Energy Balance 4 Kids with Play: Results from a Two-Year Cluster-Randomized Trial

Abstract

Background:

Identifying sustainable approaches to improving the physical activity (PA) and nutrition environments in schools is an important public health goal. This study examined the impact of Energy Balance for Kids with Play (EB4K with Play), a school-based intervention developed by the Academy of Nutrition and Dietetics Foundation and Playworks, on students' PA, dietary habits and knowledge, and weight status over 2 years.

Methods:

This cluster-randomized, controlled trial took place in four intervention and two control schools over 2 years (n=879; third- to fifth-grade students). PA (fourth and fifth grades only), dietary knowledge and behaviors, school policies, and BMI z-score were assessed at baseline (fall 2011), midpoint (spring 2012), and endpoint (fall 2012 for accelerometers; spring 2013 for all other outcomes).

Results:

At endpoint, there were no group differences in change in PA or dietary behaviors, although BMI z-score decreased overall by −0.07 (p=0.05). Students' dietary knowledge significantly increased, as did the amount of vegetables schools served. Post-hoc analyses stratified by grade revealed that, relative to control students, fourth-grade intervention students reduced school-day sedentary time by 15 minutes (p=0.023) and third-grade intervention students reduced BMI z-score by −0.2 (0.05; p<0.05). There were no signicifant differences for older students.

Conclusions:

EB4K with Play, which leverages the existing infrastructure of two national programs, increases children's dietary knowledge and may improve weight status and decrease sedentary behaviors among younger children. Future iterations should examine programming specific for different age groups.

Introduction

Obesity continues to affect nearly 1 in 5 youth in the United States.¹ Compared to their normal weight peers, obese children are at increased cardiovascular risk (CVR),^2,3 with low-income youth at far greater risk than their higher-income peers.⁴ Early interventions to prevent and stem the development of obesity among children are critical.

Reducing childhood obesity on a national scale requires effective multicomponent strategies for increasing physical activity (PA), decreasing sedentary behaviors, and improving diets. Dietary improvements are associated with reductions in cardiometabolic risk factors, as well as improved weight outcomes.⁵ Greater sedentary behavior predicts obesity in children,⁶ whereas increasing moderate-to-vigorous PA (MVPA) is associated with significant reductions in CVR.⁷ Most youth, however, fail to achieve the national recommendation of 60 minutes of daily MVPA⁷ or the recommended dietary allowance of nutrients.⁸

Schools, which reach youth from all backgrounds, are increasingly the focus for interventions to improve youths' dietary and physical behaviors.^8–15 Identifying replicable, sustainable school-based interventions will be important. Many interventions funded by research grants have no infrastructure support once the grant ends. Interventions that partner with existing organizations that promote healthy diets and PA in schools have a greater chance of long-term sustainability, but little research exists on the impact of existing programs.

Energy Balance for Kids with Play (EB4K with Play)¹⁶ is a multicomponent, school-based obesity prevention intervention developed by the Academy of Nutrition and Dietetics Foundation in partnership with Playworks, a national organization focused on developing safe, healthy play environments for elementary school students. EB4K with Play was designed to improve the nutrition and PA environments of schools through integrated education lessons delivered by a registered dietitian (RD coach) and organized activities led by a Playworks coach. The aims of the present study were to evaluate the impact of the EB4K with Play program on students' PA levels, dietary behaviors and knowledge, and adiposity over 2 school years.

Methods

Study Design

A cluster-randomized, controlled design was employed involving six low-income elementary schools in Northern California. University of California Berkeley's Committee for the Protection of Human Subjects approved this research.

Participants and Intervention

Recruitment and randomization efforts have been described previously.¹⁶ In summary, four schools were randomized to the EB4K with Play intervention for the 2011–2012 and 2012–2013 school years; two schools were randomized to a wait list control condition and received the EB4K with Play program during the 2013–2014 school year. All students enrolled in third, fourth, and fifth grade in September 2011 were eligible to participate in the evaluation. Five hundred eighty-three students (77%) in intervention schools and 296 (75%) in control schools provided parental consent and enrolled in the study. Researchers collected data from this longitudinal sample at baseline (fall 2011) and endpoint. PA endpoint measures were collected in fall 2012 to match baseline collection and account for potential seasonal effects on PA. Endpoint data for fitness and dietary outcomes were collected in spring 2013.

