Cover Image

PAPERBACK
$55.00



View/Hide Left Panel

4
Process for Developing the Nutrient Targets

The committee developed Nutrient Targets to serve as a guide for setting the standards for menu planning. It did so for 24 nutrients and other dietary components. The full range of nutrients needed to be considered to be certain that the standards for menu planning would be developed appropriately. The intent is not to use the Nutrient Targets themselves for menu planning.

In developing the Nutrient Targets, the committee took several different approaches that depended on the type of nutrient. This chapter describes the approaches used to set preliminary targets for (1) calories, (2) fats and cholesterol, (3) nutrients with Estimated Average Requirements, and (4) nutrients with Adequate Intakes. The term nutrient target is used to denote each preliminary value. Chapter 5 covers the process for using nutrient targets in developing recommended Meal Requirements, Chapter 6 covers the iterative process that led to the final recommendations, and Chapter 7 presents the recommended Nutrient Targets and Meal Requirements.

PRELIMINARY CALORIE TARGETS FOR MEALS

Method Used to Set Calorie Targets for Breakfast and Lunch for the Three Age-Grade Groups

As described in Chapter 2, the committee set mean daily calorie levels for each of the three age-grade groups (combining means for males and females) and then rounded these values to have them correspond with My-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 69
4 Process for Developing the Nutrient Targets The committee developed Nutrient Targets to serve as a guide for set- ting the standards for menu planning. It did so for 24 nutrients and other dietary components. The full range of nutrients needed to be considered to be certain that the standards for menu planning would be developed appropriately. The intent is not to use the Nutrient Targets themselves for menu planning. In developing the Nutrient Targets, the committee took several different approaches that depended on the type of nutrient. This chapter describes the approaches used to set preliminary targets for (1) calories, (2) fats and cholesterol, (3) nutrients with Estimated Average Requirements, and (4) nutrients with Adequate Intakes. The term nutrient target is used to denote each preliminary value. Chapter 5 covers the process for using nutrient targets in developing recommended Meal Requirements, Chapter 6 covers the iterative process that led to the final recommendations, and Chapter 7 presents the recommended Nutrient Targets and Meal Requirements. PRELIMINARY CALORIE TARGETS FOR MEALS Method Used to Set Calorie Targets for Breakfast and Lunch for the Three Age-Grade Groups As described in Chapter 2, the committee set mean daily calorie levels for each of the three age-grade groups (combining means for males and females) and then rounded these values to have them correspond with My- 

OCR for page 69
0 SCHOOL MEALS TABLE 4-1 Mean and Rounded MyPyramid Calorie Levels by Age- Grade Group Mean Calorie Level for Rounded Calorie Level Males and Femalesa Age-Grade Group for Males and Females Ages 5–10 y, Kindergarten–Grade 5 1,830 1,800 Ages 11–13 y, Grade 6–8 2,015 2,000 Ages 14–18 y, Grade 9–12 2,365 2,400 NOTE: y = years. aThese requirements were obtained from the mean Estimated Energy Requirement calcula- tions for the age-grade-gender group. Pyramid calorie levels for meal patterns. The original and rounded mean values appear in Table 4-1. To determine target calorie levels for school breakfast and lunch, the committee reviewed data from the third School Nutrition Dietary Assess- ment study (SNDA-III). These data (shown in Appendix G, Table G-1) in- dicated that, compared with a single value, a range would more accurately represent the proportion of calories obtained by school-aged children from meals and snacks. The children who participated in the School Breakfast Program obtained 19 to 24 percent of their total calorie intake from break- fast. The children who participated in the National School Lunch Program obtained approximately 30 to 34 percent of their total calorie intake (over 24 hours) from lunch. Findings were comparable for school-aged children overall and for low-income children (those approved for free or reduced- price meal benefits) (data not shown). The committee also reviewed data from the National Health and Nutrition Examination Survey (NHANES) 1999–2004 and found that the distribution of calories among breakfast, lunch, dinner, and snacks was consistent with that found using SNDA-III data. The committee agreed to set a maximum target for calories to help limit excessive calorie intake at breakfast and lunch. Having both a minimum and a maximum value helps ensure adequate calories while giving school food operators some flexibility when planning menus. The means of the values used for the minimum and maximum calories (21.5 percent for breakfast and 32 percent for lunch) were used in setting selected Nutrient Targets, as described later in this chapter. The committee applied the information about the proportion of calo- ries that children typically obtain from breakfast and lunch meals to the rounded calorie levels established for the three age-grade groups. For ex- ample, for children ages 5–10 years, the lower end of the calorie range was calculated as follows: lunch calories = 1,800 calories × 0.3

OCR for page 69
 PROCESS FOR DEVELOPING THE NUTRIENT TARGETS To provide calorie targets that would be practical for school food operators, the committee agreed to use rounded values to establish the target calorie values for each meal. They were rounded to the nearest 50 while retaining at least a 100-calorie range within an age-grade group, with the exception of lunch for children ages 14–18 years. In the latter age group, the commit- tee chose to round up to the nearest 50 to accommodate the caloric needs of high school males. Results and Discussion Tables 4-2 and 4-3 show the values used to set the preliminary target minimum and maximum calorie values for school breakfast and school lunch, respectively, and the rounded target calorie values. These values apply to the average daily calorie content of meals offered across a 5-day school week. The calorie content of the meals offered on a single day could be below the minimum or above the maximum as long as the average for the week falls within the range. TABLE 4-2 Values Used to Set Preliminary Target Calorie Minimum and Maximum for School Breakfast and Preliminary Target Calories, by Age- Grade Group Mean Preliminary Daily Minimum: Maximum: Target Minimum Age-Grade Group Calories 19% of Daily 24% of Daily and Maximum Ages 5–10 y, Kindergarten– 1,800 342 432 350–450 Grade 5 Ages 11–13 y, Grades 6–8 2,000 380 480 400–500 Ages 14–18 y, Grades 9–12 2,400 456 576 450–600 NOTE: y = years. TABLE 4-3 Values Used to Set Preliminary Target Calorie Minimum and Maximum for School Lunch and Preliminary Target Calories, by Age- Grade Group Mean Preliminary Daily Minimum: Maximum: Target Minimum Age-Grade Group Calories 30% of Daily 34% of Daily and Maximum Ages 5–10 y, Kindergarten– 1,800 540 612 550–650 Grade 5 Ages 11–13 y, Grades 6–8 2,000 600 680 600–700 Ages 14–18 y, Grades 9–12 2,400 720 816 750–850 NOTE: y = years.

