Nutrition Standards and Meal Requirements for National School Lunch and Breakfast Programs: Phase I. Proposed Approach for Recommending Revisions (2008)

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.

5 Proposed Planning Model for Establishing Nutrition Standards for School Meals The term planning model refers to the committee’s overall approach and rationale for establishing the Nutrition Standards for the school meal programs. The planning model used to develop the current Nutrition Standards and the related Meal Requirements was based on legislation (USDA, 1995) that provided specifications for the use of the 1995 Dietary Guidelines for Americans (HHS/USDA, 1995) and the 1989 Recommended Dietary Allowances (RDAs) (NRC, 1989). Existing law requires that the meals provide one-third and one-fourth of the RDA for lunch and breakfast, respectively, and reflects an approach that was put in place before the development of the Dietary Reference Intakes (DRIs) and guidance on their related applications. In recognition of this, the committee will describe the planned approach to use the DRIs in this Phase I report. The Phase II report will compare differences between basing standards on the RDA approach and basing the standards on the DRI methodology. The proposed planning method addresses both food and nutrients and incorporates many different factors. Importantly, the method considers that the ultimate goal is to improve children’s diets by reducing the apparent prevalence of inadequate and excessive food and nutrient intakes (see criterion 1 in Chapter 3). This chapter considers (1) the setting of nutrient intake targets for school meals, (2) the setting of food intake targets for school meals, and (3) the use of a combined approach of setting both nutrient and food intake targets. This chapter first presents the assumptions that underlie its decision making (see Box 5-1). The setting of nutrient intake targets is covered first because it provides the basis for the content related to food intake targets. For the purposes of this chapter, the term target is used in a generic sense and represents a major but preliminary part of the process of setting Nutrition Standards and Meal Requirements for the school meal programs. 89 

90 NUTRITION STANDARDS AND MEAL REQUIREMENTS BOX 5-1 Assumptions Related to the Proposed Planning Models General assumptions of the proposed planning model 1. On average, foods offered will be equal to foods consumed; it is inappropriate to inflate the Nutrition Standards or the Menu Requirements, or both, to account for food offered but not consumed. 2. Children with higher or lower energy requirements relative to the Estimated Energy Requirement will adjust their intakes to meet their needs (by altering their intakes at other eating occasions or by eating less of the food served at the school meals). 3. Energy, nutrient, and food intakes can be partitioned into meals. 4. Changes in school meals alone cannot fully eliminate dietary inadequacies. To meet daily goals for food and nutrient intakes, the quality of the non-school foods will also need to improve. Specific assumptions of the proposed planning model for nutrients 1. The current nutrient intake distributions can be accurately estimated. 2. The assumptions of the EAR cut-point approach are met: a. Intake is independent of requirement (assumed true for all nutrients except energy) b. The requirement distribution is symmetric (not true for iron) c. The variance of intake is greater than the variance of the requirement (assumed to be true for children’s intakes) d. Observed intake distributions have been adjusted to reflect usual intakes 3. Unless efforts are made to target children in the tails of a nutrient intake distribution, the shape of the distribution will not change if the median intake is changed. Specific assumptions of the proposed planning model for foods 1. Intakes that adhere to the MyPyramid recommendations will achieve a low prevalence of dietary nutrient inadequacy and a low prevalence of potentially excessive nutrient intakes. 2. On average, the selection of school foods offered within a food group will match the foods that were used to develop the nutrient profiles of the MyPyramid food groups. 

