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.
6 Special Con si cleration s an c! Act justments SUMMARY This chapter provides a discussion of the process and criteria used to establish the Dietary Reference Intakes (DRIB) in order to help users make informed judgments cluring the clietary planning process. The limitations in the ciata used to develop the DRIs and the relationship between clietary nutrient inacloquacy and inacloquate nutritional status are important considerations when planning diets. This chapter also aciciresses factors such as nutrient bioavailability and physiological, lifestyle, and health factors that may alter nutri- ent requirements and leaci to adjustments in the DRI values when planning clietary intakes for individuals and Croons INTRODUCTION - a-~--r- It is well established that biological variability exists among indi- vicluals with regard to both nutrient requirements and susceptibility to adverse effects from excessive nutrient intakes. These individual differences, when known, in the normal, apparently healthy popula- tion have already been considered in establishing the Dietary Refer- ence Intakes (DRIs). Specifically, variability in incliviclual requirements around the Estimated Average Requirement (EAR) is consiclereci in setting the Recommencleci Dietary Allowance (RDA), the intake recommendation for inclivicluals. The Acloquate Intake (AI) is set at a level thought to meet or exceed the neecis of almost all inclivicluals 133
134 DIETARY REFERENCE INTAKES of a given life stage and gender group. The Tolerable Upper Intake Level (UL) is set at an intake at which all but the most sensitive members of a population would not be expected to experience adverse effects. Thus, most normal sources of variability have already been considered in setting the DRI values, as they apply to the typical diets of apparently healthy people in the United States and Canada. However, there are other identifiable factors that may alter nutrient requirements systematically such that the DRI values may need to be adjusted when planning nutrient intakes for certain indi- viduals or groups. These factors, discussed below, include charac- teristics of the nutrient source that influence nutrient bioavailability, as well as physiological, health, or lifestyle characteristics of indi- viduals that may require tailoring of requirement estimates. INFLUENCE OF THE NUTRIENT SOURCES Bioavailability Information on the bioavailability of nutrients from foods, forti- fied foods, and supplemental nutrient sources has been used in developing the Dietary Reference Intakes (DRIB) and must also be considered in applying the DRIs to dietary planning. Issues regard- ing bioavailability for each nutrient are discussed briefly below and in greater detail in the individual DRI nutrient reports (IOM, 1997, 1998a, 2000b, 2001, 2002a). Different sources of a nutrient can vary in chemical or physical form, which can affect bioavailability. Thus, in planning diets for individuals or groups, consideration may need to be given to whether the nutrient is supplied in its natural food matrix, as a fortificant to a food source, or in a supplemental form not associated with food. For example, U.S. Department of Agriculture food composition data have only recently been modified to reflect the different bioavail- ability between natural food sources of folate (1 dietary folate equiv- alent tDFE] = 1 ,ug of folate found naturally in food) and folate added as a fortificant to foods (1 DEE = 0.6 fig). Accordingly, in planning to increase an individual's folate intake by about 100 DFEs to meet the Recommended Dietary Allowance (RDA), it would be necessary to consider whether to increase the intake of fruits and vegetables or fortified grain products (or both). An increase of 100 DFEs would require 100,ug of folate from fruits and vegetables, but only 60,ug from fortified grain products. However, if the food com- position data were reported in DEE units, differences in bioavail- ability would already have been taken into account.