OCR for page 69
 SCHOOL MEALS The committee recognizes that some children with limited access to food or with substantially higher calorie needs might benefit from school meals that provide significantly more calories (and nutrients). It believes, however, that this situation does not provide the basis for an increase in the maximum calorie levels for school meals. Instead, school food authorities and community organizations have additional mechanisms to help ensure that children have access to sufficient food during the day. SETTING THE MAXIMUM FOR SATURATED FAT AND CHOLESTEROL, THE RANGES FOR TOTAL FAT, AND ADDRESSING TRANS FAT The committee relied on recommendations from Dietary Guidelines for Americans to set a target maximum intake for saturated fat and cholesterol (substances in food that are not essential nutrients) and the range for total fat. It considered Dietary Guidelines plus supplementary information to ad- dress whether it would be possible to set a target maximum for trans fat. Reasons for Limiting Intakes of Fats and Cholesterol Limiting the intakes of saturated fat, cholesterol, and trans fat helps support healthful blood lipids. Avoiding excessive total fat intake helps control saturated fat intake and helps avoid the intake of excessive calories. Adequate fat intake helps ensure adequate intake of vitamin E and essential fatty acids, helps support a normal pattern of growth, and may help avoid unfavorable changes in certain blood lipids (HHS/USDA, 2005). Preliminary Targets for Total Fat, Saturated Fat, and Cholesterol For children, Dietary Guidelines for Americans (HHS/USDA, 2005) recommends a fat intake of 25 to 35 percent of total calories, less than 10 percent of calories from saturated fatty acids (which are abundant in the fat in dairy products and meat), and a maximum of 300 mg of cholesterol per day for all individuals over the age of 2 years. The committee used these values as the basis for the preliminary fat targets for school meals. trans Fat Dietary Guidelines for Americans (HHS/USDA, 2005) recommends that trans fat intake be kept as low as possible, but it does not specify a maximum level of intake. In turn, no data exist on which to base a maxi- mum level for trans fat in school meals, even though the goal is essentially zero grams. Nonetheless, a practical method can be used to keep the trans

OCR for page 69
 PROCESS FOR DEVELOPING THE NUTRIENT TARGETS fat content of school meals to a minimum. In particular, this is achievable by specifying that, for any food included on the school menu (including any ingredient used by schools to prepare the food), 0 g of trans fat per serving would be the maximum amount of trans fat listed on the nutrition label or in manufacturer’s specifications. This method is not always applicable because some products, such as bakery items produced by manufacturers who qualify as small businesses, are exempted from nutrition labeling, and thus the trans fat content of the product may not be specified. The com- mittee notes that foods labeled as containing 0 g of trans fat may actually contain a small amount (< 0.5 g) of trans fat per serving. The rounding rules for declaring 0 g of trans fat are established based on analytical vari- ance for the substance, and any amount that is rounded down to zero is considered “dietarily insignificant” by the Food and Drug Administration (FDA, 2008). Nonetheless, relying on label declarations is the only practi- cal approach to keeping the trans fat content of school meals as close to zero as possible. SETTING TARGETS FOR PROTEIN, VITAMINS, AND MINERALS The report Dietary Reference Intakes: Applications in Dietary Plan- ning (IOM, 2003) devotes a chapter to methods for planning daily diets for groups and discusses how to plan for a target usual nutrient intake distribution. In setting the preliminary nutrient targets for school meals, the committee followed these guidelines, adapting and modifying them as necessary to meet challenges described in Chapter 2. The work of this committee represents one of the first uses of the proposed dietary plan- ning approach for a large national program and thus extends what was a theoretical approach to an important practical application. The challenges, and solutions, presented below should provide useful guidance to others wishing to set nutrient targets for similar purposes. They also indicate the need for further evaluation of the process, as indicated in the section “Rec- ommendations for Evaluation” in Chapter 10. Overview of the Target Median Intake Approach The overall goal of planning intakes for groups of people, such as schoolchildren, is to achieve usual daily intakes within the group that meet the requirements of most individuals but are not excessive (IOM, 2003). This goal is accomplished by combining information on the group’s usual nutrient intakes with information on the group’s nutrient requirements (ex- pressed as either Estimated Average Requirements or Adequate Intakes) and Tolerable Upper Intake Levels. The target nutrient intake distribution that is chosen aims to achieve the combined goal of a low predicted prevalence