PROPOSED PLANNING MODEL 91 SETTING NUTRIENT INTAKE TARGETS FOR SCHOOL MEALS Background The DRIs are the current reference values used in the United States and Canada (IOM, 1997, 1998, 2000a, 2001, 2002/2005, 2005). The DRIs replaced the former RDAs (NRC, 1989). As discussed in Institute of Medicine publications concerning applications of the DRIs for the purposes of assessing and planning diets (IOM, 2000b, 2003), the DRIs should now be the basis for assessing and planning the nutrient intakes by population groups, such as schoolchildren. An appropriate planning model for the derivation of the Nutrition Standards for school meals considers the nutrition needs of the entire population of schoolchildren rather than the needs of specific individuals (IOM, 2003). This approach involves consideration of the distribution of usual intakes when dietary intakes are assessed (as was done in Chapter 4). The planning goal would be to shift this distribution in a more desirable direction (IOM, 2003). The twin goals of planning intakes for groups are to (1) reduce the prevalence of inadequate intakes and (2) reduce the prevalence of intakes at risk of being excessive. To accomplish these goals in the setting of nutrient intake targets for groups of school children, several types of DRIs may be useful: • Estimated Energy Requirement (EER), • Acceptable Macronutrient Distribution Range (AMDR), • Estimated Average Requirement (EAR), • Adequate Intake (AI), and • Tolerable Upper Intake Level (UL). Values for all of these DRIs for schoolchildren are given in Appendix K. The RDA, which is also a type of DRI reference value, is not an appropriate target for planning the intakes by population groups because median group intakes at the RDA level are not likely to result in an acceptable prevalence of inadequate intakes within the group (IOM, 2003). See the section Applying the Dietary Reference Intakes to Reduce the Prevalence of Inadequacy for more details on the appropriate procedures performed by use of the EAR cut- point method. 

92 NUTRITION STANDARDS AND MEAL REQUIREMENTS Age-Grade Groups for Setting Nutrient Intake Targets The current age-grade groups specified by the U.S. Department of Agriculture (USDA) for meal programs are shown in Table 5-1. TABLE 5-1 Current USDA Age-Grade Groups Groups in National School Groups in School Breakfast Lunch Program Program Preschool Preschool Ka–3 K–12 K–6 7–12 4–12 7–12 a K = kindergarten. SOURCE: USDA, 2000b, 2008h. In general, age-grade groups are driven by existing school grade configurations. The age groups used to determine reference values, however, are based on biological and physiological development. Therefore, some of the children in an age-grade group will fall into a different age group for the nutrient reference value. A calculation must be carried out to determine what amount of a nutrient should be assigned as the target for the age-grade groups that correspond to school grade configurations. During Phase II of this study, the committee will consider the appropriateness of the current age-grade groups, especially in view of the application of DRI values to common school grade configurations. Alternative approaches for various configurations may also be addressed. The committee recognizes that the school grade configurations themselves may change over time and that the configurations may vary among districts. According to the U.S. Department of Education (2000b), a growing number of children attend middle schools, most of which encompass grades 6 through 8, whereas many high schools comprise grades 9 through 12. For the purposes of illustration relative to the proposed approach described in this Phase I report, however, the committee has used the following grade and age categories. The committee recognizes that the current census data may not support these specific categories: • Prekindergarten–grade 4: ages 4–8 years • Grades 5–8: ages 9–13 years • Grades 9–12: ages 14–18 years Planning Nutrient Intakes for Heterogeneous Even when the nutrient needs of schoolchildren are divided according to the age-grade categories listed above, the nutrient needs differ for the children within the two older groups because requirements differ by gender after age 8 years. The DRI age and gender groups are 4–8 years of age for both males and females, ages 9–13 years for males, ages 9–13 years for females, ages 14–18 years for males, and ages 14–18 years for females. 