SPECIAL CONSIDE~TIONS AND ADJUSTMENTS 1 35 The source of a nutrient can also affect the potential risk of nutri- ent intakes that exceed the Tolerable Upper Intake Level (UL). For several nutrients, there is no known risk of excessive intake from natural foocis. Accordingly, the UL for nutrients such as magne- sium, folate, niacin, and vitamin E are baseci only on chemical or synthetic forms obtained from supplements or acicleci to foocis (IOM 1997, 1998a, 2000b). Excessive intakes for other nutrients such as calcium, selenium, iron, and vitamins C and D are baseci on the combination of intakes from food and supplements (IOM 1997, 2000b, 2001~. For some nutrients, the chemical form varies within natural foocis, as well as between natural and synthetic sources. For instance, heme iron, the form of approximately 40 percent of the iron in meat, poultry, and fish (Monsen et al., 1978), is generally better absorbed than the remaining (nonheme) form of iron in foocis. This cliffer- ence between heme and nonheme iron absorption, which is one factor that can contribute to the lower iron absorption seen in plant- baseci cliets, has been aciciresseci by recommencling intakes for vege- tarians that reflect the lower average absorption. These differences between sources of a nutrient can be of such importance that, in some cases, it is specified which source should be used to meet nutrient intake recommendations. For example, because about 10 to 30 percent of older adults have recluceci gastric acidity, they may not readily absorb the protein-bounci form of vita- min BE that is found naturally in food sources. To ensure that adequate vitamin BE is absorbed when planning for individuals or groups where the average age is over 50, planners are encouraged to include foods fortified with vitamin BE or a supplement contain- ing vitamin BE since the synthetic form of the vitamin is absorbed effectively even in those with low gastric acid secretion. Another example relates to planning for inclivicluals or groups where women are in their childbearing years. In this case, the diet plan should include 400 ,ug of folate from fortified foocis or supplements in aciclition to the food folate contained in a varied cliet since studies that showed reduced risk of neural tube defects were conducted with 400,ug of folate as supplements. Interactions with Other Nutrients, Food Components, and Properties of the Dietary Matrix In aciclition to the bioavailability factors cliscusseci above, nutrient utilization can be influenced by interactions with other nutrients or food constituents. Examples include enhancement of nonheme iron
136 DIETARY REFERENCE INTAKES absorption by ascorbic acid; inhibition of calcium, iron, and zinc absorption by physic acid from whole grains, nuts, and legumes; enhancement of the absorption of the fat-soluble vitamins A, D, E, and K by clietary fat; improved absorption of ,3-carotene in some vegetables after cooking and blencling; and competitive imbalances of minerals such as calcium, iron, zinc, and copper (Mertz et al., 1994~. Excessive intake of one nutrient may interfere with absorp- tion. excretion. transport, storage, function, or metabolism of another. Specific nutrient interactions with food components and drugs have also been iclentifieci (IOM, 1997, 1998a, 2000b, 2001, 2002a). Because of quantitative and bioavailability differences, nutrient- nutrient interactions are of particular concern in cliet planning when nutrients are provicleci by supplementation or fortification rather than by food sources. Such interactions have been consici- ereci in setting the DRIB, including the establishment of ULs that may be specific for nutrients used in fortification or taken as sup- plements. Accordingly, in most cases planners do not need to make adjustments to DRIs baseci on nutrient-nutrient interactions. Special Considerations for Vegetarian Diets Well-planneci vegetarian cliets are associated with good health (Messina and Burke, 1997~. However, not all vegetarian cliets are the same. Depending on the foocis inclucleci or exclucleci from the cliet, careful planning may be required to meet recommendations for various nutrients. For example, vitamin BE is found only in foocis cleriveci from animal sources or in those foocis to which it is acicleci cluring fortifi- cation. Inclivicluals following vegan cliets (exclusively composed of plant foocis) will neeci to either use a vitamin BE supplement or consume fortified foocis containing sufficient amounts of synthetic vitamin Bit. Vegetarians who do not use fluici milk are likely to have low vitamin D intakes, especially those living in northern latitudes where exposure to ultraviolet light floes not occur cluring winter months (Laclizesky et al., 1995; Webb et al., 1988~. Populations who do not use milk and milk products are likely to neeci aciclitional sources of calcium in their cliets. This can be achieved with the judicious selection of plant sources or the use of calcium-fortifieci foods and beverages. Inclivicluals or groups who follow vegetarian cliet plans that omit all animal products are likely to be at risk for inacloquate intakes of iron and zinc, which also neecis to be taken into account when
