| | The Third School Nutrition Dietary Assessment Study: Summary and ImplicationsThe US Department of Agriculture (USDA) National School Lunch Program and School Breakfast Program provide subsidized meals to children in school, and provide these meals free or at a reduced price to children from low-income families. The National School Lunch Program was created in 1946 and currently operates in nearly all public and many private schools. The School Breakfast Program, which became a permanent federal program in 1975, is offered in a somewhat smaller number of schools and serves fewer children per school. On an average school day in 2007, 30 million children ate a school lunch and 10 million ate a school breakfast (1, 2). A majority of these meals—59% and 80%, respectively—were served to low-income children. The Food and Nutrition Service of USDA sponsored the third School Nutrition Dietary Assessment Study (SNDA-III) to provide up-to-date information on the school meal programs, including the food and nutrient content of meals offered to and selected by children, the contributions of school meals to children's diets, and the availability and consumption of competitive foods (foods and beverages available for purchase outside of school meals from à la carte programs, vending machines, school stores, and fundraising activities). SNDA-III data were collected in the second half of school year 2004-2005. The study builds on two previous studies: SNDA-I in school year 1991-1992 and SNDA-II in school year 1998-1999, which have been influential in shaping school nutrition policy (3, 4). The School Meals Initiative for Healthy Children (SMI), implemented in 1995 after SNDA-I results showed that school lunches were high in fat and saturated fat, required schools to offer meals that provide ≤30% of total energy from fat and <10% from saturated fat, while providing adequate levels of target nutrients (defined as one quarter of daily needs at breakfast and one-third at lunch, on average). SNDA-II showed that schools had reduced fat and saturated fat levels in school meals by school year 1998-1999, while maintaining levels of target nutrients. However, school meals were still high in fat and saturated fat, relative to the standards established under SMI. SNDA-III provides information on how the school meal programs were operating roughly 8 years after SMI implementation. SNDA-III was based on a nationally representative sample of 130 public school food authorities (districts that offer federally subsidized school meals), 398 schools within those districts, and 2,314 public school students in grades 1-12 in 287 of these schools (5). Data were collected from district foodservice directors and their staff, school foodservice managers, principals, students, and their parents. In addition, field interviewers who were collecting data from students and parents observed and recorded the types of competitive foods available in visited schools. Study analyses incorporate the SMI nutrient standards, the Dietary Guidelines for Americans 2005 (6), and the Dietary Reference Intakes (DRIs), which recommend nutrient intake levels needed for individuals to achieve a healthful diet and prevent disease (7). The articles in this supplement present findings from the USDA-sponsored SNDA-III analyses and extensions of those findings, as well as research from other studies using SNDA-III data to examine the school food environment, children's dietary behaviors at school and outside of school, and child overweight/obesity. This article provides a summary of key findings and of implications for future policy and research. School Meals and the School Food Environment  Nutrients Offered in School Meals School Lunch As offered, school lunches generally met SMI requirements for vitamins and minerals (88% or more of schools met each of the requirements) (8). Lunches offered in most secondary schools (61% of middle schools and 77% of high schools) provided less energy than SMI standards require; lunches offered in 40% of elementary schools provided less energy than required. Most schools offered lunches that exceeded SMI standards for energy from fat and saturated fat (81% of schools exceeded the standard for total fat, 72% for saturated fat). On average, schools offered lunches that provided 34% of energy from fat and 11% from saturated fat, compared with SMI standards of ≤30% and <10%, respectively. The percentage of schools that offered lunches meeting all of the SMI standards (ie, for energy, fat, saturated fat, protein, vitamin A, vitamin C, calcium, and iron) was only 6%; most schools failed to meet the energy and/or fat standards. Essentially no schools offered lunches that, on average, contained less than one third of the maximum 2005 Dietary Guidelines recommendation for sodium. Very few offered lunches that met the recommendation for fiber of 14 g/1,000 kcal (<8%), although the recommendation is based on total fiber, whereas the USDA nutrient database measures dietary fiber only, so that estimate is a lower bound. School Breakfast Among schools that offered breakfast (about 85% of public schools that offer the school lunch program), breakfasts generally met SMI vitamin and mineral requirements (8). At least 90% of schools met each requirement. On average, most schools offered breakfasts that met regulatory standards for fat and saturated fat (88% met the fat standard, 