PROPOSED PLANNING MODEL 93 To derive revised calculated Nutrition Standards (see the box in Figure 2-1 in Chapter 2), two approaches have been considered for heterogeneous groups: (1) an approach based the weighted average of the DRI and (2) an approach based on nutrient density. 1. Weighted average The current Nutrition Standards for school meals use the weighted average of the RDA for the relevant age-grade group. Calculations based on the RDAs for the ages and genders represented within the school age-grade category are used to obtain the nutrient reference values used for each age-grade category. The weights are determined by using the approximate proportions of the children that fall within each of the reference value age-gender categories. Under this approach, if the new DRIs were used to develop new Nutrition Standards for the illustrative age-grade groups identified earlier, the determinations would be as follows: • Prekindergarten–grade 4 (ages 4–8 years): the target DRIs for ages 4–8 years would not be weighted because the DRIs apply to both genders. • Grades 5–8 (ages 9–13 years): the target DRIs would be a weighted average of the reference values for 9- through 13-year-old males and females. • Grades 9–12 (ages 14–18 years): the target DRIs would be a weighted average of the reference values for 14- through 18-year-old males and females. 2. Nutrient density A different approach could be used on the basis of consideration of the nutrient density of the DRI for each of the gender groups within each of the age-grade categories. For each of the gender groups, a target nutrient intake per 1,000 calories would be calculated for each nutrient of interest. The value for the gender with the higher target, expressed as a nutrient density (that is, that value required for the most vulnerable group), would then be used in the process of setting the Nutrition Standards. For example, females may require more nutrients per 1,000 calories because their DRIs are often the same as those for males, but females’ calorie intakes are usually lower. If the school meals were designed to provide an amount of a nutrient per 1,000 calories that would meet the needs of most individuals in the more vulnerable group (e.g., females), then the meals would likely meet the nutrient needs of almost everyone in the less vulnerable group (e.g., males). The Nutrition Standards could potentially be higher by this approach than by the weighted-average approach, but the result could be meals that meet the needs of a higher percentage of the students. With either approach, it is important to examine the projected distribution of daily nutrient intakes around the target DRIs to ensure that very few, if any, would be above the UL, and thus at risk of being excessive (as discussed later in the section, Applying the Dietary Reference Intakes to Reduce the Prevalence of Intakes at Risk of Being Excessive). The gender group with the higher energy intake (e.g., males) would be more likely to exceed the UL if the value is the same for males and females. Applying the Dietary Reference Intakes for Energy Major decisions to be made by the committee include the determination of calorie levels—in particular, the percentage of the daily calorie intake to be provided by the National School Lunch Program (NSLP) and the School Breakfast Program (SBP) and how this translates to the calories provided in school meals. Consideration of calories is the first step in the committee’s proposed 

94 NUTRITION STANDARDS AND MEAL REQUIREMENTS plan for setting nutrient intake targets, which is outlined later in this chapter. Elements of the committee's approach to determining calorie levels appear below. Determining the Calorie Content of School Meals The EER is the appropriate reference value for use for the planning of energy intakes. Intakes should be planned to meet but not exceed the EER for a DRI age-gender-activity level (IOM, 2003). The average energy requirements determined in the School Nutrition Dietary Assessment Study–III (USDA, 2007a), assuming a low-active level of physical activity, are approximately the following, as explained in Chapter 4: • Prekindergarten–grade 4 (ages 4–8 years): 1,600 calories/day • Grades 5–8 (ages 9–13 years): 2,000 calories/day • Grades 9–12 (ages 14–18 years): 2,400 calories/day To illustrate the proposed planning approach presented in this Phase I report, the committee used the age-grade groups and calorie levels shown above. Although many children in grade 4 are 9 years of age, the DRI age group for 4–8 years is the age category that is most appropriate for prekindergarten through grade 4. Likewise, the age group of 9–13 years likely covers most configurations for the children in middle schools. However, given the variability of calorie needs within these categories, during Phase II the committee will also consider specifying a range of energy intakes within these age-grade categories. To gain perspective on the current calorie standards, the committee estimated the extent to which the current standard for the calorie content of school lunches (see Appendix H) contributes to children’s estimated requirements for calories. Because the school lunch is required to provide one-third (approximately 33 percent) of the day’s calorie intake, the committee estimated a full- day’s intake by multiplying the current standard for age-grade groups by three. For the kindergarten through grade 3 group (ages 5–8 years), for example, the calculation and comparison are as follows: Current meal standard 633 calories/lunch meal Multiplier (based on a lunch providing one- third of the energy requirement for the day) ×3 Calculated daily energy intake 1,899 calories/day Average daily energy requirement for age- grade group 1,600 calories/day Current meal standard as a percentage of the average daily energy requirement for the age-grade group 40% Thus, the current energy standard for the lunch meal provides about 40 percent of the day’s energy needs for this group, rather than 33 percent. Stated differently, for the youngest age group, the calculated daily amount is almost 300 calories per day higher than the average amount 