SPECIAL CONSIDERATIONS AND ADJUSTMENTS 1 37 planning cliets. Hunt and Rougheaci (1999) clemonstrateci that iron absorption from vegetarian cliets was recluceci compared with an omnivorous cliet. In similar studies, zinc absorption was approxi- mately 35 percent less from a lactoovovegetarian cliet as compared with an omnivorous cliet (Hunt et al., 1998~. The description of the recommencleci intakes for iron and zinc further reviews the evidence of lower bioavailability of these nutrients from plant sources and recommencis iron intakes for vegetarians that are higher than the RDAs for the general population (IOM, 2001~. Another nutrient of potential concern for vegetarians is protein. Because protein intakes of vegetarians are typically lower than intakes of those following omnivorous cliets, the issue of protein quality becomes particularly important. In the past there were no recom- mencleci intakes for indispensable amino acids, and it was assumed that inclivicluals consuming a mixed cliet (animal and vegetable pro- teins with a biological value of 75 percent) with the recommencleci amounts of protein would obtain the neecleci amounts of inclispens- able amino acids. Now that both Estimated Average Requirements (EARs) and RDAs have been provicleci for indispensable amino acids, it is important to reexamine this issue. . . It appears that cliets acloquate in total protein may not be neces- sarily acloquate in all the indispensable amino acids, at least for lysine. Data in Table 6-1 compare the amino acid composition of various protein sources to the Food and Nutrition Boarci/Institute of Medicine amino acid scoring pattern (IOM, 2002a). The scoring pattern indicates the amounts of each indispensable amino acid per gram of protein neecleci to meet the EAR for the indispensable amino acid when total protein intake equals the EAR. A single scoring pattern has been aclopteci because there are relatively small differences between the amino acid requirements of children and adults when the requirements are expressed relative to total protein requirements. The ciata suggest that although most protein sources provide recommended amounts of threonine, tryptophan, and sulfur- containing amino acids, this may not be true for lysine. Animal protein sources provide relatively high amounts of lysine, so incli- vicluals who do not consume animal protein sources (or who con- sume limited amounts) may be unlikely to obtain the recommencleci amounts of lysine when total protein intake is limited to the RDA, unless beans are the primary protein source in their cliet. Even then, cliets may be marginal, as the ciata in the table are not acljusteci for the lower digestibility often seen in plant protein sources. There- fore, in addition to planning total protein intakes, it may be neces- sary to plan for intakes of lysine in vegan diets.
138 DIETARY REFERENCE INTAKES TABLE 6-l Selected Indispensable Amino Acici Content of Protein Sourcesa Compared to Recommencleci Levels Indispensable Amino Acid (mg/g protein) Methionine Lysine Threonine Tryptophan + Cysteine FNB/IOM scoring patterns 51 27 7 25 Beef, lean 83 44 11 37 Cheddar cheese 76 33 12 29 Egg 70 49 16 56 Tofu 66 41 16 27 Soymilk 65 41 16 32 Garbanzo beans 67 37 10 26 Almonds 29c 32 15 25 Peanut butter 36c 34 10 33 Brown rice 38c 37 13 35 Cornmeal 28C 38 7 39 Wheat bread 28C 30 13 39 a USDA Nutrient Database for Standard Reference, Release 15, August 2002. b From IOM (2002). The scoring patterns indicate the amounts of essential amino acids per gram of protein needed to meet the Estimated Average Requirement (EAR) for the essential amino acid when total protein intake equals the EAR for protein. c The protein source would not provide recommended amounts of the indispensable amino acid if it were the only source of protein in the diet. The neeci to plan intakes of lysine is likely of greatest importance for individuals whose diets emphasize plant foods and are relatively low in total protein. For example, the RDA for total protein for the reference 57-kg woman is 46 g/ciay. If she followoci a plant-baseci cliet and ate no more than the RDA of 46 