75% met the saturated fat standard). At the same time, 73% of schools offered breakfasts that provided less energy than required by SMI regulations. Less than half of School Breakfast Program schools (47%) offered breakfasts containing less than one fourth of the upper limit for sodium, and almost none of the schools offered breakfasts that met one fourth of the 2005 Dietary Guidelines recommendation for fiber. Foods Offered in School Meals School Lunch Nearly all daily menus (96%) offered one or more vegetable options, in addition to any vegetables that were part of entrées (9). Almost half of high school lunch menus included french fries or similar potato products, compared to one fifth of elementary school menus. Fruit appeared on 94% of menus, with fresh fruit offered less often than canned fruit (on 50% of menus vs 62%). Menus rarely offered whole-grain breads or legumes as separate items (on 5% and 10% of menus, respectively). Menus almost always offered flavored milk (99%), which was usually 1% low-fat or skim. Desserts and other snack-type items were offered in almost half (47%) of high school lunch menus, 41% of middle school menus, and 37% of elementary school menus. Combination entrées were major sources of fat, saturated fat, sodium, and energy (8). Examples of combination entrées include sandwiches, such as hamburgers, peanut butter, or chicken patty; mixed dishes such as pizza, lasagna, or tacos; and entrée salad bars. Whole milk and 2% milk were also major sources of saturated fat. School Breakfast Breakfast menus provided fruit mostly as juice (88% of menus) rather than fresh fruit or canned (26% and 14% of menus, respectively); they rarely included vegetables (9). The most common grain item included on breakfast menus was ready-to-eat cereal (78% of menus), most often a presweetened variety (72% of menus vs 27% for unsweetened). Changes since SNDA-II SNDA-II assessed the food and nutrient content of school meals in school year 1998-1999, soon after implementation of SMI. Comparisons between SNDA-II and SNDA-III indicate that little change occurred in the food and nutrient content of school meals offered to children between 1998-1999 and 2004-2005 (10). Changes in school lunches were as follows: •Significantly more schools met the standard for saturated fat in school year 2004-2005 than in school year 1998-1999—compliance increased from 15% to 34% for elementary schools, and from 13% to 24% for secondary schools. The percentage of schools meeting the total fat standard did not change significantly over time (approximately 23% for elementary schools and 13% for secondary schools in both periods). •Fewer school lunch menus included whole milk during school year 2004-2005 than in school year 1998-1999 (availability decreased from 50% of daily menus to 31%), and more menus offered flavored skim milk (availability increased from 17% of daily menus to 31%). •Fresh fruit was offered more often in school year 2004-2005 menus, but availability of fresh fruit was still relatively limited (50% of lunch menus in 2004-2005 vs 41% in 1998-1999). Changes in school breakfasts were similar to those in lunches: •More schools met the standards for fat and saturated fat in 2004-2005 than in 1998-1999. In both years, more than half the schools met the standards. •As with lunch, fewer school breakfast menus offered whole milk in school year 2004-2005 (fell from 49% to 29%); more offered flavored skim milk (increased from 11% to 29%). •More breakfast menus offered fresh fruit (increased from 17% in school year 1998-1999 to 26% in school year 2004-2005). Competitive Food Sources Available At least one source of competitive foods—foods and beverages that compete with USDA-sponsored meals—was available in 73% of elementary schools, 97% of middle schools, and all high schools—and this does not count offerings in class or through fundraisers (11). School foodservices frequently offered à la carte items at all levels, particularly at lunch, when two thirds of elementary schools and more than 90% of secondary schools made them available. (À la carte here does not include schools where nothing but milk was available à la carte.) Vending machines were almost universal in secondary schools, but were available to students in less than a quarter of elementary schools. School stores and snack bars were less common, and found almost entirely in middle and high schools; for example, 20% of high schools had stores and 14% had snack bars. Schoolchildren's Dietary Intakes Assessment of schoolchildren's energy and nutrient intakes was a main focus of SNDA-III. SNDA-III collected one day of 24-hour dietary recall data from the full sample of children in grades 1 through 12. A second day of dietary recall data were collected for a subsample of these children to permit estimation of the distributions of the usual daily intakes of key nutrients and the proportions of children whose usual nutrient intakes were inadequate or excessive, using a statistical approach recommended by the Institute of Medicine (12). In comparing the usual intakes of school meal program participants to those of a matched comparison group of nonparticipants, the analysis controlled for a wide range of school and student characteristics. In contrast, SNDA-III did not attempt to assess the usual intakes of specific foods or (unless otherwise noted) to control for other characteristics