PROPOSED PLANNING MODEL 95 of energy required for children ages 4–8 years. Comparable calculations for the other age-grade groups show that the current energy standards for the lunch meal also substantially exceed one- third of the estimated daily requirement for calories. Determining the Percentage of Calorie Intake to Be Supplied by School Meals By law, the current school meals must supply, on average and at a minimum, one-third of the day’s energy requirements in a school lunch and one-fourth in a school breakfast. The School Nutrition and Dietary Assessment Study-III found that the school lunch typically supplies approximately 30 percent of the day’s energy intake for schoolchildren, whereas the school breakfast supplies approximately 20 percent of the day’s energy intake (USDA, 2007a). For the purposes of this Phase I report, the committee used 30 percent and 20 percent of intake as the amounts of energy to be supplied by school lunch and school breakfast, respectively. During Phase II, however, the committee will examine other sources (such as the National Health and Nutrition Examination Survey) that provide data on typical calorie intakes at lunch and breakfast as a percentage of the total day’s intake to determine the percentage of calorie intake to be supplied by school meals. It appears possible that the use of percentages that are lower than the current calorie standards may be more realistic for planning purposes for lunch and breakfast, especially in view of the observation that many children snack throughout the day. The use of lower percentages might also be less likely to contribute to the overconsumption of energy than the use of the current percentages prescribed by law. Applying the Dietary Reference Intakes for Macronutrient Ranges The AMDRs specify the desirable ranges of macronutrient intakes as a percentage of energy intake. AMDRs for schoolchildren have been set for five macronutrients: total fat, linoleic acid, alpha-linolenic acid, carbohydrate, and protein. The desirable ranges of values are given in Appendix K. Ideally, the amounts of macronutrients provided in school meals should minimize intakes that fall outside these ranges. Applying the Dietary Reference Intakes to Reduce the Prevalence of Inadequacy Nutrients with an Estimated Average Requirement For most nutrients for which an EAR is used, the current prevalence of nutrient inadequacy may be estimated by the EAR cut-point method (see Chapter 4). If the prevalence of inadequacy is too high, then one goal of the planning process is to reduce the prevalence of inadequacy to an acceptable level. For the purposes of this Phase I report, a 5 percent prevalence of inadequacy is used as an example. The committee may consider other levels, based on the feasibility of implementation and other relevant factors. Furthermore, if some children’s eating patterns differ substantially from those of the majority of their peers, then other approaches may be needed to reach children at the lower end of the intake distributions. Finally, the cost of providing the extra amounts of nutrients, which may be revealed through the committee’s future analyses, may influence the committee’s recommendations for some nutrients. One assumption that could be made in developing recommendations for revisions to the Nutrition Standards is that any changes in the school meals will be directly reflected in the total daily intake. If this assumption is correct, then alteration of the nutrient content of the school meals would shift the distribution of usual nutrient intakes without changing the shape of the 

96 NUTRITION STANDARDS AND MEAL REQUIREMENTS distribution. Under this assumption, it is possible to calculate how far the distribution needs to be moved (IOM, 2003), as follows: • If the goal is to have 5 percent of usual intakes below the EAR, then the 5th percentile of the intake distribution should be positioned at the EAR. • If the shape of the distribution does not change, then each point on the distribution, including the median, would change by the same amount. • The new median of the distribution could be calculated as the old median plus the amount of the change. • The new median would be considered the target median intake (TMI) and could be used as the basis for the establishment of Nutrition Standards for school meals. For example, assume that the EAR for a hypothetical nutrient is 10 milligrams (mg) per day and that 30 percent of the group has usual intakes below the EAR. To reduce the estimated prevalence of inadequacy from 30 percent to 5 percent, it would be necessary to increase the intake at the 5th percentile to 10 mg/day. If the intake at the 5th percentile was currently 4 mg/day, for example, then the distribution would need to be shifted by +7 mg/day (Figure 5-1). If the intake median (the intake at the 50th percentile) was currently 26 mg/day, then the TMI would be 33 mg/day. 

PROPOSED PLANNING MODEL 97 FIGURE 5-1 Concept of setting a TMI. (A) Baseline usual nutrient intake distribution, in which the prevalence of inadequate intake (percentage below the EAR) is about 30 percent. Shifting the baseline distribution up so that the prevalence of inadequate intakes reflects the planning goal (in this example, 5 percent below the EAR) attains the target usual nutrient intake distribution around the median (the TMI) (B). SOURCE: Adapted from IOM, 2006b. 