OCR for page 69
 SCHOOL MEALS of nutrient inadequacy and a low predicted prevalence of excessive intakes. The median of this intake distribution is the Target Median Intake. The Target Median Intake is the starting point for the committee’s calculations to derive the Nutrient Targets for school meals. The initial Target Median Intakes for nutrients, which are discussed in the following two sections, appear in Appendix Table J-1. Setting Targets for Nutrients with an Estimated Average Requirement Oeriew For most nutrients with an Estimated Average Requirement (EAR), the current prevalence of inadequacy may be estimated using the EAR cut-point method (IOM, 2006). 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. Thus, one of the steps in planning for the nutrient intake of groups is to select the target prevalence of inadequacy. The com- mittee set 5 percent rather than the more conservative 2 to 3 percent that has been suggested as an acceptable level of inadequacy (IOM, 2003) for three reasons: 1. The intake distributions for school meal participants come from SNDA-III (USDA/FNS, 2007a). Although the overall sample of children was large, the number of children in specific age and gender groups was relatively small (approximately 200 to 300), and the standard errors in the tails of the distribution were large. Estimates at the 5th percentile were more stable than those at the 2.5th percentile and less likely to be affected by outliers. 2. Nutrient intakes collected using the 24-hour diet recall are likely to be underreported, especially by adolescent girls. Intakes in the bottom 2.5 percent of the distribution are very likely to be underestimates. As a result, using the 2.5th percentile as the basis for setting the Nutrition Standards might result in unnecessarily high standards. 3. Data were unavailable on the effect of changes in the school meals on the rest of the day’s intake. The EARs used to determine the Target Median Intakes for school- children 6–10 years of age are weighted averages of two age groups. The use of weighted averages was necessary because the proposed elementary school group for school meals spans part of two Dietary Reference Intake (DRI) age groups (ages 4–8 and 9–13 years). The weighting factor was the proportion of the 5-year age span: three-fifths for ages 6–8 years and two- fifths for ages 9–10 years.

OCR for page 69
 PROCESS FOR DEVELOPING THE NUTRIENT TARGETS Description of the Method To achieve a target nutrient intake distribution with approximately a 5 percent prevalence of inadequacy, it is necessary to alter the current distri- bution of children’s intakes for many nutrients. Using the method recom- mended by the Institute of Medicine (IOM, 2003), the committee shifted each current intake distribution upward or downward until approximately 5 percent of the group’s intakes were below the EAR. This method for determining the target distribution assumes that a change in the nutrient content of the daily diet would apply to everyone, and thus the distribution of usual nutrient intakes would shift without changing the shape of the distribution.1 Under this assumption, the appropriate change in the nutrient intake distribution was calculated as follows: • The 5th percentile of the intake distribution was positioned at the EAR. • The new median of the distribution was calculated as the original median plus the difference between the intake at the 5th percentile and the EAR. If current intake at the 5th percentile of the current intake distribu- tion is above the EAR, the new median would be below the current median. The new median is the Target Median Intake for the day. The same method was used for all vitamins and minerals with an EAR, except for iron (see discussion of iron below). It was also used to determine a protein Target Median Intake in grams per kilogram of body weight (the units of the EAR for protein). To convert the value to grams of protein per day, it is necessary to assume a body weight for the children in each age-grade group. The committee used the SNDA-III body weights shown in Appendix Tables F-1 and F-2 for the midpoint ages in each age-grade group and averaged the weights for males and females. Although energy needs were based on body weights from the CDC growth charts because they are the reference standards for healthy children, the committee decided to base protein needs on the actual reported body weights from SNDA-III. Because the SNDA-III weights are higher than the CDC body weights, this method ensures that the protein targets cover almost all schoolchildren. The resulting average body weights were 29.3 kg for the kindergarten through grade 5 group, 51.1 kg for the grade 6 through 8 group, and 67.0 kg for the grade 9 through 12 group. 1 The committee recognizes weaknesses of this assumption; however, the method provides useful estimates, and a superior alternative method has not been developed.

OCR for page 69
 SCHOOL MEALS Example To illustrate the method, the vitamin C Target Median Intake for high school students is used as an example. 1. The SNDA-III data show that vitamin C intakes at the 5th percen- tile are a. 32 mg per day for males and b. 19 mg per day for females ages 14–18 years (Appendix Table I-1). 2. The EARs for vitamin C are 63 and 56 mg per day, respectively. Thus, • the intake of the males needs to increase by 31 mg per day (63 mg minus 32 mg equals 31 mg), • the intake of the females needs to increase by 37 mg per day (56 mg minus 19 mg equals 37 mg). As a result, 3. The Target Median Intake for the males would be 121 mg per day (90 mg [the current median intake] plus 31 mg equals 121 mg per day). 4. The Target Median Intake for the females would be 104 mg per day (67 mg [the current median intake] plus 37 mg equals 104 mg per day). Iron as a Special Case Because iron requirements are not normally distributed for menstruat- ing females, the EAR cut-point method is not appropriate for calculating the Target Median Intakes for iron for females ages 11–13 and 14–18 years. Instead, the committee used a modeling approach based on the probability method (IOM, 2001b, pp. 205–208) for females in these two age groups (see Appendix I). The resulting Target Median Intakes were 15.5 mg per day for females ages 11–13 years and 15.9 mg per day for females ages 14–18 years. Nutrients with an Adequate Intake General Approach Some nutrients have an Adequate Intake (AI) rather than an EAR. Based on guidance from the Institute of Medicine (IOM, 2003), the com- mittee assumed that a low prevalence of inadequacy would result if the median of the usual intake distribution was at least equal to the AI. Thus,