g of protein tinily, she would be unlikely to meet her RDA for lysine (2.2 g/ciay) unless 50 percent or more of her clietary protein was provicleci from beans or tofu (rich sources of lysine). To be specific, 23 g of protein from beans and tofu would provide about 1.5 g of lysine, and 23 g of protein from other sources, such as wheat, rice, and nuts, would provide about 0.7 g of lysine. However, if her total protein intake was greater (e.g., about 63 g/ciay, or similar to the meclian protein intake reported by women in the 1994-1996 Continuing Survey of Food Intakes by Inclivicluals tUSDA/ARS, 19974), she could meet her RDA for lysine with much smaller amounts of beans and tofu (providing about 10 percent of her total dietary protein). Thus,
SPECIAL CONSIDE~TIONS AND ADJUSTMENTS 1 39 planning for inclivicluals who consume only plant sources of protein should involve careful review of lysine intakes. If their total protein intake is limited to the RDA for protein, beans and legumes should be emphasized as the major source of clietary protein. INDIVIDUAL CHARACTERISTICS THAT INFLUENCE DIETARY REQUIREMENTS Recommencleci Dietary Allowances (RDAs) and Acloquate Intakes (AIs) are used as goals for nutrient intakes to meet the known nutri- ent requirements of almost all healthy inclivicluals in various life stage and gentler groups. As cliscusseci below, the Dietary Reference Intake (DRI) process has already accounted for normal incliviclual variability, and incliviclual adjustments for factors such as age, nutri- ent status, genetic variation, or body size are generally not required. In other instances, adjustments may be warranted for inclivicluals with lifestyle differences or who are ill. Nutrient Status Nutrient absorption, excretion, and utilization can all be substan- tially affected by the nutrient status of the incliviclual (e.g., low, moderate, or high tissue concentrations). Inclivicluals with lower body stores or who have aciapteci to lower intakes of a nutrient are likely to have greater rates of absorption and lower rates of excre- tion. These relationships have probably been best characterized in humans for iron. However, the Estimated Average Requirement (EAR) and resulting RDA are baseci on 18 percent iron absorption by people with minimal iron stores (defined as a serum ferritin level of 15 ,ug/L) and have already been acljusteci for incliviclual variation in iron status; thus, no further adjustments are required. Genetic Variation Rapidly expanding information on the human genome indicates many possible interactions between incliviclual genetic traits and nutrient requirements. Examples of genetic clisorclers requiring nutritional treatment include classical inborn errors of metabolism such as phenylketonuria, lipoprotein lipase deficiency, and vitamin D-clepenclent rickets. More subtle genetic differences may contribute to variability in requirements within populations generally regarded as normal and healthy. For example, a genetic polymorphism under current investigation adversely affects homocysteine concentrations
140 DIETARY REFERENCE INTAKES (and thus potential heart disease risk) in subjects with relatively poor folate status Jacques et al., 1996~. The continuing discovery and evaluation of genetic influences on nutritional requirements may lead to more specific recommendations for subgroups of the population. In the meantime, however, the RDAs are expected to meet the needs of almost all individuals, which should include many who may have higher than average requirements. Unusual Body Size or Composition, Energy Expenditure, or Physical Activity By establishing EARs and using the estimated variability of the requirement distribution to set RDAs to include 97 to 98 percent of all individuals in a life stage and gender group, these recommended intakes already account for typical variation in body size or energy expenditure in a specific group. Depending on the function and tissue distribution of the nutrient, such variation may be associated with skeletal mass, lean body mass, body water, or total body mass (IOM, 1997, 2000b). Larger individuals would be expected to have greater requirements based on larger body nutrient pools or func- tional compartments. Although reference body sizes (IOM, 1997) have been considered in deriving recommended intakes for specific life stage and gender subgroups, information on most nutrients is inadequate to precisely set recommendations in relation to an indi- vidual's body size or energy expenditure. While there was insufficient evidence to define a relationship between energy requirements or body size and the requirements for thiamin, riboflavin, and niacin (IOM, 1998a), the functions of these nutrients are known to be directly related to energy metabo- lism. If, when planning diets, professionals choose to make an upward adjustment of B vitamin recommendations for individuals with unusually high energy requirements, the conservative approach (in terms of making recommendations to minimize the possibility of dietary inadequacy) would be to assume that vitamin require- ments increase in direct proportion to energy requirements. An example of how these adjustments should be made has been oro- vided in the DRI assessment report (IOM, 2000a). . . Research on the impact of physical activity on nutrient require- ments was evaluated as part of the DRI process, especially in rela- tion to the requirements for B vitamins, vitamins with antioxidant properties such as vitamins C and E, and protein. For most nutri- ents, the data were considered insufficient to recommend specific alterations in the EARs or RDAs related to physical activity or athletic
SPECIAL CONSIDERATIONS AND ADJUSTMENTS 141 performance. An exception is iron. Body iron losses appear to increase with vigorous exercise, perhaps because of increased gastrointestinal blood losses or because of erythrocyte rupture within the foot clur- ing running (IOM, 2001~. Consequently, athletes engaged in regu- lar intense exercise may have average requirements for iron that range from 30 to 70 percent above those of normally active incli- vicluals. Aciclitionally, athletes with extremely high energy intakes Exceeding 6,000 kcal/ciay) may have clietary phosphorus intakes that exceed the Tolerable Upper Intake Level (UL), but this is not thought to be harmful (IOM, 1997~. Age and Physio11og~ca11 Stage Children Recommencleci intakes change considerably across some age boundaries in children. For example, the RDA for magnesium for children ages 4 to 8 years is 130 mg/ciay, whereas the RDA for children ages 9 to 13 years is 240 mg/day. Clearly, magnesium needs do not change abruptly on a child's ninth birthday. Although it might appear reasonable to speculate that those at the higher end of an age range would have higher requirements than those at the lower end of the age range, in most cases knowledge of exactly how a chilcl's nutrient requirements change with age is imprecise. For this reason, adjustment of recommencleci intakes within an age range is not recommencleci. Adjustments in recommencleci intakes may be appropriate when relevant physiological changes can be iclentifieci for inclivicluals. An example is the onset of menarche in girls. The RDA for iron for girls 14 to 18 years of age allows for iron losses in menses. If menarche occurs prior to age 14, an aciclitional amount, about 2.5 mg of iron/clay, would be neecleci to cover menstrual blood losses. Conversely, girls ages 14 and above who have not reached menarche can subtract 2.5 mg from the RDA for this age group. When boys or girls can be iclentifieci as undergoing the growth spurt of adolescence, the RDA for iron can be further adjusted by increas- ing daily intakes by 2.9 and 1.1 ma, respectively (IOM, 2001~. Women of Reproductive Age To reduce the risk of neural tube defects it is recommended that all women capable of becoming pregnant obtain 400 ,ug of folate from fortified foocis or supplements on a ciaily basis in aciclition to
142 DIETARY REFERENCE INTAKES folate from a varied cliet. For most women, a straightforward way to do this is to use a multivitamin supplement containing 400 ,ug of folate. Folate is also acicleci to grains and cereals, but unless a highly fortified breakfast cereal is consumed, it would take unusually large amounts of some of these foocis to obtain 400 ,ug. For example, a slice of breaci contains 20 ,ug of acicleci folate (the required level of folate fortification of breaci). Major differences in menstrual iron losses are an example of iclen- tifiable incliviclual characteristics that mollify nutrient requirements. These losses can be substantially moclifieci by physiological changes such as menopause or hormonal therapy. The RDA for women ages 31 to 50 is intencleci to cover losses associated with menstruation, while the RDA for women over age 50 assumes that menopause has occurred. Menopause, then, rather than turning 50, is the physio- logically significant event related to iron requirements. A woman who experiences menopause before age 50 (anci who floes not com- mence cyclic hormone treatment that results in the partial return of menstrual blood losses) could safely aim for an iron intake of 8 ma, the RDA for women over age 50. Conversely, a bl-year-olci woman who is still menstruating regularly should aim for an iron intake of 