98 NUTRITION STANDARDS AND MEAL REQUIREMENTS By using data from the School Nutrition and Dietary Assessment Study-III study on the distribution of schoolchildren’s usual intakes, a TMI can be calculated for the nutrients of interest in schoolchildren’s diets. As an example, Table 5-2 presents the median intakes of selected nutrients reported by children ages 14–18 years and compares the TMIs associated with a 5 percent prevalence of inadequacy to those median intakes. Vitamin E is an example of a nutrient for which the proposed TMI would be much higher than the current median intake. As a result, the target may need to be revised because of concerns about feasibility and acceptability. Table 5-2 also presents TMIs for nutrients that do not have an EAR. For a discussion of how those TMIs were set, see the next section, Nutrients with an Adequate Intake. TABLE 5-2 Basis for TMI Values for Children Ages 14–18 Years, Selected Nutrients Reported 5th Percentile of Median Current Intake Nutrient Intake Distribution EARa TMIb Protein (g/kg/d) 1.3 0.8 0.72 1.22c Vitamin A (μg RAE/d) 580 243 558 895 Vitamin E (mg αT/d) 6.5 3.9 12 14.6 Vitamin C (mg/d) 76 26 60 110 Thiamin (mg/d) 1.7 0.9 1.0 1.8 Riboflavin (mg/d) 2.3 1.2 1.0 2.1 Niacin (mg/d) 22.9 13.8 11.5 20.6 Vitamin B6 (mg/d) 1.8 1.0 1.1 1.9 Folate (μg DFE/d) 550 284 330 596 Vitamin B12 (μg /d) 5.2 2.4 2.0 4.8 Calcium (mg/d) 1,073 510 N/A 1,300 Phosphorus (mg/d) 1,440 823 1,055 1,672 Magnesium (mg/d) 258 156 320 422 Iron (mg/d) 15.3 8.7 7.8 N/Cd Zinc (mg/d) 12.2 6.8 7.9 13.3 Sodium (mg/d) 3,677 2,361 N/A 2,300e Potassium (mg/d) 2,625 1,622 N/A 4,700 Fiber (g/d) 13.9 7.8 N/A 32 NOTE: αT = α-tocopheral; d = day; DFE = dietary folate equivalents; g = gram; mg = milligram; N/A = not applicable because an AI rather than an EAR has been set for these nutrients; N/C = not calculated; RAE = retinol activity equivalents; μg = microgram. a Average EAR of males and females (IOM, 2006b). b TMI is defined as the median of the target usual intake distribution; for illustrative purposes, the TMIs shown here are calculated as reported median intake + (EAR – intake at 5th percentile); for nutrients with an AI, the TMI is equal to the AI; all TMIs are the average for males and females. c Approximately 74 g/d for a boy weighing 61 kg and 66 g/d for a girl weighing 54 kg. d A TMI for iron cannot be calculated by the EAR cut-point method because iron requirements are not symmetrically distributed. e To foster palatability, and for consistency with the 2005 Dietary Guidelines for Americans, the UL is used as the TMI for sodium. SOURCE: USDA, 2007a. 

PROPOSED PLANNING MODEL 99 Although an EAR has been set for iron, a TMI for iron is not included in Table 5-2. The EAR cut-point method cannot be used for iron because the distribution of the requirement is not symmetric. The prevalence of inadequacy must be estimated from probability tables (IOM, 2001), and this may be done during Phase II of the study. Otherwise, the same approach as described above may be used to calculate the TMI for iron. Because intake at the 5th percentile (8.7 mg/day) approximates the EAR (7.8 mg/day) in the example shown in Table 5-2, it is likely that the TMI would be close to the current median intake (15.3 mg/day). Nutrients with an Adequate Intake An AI rather than an EAR is used for some nutrients. For these nutrients, a prevalence of inadequacy cannot be determined. If the median of the usual intake distribution is equal to the AI, however, a low prevalence of inadequacy can be assumed. Thus, for nutrients with an AI, the TMI can be set equal to the AI (as shown for potassium and fiber in Table 5-2). Although an AI has been set for sodium, the committee proposes to use a higher value (the UL) as the TMI. The higher value would be more consistent with the acceptability of the school meals to students. The use of the UL is consistent with the 2005 Dietary Guidelines for Americans. For sodium, the goal would be to reduce the median intake to the TMI. The derivation of the AI differs substantially for different nutrients and for different age- gender groups. Nonetheless, for nutrients with an AI, the AI is still the most appropriate target to use for the planning of school meals. For the AIs that were not set specifically on the basis of the current intakes by a healthy population, however, it is possible that median usual intakes at the AI level will not always be associated with a low prevalence of inadequate intakes. Potential Applications of Target Mean Intakes As shown in Table 5-2 for schoolchildren ages 14–18 years, the median usual intakes of some nutrients exceed the estimated TMI and the median usual intakes of many nutrients are well below the estimated TMI. Children in the younger age groups have more favorable intakes of many nutrients. For children ages 6–8 years, for example, the median usual intakes of 13 nutrients 1 exceed the estimated TMI (data not shown). Thus, if the revised Nutrition Standards for school meals are based on a TMI, the resulting daily intakes are likely to be below the current usual intakes for several nutrients, especially for the younger age groups. Because school meals are not intended to provide 100 percent of the daily nutrient intake, it is reasonable to set the Nutrition Standards on the basis of a percentage of the TMI. For example, the Nutrition Standard for a nutrient might be 20 percent of the nutrient’s TMI for breakfast and 30 percent for lunch. The percentage of a nutrient’s TMI that is used as the school meal target may vary across the nutrients. For example, the percentage goals for nutrients in the inadequate intake column of Table 4-12 in Chapter 4 might be higher than for those whose daily reported intakes are adequate. However, it is unlikely that school meals can fully compensate for all nutrient inadequacies. Thus, improvements in dietary intake at other eating occasions will be necessary to meet all of the daily intake goals. 1 These nutrients exclude sodium. 