OCR for page 69
 PROCESS FOR DEVELOPING THE NUTRIENT TARGETS for five nutrients with an AI (calcium, potassium, fiber, linoleic acid, and α-linolenic acid), the Target Median Intake would be set at the AI. As was done with the EARs, weighted averages were used for the AIs for the youngest age group (6–10 years). Although the derivation of the AI differs substantially among these nutrients and among different age-gender groups, the AI is still the most appropriate type of DRI to use to set the Target Median Intake. Exceptions Sodium The approach used to address sodium did not involve setting a Target Median Intake. Instead, the committee agreed to set maximum daily targets for sodium that are based on the age-specific ULs for sodium. This decision was made for several reasons. The AI for sodium is 1.2 g per day for children ages 5–8 years and 1.5 g per day for older children—far less than children consume on average. Recognizing that sodium intake in the United States far exceeds the AI and also the Tolerable Upper Intake Level (UL), the sodium recommendation in Dietary Guidelines for Americans is 2.3 g per day—the value of the UL for persons ages 14 years and older. (The ULs for children younger than 14 years are slightly lower than 2.3 g per day.) Basing the sodium target on the UL rather than the AI is more consistent with achieving meals that are palatable and thus acceptable to U.S. schoolchildren. For sodium, the goal would be to reduce the median intake to the UL. Vitamin D A Target Median Intake was not calculated for vitamin D (which has an AI) because of a lack of reliable data on the vitamin D content of foods and on vitamin D intakes. Vitamin D intakes were not assessed for SNDA-III. Although vitamin D intakes have recently been estimated for the What We Eat in America survey (NHANES 2005–2006) (USDA/ARS, 2009a) and were found to be low, especially for adolescent females, the age groups for the reported data could not be used to calculate Target Median Intakes for the age groups in this report. Although exposure to sunshine reduces the need to ingest vitamin D, this vitamin D source is highly variable and is not under the control of school meal programs. Thus the role of sunshine in providing vitamin D was not considered by the committee. As described in the “Supportive Findings” section of Chapter 3, the committee is aware of the pending Institute of Medicine report on the requirements and upper levels of intake for vitamin D and acknowledges the appropriateness of using that report in the future to inform decisions that may be made about the vitamin D levels in school meals.

OCR for page 69
 SCHOOL MEALS Method and Rationale for Calculating the School Meal-Target Median Intakes To incorporate the Target Median Intake concept into the setting of the Nutrient Targets for school meals, the committee first addressed the fact that nutrient needs differ substantially between males and females within the age-grade groups. Its aim was to calculate targets for total daily intake that would best reflect these differences in nutrient needs. The committee used three methods of calculation (described below) to obtain candidate values for School Meal-Target Median Intakes (School Meal-TMIs). Re- gardless of the method used, the committee deemed the differences in re- quirements too small to consider both gender- and age-specific requirements within the grade group that encompasses kindergarten through grade 5. Thus, in examining the three approaches to setting School Meal-TMIs, only gender was considered within the kindergarten through grade 5 group. Calculation Method Used The committee used the following three methods to combine the Target Median Intakes by gender for each of the three grade groups. 1. Aerage Target Median Intake. Calculate the values for males and females separately within each of the grade groups (see Appendix Table J-1), and calculate the aerage for the grade group as the candidate School Meal-TMI. 2. Highest Target Median Intake. Calculate the values for males and females separately within each of the grade groups, and use the higher one for the grade group as the candidate School Meal-TMI. 3. Simple Nutrient Density2 Target Median Intake. Calculate the nutrient density (the ratio of the gender-specific Target Median Intake to the gender-specific Estimated Energy Requirement shown in Table 2-4) for males and females separately within each of the grade groups. Then mul- tiply the higher density times the mean Estimated Energy Requirement for the grade group to obtain the candidate School Meal-TMI. The simple nutrient density method (#3 above) had been specifically designed for this purpose (IOM, 2003). Although other approaches have 2 The term nutrient density has been used in different ways in the literature. The usage pre- sented here is the one presented in Dietary Reference Intakes: Applications in Dietary Planning (IOM, 2003). This usage applies to setting a target for daily intake of each nutrient relative to daily calorie needs. In other sections of this report, the committee refers to the nutrient density of foods and has adopted the definition that it considers most useful and understand- able: namely, the nutrient density of foods refers to the amount of a specific nutrient in a food per 100 calories of that food.

OCR for page 69
 PROCESS FOR DEVELOPING THE NUTRIENT TARGETS been proposed, they were considered unnecessarily complex for setting school meal targets. Nutrient density may be expressed in several ways; the approach described here considers each nutrient’s requirement for a group of children relative to the energy requirement for the same group—that is, the ratio of the amount of a nutrient to the energy provided by the diet (IOM, 2003, p. 14). Comparison of Results The differences in the resulting candidate School Meal-TMI values from the three methods were not large, ranging up to 11 percent across 20 nutrients within the kindergarten through grade 5 group, up to 23 percent (for iron) within the grade 6 through 8 group, and up to 19 percent (for iron) within the grade 9 through 12 group (Appendix Table J-2). In general, the following conclusions can be drawn by comparing the results of the three methods. • Calculating the average Target Median Intake usually resulted in the lowest value. This level of total daily intake, if achieved, should result in approximately a 5 percent oerall prevalence of inadequacy for the grade group, but a higher than 5 percent prevalence for one of the gender groups. For example, females might have a higher than 5 percent prevalence of inadequacy, and males might have a prevalence of inadequacy of less than 5 percent. • By comparison, the use of a School Meal-TMI based on the highest Target Median Intake would result in a maximum prevalence of inadequacy of 5 percent for either of the gender groups within the grade group. For example, females might have a 5 percent prevalence of inadequacy, but males would have a prevalence of less than 5 percent. • The nutrient density method often results in the highest values, particularly for the two older grade groups. It sets the concentration of the nutrient high enough to result in a maximum of a 5 percent prevalence of inadequacy even if one of the gender groups has a lower energy requirement and thus consumes less food. Because energy requirements are similar for males and females in the kindergarten through grade 5 group, this method yields a value that is similar to the values obtained using the other methods for this age-grade group. Example to Illustrate the Nutrient Density Method Continuing with the vitamin C example above, the nutrient density of the requirements is calculated as the Target Median Intake divided by the

