Impact of the Updated USDA School Meal Standards, Chef-Enhanced Meals, and the Removal of Flavored Milk on School Meal Selection and Consumption

Abstract

Background

In Fall 2012, updated USDA school meal standards went into effect and did not result in increased food waste overall. However, consumption of school foods, especially fruits and vegetables, remain low. Therefore strategies to improve school meal consumption are necessary.

Objectives

This study evaluated the combined impact of the updated school meal standards and chef-enhanced, healthier meals and the removal of flavored milk on students’ school food selection and consumption compared with students in control schools.

Design

Project Modifying Eating and Lifestyles at School (MEALS) Study was a cafeteria-based quasi-experimental intervention conducted during the 2012–13 school year.

Participants/Setting

Student (n=1309) in grades 3–8 attending four intervention and four control schools in two low-income, urban school districts.

Intervention

Chef-enhanced school meals and the removal of flavored milk combined with the updated USDA school meal standards.

Main Outcome Measures

Changes in school meal selection and consumption were examined using plate waste methodology.

Statistical Analyses Performed

Logistic regression and mixed-model ANOVA adjusting for student demographics and schools/students as a random effect (students nested within schools) were used to examine differences in selection and consumption before (Fall 2012) and after (Spring 2013) a chef-based intervention with the updated school meal standards.

Results

After the chef-based intervention was implemented, there were no significant differences in entrée, vegetable, or fruit selection. Significantly fewer students selected milk compared with students in control schools (56.8% vs 94.0%;p<0.0001) and milk consumption was significantly lower (54.8% vs 63.7%;p=0.004). However, consumption was significantly greater for vegetables (62.2% vs 38.2%;p=0.005) and fruits (75.2 vs 59.2%;p=0.04) in the intervention schools compared with control schools. There were no significant differences in entrée consumption.

Conclusions

Schools collaborating with chefs can be an effective method to improve the consumption of fruits and vegetables with the updated USDA school meal standards. Further research should examine the longer-term impact of the removal of flavored milk from schools to determine if the lower selection and consumption rates persist.

In the United States, approximately 30 million children nationwide receive a school lunch daily.¹ The United States Department of Agriculture (USDA) standards for the types and quantities of foods available through the National School Lunch Program were updated through the Health, Hunger Free Kids Act of 2010, which went into effect in the Fall of 2012 and revised in 2018.^{2, 3} These standards included requirements aimed to increase the availability and consumption of whole grains, fruits, and vegetables; increased portion sizes for fruits and vegetables; sodium reductions; and age-specific calorie limits. Despite initial concerns that these healthier food standards would negatively impact students’ diets, initial studies found students were still selecting and consuming the school meals, although this did vary in some studies by grade and/or food type such as fruits and vegetables.^4–6 These studies also found that food waste was an issue in schools, both before and after the updated school meals standards went into effect, and therefore the high waste levels in schools appear to be consistent over time.^{4, 5} Therefore, initiatives that improve school meal selection and consumption are necessary to ensure students are eating the healthier school meals.

Some initial research has found that programs that enhance the palatability of school meals can be an effective method to reduce food waste in schools. A pilot study in Boston Public Schools found that a chef-based initiative led to increases in whole grain selection and vegetable consumption among students.⁷ A second study that compared chef-enhanced meals with Smarter Lunchroom strategies (i.e. nudges to encourage selection of the healthier school meal options) found that only the chef intervention successfully increased both the selection and consumption of school meals.⁸ Lastly, a study in two elementary schools that provided two chef-enhanced vegetables (cauliflower and sweet potato fries) in the presence of the updated school meals standards found increases in their consumption.⁹ However, these studies were either conducted prior to implementation of the updated USDA school food standards^{7, 8} or were limited in scope,⁹ and therefore the effect of a chef-based program is unclear in the presence of healthier meal standards compared with schools with just the updated USDA food standards.

To address these gaps, the objectives of this study were to examine the impact of healthier, chef-enhanced meals and the removal of flavored (i.e. chocolate and strawberry) milk on school food selection and consumption with the updated school meal standards. This study was conducted during the 2012–2013 school year among elementary and middle school students within in two large urban school districts participating in the Project Modifying Eating and Lifestyles at School (MEALS) study.