when comparing the food intakes of participants and nonparticipants. Any differences resulting from such comparisons may not reflect effects of the school meal programs. Usual Nutrient Intakes Overall and by School Level The SNDA-III data showed similar distributions of usual intakes among schoolchildren as were found using National Health and Nutrition Examination Survey data (12). Most children had adequate usual intakes of most vitamins and minerals (13). Energy levels were more difficult to assess, given indications of overreporting among young children and underreporting among older children, as well as the lack of a direct measure of physical activity. In general, the prevalence of inadequate intakes (intakes less than the Estimated Average Requirement) of most vitamins and minerals was low (<3%) for school-aged children. However, the prevalence of inadequacy was 15% or higher for a few vitamins and minerals, including vitamin A, vitamin C, vitamin E, magnesium, phosphorous, and zinc. For almost all vitamins and minerals examined, the prevalence of inadequate intakes was lowest among elementary school children and highest among high school–aged children, particularly female high school students. Inadequacy could not be directly assessed for nutrients without an Estimated Average Requirement, but mean intakes were instead compared to the recommended Adequate Intake (AI) level. (Because the AI is set well above average requirements, mean intakes less than the AI are not evidence of inadequate intakes, although mean intakes above the AI imply that risk of inadequacy is low.) Mean calcium intakes of middle and high school children were less than the AI, and mean potassium intakes were less than the AI for elementary, middle, and high school children. Mean fiber intakes were about half of the AI for total fiber; as noted above, only intakes of dietary fiber were assessed. Excessive intakes of some dietary components were found, particularly among high school–aged children. Usual daily intakes of sodium were excessive (exceeding the Tolerable Upper Intake Level) for more than 90% of school children. About one in five children overall had intakes of total fat above the Acceptable Macronutrient Distribution Range, but 31% of high school children did. Nearly 80% of children had saturated fat intakes that exceeded 10% of energy, with little variation by age. Cholesterol intakes above the recommended maximum were much less common (about 13%). Role of the School Meal Programs in Children's Dietary Intakes Foods Consumed at Lunch Not controlling for other factors, school lunch participants were almost four times more likely than nonparticipants to drink milk at lunch (75% vs. 19%). Participants were more likely to consume fruit—especially canned fruit—and all types of vegetables—including french fries—for lunch (9). Nonparticipants were more likely than participants to consume desserts, snack foods, and beverages other than milk or 100% juice (eg, juice drinks and carbonated soft drinks) at lunch. Foods Consumed at Breakfast School breakfast participants were significantly more likely to consume milk and fruit at breakfast than nonparticipants; among both groups, most fruit was consumed as 100% fruit juice (9). (These comparisons were also not adjusted for other factors that may differ between the two groups.) Usual Daily Intakes of School Lunch Participants and Nonparticipants Comparisons of means or distributions of usual daily intakes of participants and nonparticipants were the result of a two-step process. First, statistical matching techniques were used to select a matched sample of nonparticipants similar to school meal participants on a wide range of characteristics. Second, the usual intake distributions of key nutrients (and standard errors) were estimated for each group, using procedures that relied on the second day of 24-hour dietary recall data for part of the sample to adjust for day-to-day variation in nutrient intakes (13). School lunch participants had daily energy intakes that were greater than those of matched nonparticipants by about 130 kcal, on average, controlling for other factors. These differences were statistically significant overall and for elementary and high school students. Participants were less likely than matched nonparticipants to have inadequate usual intakes of vitamin A, vitamin B-6, vitamin C, folate, thiamin, magnesium, and phosphorous at the middle and/or high school levels, although there were no significant differences between participants and matched nonparticipants at the elementary school level in adequacy of these nutrients. Middle and high school participants had significantly higher mean intakes of calcium and potassium relative to the AI than nonparticipants, with both groups below the AI for both nutrients. High school participants were more likely to have excessive sodium intakes than nonparticipants (96% vs 78%). Participants and nonparticipants consumed similar percentages of energy from fat and saturated fat. School lunch participants had significantly higher mean intakes of fiber than nonparticipants at all three grade levels. Usual Daily Intakes of School Breakfast Participants and Nonparticipants There were few statistically significant differences in usual nutrient intakes between school breakfast participants and matched nonparticipants, controlling for many background