OCR for page 69
0 SCHOOL MEALS Estimated Energy Requirement (Table 2-4), as shown in equations i and ii below: i. 120 mg vitamin C divided by 2,686 calories = 0.0448 mg/calorie for males and ii. 104 mg divided by 2,044 calories = 0.0511 mg/calorie for females. Because the nutrient density for vitamin C is higher for the females, the Target Median Intake based on nutrient density would be shown by equa- tion iii: iii. 0.0511 (the nutrient density) times 2,365 calories/day (the average Estimated Energy Requirement for males and females) = 121 mg per day. As shown in equation iii above, the nutrient density is multiplied by the average Estimated Energy Requirement for males and females because the calories provided by the school meals reflect the average calorie needs of both genders. However, the nutrient density of the foods consumed should be high enough to cover the needs of the females in the likely event that their calorie intake is below this average. If the committee had assumed a sedentary level of activity rather than a light-active level for the older females, their lower Estimated Energy Requirements would have led to nutrient density Target Median Intakes that would be unrealistically high.3 For youth ages 14–18 years, the vitamin C nutrient density Target Median Intake is similar to the values obtained using methods to calculate both the average and the highest Target Median Intake (see Appendix Table J-2), but that is not the case for a number of other nutrients (e.g., potassium). Selection of the Nutrient Density Method The committee chose the nutrient density method of setting the School Meal-TMIs. The committee notes that the nutrient density method aligns well with the emphasis placed on nutrient density by the Dietary Guidelines, where the focus is on selecting foods that provide substantial amounts of vitamins and minerals but relatively few calories. Using the simple nutrient 3 Because calories enter the equations used for the nutrient density method, it may be helpful to recognize how the committee’s early decision about calories would affect the results. In par- ticular, what difference does it make when calorie needs for female adolescents are estimated using a low-active rather than a sedentary level of activity? In that case, the divisor in formula i would be smaller, meaning that the resulting nutrient density would be higher. Although it would be multiplied by a slightly lower average number of calories, the result would be a higher value for the School Meal-TMI.

OCR for page 69
 PROCESS FOR DEVELOPING THE NUTRIENT TARGETS density approach to set targets for each nutrient is likely to provide a basis for menus that correspond closely to the goals of the Dietary Guidelines. Although the resulting School Meal-TMIs were often somewhat higher than those obtained from the other two methods, they represent the daily targets most likely to result in a low prevalence of inadequacy (5 percent or less) among the more vulnerable gender group (typically, the females). That is, the nutrient density method is designed to achieve a 5 percent prevalence of inadequacy for females even if the females’ daily calorie intake is lower than the mean value set for the grade group. Limitations of the Target Median Intake Methods The Target Median Intake methodology makes several assumptions. An important one is that the additional amounts of nutrient in the diet will be consumed by everyone. That is, the shape of the intake distribution will not change. Although this assumption may not be correct, there is almost no evidence on which to base a different assumption. The research recommendations in Chapter 10 recognize this limitation and call for more research on this topic. Likewise, although the Target Median Intake ap- proach is designed for setting daily nutrient targets, the school meals can only alter intakes at specific meals. The impact on the rest of the day’s in- take is unknown. Moreover, the students themselves determine how much of the school meal they will consume. Thus, it is not possible to conclude that Nutrient Targets based on the selected School Meal-TMI will result in a low prevalence of nutrient inadequacy for the total day’s intake. However, the nutrient density School Meal-TMI is based on the methodology recom- mended in the DRI planning report (IOM, 2003), and the resulting nutrient targets represent a step forward in applying the DRIs to planning intakes for groups so as to reduce the prevalence of inadequate nutrient intakes. DAILY SCHOOL MEAL-TARGET MEDIAN INTAKES COMPARED WITH MYPYRAMID FOOD PATTERNS The final School Meal-TMIs are the values obtained using the nutrient density approach. For these values to be useful, they need to correspond well with daily food patterns that meet Dietary Guidelines. To address this, the committee compared the daily School Meal-TMIs with the nutrient con- tent of the corresponding MyPyramid food patterns (Table 4-4). For almost all nutrients, the School Meal-TMI value was lower than the amount of the nutrient that would be obtained by following the MyPyramid pattern. This means that MyPyramid food patterns provide a sound basis for developing standards for menu planning. For the youngest age group (ages 5–10 years), vitamin E and potassium are the only nutrients that would be provided

