Comparison of Consumed Lunch Nutrient Profiles of 6th-Grade Students Before and After Implementation of the West Virginia School Nutrition Standards

Please note that this study was published before the implementation of Healthy, Hunger-Free Kids Act of 2010, which went into effect during the 2012-13 school year, and its provision for Smart Snacks Nutrition Standards for Competitive Food in Schools, implemented during the 2014-15 school year. As such, certain research may not be relevant today.

Childhood is an opportune time to establish lifelong diet and exercise habits, and schools are an ideal environment in which healthful lifestyle habit examples can be modeled and learned. The

American Dietetic Association (ADA), the Society for Nutrition Education (SNE), and the American School Food Service Association (ASFSA) recommend that comprehensive school- based nutrition programs be provided to all elementary and secondary school students in the United States, and that such health programs include education in foods and nutrition, an environment that promotes healthful eating and physical activity, and involvement of parents and the community (Position of ADA, SNE & ASFSA, 1995). Likewise, national health objectives include recommendations for improvements in diet and an increase in regular physical activity among children and adolescents; these are important for the prevention of obesity and adulthood chronic diseases such as coronary heart disease (CHD), diabetes, and cancer (U.S. Department of Health and Human Services [HHS], 1990 & 2000).

In the early 1990s, West Virginia (WV) had been reported to lead the nation in heart disease mortality (West Virginia Department of Health & Human Resources [WVDHHR], Bureau of Public Health, 1994). Data from a survey of 45 states indicated that WV remained among the top 10 states with the highest percentage of persons >18 years of age who were overweight (24.8 percent) and who smoked (26.6 percent) (Centers for Disease Control and Prevention [CDC], 1991). In response to these dire health statistics and with a desire to be proactive regarding health promotion, the West Virginia Board of Education (WVBOE) designed its Standards for School Nutrition (WVSNS) in 1993 in an effort to “ensure a healthy school environment that provides for the nutritional well-being for students” (WVBOE, 1994; Stuhldreher et al., 1998). These independently developed standards were implemented before the most recent federal mandates (U.S. Department of Agriculture [USDA], 1995). The objectives of the WVSNS were to: 1) ensure nutrient adequacy of all meals; 2) provide guidelines that target nutrients identified as critical to health promotion and disease prevention; and 3) adhere to standards defined in the Dietary Guidelines for Americans (DGAs) (USDA, 1990).

The overall goal of the West Virginia Healthy Schools Nutrition Study (WVHSNS) was to determine if the new standards influenced the dietary intake of students in rural areas, which are comprised of a higher percentage of students who rely on school meals to provide the majority of their daily nutrient intake. The specific aims of this report were to:

1. Describe the school lunch intake of a cohort of WV students before implementation and evaluate it against the WVSNS;
2. Compare the lunch intake of a cohort of WV students after implementation of WVSNS with that of a cohort of WV students before implementation; and
3. Compare the lunch intake of the WV after cohort with intakes of a cohort of students from PA during the same time period, which could control for changes in products in the marketplace and dietary influences that might have been attributed to the passage of

Methodology

Terminology

For the purposes of this study, the term “standard” refers to the target level of nutrients that the lunch is supposed to offer, according to the WVSNS. These standards will be the benchmarks upon which all cohorts’ nutrient profiles are judged for adequacy. For protein, vitamins, and minerals, the lunch standard was one-third of the gender-specific 1989 Recommended Dietary

Allowance (RDA) value for 11- to 14-year-olds (National Research Council [NRC], 1989). For nutrients included in the DGAs, the lunch standard refers to the following percentages:

• ≤30 percent of calories from fat;
• <10 percent of calories from saturated fat;
• ≤100 mg of cholesterol;
• ≤6 grams of fiber; and
• ≤1,100 mg of

The term “WV before cohort” refers to dietary intake data collected from students prior to standard implementation. The term “WV after cohort” refers to dietary intake data collected from students during the time period the WVSNS were in place.

Study Design and Population

The WVHSNS was a community intervention study that compared nutrient profiles of cohorts of 6th-grade students before and after implementation (approximately 3 years apart) of WVSNS. Two poverty-stricken counties were chosen to participate, areas from which approximately 70% of students received free or reduced-price meals (WVBOE, 1992).

Because changes in food products and consumer practices could confound the results over time, especially regarding dietary fat, a comparison cohort from rural Pennsylvania (PA) schools was assembled. At the time, federal school lunch guidelines did not require specific levels for fat, saturated fat, cholesterol, sodium, and fiber. The comparison cohort from PA was drawn from areas that would provide a population similar to the WV cohort. In order to obtain a sample size comparable to the WV cohort, the PA sample was drawn from four rural schools.