18 ma, the RDA for women ages 31 to 50. Dietary iron neecis are lower for women using oral contraceptives clue to recluceci menstrual blood loss. Accordingly, the recommencleci intake for iron is adjusted down to 11.4 mg/day for adolescent girls and clown to 10.9 mg/ciay for premenopausal women using oral contraceptives (IOM, 2001~. Although a number of reports suggest some changes in riboflavin, B6, or folate status for women using oral contraceptives, the available evidence floes not indicate any neeci for adjustment in the RDAs for these nutrients. Gestation of Mu;ttip;te Fetuses The RDAs and AIs for pregnancy and lactation have been clevel- oped for singleton pregnancies and the production of sufficient breast milk to nourish one infant. During pregnancy and lactation of multiple births, the intakes recommencleci for singletons may not be appropriate. To experience good pregnancy outcomes, women who are preg- nant with two or more fetuses neeci to gain more weight than has been associated with good outcomes for singleton pregnancies, and guidelines for weight gain during multiple pregnancies have been clevelopeci (IOM, 1990~. At this point, however, average nutrient requirements for women pregnant with multiple fetuses are not
SPECIAL CONSIDERATIONS AND ADJUSTMENTS 143 known and specific recommencleci intakes have therefore not been cleriveci. It has been noted, though, that intakes of some nutrients, such as protein, should be higher for women pregnant with two or more fetuses than for women pregnant with one (IOM, 2002a). For lactating women, recommencleci intakes for many nutrients are clevelopeci, at least in part, on the basis of the amount of the nutrient secreted in breast milk. Women nursing two or more infants secrete greater volumes of breast milk (Saint et al., 1986~; thus, it is reasonable to assume that their nutrient neecis are also higher. The increased amount of energy required to nurse multiple infants will likely be met by natural appetite adjustments, and energy balance can be evaluated by monitoring body weight for mother and infants. If this increase in maternal energy intake emphasizes nutrient-clense food selections, then consumption of a variety of nutrients will be proportionally increased. Similar to pregnancy, however, specific recommendations for women nursing more than one infant have not been established. Adu;tts Over Age 50 For some nutrients, requirements (anci thus recommenciations) change in association with physiological changes that are expected to occur with aging. For example, the AI for vitamin D is higher for adults over age 50 years than for those uncler age 50 years. The AI for vitamin D increases from ~ ,ug for inclivicluals through age 50 years to 10 ,ug for those ages 51 to 70 years, and to 15 ,ug for those over age 70 years (IOM, 1997~. Because vitamin D is not wicle- ly clistributeci in the food supply (it occurs naturally in liver, fatty fish, and egg yolk, and is routinely acicleci to fluici milk, cirieci skim milk powder, and margarine), it is easy to envision cliets that would not provide vitamin D in amounts recommencleci for older adults. Special attention to intakes of this vitamin is thus warranted for individuals in this category, particularly because endogenous syn- thesis is less efficient with advancing age (MacLaughlin and Holick, 1985~. Use of a supplement containing vitamin D could be consid- ered, particularly by those living in northern latitudes or who rarely receive sun exposure and do not regularly drink milk. It has been estimated that from 10 to 30 percent of inclivicluals over the age of 50 have low levels of gastric acidity, resulting in insufficient release of vitamin BE from the protein to which it is bounci in foocis, and thereby resulting in recluceci absorption of the vitamin. For this reason it is recommencleci that adults over the age
144 DIETARY REFERENCE INTAKES of 50 obtain most of their RDA for vitamin BE from synthetic sources (either in a supplement or in fortified foods) (IOM, 1998a). LIFESTYLE FACTORS THAT AFFECT REQUIREMENTS Alcohol Abuse Alcoholism or alcohol abuse is associated with reduced food and nutrient intakes and a greater frequency of nutrient deficiencies, especially thiamin, niacin, vitamin Be and folate (IOM, 1998a). Chronic, excessive alcohol intake results in damaging physiological effects that may affect absorption, plasma concentrations, metabolism, and excretion of nutrients such as vitamin B6 and folate. Specific nutrient requirements have not been established in relation to levels of alcohol consumption. The importance of assuring adequate intakes of micronutrients in situations of alcohol abuse is emphasized