OCR for page 69
TABLE 4-4 Daily SM-TMIsa for Different Age Groups Compared to MyPyramid Food Intake Patterns  6–10 y, 1,800-Calorie Intake Pattern 11–13 y, 2,000-Calorie Intake Pattern 14–18 y, 2,400-Calorie Intake Pattern 1,800 MyPyramid 2,000 MyPyramid 2,400 MyPyramid calories/d Nutrient calories/d Nutrient calories/d Nutrient MyPyramid as % of MyPyramid as % of MyPyramid as % of Nutrient SM-TMI Pattern* SM-TMI SM-TMI Pattern* SM-TMI SM-TMI Pattern* SM-TMI Protein(g/d)b 47.4 87 184 100.6 91 90 101.6 105 103 601 1011 168 753 1,052 140 867 1126 130 Vitamin A (µg RAE/d) Vitamin C (mg/d) 74 130 176 93 155 167 121 163 135 9.3 8.6 92 12.5 9.5 76 17 10.7 63 Vitamin E (mg αT/d) Thiamin (mg/d) 1.16 1.9 164 1.48 2.0 135 1.74 2.4 138 Riboflavin (mg/d) 1.45 2.7 186 1.9 2.8 147 2.08 3.1 149 Niacin (mg/d) 14.7 20.8 141 18.8 22 116 22.7 27.3 120 1.24 2.3 185 1.69 2.4 142 1.97 2.9 147 Vitamin B6 (mg/d) 425 668 157 528 695 132 640 822 128 Folate (µg DFE/d) 3.7 8.0 216 4.2 8.3 198 5.1 9.2 180 Vitamin B12 (µg /d) Iron (mg/d) 10.5 17 162 16.4c 17.5 107 18.4c 21.5 117 Magnesium (mg/d) 226 363 161 306 390 127 459 440 96 Zinc (mg/d) 9.1 13.7 151 11.6 14.3 123 13.5 16.7 124 Calcium (mg/d) 1,037 1,302 126 1,375 1,316 96 1,504 1,388 92 Phosphorus (mg/d) 1,127 1,691 150 1,682 1,740 103 1,787 1,961 110 Potassium (mg/d) 4,229 3,784 89 4,760 4,044 85 5,438 4,416 81 Linoleic Acid (g) 10.4 15.9 153 11.4 18 155 14.1 20.9 148 0.97 1.6 165 1.14 2 149 1.41 2 143 α-Linolenic Acid (g) Fiber (g/d) 26.5 29 109 29.4 31 105 33.5 37 110 NOTES: αT = α-tocopherol; d = day; DFE = dietary folate equivalent; g = gram; kg = kilogram; mg = milligram; RAE = retinol activity equivalents; SM-TMI = School Meal-Target Median Intake; µg = micrograms; y = years. aSodium is excluded from this table because the SM-TMI approach was not used to set the target for this nutrient. bAssumes body weights of 29.3 kg for children ages 6–10 years, 51.1 kg for children ages 11–13 years, and 67.0 kg for children ages 14–18 years (average body weights from Tables F-1 and F-2 in Appendix F). cIron values were based on results of calculations that used the probability method. Details appear in Appendix I. SOURCE: *Britten et al., 2006. Reprinted from Journal of Nutrition Education and Behaior, 38/6 Supp., P. Britten, K. Marcoe, S. Yamini, and C. Davis, Development of Food Intake Patterns for the MyPyramid Food Guidance System, pp. S78–S92, Copyright (2006), with permission from Elsevier.

OCR for page 69
 PROCESS FOR DEVELOPING THE NUTRIENT TARGETS by the MyPyramid pattern in amounts below the School Meal-TMI. For the middle school group (ages 11–13 years), these same two nutrients, as well as calcium and protein intakes, would be somewhat below the School Meal-TMI. For the high school group (ages 14–18 years), the amount of vitamin E provided by the MyPyramid pattern would be only 63 percent of the School Meal-TMI; potassium would be about 80 percent of the target; calcium and magnesium would be slightly below the targets. CONVERTING DAILY SCHOOL MEAL-TARGET MEDIAN INTAKES TO BREAKFAST AND LUNCH NUTRIENT TARGETS School Meal-TMIs are for daily intake, but school meals provide only a portion of the day’s intake. As described earlier in this chapter, the com- mittee set a preliminary range of calories for school breakfast (19 to 24 percent of the day’s total) and for school lunch (30 to 34 percent of the day’s total). When developing recommendations for the Nutrient Targets for school meals, the committee multiplied the School Meal-TMIs, the maximum for cholesterol, and the sodium ULs by the midpoint of those percentages to obtain preliminary nutrient targets. That is, the targets for breakfast represent 21.5 percent of the School Meal-TMIs, and the targets for lunch represent 32 percent. Preliminary nutrient targets for school meals appear in Table 4-5. The committee recognizes that school food authorities have no way to ensure that students will achieve the target nutrient intake distribution for the day or even the Nutrient Targets for school meals. The target nutrient intake distribution would be achieved only if students’ intake from school meals were accompanied by similar changes in the nutrient intakes from foods consumed outside the school meal setting. That is, the recommended amounts of nutrients from the school meals would need to be consumed, and comparable intakes would have to be sustained across the full day’s intake in order to meet the School Meal-TMI and achieve a 5 percent preva- lence of inadequacy. Nonetheless, it is desirable to set Nutrient Targets for school meals to provide a scientific basis for standards for menu planning and also to serve as a model for the meals and snacks served outside the school meal setting. CONSIDERATION OF THE TOLERABLE UPPER INTAKE LEVEL IN THE SETTING OF NUTRIENT TARGETS The committee examined the possibility that, for some nutrients, the prevalence of intakes above the UL would be undesirably high if the School Meal-TMIs were achieved for the full day’s intake. Data from SNDA-III for children ages 6–18 years were used for this purpose. An “adjusted” intake