Human Subjects Approval (IRB) from West Virginia University and the University of Pittsburgh was granted for parental consent and participant assent. Guardians of 6th-graders were sent a letter, which described the study and included consent forms that would grant their children permission to participate. If parents gave consent, students were read assent forms that indicated their willingness to participate. If parents gave consent, students were read assent forms that indicated their willingness to participate. Researchers witnessed each student’s signature. This study had approval from the WV Board of Education and each participating school district in WV and PA.

Dietary Assessment

Identical dietary assessment methods were used for WV after and PA cohorts (Stuhldreher et al., 2000). Briefly, dietetic student researchers were given a one-day training session that included directions on following a standard protocol, which would ensure that everyone used the same format when conducting the 24-hour recalls. Researchers practiced conducting recalls using food models to ascertain portion size. Researchers also were given specific directions regarding their observations of students during school lunch. They were briefed on the protocol for labeling lunch trays, instructed on how to discreetly observe the students eating during lunch, and practiced measuring plate waste.

During the actual data collection, participants in the study were instructed to leave their trays on the tables so that researchers could record the amounts consumed. Trays were labeled at the beginning of the serving line, and observation was performed for every student participant at least once. Data from the observations were compared with the dietary recall of that same lunch to assess the quality of the recall; these data are explained elsewhere (Stuhldreher et al., 2000). Researchers used school menus and the National Dairy Council’s standardized food models to assist students during diet recalls. Use of 24-hour dietary recalls has been cited as an acceptable method to use with children (Goran, 1998).

Dietary variables selected for analysis included calories, protein, vitamin A, vitamin C, thiamin, vitamin B6, calcium, iron, zinc, cholesterol, sodium, and dietary fiber. The percentage of calories from total fat and saturated fat also were examined. Although vitamin B6 and zinc were not listed as nutrients to be monitored in either the WVSNS or in the current School Meals Initiative (USDA, 1995), these nutrients were included due to the results of the School Nutrition Dietary Assessment Study that indicated the vitamin B6 and zinc content of school meals as offered and consumed was below 33% of the RDA for select age groups (Burghardt et al., 1993).

Dietary data were analyzed using Nutritionist IV (software version 4.0 for Windows, 1994, First Databank, San Bruno, CA). To ensure accuracy, data were coded and entered by one researcher and verified by another in a blind fashion. Coding records of all data assumptions were kept to provide consistency of data entry procedures.

Statistical Analysis

Pearson product moment correlations were used to compare observed and recalled lunch intake. For most nutrients, the correlation between the observed and recalled lunches was acceptable (r>0.70), thus, dietary recalls were judged acceptable.

Differences in the mean dietary intake of nutrients among the cohorts were tested using Analysis of Variance (ANOVA) separately for males and females. Comparison of the percent of students from each cohort who were in compliance with the standards was evaluated using Chi-square analysis. All analyses were performed using SPSS (SPSS, Inc., version 10.1, Chicago, IL).

Statistical significance was defined as p>0.05.

Results And Discussion

Participant Characteristics

Of those students enrolled in the two WV and four PA schools from which the sample was drawn, there was a total of 274 WV and 419 PA students enrolled in the 6th-grade. The school foodservice directors estimated that of these students, approximately 60 to 85% usually consumed school meals. This report is based on 71 WV (43 female) students from the initial study (WV before cohort), as well as 98 WV (53 female) and 70 PA (36 female) students from the after observation (WV after cohort and PA cohort). All students were 11 or 12 years of age.

Recalled Lunch Intake in the WV cohorts: Before and After Guideline Implementation Comparisons between the lunch intakes of WV students before and after implementation of the WVSNS are shown in Table 1. Caloric intake was lower for the WV after cohort and was less than the standard. Although mean protein intakes met the standards, the mean level was lower in the WV after cohort. Among females, mean levels were lower in the WV after cohort for thiamin, vitamin B6, calcium, iron, and zinc. Among males, mean levels were lower in the WV after cohort for thiamin, iron, and zinc. A comparison of the mean as a percent of standard also is listed as an indicator of the magnitude of that contribution. The most noticeable decreases among the WV after cohort were for calories, iron, and zinc, and among females, for vitamin B6 and calcium.

There was an improvement (lowered levels) in cholesterol intake for both males and females in the after cohort, and females consumed a lower mean sodium. Mean total fat as a percent of calories (mean percent of fat) decreased in the WV after cohort, as well. However, no difference in the percent of calories from saturated fat (percent of SFA) between the two time periods was found for either gender. Mean fiber intakes were lower in the WV after cohort for females.