characteristics (13). Breakfast participants were less likely to have inadequate intakes of vitamin A and phosphorous, both of which may be linked to the higher milk intakes noted above. School breakfast participants were also more likely than matched nonparticipants to consume excessive levels of sodium. Consumption at School vs Home and Other Locations Because previous research has linked consumption of low-nutrient, energy-dense foods and beverages to risk of obesity, SNDA-III data were used to study schoolchildren's consumption of low-nutrient, energy-dense foods and sugar-sweetened beverages and the energy obtained from these foods over a typical school day (14). These descriptive comparisons were based on 1 day of 24-hour dietary recall data, and controlled only for school level. Among both elementary and secondary schoolchildren, school lunch participants were significantly less likely to consume sugar-sweetened beverages at school than nonparticipants (9% vs 38% for elementary; 29% vs 38% for secondary) (14). However, secondary school participants were more likely to consume low-nutrient, energy-dense foods—particularly french fries and baked goods—at school than nonparticipants. (Most likely, these foods were obtained as part of reimbursable meals or à la carte.) Nonparticipants were more likely to consume energy-dense chips and higher-fat salty snacks than participants, but still consumed significantly less energy on average than participants from low-nutrient, energy-dense solid foods at school. Participants and nonparticipants did not differ significantly in their consumption of low-nutrient, energy-dense foods or beverages at home. Interestingly, secondary school lunch participants, who were less likely to consume sugar-sweetened beverages at school, were also less likely to consume sugar-sweetened beverages at other locations than school or home. Although secondary school participants were more likely to consume low-nutrient, energy-dense solid foods at school, they were less likely to consume low-nutrient, energy-dense foods at locations other than home or school than nonparticipants. In addition, children who consumed sugar-sweetened beverages at school were no more likely to consume sugar-sweetened beverages at home or other locations than children who had not drunk sugar-sweetened beverages at school. These results need to be interpreted with caution, particularly because other characteristics of the children were not controlled for in the analysis. Nonetheless, these cross-sectional descriptive data suggest that restrictions on the availability of low-nutrient, energy-dense foods or beverages in school are not associated with children consuming more sugar-sweetened beverages or low-nutrient, energy-dense foods at home. Role of the School Food Environment, Policies, and Practices in Food Choices and Consumption of “Empty Calories” SNDA-III data showed that about 40% of schoolchildren consumed one or more foods from competitive sources (11). Children who ate a school lunch were significantly less likely than children who did not eat a school lunch to consume competitive foods (36% vs 45%); however, low-nutrient, energy-dense foods were the leading competitive food choices for both groups. On average, competitive food consumers who ate school lunches obtained 159 kcal from low-nutrient, energy-dense competitive foods, compared with 201 kcal for competitive food consumers who did not eat school lunches. The importance of low-nutrient, energy-dense foods and beverages in children's diets has brought to national attention policies related to these foods in school meals and competitive sources. Three dietary behaviors that previous research has associated with children's weight—consumption of sugar-sweetened beverages, consumption of solid low-nutrient, energy-dense foods, and consumption of fresh fruits and raw vegetables—may be influenced by the school food environment and related policies and practices. School food environment and related policies and practices were examined in three domains: wellness and nutrition education policies, competitive food practices, and school lunch characteristics (15). Based on multivariate models that controlled for all three policy/practice domains, as well as child and school background characteristics, competitive foods restrictions were found to reduce sugar-sweetened beverage consumption in middle schools and high schools. Specifically, not having a school store or snack bar, and, in middle schools, eliminating pouring rights contracts and low-nutrient, energy-dense foods sold à la carte were each associated with consuming fewer calories from sugar-sweetened beverages. Competitive foods practices were not significantly related to consumption of sugar-sweetened beverages at the elementary school level, because sugar-sweetened beverages were much less available there. More healthful school lunch practices were significantly associated with reduced consumption of low-nutrient, energy-dense solid foods among children in elementary schools. In particular, offering french fries less than once a week and offering fresh fruit or raw vegetables daily were both associated with consuming less energy from low-nutrient, energy-dense foods. Competitive foods restrictions, including the absence of school stores or vending machines, were associated with increases in energy consumed from