OCR for page 69
 SCHOOL MEALS TABLE 4-5 Preliminary Nutrient Targets for Selected Nutrients, by Meal and Age Group Breakfasta Lunchb Nutrient (unit) 5–10 y 11–13 y 14–18 y 5–10 y 11–13 y 14–18 y Calories (kcal) 350–450 400–500 450–600 550–650 600–700 700–800 < 65 < 65 < 65 < 96 < 96 < 96 Cholesterol (mg)* Total fat (% of kcal)* 25–35 25–35 25–35 25–35 25–35 25–35 < 10 < 10 < 10 < 10 < 10 < 10 Saturated fat (% of kcal)* NAc NAc NAc NAc NAc NAc trans fat (g/d) Linoleic acid (g/d) 2.2 2.5 3.0 3.3 3.6 4.5 α-Linolenic acid (g/d) 0.21 0.25 0.30 0.31 0.36 0.45 Protein (g/d) 10.2 21.6 21.8 15.2 32.2 32.5 Vitamin A (µg RAE/d) 129 162 186 192 241 277 Vitamin C (mg/d) 16 20 26 24 30 39 Vitamin E (mg αT/d) 2.0 2.7 3.7 3.0 4.0 5.4 Thiamin (mg/d) 0.2 0.3 0.4 0.4 0.5 0.6 Riboflavin (mg/d) 0.31 0.41 0.45 0.46 0.61 0.67 Niacin (mg/d) 3.2 4.0 4.9 4.7 6.0 7.3 Vitamin B6 (mg/d) 0.3 0.4 0.4 0.4 0.5 0.6 Folate (µg DFE/d) 91 114 138 136 169 205 Vitamin B12 (µg/d) 0.8 0.9 1.1 1.2 1.3 1.6 Iron (mg/d) 2.3 3.5 4.0 3.4 5.2 5.9 Magnesium (mg/d) 49 66 99 72 98 147 Zinc (mg/d) 2.0 2.5 2.9 2.9 3.7 4.3 Calcium (mg/d) 223 296 323 332 440 481 Phosphorus (mg/d) 242 362 384 361 538 572 Potassium (mg/d) 909 1,023 1,169 1,353 1,523 1,740 ≤ 434 ≤ 473 ≤ 495 ≤ 636 ≤ 704 ≤ 736 Sodium (mg/d)d Fiber (g/d) 5.7 6.3 7.2 8.5 9.4 10.7 NOTES: αT = α-tocopherol; d = day; DFE = dietary folate equivalent; g = gram; kg = kilogram; mg = milligram; RAE = retinol activity equivalent; µg = microgram; y = years. aTargets based on 21.5 percent of the daily School Meal-Target Median Intake for the age- grade group. bTargets based on 32 percent of the daily School Meal-Target Median Intake for the age- grade group. cZero grams of trans fat per serving as listed on the nutrition label or in manufacturer’s specifications, for any food included on the school menu. dTargets for sodium, which are based on the Tolerable Upper Intake Level, are for the year 2020. SOURCE: *HHS/USDA, 2005. at the 95th percentile was calculated assuming that the median intake of a nutrient changed to be equal to the School Meal-TMI and that the whole distribution (including the 95th percentile) would change by the same amount. Calculations were performed separately for males and females

OCR for page 69
 PROCESS FOR DEVELOPING THE NUTRIENT TARGETS within each grade group. This same method was used for nutrients with an EAR and for nutrients with an AI. For each of the three age-grade groups covered by the SNDA-III data (6–10, 11–13, and 14–18 years), the adjusted intake at the 95th percentile was compared to the UL. (Magnesium was excluded because the UL is only for pharmacological agents. The UL does not apply to magnesium in foods [IOM, 1997].) For children ages 6–10 years, the UL for the younger children (ages 6–8 years)—that is, the most conservative value—was used. For several nutrients, the ULs are considerably lower for children ages 8 years or younger than for the older children. The results are shown in Table 4-6. For each grade group, there were some nutrients with the adjusted 95th percentile of intakes above the UL, meaning that at least 5 percent of the children would have intakes above the UL if the median intake was at the School Meal-TMI, as follows • 6–10-year-olds: vitamin A, niacin, folate, and zinc for males and females • 11–13-year-olds: niacin and folate for males and females • 14–18-year-olds: niacin and folate for males and females; males’ 95th percentile of intake would be slightly above the calcium UL It is worth noting that in all these cases except calcium, current intakes at the 95th percentile also exceed the UL. As would be expected, at the 95th percentile of intake, all values for sodium are well above the UL. For most nutrients, intakes above the UL are not likely to be a concern. This is largely because the ULs only apply to certain forms or sources of nutrients, whereas the intake estimates are for the total diet. The degree of concern about intakes above the UL is summarized for six nutrients below. Probably a concern • Sodium 50 percent of schoolchildren would have intakes above the sodium UL. See Chapter 3 regarding effects of sodium on blood pressure. Nonetheless, setting the School Meal-TMI to reduce intakes to less than the UL is a reasonable goal. • Probably not a concern Vitamin A The UL applies only to preformed vitamin A (retinol). Dairy products and eggs are the most common sources of preformed vita- min A in children’s diets. It would take approximately 1.5 quarts of milk to exceed the UL for children ages 5–8 years, and much more than that for the older children. Calcium Although the adjusted intakes for the older males might result in 5 percent with intakes above the calcium UL, the committee agreed

OCR for page 69
TABLE 4-6 Tolerable Upper Intake Level (UL) for Nutrients with a UL,* Reported Intakes,**, a Predicted Adjusted  Intakes at the 95th Percentile, and Adjusted Intakes as a Percentage of the UL, by Age-Gender Group Intake Adj. Intake Adj. Intake Intake Adj. Intake Adj. Intake Nutrient (unit) ULb at 95th at 95th as % of UL ULb at 95th at 95th as % of UL Males Females 6–10 years 6–10 years 900 1,025 1,010 112 900 1,022 1,024 114 Vitamin A (µg RAE/d) Vitamin C (mg/d) 650 169 162 25 650 161 146 22 300 7.5 11.1 4 300 8.9 13.3 4 Vitamin E (mg αT/d) Niacin (mg/d) 15.0 29.3 23.9 159 15.0 30.9 26.1 174 40 2.4 2.0 5 40 2.6 2.3 6 Vitamin B6 (mg/d) 400 951 834 209 400 988 887 222 Folate (µg DFE/d) Iron (mg/d) 40 24 20 50 40 24 21 52 Zinc (mg/d) 12 17.4 15.5 129 12 16.7 15.9 133 Calcium (mg/d) 2,500 1,927 1,863 75 2,500 1,579 1,571 63 Phosphorus (mg/d) 3,000 2,027 1,753 58 3,000 1,708 1,531 51 Sodium (mg/d) 1,900 4,661 3,321 175 1,900 4,405 3,249 171 11–13 years 11–13 years 1,700 1,225 1,314 77 1,700 1,006 1,254 74 Vitamin A (µg RAE/d) Vitamin C (mg/d) 1,200 181 185 15 1,200 164 187 16 600 10.4 16.8 3 600 11.5 19 3 Vitamin E (mg αT/d) Niacin (mg/d) 20.0 35.0 31.9 160 20.0 31.6 31.4 157 60 3.5 3.4 6 60 2.7 2.9 5 Vitamin B6 (mg/d) 600 1,081 986 164 600 994 1,062 177 Folate (µg DFE/d) Iron (mg/d) 40 25 25.2 63 40 25 27.7 69 Zinc (mg/d) 23 18.2 17.8 77 23 16.3 18.4 80 Calcium (mg/d) 2,500 1,724 1,926 77 2,500 1,612 2,128 85 Phosphorus (mg/d) 4,000 1,984 2,236 56 4,000 1,925 2,489 62 Sodium (mg/d) 2,200 5,180 3,599 164 2,200 5,028 3,970 180