100 NUTRITION STANDARDS AND MEAL REQUIREMENTS Applying the Dietary Reference Intakes to Reduce the Prevalence of Intakes at Risk of Being Excessive The UL should be used to evaluate whether planned meals are likely to result in a risk of excessive intakes. Ideally, the predicted prevalence of intakes above the UL should be close to zero for all age-gender groups. If the prevalence of intakes above the UL is not low, it may be necessary to adjust the target nutrient intake distribution so that the prevalence of inadequacy and the prevalence of intakes above the UL are acceptably balanced. For some intake distributions, it may be desirable to investigate ways to change the shape of the nutrient intake distribution, rather than trying to move the current distribution up or down. There may be nutrients for which it would be desirable to reduce the prevalence of very high or very low intakes, without changing intakes in the middle of the distribution. For example, if highly fortified foods are contributing to very high nutrient intakes for a few individuals, these foods might be removed from the menu. Likewise, if some children avoid foods (such as milk) that are rich sources of certain nutrients, then including acceptable substitutes in the menu might increase their intakes of those nutrients. However, detailed examinations of the impact of interventions (such as feeding programs) on the shape of an intake distribution are almost nonexistent (IOM, 2003, p. 88). Proposed Method for Setting Nutrient Intake Targets for School Meals On the basis of earlier guidance from the Institute of Medicine (IOM, 2003), the committee has proposed the following seven steps for the setting of nutrient intake targets for school meals. 1. For each age-gender group, determine the target daily energy intake and the goals for the percentages of energy to be provided by breakfast and lunch. 2. For nutrients with an EAR: a. Determine the acceptable prevalence of inadequacy and the acceptable prevalence of excessive intakes. b. Determine a target nutrient intake distribution to achieve these goals. The median of this distribution is the TMI for the age-gender group. c. If necessary, adjust the target nutrient intake distribution so that the prevalence of inadequacy and the prevalence of intakes above the UL are acceptably balanced. 3. For nutrients (other than sodium) with an AI: a. Set the TMI equal to the AI for the age-gender group. b. If necessary, adjust the TMI to reduce the prevalence of intakes above the UL. 4. For sodium, set the TMI equal to the UL for the age-gender group. 5. For each nutrient, apply the age-gender TMIs to develop a TMI for each grade category using either a weighted average or a nutrient-density approach. 6. For each nutrient, determine the goal for the percentage of a day’s intake to be provided by school breakfast and school lunch, and apply the percentage to the TMI to obtain the school meal target. 7. Evaluate the proposed school meal targets in terms of feasibility, cost, and acceptability. Revise the targets as needed to provide an acceptable balance of adequacy, avoidance of excess, feasibility, cost, and acceptability. Sodium is an example of a nutrient for which the proposed TMI may need to be revised because of concerns about feasibility and acceptability. 