Figure 1 shows a comparison of the percent of students in the before and after cohort who met the standards. The percentage of students in the WV after cohort who met the fat, cholesterol, and sodium standards all increased. However, the percent that met the standard for vitamins and minerals, except vitamin C, dropped. In fact, fewer than 25% of the WV after cohort met the standards for vitamin A, thiamin, vitamin B6, calcium, iron, and zinc. The frequency of those compliant for these nutrients dropped and this decrease was disturbing, especially for iron (34% to 7%), zinc (22% to 3%) and thiamin (62% to 20%).

Critics opposed to reducing the fat content of children’s diets have raised the concern that such reductions typically are accompanied by a reduction in nutritional quality (Joint Working Group, 1993; Lifshitz & Tarim, 1996; Olson, 1995). A study conducted during the time in which the WV standards were drafted reported that students who self-select diets lower in fat have shown a reduction in intakes of micronutrients (Nicklas et al., 1992). Stuhldreher et al. (2001) examined the nutritional intake of children living in low-income areas of southern WV. These researchers reported that while most children met nutritional guidelines for cholesterol and vitamin C, less than adequate intakes were noted for fiber, calcium, and iron. Furthermore, excesses were found for most children in total and saturated fat intake.

Examination of changes in food selections may give a plausible explanation for the reduction in nutritional quality. For example, the standards require that five servings of bread and bread products per week contain at least 33% whole grain. In one of the districts in the study, the foodservice staff baked fresh rolls in compliance with these regulations. In spite of the appetizing appeal of fresh baked bread, acceptance of the part-whole-grain product was not great, and could explain the low compliance for thiamin, vitamin B6, and zinc. A study by Shanklin and Wie (2001) found that the grain/bread component of the school lunch provided the most energy [calories], sodium, iron, total fat, and saturated fat per penny. When nutrient density was compared with cost, these researchers reported this group provided the most energy, protein, and carbohydrate per penny.

Keeping in mind that the WVHSNS study population came from very needy families, it is plausible that it is more economical for those families to serve refined bread products in the home; thus, whole-grain products are unfamiliar to these students. This also would explain the lack of an increase in dietary fiber intake over time. Furthermore, the changes in items served from the meat group may account for the lower mean and number of students meeting the standards in the after cohort. Less of the iron-rich meats, such as ground beef and other more fatty red meats, would be on the menus after implementation of the standards. This, plus the lower acceptance of whole-grains, could explain the reduction in iron, vitamin B6, and even zinc intakes seen in the WV after cohort.

The reduction in mean calcium intake among girls and the lower percent meeting this standard in the WV after cohort could be explained by a failure to consume lower fat (2%) and skim milk and by a reduction in the amount of cheese used in school meals. These findings would be consistent with the results of the Bogalusa Heart Study, which indicated that groups of students with higher fat intakes consumed more meat and dairy products than those with lower fat intakes (Nicklas et al., 1992). A study by Johnson et al. (1998) found calcium intakes were achieved only when milk was included at the noon meal.

Comparison of West Virginia Versus Pennsylvania

Comparisons between the gender-specific mean school lunch nutrient intake of the WV and PA after cohorts also are shown in Table 1. Mean calorie and protein levels were higher in the PA cohort than the WV after cohort. Although neither gender consumed a sufficient number of calories at lunch to meet the standard, protein intake did surpass the school lunch standard.

Higher mean values for thiamin, calcium, and iron were observed among the PA cohort. Mean zinc intakes were higher for the PA cohort among females. The PA cohort consumed greater than 33% of the RDA for vitamin C, thiamin, and calcium but not for vitamin A, vitamin B6, iron, or zinc. WV after cohort means fell short of the lunch standard for thiamin, vitamin B6, calcium (females only), iron, and zinc.

Examination of the means as percent of standard showed that intakes of protein, vitamin A, vitamin C, thiamin, and calcium were at or above 100% of the standard in the PA cohort. In the WV after cohort, male mean calcium intake met the lunch standard. Lower values of the means as percent of standard are seen for thiamin, iron, and zinc in the WV after cohort and among females for calcium. As shown in Figure 2, a greater percentage of the PA cohort met the goals for calories, thiamin, calcium, iron, and protein than the WV after cohort. Conversely, a greater percent of the WV after cohort met the standard for vitamin C.

A comparison of the nutrients with standards derived from the DGAs revealed that male students in the WV after cohort consumed a higher percentage of calories from total fat and saturated fat than males in the PA cohort. No significant differences in the percent meeting the standards derived from the DGAs were observed (Figure 2).

It may be possible that the WV after cohort was less familiar with lowfat and whole-grain foods; thus they were less accepting of them. It also was noted that there are unique differences in PA and WV food patterns. The acceptance of cornbread and beans is typically seen in southern WV, whereas the PA Dutch influence is evidenced in PA schools. Further exploration of these influences may yield clues suggesting the differences in nutrient values are a result of diversity in food acceptance patterns.