low-nutrient, energy-dense foods at the elementary level, in part because of younger children's higher levels of participation in school meals. School lunch practices were not significantly related to energy consumed from low-nutrient, energy-dense foods in middle and high schools. Elementary school children's consumption of fruits and vegetables other than fried potatoes was positively associated with schools that offered french fries less than once a week and offered fresh fruit and raw vegetables daily (15). Among middle schools, schools with no low-nutrient, energy-dense foods sold à la carte had students who consumed more vegetable servings (excluding fried potatoes), whereas in high schools, not having an open campus was associated with students' consuming more servings of vegetables (again, excluding fried potatoes). Prevalence of Overweight and Obesity Students' height and weight were directly measured using standardized procedures for SNDA-III. Each student's body mass index (BMI), defined as weight in kilograms divided by the square of height in meters, was compared to standard percentiles adjusted for age and sex. SNDA-III data indicate that about 40% of children attending public schools were overweight (BMI between 85th and 95th percentile of the age/sex-adjusted distribution) or obese (BMI >95th percentile). Although these figures are somewhat higher than in recent National Health and Nutrition Examination Surveys, differences appear to be related to differences in the sample frames for the two surveys, as, by design, SNDA-III sampling was school-based and limited to students in public schools that offer the school lunch program (16, 17). Role of the School Food Environment, Policies, and Practices Specific school food and environment policies and practices affect foods consumed at school (15). Foods consumed, particularly, low-nutrient, energy-dense foods, have been associated with higher BMI in previous studies. SNDA-III data offered a rare chance to examine the associations between school food environment and related policies and practices and schoolchildren's BMI, controlling for a wide range of child and school characteristics, including whether a child usually participates in the school lunch and school breakfast programs. Among elementary school students, those in schools that offered french fries in school lunches more than once a week and in schools that offered desserts more than once a week were more likely to be obese (18). Among middle school students, attending a school with low-nutrient, energy-dense foods in vending machines was associated with a higher BMI. Offering low-nutrient, energy-dense foods à la carte was associated with a lower BMI. School food environment and related policies and practices were not significantly associated with BMI or the likelihood of obesity for high school students. Role of the School Meal Programs Neither school lunch nor breakfast (usual participation) was associated with a higher BMI or a higher incidence of overweight or obesity among school-aged children, when controlling for a wide range of background characteristics (17). School breakfast participation was associated with a lower BMI, which may be a proxy for eating breakfast in general; the effect was smaller and not significant after controlling for whether a child ate any breakfast. Eating breakfast was significantly associated with a lower BMI, a result also found in other studies (19, 20, 21). Implications for Policy and Future Research Updating Nutrient Standards and Menu Planning Approaches Nutrient Standards SNDA-III assessed the energy and nutrient content of school meals using the 1995 SMI nutrient standards—the federal regulatory standards in place at the time of the study. Since 1995, scientific knowledge of nutrient requirements for good health has changed, as reflected in the DRIs and the 2005 Dietary Guidelines. For example, instead of no more than 30% of energy from total fat (required under SMI), the recommended range for total fat for children is 25% to 35% of energy. Three times as many schools provided lunches within this range as met the SMI requirement. At the same time, use of the DRIs to assess school-aged children's diets has identified nutrients of concern (eg, magnesium, vitamin E, potassium, fiber, and sodium) that may not be adequately captured by the nutrients included in SMI standards, which were intended to serve as proxies for adequate intakes of other nutrients. To ensure that school meals make an important and positive contribution to schoolchildren's diets, quantitative targets for energy and nutrient content need to be updated to reflect the 2005 Dietary Guidelines and the DRIs. Once new targets have been established, they can be translated into policy and guidance for school foodservice professionals. Menu Planning Systems Changes to food-based standards are also needed if schools are to offer meals consistent with the 2005 Dietary Guidelines and USDA's MyPyramid food guidance (6, 22). SNDA-III analyses of the types of foods offered by schools and consumed by schoolchildren suggest more emphasis could be placed on daily offerings of fresh fruits and whole grains, and on the frequency of offering dark green/orange and other nutrient-dense vegetables as well as legumes. To meet standards for fat and sodium, the food-based meal patterns may need to allow for more meatless entrées and/or