PROPOSED PLANNING MODEL 101 The resulting nutrient intake targets would be consistent with the goals of planning school meals to reduce the prevalence of inadequacy and to reduce the risk of excessive intakes among schoolchildren. However, the impact of changes in the Nutrition Standards and Meal Requirements on children’s daily intakes cannot be completely predicted. An intake assessment performed after changes are implemented by USDA would be needed to determine the impact. SETTING FOOD INTAKE TARGETS FOR SCHOOL MEALS Background To develop food intake targets that are based on the current Dietary Guidelines for Americans (at present, the year 2005 edition), the committee proposes to incorporate the MyPyramid food intake patterns as shown in Appendix J. The use of food intake targets that are consistent with the MyPyramid food intake patterns would be consistent with adherence to the Dietary Guidelines for Americans. As described in Chapter 4, the committee used MyPyramid to assess food intakes for children ages 4–8 years, 9–13 years, and 14–18 years. MyPyramid was designed by using the most recent RDAs or AIs for 17 micronutrients and fiber and the AMDRs (Britten et al., 2006). In the development of MyPyramid, the nutrient profiles of each of the food groups were used to ensure that the specified number of servings from the MyPyramid food groups would meet those nutrient intake targets (Marcoe et al., 2006). Comparison of Food and Nutrient Intake Targets Because the RDAs are not an appropriate target for use for the planning of the intakes by groups (IOM, 2003), the committee used a new approach to investigate the applicability of basing food intake targets on MyPyramid. In particular, the committee compared the amount of nutrients provided by MyPyramid intake patterns to TMIs, which, as described above, would reduce the predicted prevalence of nutrient inadequacy to an acceptable level. Table 5-3 makes this comparison by using the TMIs from Table 5-2. For children ages 14–18 years, the amount of nutrients provided by the 2,400-calorie MyPyramid food intake pattern would exceed the TMI for all but two nutrients (vitamin E and potassium), and all nutrients would be provided at 73 percent or more of the respective TMIs (see the rightmost column of Table 5-3). For the children ages 6–8 years, the amount of nutrients provided by the MyPyramid intake patterns would exceed the TMI for all nutrients except vitamin E and potassium (data not shown). For children ages 9–13 years, vitamin E, which would be provided at only 73 percent, was the only nutrient that would not exceed the TMI (data not shown). 

102 NUTRITION STANDARDS AND MEAL REQUIREMENTS TABLE 5-3 Contents of Selected Nutrients for a 2,400-Calorie MyPyramid Food Intake Pattern, Absolute Value and Percentage of the TMI Nutrient content of MyPyramid food intake pattern for 2,400 MyPyramid Nutrients as % of Nutrient calories/day TMIa Protein (g/d) 105 128b Vitamin A (μg RAE/d) 1,126 126 Vitamin E (mg αT/d) 10.7 73 Vitamin C (mg/d) 163 148 Thiamin (mg/d) 2.4 133 Riboflavin (mg/d) 3.1 148 Niacin (mg/d) 27.3 133 Vitamin B6 (mg/d) 2.9 153 Folate (μg DFE/d) 822 138 Vitamin B12 (μg/d) 9.2 192 Calcium (mg/d) 1,388 107 Phosphorus (mg/d) 1,961 117 Magnesium (mg/d) 440 104 Iron (mg/d) 21.5 141c Zinc (mg/d) 16.7 126 Sodium (mg/d) 2,136 142 Potassium (mg/d) 4,523 96 Fiber (g/d) 37 116 NOTE: αT = α-tocopheral; d = day; DFE = dietary folate equivalents; g = gram; mg = milligram; RAE = retinol activity equivalents; μg = microgram. a TMIs are based on the values in the last column of Table 5-2. b Based on an average weight of 67 kg (reference weight for an 18 year-old male (IOM, 2002/2005, p. 137))—the age and gender with the highest TMI within this age group. c For illustrative purposes, this assumes that the iron TMI is equal to the current median intake of 15.3 mg/day. SOURCE: Britten et al., 2006. Reprinted from the Journal of Nutrition Education and Behavior, Volume 38, P. Britten, K. Marcoe, S. Yamini, and C. Davis, Development of Food Intake Patterns for the MyPyramid Food Guidance System, pages S78–S92, Copyright (2006), with permission from Elsevier. Comparison of MyPyramid Intake Patterns with Current School Meal Requirements To compare the amounts of food specified by the MyPyramid food intake pattern with the current school meal requirements for a school lunch, the committee divided the total daily MyPyramid amounts for selected calorie levels by three. (By law, school lunch must provide one-third of a day’s intake of calories and nutrients.) Based on the committee’s calculations, the amount of food recommended in the MyPyramid food intake patterns exceeds the current minimum amounts required for the majority of food groups in NSLP meal patterns. A table illustrating the MyPyramid food amounts for 800 calories (2,400 calories/3, the calorie level that might apply to school lunches for children ages 14–18 years) is located in Appendix L. For comparison, Appendix L also shows the current minimum food group amounts for both a traditional and an enhanced meal in the NSLP for children in grades 7–12 and the percentage of the MyPyramid food intake pattern amount represented. 

