some nutrients (notably iron), prioritization of the neecis of the subgroup with the highest requirement relative to energy can result in the selection of a target meclian intake that far exceeds the neecis of all other subgroups. Uncler these circumstances, planners must consider the risk that members of subgroups with lower nutri- ent requirements relative to energy may achieve intake levels in excess of the UL. In such situations, it may be preferable to target the vulnerable subgroups through education or supplementation. Because the simple approach floes not consider the distribution of nutrient densities, and the target nutrient density distribution approach is currently untested, it is particularly important to assess nutrient intakes as a final step in the process of planning for groups. SPECIAL CONSIDERATIONS When using the Dietary Reference Intakes (DRIB) for planning clietary intakes, it is helpful to consider the process and criteria used for developing the DRIs for specific nutrients. Special consici- erations for planning include factors that affect nutrient bioavail- ability, such as the source, chemical form, and clietary matrix, as well as the physiological, lifestyle, and health factors that may alter nutrient requirements and therefore recommencleci intakes. These factors neeci to be consiclereci whether planning cliets for inclivicluals or for groups. Both planning and assessment often rely on self-reported intake, and thus it is important to consider the well-clocumenteci issue of underreporting of energy intakes and its effects on the accuracy of self-reported nutrient intakes. If intakes are underreported, then the planner may start the planning process with incorrect ciata on cur- rent intakes and may also incorrectly assess the results of the plan- ning process. Unfortunately, well-accepted, validated methods to statistically correct for the effects of underreporting the estimated distribution of usual intakes are presently lacking. If planners have the means to measure intakes (e.g., by observing foods selected and
16 DIETARY REFERENCE INTAKES food waste ci by patients in a nursing home), the results of the plan- ning and subsequent assessment will be more valid than self-report- eci intakes for almost all groups. RESEARCH IMPLICATIONS AND RECOMMENDATIONS Several crucial areas have been iclentifieci where ciata and tech- niques do not exist or aciclitional knowledge is neecleci. These neecis are synthesized and prioritized in several key areas, including re- search on clietary planning for groups, improving the quality of clietary intake ciata, providing guidance for clietary planning, and improving estimates of nutrient requirements. These areas are sum- marizeci below. Implementing Dietary Planning for Groups · Pilot test the approaches to clietary planning for groups that are proposed in this report. The approach to group planning proposed in this report is a new paradigm, and should be tested in pilot studies before being implemented on a larger scale. · Determine how different nutrition interventions affect intake distributions. Examination and publication of intake distributions before and after an intervention, with a systematic collection of this type of ciata, would allow a more informed selection of methods for planning a clietary intervention. · Determine the intake distributions of specific population groups. Although data on dietary intakes may be available either from national population surveys or surveys of large groups, often such information has not been reported in a manner that facilitates the estimation of variations in the usual intake of inclivicluals. · Determine the relationship between foocis offered and nutrient intake in the context of group planning. Research is neecleci to determine how food offerings relate to food and nutrient intakes, and how the relationship between food offered and intake varies according to planning context. · Develop and evaluate clietary planning strategies for heteroge- neous groups, including a nutrient density approach to dietary plan- ning. Research is neecleci to determine the practical usefulness of planning for a target nutrient density, determine if the applicability of the nutrient density approach is limited to situations with pre- determined food allocations or restricted food choices (e.g., emer-
SUMMARY 17 gency relief rations), and determine if this approach would be prac- tical in situations offering a wicle variety of food choices, where the nutrient density is more clepenclent on food selection than on total food access to meet energy neecis. Improving the Quality of Dietary Intake Data Much has been written about ways to improve the quality of the intake ciata on which clietary assessment and planning are based; a number of these issues were cliscusseci in a previous report (IOM, 2000a) and are reiterated here. · Develop and validate statistical procedures to identify and cor- rect for both uncler- and overreporting in self-reporteci intake ciata for energy and other nutrients. · Identify and validate better ways to quantify the intake of sup- plements. · Update food composition databases to include the forms and units that are specified by the DRIB. . . Developing Approaches to Providing Guidance for Dietary Planning · Review and, where necessary, revise existing food guides. · Develop technical tools for the professional. · Educate nutrition professionals about correct uses of the DRIB. · Assess application of the DRIs for food and supplement labeling. · Develop and evaluate food guides for group planning. Improving Estimates of Nutrient Requirements · Improve existing estimates of the Estimated Average Require- ment (EAR) and Recommencleci Dietary Allowance. · Provide better information on requirements so it becomes pos- sible to establish an EAR for nutrients that currently have Acloquate Intakes. · Improve estimates of the distribution of requirements so that the appropriate method for assessing the prevalence of inacloquacy for groups can be determined (cut-point method versus probability approach). · Identify the factors that can alter the upper intake levels that can be tolerated biologically.