flexibility in the minimum quantities required for meat/meat alternates. SNDA-III findings on the diets of participants confirm the importance of fluid milk and fruits and vegetables as required components in school meals. However, to reduce fat and saturated fat while maintaining levels of other nutrients, meal patterns could be updated to allow only 1% low-fat or skim milk and to limit the frequency of offering french fries and similar potato products. These types of changes in food-based requirements for school meals are likely to have the greatest potential for improving the healthfulness of school meals. Most schools during school year 2004-2005 continued to use one of the food-based approaches to menu planning—48% used Traditional Food-Based and 22% used Enhanced Food-Based Menu Planning (30% used Nutrient Standard Menu Planning, up from 24% during school year 1998-1999 [SNDA-II]) (4, 5). Implementing Strategies to Bring More Healthful Foods into School Meals Bringing school meals in line with DRIs and the 2005 Dietary Guidelines will involve more than changing nutrient standards and revising menu planning regulations. SNDA-III provides evidence that, despite the hard work of many individuals, schools, and organizations, SMI goals were not fully achieved. Strategies are needed to enhance the ability of schools to offer more healthful meals. In developing these strategies, it will be important to consider the potential effects on food and labor costs, acceptability by students, and the unique features of each school environment—for example, extent of cooking facilities and skill level of foodservice staff. Technical assistance and training will be needed at all levels—for state agencies, school food authorities, and school foodservice professionals. There is considerable room for increasing the variety and frequency of fruit and vegetable offerings and whole grains in school menus. Innovative preparation methods or improvements in the presentation of fruits and vegetables could make these items more appealing to children. In addition to promoting individual whole-grain items, such as breads and rolls, whole grains could be incorporated into combination entrées. For example, because a large proportion of children consume pizza at lunch, schools that prepare dough from scratch could begin to gradually use whole-grain flours to develop a whole-grain pizza crust. Whole-grain pasta could be used to make pasta-based entrées, perhaps mixed with regular pasta. Requirements for the whole-grain content of ready-to-eat cereals could help schools substitute cereals that that will improve children's daily intake of whole grains for current cereal offerings. Data from SNDA-III indicate that many schools relied on the convenience and lower cost of preprepared fast-food–like items, such as pizza, breaded chicken, hamburgers, and french fries. However, these products provided most of the energy, fat, saturated fat, and sodium in school lunches. To obtain more healthful products, states and school foodservice personnel would benefit from guidance on appropriate nutrient-based specifications that they can use with vendors for commodity processing and food purchasing. The school food industry can help by developing products that meet these specifications as well as new foods, such as meatless or vegetarian entrées, and testing them for acceptability by students. Strategies to reduce fat and sodium in school meals also include decreasing the frequency or eliminating whole and 2% milk from menus and substituting low-fat or fat-free salad dressings and spreads (eg, mayonnaise and sour cream) for full-fat versions. Portion control on condiments such as ketchup, mustard, and barbeque sauce may help reduce the amount of sodium available to children who consume school meals. Using State and District Wellness Policies to Limit Competitive Foods and “Junk Foods” School district wellness policies must include policies concerning all foods available on school campuses. SNDA-III data confirm previous research that has shown that competitive food sources—sources that compete with the federally subsidized school meal programs, such as vending machines, à la carte lines, or school stores—were important sources of low-nutrient, energy-dense foods and beverages. SNDA-III results also suggest that restrictions on competitive foods were sometimes associated with consuming less energy from low-nutrient, energy-dense foods and/or with lower BMI values. Although the cross-sectional nature of SNDA-III implies that other factors may be at work, many analyses controlled for a very wide range of other factors (15, 18). Thus, the research provides preliminary evidence supporting policies that do not permit sugar-sweetened beverages to be sold from competitive foods sources, and that limit the availability of low-nutrient, energy-dense foods à la carte and in vending machines. School lunch participants actually consumed more energy from low-nutrient, energy dense foods in school than nonparticipants, and these foods appear to have largely come from the school meals, which regularly included low-nutrient, energy-dense foods such as french fries and cakes and cookies (15). (Low-nutrient, energy-dense foods consumed may also have been à la carte selections sold in the same line as school meals.) Thus, limitations on low-nutrient, energy-dense foods within school meals, particularly