by the greater frequency of nutrient deficiencies in alcoholics, an example of which is the irreversible consequences of the Wernicke-Korsakoff syndrome of severe thiamin deficiency. For uncontrolled alcoholics who are unable to correct their poor food intake habits, a nutrient supplement may be helpful in meeting their requirements for micronutrients. Cigarette Smoking Although blood folate concentrations have been reported to be lower in smokers than in nonsmokers (IOM, 1998a), data suggest that a low intake (Subar et al., 1990) rather than an increased requirement may account for the poorer folate status of smokers. In contrast, there is substantial evidence that smoking increases oxidative stress and metabolic turnover of vitamin C, thus recom- mended intakes of vitamin C are increased by 35 mg/day for smokers (IOM, 2000b). DIETARY PLANNING FOR PEOPLE WHO ARE ILL Just as is the case with healthy persons, planning diets for those who are ill first involves setting nutrient goals that are appropriate for their health status and nutrient needs. The Recommended Dietary Allowance (RDA), the Adequate Intake (AI), and the Toler- able Upper Intake Level (UL) are appropriate Dietary Reference Intakes (DRIB) for dietary planning for healthy individuals. How- ever, some individuals who are ill have conditions that affect the
SPECIAL CONSIDERATIONS AND ADJUSTMENTS 145 absorption, storage, metabolism, or excretion of one or more nutri- ents and, as a result, the DRIs for these nutrients must be moclifieci to take these clisease-relateci factors into account. This section describes a general approach for using the DRIs in these situations. Once appropriate therapeutic goals are cletermineci, they too must be converted into a cliet that the incliviclual can acquire, afford, and will eat. Most diseases and conditions alter neecis for only a few nutrients, with other nutrient neecis remaining similar to those of healthy persons. In clinical practice it is usually assumed that unless there is a specific deviation of a nutrient known to be associated with the disease or condition, the incliviclual is "healthy" with regard to that nutrient and the RDAs or AIs are reasonable goals for incliviclual planning. Thus, the intake recommendation that is appropriate for the incliviclual's gentler, age, level of physical activity, and physio- logical state (e.g., pregnancy, lactation) would apply. Government agencies or other organizations frequently specify that cliets feci to patients or to institutionalized populations meet previously established RDA or Recommencleci Nutrient Intake (RNI) levels. The approaches clescribeci in this report to plan cliets for a low risk of inacloquate nutrient intakes for groups and inclivici- uals would apply in these situations. For example, patients who are not at nutritional risk, who do not require a nutrition intervention, or who receive a regular cliet, can be treated as a group unless their nutritional status changes. Individual patients with specific nutri- tion therapy plans can have their dietary intakes planned initially using the RDAs or AIs with appropriate mollifications macle for their specific conditions by a trained health care professional or dietitian. After the appropriate nutrient goals for the individual who is ill have been cletermineci, these goals must then be converted into a clietary pattern that the incliviclual will consume. Therapeutic clietary planning relies upon specialized food guidance and menu planning systems specific to the various disease states that affect nutrient needs. The DRIs will be useful in the development of diet manuals for people with special health care needs. Parenterally-fed patients require special forms of nutrients, and needs must be adjusted since bioavailability factors are not applicable and absorptive losses do not occur. Thus, the DRIs cannot be used directly to plan parenteral intakes. As an example, a uremic patient who has end-stage renal disease might be placed on a very low protein diet to decrease blood urea nitrogen and other biochemical indices of uremia and to provide
146 DIETARY REFERENCE INTAKES symptomatic relief. The cliet might also be moclifieci to restrict sodium and phosphorus. However, the RDA or AI would be used for other nutrients not known to be affected by the disease process. The DRIs are formulated to meet the neecis of the vast majority of the healthy population within specified life stage and gentler groups. However, when the absorption, metabolism, or excretion of a nutrient is known to be altered by a specific illness or disease process, the DRIs can also be used as the base for developing thera- peutic flints.