PROPOSED PLANNING MODEL 103 By quantity, the primary difference is for total fruits and vegetables. The amount of fruits and vegetables in the current Meal Requirements for a traditional lunch is less than half the MyPyramid amount of approximately 1.67 cup equivalents. Similar differences are seen for the other two age-grade categories. MyPyramid is very specific about the quality of the foods within the food groups. For example, MyPyramid provides targets for fruits and vegetables separately, and it specifies separate targets for five vegetable subgroups: dark green vegetables, orange vegetables, legumes, starchy vegetables, and other vegetables. MyPyramid also specifies that half of the grain intake should be whole grain. Such specifications are not currently included in the NSLP Meal Requirements. Importantly, MyPyramid assumes that all foods are provided in their lowest-fat form and have no added sugars. Unlike the current Meal Requirements, MyPyramid includes a value for discretionary calories, which are calories from any source that can be used flexibly (these calories are often from added sugars or solid fats or fat from foods that are not in their lowest-fat form, such as milk with 2 percent fat). However, if the MyPyramid meal plans are followed, few discretionary calorie are available: for a lunch meal that is one-third of the daily calories, the discretionary calorie levels would be 44 calories/meal for the 1,600-calorie/day plan, 89 calories/meal for the 2,000-calorie/day plan, and 121 calories/meal for the 2,400-calorie/day plan. Proposed Method for Setting Food Intake Targets for School Meals The following three steps outline a potentially useful general approach to applying current dietary guidance to the planning of school meals: 1. Select appropriate energy levels. 2. Apply the goal for the percentage of the day’s intake (e.g., 20 and 30 percent for breakfast and lunch, respectively) to the MyPyramid food intake pattern for the energy level (by using the amounts in the MyPyramid food groups shown in Appendix J, Table J-2) to obtain amounts of each food group to recommend. For some food groups, amounts may be specified per week rather than per day, to achieve practical portion sizes. 3. Consider the recommendations for discretionary calories. Staying within these recommendations may require greatly decreasing or eliminating the use of foods that are high in fats and added sugars. SUMMARY: COMBINING THE NUTRIENT INTAKE AND FOOD INTAKE TARGETS Although the committee recognizes the need for nutrient intake targets, the process for setting nutrient intake targets described in this chapter involves many assumptions. Thus, there are many uncertainties about the accuracy of the estimated TMIs. A comparison of the TMIs with the nutrients provided by the MyPyramid food intake patterns shows that adherence to MyPyramid results in diets whose nutrient contents almost always meet or exceed the TMIs. Moreover, adherence to MyPyramid results in diets that are consistent with the Dietary Guidelines for Americans. To achieve the planning objectives, the committee will consider recommending that food targets be emphasized in the development of the Nutrition Standards and Meal Requirements for school meals. In particular, the committee may begin by using the MyPyramid food plans as the basis for the school meal targets and then assess projected nutrient 

104 NUTRITION STANDARDS AND MEAL REQUIREMENTS intake distributions (using information about the shape of current intake distributions) to determine if the desired objectives are likely to be achieved. As with any method of planning school meals, it would be necessary to assess the children’s actual nutrient intakes after changes are implemented by USDA to determine if the planning objectives have been achieved. The school meal food targets would be supplemented with selected school meal nutrient targets for nutrients such as sodium, total fat, saturated fat, cholesterol, and a target for added sugars. An iterative process will be used to find the most satisfactory balance of food group targets to achieve consistency with Dietary Guidelines for Americans and DRIs. If it is workable, this approach may offer an additional advantage: the simplification of meal planning and monitoring.