french fries and desserts, also seem appropriate. Promoting Healthful Food Choices among Children The ultimate success of efforts to improve the healthfulness of school meals depends on the nation's children choosing to eat the more healthful foods. SNDA-III data demonstrate that children do not always make healthy choices. For example, the percentage of children who consumed milk at lunch decreased from elementary school to high school, even among children who consumed a school lunch (9). Moreover, fewer than half of all children who ate lunch (from school or anywhere else) consumed a vegetable or fruit. It is also important that children not substitute low-nutrient, energy-dense foods and beverages from home or other sources for those eliminated from the school environment. Data from SNDA-III indicate that children sometimes counteract a healthful food environment. For example, among elementary school students, who had limited access to competitive foods and beverages, most sugar-sweetened beverages consumed at school were brought in from home. On the other hand, secondary school children who did not consume a sugar-sweetened beverage at school consumed less energy from these beverages over the full day (ie, they did not compensate by consuming more sugar-sweetened beverages in another location). These issues point to the critical need for wellness and health promotion policies that include nutrition education at school as well as outreach to parents and the community at large to promote healthful food choices. Areas for Future Research The SNDA-III data provide a baseline for assessing changes in school meal programs and school environment policies since 2005 and in the future. For example, an Institute of Medicine expert panel is currently considering revisions to the meal patterns and nutrition standards for USDA's school meal programs. In addition, Congress will take up reauthorization of the school nutrition programs again in 2009. Future studies will look back to SNDA-III to examine how school meals and school food environments have changed after implementation of subsequent federal policy initiatives. SNDA-III data could also be used to estimate the potential effects of proposed changes in policy on schoolchildren's diets, with the caveat that the estimates are based on cross-sectional data, rather than longitudinal studies with rigorous comparison groups. Further research could support improvements in the data and methods for analysis of children's dietary intakes. Continued efforts to identify and refine data collection strategies to minimize overreporting and underreporting of energy and nutrient intakes among school-aged children are essential. To track competitive foods more precisely, it would be useful to develop methods to identify, among the foods children eat at school, those that were offered in school meals but purchased on an à la carte basis. Studies in schools with detailed point-of-sale data, for example, might illuminate how often this occurs. SNDA-III results show that policy changes are not always fully implemented. Qualitative research is needed to learn more about obstacles to improving school food environments and school meals—such as lack of kitchen facilities, food and/or labor costs, and lack of acceptance among children. An examination of practices among schools offering meals that meet current standards would help identify “promising practices” for future study. To the extent possible, future research should track children's diets and BMI/obesity before and after implementation of specific policies to obtain more definitive information about the influence of policy changes on obesity. A randomized controlled trial would be the ideal approach. Although such a study would be challenging and costly to implement using a large national sample, smaller, local studies that are well designed could also make significant contributions to the evidence base. STATEMENT OF CONFLICT OF INTEREST: The authors have no conflict of interest to report with the sponsor of this supplement article or products discussed in this article. The third School Nutrition Dietary Assessment Study was funded by the Food and Nutrition Service, US Department of Agriculture (contract nos. 43-3198-3-3716, 43-3198-4-0060, AG-3198-D-05-0071). The Robert Wood Johnson Foundation provided support for the preparation of the manuscript (grant no. 60542). The opinions expressed are those of the authors and do not necessarily represent the views or recommendations of the funding organizations or Mathematica Policy Research, Inc. The authors thank Philip Gleason, PhD, for his careful review of this article and Elizabeth Condon, MS, RD, for her assistance in its preparation. The authors also thank Mary Story, PhD, RD, for her thoughtful comments and suggestions; the authors of the articles summarized herein (Charlotte Cabili, MS, MPH; Rhoda Cohen, MS; Elizabeth Condon, MS, RD; Allison Hedley Dodd, PhD; Philip Gleason, PhD; John Hall, MS; Renée Nogales, MPubAff; Ander Wilson, and Eric Zeidman); and the school districts, schools, students, and parents who provided information for the study. References 1. 1National School Lunch Program: Participation and lunches served. US Department of Agriculture, Food and Nutrition Service Web site. http://www.fns.usda.gov/pd/slsummar.htm. 2. 2School Breakfast Program: Participation and breakfasts served. US Department of Agriculture, Food and Nutrition Service Web site. http://www.fns.usda.gov/pd/sbsummar.htm. 3. 3Burghardt J, Gordon A, Chapman N, Gleason P, Fraker T. The School Nutrition Dietary Assessment Study: School Food Service, Meals Offered, and Dietary Intakes. Alexandria, VA: US Department of Agriculture, Food and Nutrition Service, Office of Analysis and Evaluation; 1993;. 4. 4Fox MK, Crepinsek MK, Connor P, Battaglia M. School Nutrition Dietary Assessment Study–II: Final report. US Department of Agriculture Web site. http://www.fns.usda.gov/oane/MENU/Published/CNP/FILES/sndaII.pdf. 5. 5Gordon AR, Cohen R, Crepinsek MK, Fox MK, Hall J, Zeidman E. The third School Nutrition Dietary Assessment Study: Background and study design. J Am Diet Assoc. 2009;109(suppl 1):S20–S30. Abstract | Full Text |