Conclusions And Applications

The WV Healthy Schools Nutrition Study is the first evaluation of the actual nutritional intakes of students against the WVSNS. Although it would have been interesting to survey the same group of students before and after implementation of the new school nutrition standards, it would have been challenging to retain them in the study for the three-year period. In addition, the impact of the adolescent growth spurt would have introduced another source of variability.

This report is limited to a comparison of nutrient profiles from lunch intake only; therefore, the specific influence of other foods eaten during the day on nutrient intake was not assessed.

Despite these limitations, however, there are some important lessons to be learned. Although the mean percent of calories from fat decreased over time, there was no difference between the cohorts in the percent that met the saturated fat recommendations. Mean levels (as well as the mean as a percent of standard for thiamin, iron, and zinc, and among females for calcium and vitamin B6) all were lower in the WV after cohort. Correspondingly, fewer students in the WV after cohort met these recommendations.

These are not reasons to conclude that lower fat consumption compromises nutritional quality. Quite the opposite has been demonstrated in the Child and Adolescent Trial for Cardiovascular Health (CATCH), which indicated that changes made in school nutrition to lower fat and sodium content were accompanied by a retention of high levels of other nutrients, such as vitamins and minerals. CATCH study results indicate the importance of a combined effort of school- and family-based interventions (Nicklas et al., 1992). Likewise, reports from the Bogalusa Heart Study indicated that children with high fiber intakes consumed less fat, particularly saturated fat, than did children with low fiber intakes (Nicklas et al., 1995). And results from the Dietary Intervention Study in Children (DISC) found lower fat intakes were adequate for growth and normal levels of nutritional biochemical measures (Obarzanek et al., 1997). Thus, achievement of both goals–meeting the standards for protein, vitamins, and minerals, as well as those from the DGAs–is attainable.

Keeping in mind that there were no competitive foods sold during the instructional day in the WV after cohort, and that a high percentage of students in all the cohorts received financial assistance for lunch, it is highly probable that lunch intake makes an important contribution to the overall dietary quality of these students. On a positive note, the mean values for vitamins A and C and the percent in compliance with the standard suggest that meals were heading in the right direction. The results for vitamin A were not significant, probably because of the wide variation (seen in the standard deviation) in intake levels. The values for vitamin C may have been positively affected by the changes attributed to the increase in the number of fresh fruits and vegetables required in the WV standards. The role of fresh fruits and vegetables on dietary quality is consistent with findings from a study of three schools in metropolitan Atlanta (Baranowski et al., 1997).

As with any new lifestyle practice, change takes time. The WVBOE implemented these standards in tiers (Stuhldreher et al., 1998). However, the time period (approximately three years) may not have provided sufficient time for students to adapt to the changes, especially if these dietary patterns were not reinforced at home. This report should be viewed as an opportunity for further nutrition education and the development of strategies to encourage students to try and learn to accept healthful foods. Perhaps an examination of the barriers to new food acceptance or the involvement of parents may help tailor nutrition education programs that would encourage acceptance of different foods. Helpful suggestions, such as making healthful food taste and look better and changing social norms to make it “cool” to eat healthfully, have been reported from adolescent focus-group research (Neumark-Sztainer et al., 1999).

The results of this WV study should not thwart the efforts to continually improve the nutritional quality of school meals. It is important that schools not only teach patterns of good nutrition in the classrooms, but also offer such patterns in their school cafeterias.

Translation of nutrition knowledge into behavior is complex. Guidelines for school health programs to promote lifelong healthful eating (CDC, 1996) included seven recommendations, two of which are applicable to this report. The first recommendation is a “policy that promotes healthy eating through classroom lessons and a supportive school environment.” The sixth recommendation, which stresses the importance of family and community involvement to reinforce nutrition education, underscores the vision and significance of efforts such as those in WV.

WV was one of the first states to prohibit sales and service of non-nutritious foods and beverages to students during the school day (School Nutrition Policy Committee, 1992). Implementation of these new standards long before the changes were required by the federal government is an indicator of the WVBOE’s concern for the health of students. Development and passage of such standards was only possible by assembling a team of experts in education, nutrition, and health, plus food industry partners. These results should be compared with those from other school nutrition programs in an attempt to find successful strategies that improve nutritional intake. The WVBOE and other WV nutrition support personnel should be commended for their leadership in school nutrition. The researchers recommend replicating this study to determine if the acceptance has improved with the passage of time.

Acknowledgments

The authors thank the Child Nutrition Foundation for providing funding for the West Virginia Healthy Schools Nutrition Study through the Hubert Humphrey Research Grant.