OCR for page 69
14–18 years 14–18 years 2,800 1,238 1,509 54 2,800 914 1,380 49 Vitamin A (µg RAE/d) Vitamin C (mg/d) 1,800 207 244 14 1,800 186 246 14 800 12 22.6 3 800 10.6 23.1 3 Vitamin E (mg αT/d) Niacin (mg/d) 30.0 40.1 36.7 122 30.0 30.2 35.7 119 80 3.2 3.1 4 80 2.6 3.3 4 Vitamin B6 (mg/d) 800 1,216 1,238 155 800 948 1,175 147 Folate (µg DFE/d) Iron (mg/d) 45 31.3 31.8 71 45 23.6 30.2 67 Zinc (mg/d) 34 24.1 24 7 34 16 21 62 Calcium (mg/d) 2,500 2,134 2,519 101 2,500 1,482 2,251 90 Phosphorus (mg/d) 4,000 2,530 2,775 69 4,000 1,809 2,589 65 Sodium (mg/d) 2,300 6,280 4,072 177 2,300 4,558 3,690 160 NOTES: αT = α-tocopherol; Adj. = Adjusted; d = day; DFE = dietary folate equivalent; g = gram; kg = kilogram; mg = milligram; RAE = retinol activity equivalent; µg = microgram. Bold font indicates intake values above the UL. aIntakes exclude contributions from nutrient supplements. Magnesium is excluded from the table because the UL for magnesium applies only to pharmacological agents. bThe UL listed is the lowest UL value within the age-grade group. SOURCES: *IOM, 2006; **USDA/FNS, 2007a. 

OCR for page 69
 SCHOOL MEALS that these very high calcium intakes were not likely to be a result of intakes from school meals and thus would not be a concern when setting the School Meal-TMI for calcium for the oldest grade group. Zinc. The UL for children ages 4–8 years is very low, and it may be more applicable to children ages 4–5 years than to children ages 6–8 years (Zlotkin, 2006). • Unknown concern Niacin The UL applies only to niacin from supplements and from foods that are fortified with niacin. The committee notes that it is not known if highly fortified foods (such as those that provide 100 percent of the Daily Value for niacin [20 mg] in a single serving) pose a risk for young children. Although this amount exceeds the UL for niacin for the youngest children and equals the UL for children ages 11–13 years, many children’s intakes are already at this level. The ULs for children were based on limited evidence that some adults experienced flushing as a short-term response to the ingestion of high levels of nicotinic acid (a form of niacin that does not occur naturally in foods and that differs from niacinamide, which is the substance used to fortify foods) (IOM, 1998). Folate Current intakes at the 95th percentile exceed the folate UL for all grade and gender groups. The adjusted intake distributions would result in intakes that exceed the UL for three of the age-gender groups, espe- cially the youngest grade groups; but intakes for the other three age-gender groups would probably be below the UL (Table 4-6). The UL applies only to synthetic forms of folic acid (e.g., the folic acid added to fortify enriched grains, not the folate that occurs naturally in foods). The 95th percentile intakes, however, would be almost twice the UL for the youngest children. It is not known if highly fortified foods (such as those that provide 100 percent of the Daily Value for folate [400 µg] in a single serving; an amount that equals the UL for the younger children) pose a risk for young children. As is the case with niacin, the ULs for folate for children were based on limited evidence from studies with adults; but, in this case, they were long- term studies on folic acid ingestion (IOM, 1998). CHAPTER SUMMARY The Nutrient Targets are intended to serve as a guide for setting stan- dards for menu planning, not for direct use in menu planning. This made it reasonable for the committee to develop targets for 24 nutrients. The committee used a data-based method to set preliminary minimum and maximum target calorie levels for school breakfast and lunch for the three age-grade groups, rounding the values for ease of implementation. Setting both a minimum and a maximum level has the advantages of providing adequate intake without encouraging the overconsumption of calories,

OCR for page 69
 PROCESS FOR DEVELOPING THE NUTRIENT TARGETS while still allowing some flexibility to school food service operators. The committee based its preliminary targets for saturated fat, cholesterol, and total fat on Dietary Guidelines for Americans and used a labeling approach to address trans fat. In setting the preliminary nutrient targets for protein, vitamins, and minerals, the committee used methods recommended by the Institute of Medicine for using the DRIs in planning for groups. The use of the nutri- ent density method results in nutrient targets that recognize that females have nutrient needs that ordinarily are higher than those of males relative to their calorie needs. Thus, the resulting Nutrient Targets should provide a sound basis for planning menus that are appropriate for both males and females in the age-grade group. Although the resulting intakes at the 95th percentile may exceed the UL for some nutrients, especially for the young- est children, it is unlikely that the amounts provided by the school meals pose a health risk.

OCR for page 69