METHODS

Study design and participants

Project MEALS was a collaboration between the non-profit, anti-hunger organization Project Bread (www.ProjectBread.org) and the Harvard T.H. Chan School of Public Health (HSPH) to examine the impact of healthier, chef-enhanced meals (i.e. “Chef” schools) on students’ school food selection and consumption. Beginning in 2011, Project Bread hired a professional chef to collaborate with cafeteria staff in a low-income, urban school district in Massachusetts to improve the taste and nutrient quality of the school meals. Both elementary and middle schools were recruited to examine the impact on a more generalizable student population. In the first phase of the Project Meals study, which was conducted during the 2011–12 school year, n=4 schools received the Chef intervention and n=4 schools were controls from within the same school district.⁸ The following year (2012–13 school year), the second phase of the study went into effect with the four control schools from the initial phase receiving the Chef intervention (i.e. delayed intervention). The updated USDA school meal standards also went into effect nationwide at the beginning of the school year. Four matched control schools from another low-income urban district were also recruited based on student demographics using data from the Massachusetts Department of Elementary and Secondary Education and received only the updated school meal standards. The present analyses focuses on the data collection in the second phase.

All students in grades 3–8 in the participating schools were recruited for the study. Parents also filled out a survey with demographic information including the child’s gender, race/ethnicity, and age. A total of n=864 students at the Chef schools and n=445 students at the control schools provided parental/student consent (27% and 20% of those eligible, respectively). Students were included in the study if they attended lunch on a study day and received a school lunch. On average, approximately 93% of the students in the participating intervention schools and 95% of the students in the intervention schools were eligible for free or reduced-priced meals. The participating students had socio-demographic characteristics similar to the overall school populations.

Intervention

The Chef intervention consisted of a professional chef who collaborated with the intervention schools 2–3 days per week throughout the school year. The chef worked with the existing staff on their culinary skills, developed new recipes to improve the palatability of the school meals, and was present during the lunch periods to assist the staff and ensure compliance with updated menus. The recipes used the foods already available to schools, utilizing the cost-effective commodity foods, providing whole grains for all grain products, as well as emphasizing fresh/frozen produce, healthier poly- and mono-unsaturated fats instead of saturated fats, and seasonings without added salt or sugar (recipes are available at http://www.projectbread.org/reusable-components/accordions/download-files/school-food-cookbook.pdf). The students were repeatedly exposed to the Chef enhanced meals 2–3 days per week during the seven month intervention period. Additionally, flavored milk was removed from the Chef schools as part of the district-wide initiative to reduce added sugars in the foods offered. The lunches served at the control schools met the requirements for the updated USDA school meal standards and provided the students with a choice of fat-free flavored milk, or plain 1% or fat-free milk.

Plate Waste Measures

Students’ selection and consumption was measured using standard plate waste methodology.^{10, 11} Plate waste was measured on two non-consecutive days in the Fall (baseline) and post-implementation in the Spring (n=4 days of data collection total) on days that chef-enhanced menus were offered in intervention schools. Prior to students entering the cafeteria, research assistants (RAs) removed the trash cans, labeled all the trays with unique identifying numbers, and weighed ten random samples of each food being served that day using a food scale (Oxo 1130800; New York NY) to determine the average baseline weight for each food. At the beginning of every lunch period, students were reminded about the study and that participation was voluntary. Students then selected their meals, and as they left the lunch line with their foods, RAs discreetly recorded their choices and tray numbers. Participating students were also asked to write their names on their trays. RAs then collected all trays at the end of the meal, and recorded the names and tray numbers of the participating students. RAs then weighed all the individual food items on the trays. Meal components were categorized into entrée (combined grains and proteins), fruit, vegetable, and milk (there were insufficient quantities of vegetables and/or fruits in the entrées to count towards those components). This study was approved by the Committee on Human Subjects at the Harvard T.H. Chan School of Public Health and all participants provided written informed consent and child assent.