Full-Text PDF (180 KB)
|
CrossRef
6. 6Dietary Guidelines for Americans 2005. US Department of Health and Human Services and US Department of Agriculture Web site. http://www.health.gov/dietaryguidelines/dga2005/document/default.htm. 7. 7Institute of Medicine, Food and Nutrition Board. Dietary Reference Intakes: Applications in Dietary Assessment. Washington, DC: National Academies Press; 2000;. 8. 8Crepinsek MK, Gordon AR, McKinney PM, Condon EM, Wilson A. Meals offered and served in US public schools: Do they meet nutrient standards?. J Am Diet Assoc. 2009;109(suppl 1):S31–S43. Abstract | Full Text |
Full-Text PDF (159 KB)
|
CrossRef
9. 9Condon EM, Crepinsek MK, Fox MK. School meals: Types of foods offered to and consumed by children at lunch and breakfast. J Am Diet Assoc. 2009;109(suppl 1):S67–S78. Abstract | Full Text |
Full-Text PDF (128 KB)
|
CrossRef
10. 10Gordon AR, Crepinsek MK, Nogales R, Condon E. School Nutrition Dietary Assessment Study-III. Vol. I: School Foodservice, School Food Environment, and Meals Offered and Served. Final report. Princeton, NJ: Mathematica Policy Research, Inc; 2007;. 11. 11Fox MK, Gordon A, Nogales R, Wilson A. Availability and consumption of competitive foods in US public schools. J Am Diet Assoc. 2009;109(suppl 1):S57–S66. Abstract | Full Text |
Full-Text PDF (247 KB)
|
CrossRef
12. 12Gordon AR, Fox MK, Clark M, Nogales R, Condon E, Gleason P, et al. School Nutrition Dietary Assessment Study–III. Vol. II: Student Participation and Dietary Intakes. Final Report. Princeton, NJ: Mathematica Policy Research, Inc; 2007;. 13. 13Clark MA, Fox MK. Nutritional quality of the diets of US public school children and the role of the school meal programs. J Am Diet Assoc. 2009;109(suppl 1):S44–S56. Abstract | Full Text |
Full-Text PDF (147 KB)
|
CrossRef
14. 14Briefel RR, Wilson A, Gleason PM. Consumption of low-nutrient, energy-dense foods and beverages at school, home, and other locations among school lunch participants and nonparticipants. J Am Diet Assoc. 2009;109(suppl 1):S79–S90. Abstract | Full Text |
Full-Text PDF (264 KB)
|
CrossRef
15. 15Briefel RR, Crepinsek MK, Cabili C, Wilson A, Gleason PM. School food environments and practices affect dietary behaviors of US public school children. J Am Diet Assoc. 2009;109(suppl 1):S91–S107. Abstract | Full Text |
Full-Text PDF (382 KB)
|
CrossRef
16. 16Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM. Prevalence of overweight and obesity in the United States, 1999-2004. JAMA. 2006;295:1549–1555.
CrossRef
17. 17Gleason PM, Dodd AH. School breakfast program but not school lunch program participation is associated with lower body mass index. J Am Diet Assoc. 2009;109(suppl 1):S118–S128. Abstract | Full Text |
Full-Text PDF (128 KB)
|
CrossRef
18. 18Fox MK, Dodd AH, Wilson A, Gleason PM. Association between school food environment and practices and body mass index of US public school children. J Am Diet Assoc. 2009;109(suppl 1):S108–S117. Abstract | Full Text |
Full-Text PDF (126 KB)
|
CrossRef
19. 19Wolfe WS, Campbell CC, Frongillo EA, Haas JD, Melnik TA. Overweight schoolchildren in New York State: Prevalence and characteristics. Am J Public Health. 1994;84:807–813. MEDLINE |
CrossRef
20. 20Fiore H, Travis S, Whalen A, Auinger P, Ryan S. Potentially protective factors associated with healthful body mass index in adolescents with obese and nonobese parents: A secondary data analysis of the third National Health and Nutrition Examination Survey, 1988-1994. J Am Diet Assoc. 2006;106:55–64. Abstract | Full Text |
Full-Text PDF (126 KB)
|
CrossRef
21. 21Timlin MT, Pereira MA, Story M, Neumark-Sztainer D. Breakfast eating and weight change in a 5-year prospective analysis of adolescents: Project EAT (Eating Among Teens). Pediatrics. 2008;121:e638–e645. 22. 22US Department of Agriculture. MyPyramid: Steps to a healthier you. MyPyramid Web site. http://www.mypyramid.gov. A. R. Gordon and M. A. Clark are senior researchers, Mathmatica Policy Research, Inc, Princeton, NJ M. K. Crepinsek and M. K. Fox are senior researchers, Mathematica Policy Research, Inc, Cambridge, MA R. R. Briefel is a senior fellow, Mathematica Policy Research, Inc, Washington, DC.  Address correspondence to: Anne R. Gordon, PhD, Mathematica Policy Research, Inc, PO Box 2393, Princeton, NJ 08543
STATEMENT OF CONFLICT OF INTEREST: See page S135. Address reprint requests to: Jackie Allen, Mathematica Policy Research, Inc, PO Box 2393, Princeton, NJ 08543-2393. E-mail: jallen@mathematica-mpr.com PII: S0002-8223(08)02059-2 doi:10.1016/j.jada.2008.10.066 Published by Elsevier Inc. | |
|
FULL TEXT