Statistical Analyses

To examine differences in meal selection, logistic regression with multilevel modeling was used (SAS PROC GLIMMIX),¹² accounting for the repeated measures of students and student nested within schools. Differences in consumption were analyzed using mixed-model analysis of variance, accounting for the repeated measures of students and student nested within schools. (SAS PROC MIXED).¹² All models adjusted for the students’ age, gender, race/ethnicity, and baseline selection or consumption of the food component (entrée, fruit, vegetable, or milk). A P-value of <0.05 was considered statistically significant. Due to the non-generalizable high consumption levels of pizza and French fries, which had uneven distribution as a choice between the participating schools, pizza and French fry consumption were excluded from entrée and vegetable analyses, respectively.

RESULTS

The characteristics of the participating students are presented in Table 1. Among the 1309 students participating in Phase 2 of Project Meals, approximately half were female and the majority were Hispanic (67.4% and 84.0% in the control and Chef schools, respectively). Students were on average between 9–10 years old (range 7–13 years).

Table 1.

Characteristics of students attending intervention (n=4) and control (n=4) elementary and middle schools in two large urban school districts participating in the Project MEALS^a Study

Characteristics	Control Schools (n=445)	Chef Schools (n=864)
Age, years, mean (range)	10 (7–13)	9 (7–13)
Grade, mean (range)	4 (3–8)	4 (3–8)
Female gender, n (%)	229 (51.5)	419 (48.5)
Race/ethnicity, n (%)
Asian	30 (6.7)	33 (3.8)
Black	51 (11.5)	20 (2.3)
Hispanic	299 (67.2)	725 (84.0)
White	41 (9.2)	42 (4.8)
Other/multi-racial	24 (5.4)	44 (5.1)

^aMEALS= Modifying Eating and Lifestyles at School

After the Chef intervention was implemented, there were no significant differences in the selection of entrées, vegetables, or fruits compared with control schools (Table 2). With the removal of flavored milk, significantly fewer students selected plain 1% or fat-free milk in the Chef schools compared with students in the control schools who had access to both flavored and plain milk (56.8% vs 94.0%; p<0.0001).

Table 2.

Meal component Selection among students attending intervention (n=4) and control (n=4) elementary and middle schools in two large urban school districts participating in the Project MEALS^a Study

Meal component	Chef Schoolsb (%)	Control Schoolsb (%)	P-valuec
Entree	99.1	97.4	0.55
Milkd	56.8	94.0	<.0001
Vegetable	49.1	55.5	0.38
Fruit	83.9	74.8	0.40

^aMEALS= Modifying Eating and Lifestyles at School

^bResults are unadjusted

^cCalculated using logistic regression, with school as a random effect and repeated measures among students, adjusting for students’ age, gender, race/ethnicity, and baseline selection of the meal component.

^dFlavored milk was removed in the Chef Schools but was available in the Control Schools.

When examining the average percent of the school foods consumed, there was significantly greater intakes of fruits (75.2 vs 59.2%; p=0.04) and vegetables (62.2% vs 38.2%; p=0.005) among students in the Chef schools compared with controls schools (Table 3). This resulted in students consuming significantly more cups of fruits (0.38 vs 0.30; p=0.04) and vegetables (0.31 vs 0.19; p=0.005) on average in the Chef schools compared with the control schools. There were no significant differences in average entrée consumption in Chef versus control schools (79.0% vs 70.2%; p=0.15). Among students who selected milk, average consumption was significantly lower in the Chef schools compared with control schools (54.8% vs 63.7%; p=0.004). Overall, compared with children who were in middle school, children in elementary school consumed on average 4.8% less of their entrée (p=0.03), 9.8% less of their milk (p=0.0005), and 14.0% less of their fruit (p=0.0003). There were no significant differences in overall vegetable consumption by grade. Female students also consumed on average 3.7% less of their entrée and 7.5% less of their milk (p=0.003) compared with male students. There were overall no significant differences in vegetable or fruit consumption by gender.

Table 3.

Meal component consumption among students attending intervention (n=4) and control (n=4) elementary and middle schools in two large urban school districts participating in the Project MEALS^a Study

Meal component	Chef Schoolsb	Control Schoolsb	P-valuec
	(% [SE])
Entree	79.0 (4.2)	70.2 (4.4)	0.15
Milkd	54.8 (2.2)	63.7 (2.2)	0.004
Vegetable	61.1 (5.4)	38.2 (5.9)	0.005
Fruit	75.2 (5.3)	59.2 (5.6)	0.04
	Servings (cups) consumed (SE)
Cups of Vegetable	0.31 (0.02)	0.19 (0.03)	0.005
Cups of Fruit	0.38 (0.02)	0.30 (0.03)	0.04

^aMEALS= Modifying Eating and Lifestyles at School

^bCalculated using least squares mean regression.

^cResults are calculated among students who selected the meal component using mixed-model analysis of variance, with school as a random effect and repeated measures among students, adjusting for students’ age, gender, race/ethnicity, and baseline consumption of the meal component.

^dFlavored milk was removed in the Chef Schools but was available in the Control Schools.

DISCUSSION

Overall, this study found that chef-enhanced school meals led to modest increases in fruit and vegetable consumption, and thus some improvements in diet quality while reducing food waste. There were no significant differences in entrée consumption despite the entrées consisting of 100% whole grains in the Chef schools. However, there were decreases in milk selection and consumption when flavored milk was removed.

This study has important implications. While evidence suggests that food waste has been an issue in schools both before and after the updated school meal standards,^{4, 5} concerns about food waste and attempts to weaken healthier meal standards persist, including recent rulings by the USDA to delay and remove future phases of sodium reductions and increases in whole grains.^{3, 13} The results from this study suggest that long-term collaborations with professional chefs can be an effective strategy to improve school meal consumption, which may be due to multiple factors including improved taste of the school foods and the presence of the chef.

Interestingly, there were no differences in fruit or vegetable selection seen in the Chef schools in the present study. Previous research examining the impact of a chef prior to implementation of the updated school meal standards found significant increases in the selection of fruits and vegetables in the intervention schools.⁸ It is possible that no differences in selection were seen in the Chef schools in the present study because of the already high selection rates of these meal components due to the updated school meal standards.⁴

The decreases in milk selection and consumption were also noteworthy. Previous research by Hanks and colleagues also found similar effects when flavored milk was limited.¹⁴ However, one longer term study did find that after two years with limited access to flavored milk, students selected and consumed just as much white milk.⁷ Future research should examine the length of time needed for students to adjust to the removal of flavored milk in school cafeterias.

This study has several limitations. Only low-income, urban school districts were examined. However, the results are likely generalizable to similar school districts. Additionally, this vulnerable population is likely to benefit the most from the updated school meals standards due to their reliance on school meals for up to half their daily energy intake.¹⁵ Future studies should examine the impact of a chef-based intervention in schools with different socio-demographic characteristics. Additionally, only elementary and middle school children were examined. Researchers should also consider examining chef-based interventions in high schools. While plate waste measurements were only collected on two days pre- and post-intervention at each school, there was no evidence to believe that consumption on these days differed from other days. Lastly, it is possible that some of the increases in fruits and vegetables observed were due to compensation from the reductions in milk consumption. This study was strengthened by the detailed consumption data collected through the plate waste measurements and large sample size.

Conclusions

To our knowledge, this is the first large-scale study to examine the impact of providing chef-enhanced meals with the updated USDA school meal standards. Overall, selection of entrées and fruits were high in both Chef and control schools in the presence of the healthier school meal standards. In the Chef schools, where there was a focus on the palatability of the healthier school foods, students consumed significantly more of their fruits and vegetables. However, when flavored milk was removed, milk selection and consumption decreased in the Chef schools.

In the presence of the updated USDA school meal standards, collaborating with a professional chef may be an effective way to increase consumption of the fruits and vegetables and reduce food waste. Rather than weaken the healthier school meal standards, this study suggests that providing schools with additional resources and support may help to address food waste in schools.

RESEARCH SNAPSHOT.

Research Question

What is the combined impact of the updated school meal standards and chef-enhanced, healthier meals and the removal of flavored milk on students’ school food selection and consumption compared with students in control schools?

Key Findings

In this school-based intervention that included 1309 students in grades 3–8 in two low-income, urban school districts, students consumed significantly more vegetables (62.2% vs 38.2%; p=0.005) and fruits (75.2 vs 59.2%; p=0.04) with chef-enhanced meals compared with control meals. There were significant reductions in milk selection and consumption. There were no significant differences in the selection of entrées, vegetables or fruit, nor in entrée consumption.