The EB4K with Play intervention has been described previously.¹⁶ Briefly, each intervention school received one part-time RD coach and one full-time Playworks coach for 2 school years. Each year, the RD coach delivered a 12-week nutrition and energy balance education curriculum that included food tastings, PA games to reinforce nutrition messages, and strategies to help students meet their nutrition and PA goals. RD coaches also worked with a team of school staff and parents to implement classroom wellness policies and make improvements in school food, including increased offerings of fruits and vegetables (FVs) to meet the Bronze-level Healthier US School Challenge criteria.¹⁷ During the second year of the program, EB4K with Play funding purchased packaging equipment for the district's central kitchen that enabled larger portions of FVs to be served in schools districtwide.

The Playworks coach structured recess activities before and during school hours to encourage active participation from all students. The Playworks coach also led a PA session with individual classes every other week. Classroom teachers were trained to implement Playworks games and classroom management strategies in their physical education (PE) sessions with students (classroom teachers were responsible for leading PE in this district). Last, Playworks coaches led four afterschool sports leagues throughout each year. Each league lasted for 5 weeks and accommodated 12 different students per team.

Measures

Physical activity

Fourth- and fifth-grade students (who became fifth- and sixth-grade students in year 2) wore an Actigraph GT1M or GT3X accelerometer (AG; Actigraph, LLC, Pensacola, FL) for 5 consecutive weekdays at each time point. Actigraph accelerometers have been validated against treadmill tests with children,¹⁸ and GT1M and GT3X accelerometers perform comparably when collecting data in uniaxial mode.¹⁹ Programmed in the uniaxial mode, accelerometers collected data in 15-second “epochs” or time intervals. Data collection occurred within a 6-week period at each time point (1 week at each school).

When analyzing accelerometer data, researchers considered 20 consecutive minutes or more of zero activity counts to be nonwear time and dropped those epochs.²⁰ Activity during each epoch was considered sedentary if activity counts were ≤25 and MVPA if activity counts were ≥574, based on Evenson's cutpoints.²¹ Researchers created a composite day by averaging activity classifications for corresponding epochs across days. The composite day approach reduces the amount of missing accelerometer data and provides a better representation of students' typical activity levels.²⁰

Primary outcomes were change in minutes of MVPA and sedentary activity during the school day. The school day lasted 6 hours and 10 minutes for all schools during both school years, with the exception of one school, which had an additional 3 minutes during the 2011–2012 school year.

Cardiorespiratory fitness

Participants completed the 1-mile run as a measure of cardiorespiratory fitness. The 1-mile run has been validated against treadmill tests and established as a reliable indicator of cardiorespiratory fitness among youth.²²

Fruit and vegetable consumption

Research staff used digital images of students' lunch trays before and after meal consumption to measure students' FV intake at school. FV consumption was determined by estimating the portion eaten (in increments of 0.125) and multiplying by the serving size. To ensure standardization between data collection points, endpoint data collection was planned on days when lunch offerings most closely matched those at baseline. Lunches brought from home were excluded from analysis. This methodology has been shown to be a reliable measure of students' dietary intakes at school.^23,24

Dietary behaviors

Survey questions adapted from the School Physical Activity and Nutrition Questionnaire²⁵ and the Child Food Consumption Questionnaire²⁶ assessed consumption of milk, sweetened beverages, and snack foods for the previous day (“yesterday”). These measures have acceptable levels of reliability and validity based on correlation with a 24-hour dietary record in a slightly older cohort of students.²⁷ Questions were adapted for the younger cohort and were group administered by trained staff using examples of foods familiar to the students.

Dietary knowledge

Dietary knowledge scores were computed from responses to six questions on the student survey tailored to the curriculum content (nutrition facts label reading, identification of nutrient-dense snacks and beverages, and messages from MyPlate²⁸).

School food offerings

On 24 randomly selected days during the second year of the evaluation (October–November 2012 and January–April 2013), district menu production records were collected, including serving sizes, quantities ordered, and quantities served. Researchers calculated mean cups of FV ordered per student.

Adiposity (anthropometric measures)

Standing height (in centimeters) and weight (in kilograms) were obtained from all participating students using National Health and Nutrition Examination Survey anthropometry standardized protocols at baseline and endpoint.²⁹ These measurements were used to calculate BMI z-scores using the 2000 CDC Growth Charts.³⁰

Statistical analyses

Mixed-effects linear regression models were used with group as the predictor, change in each outcome as the dependent variable, and school as a random effect to account for clustering. Adjusted models controlled for sex, race, age, and the baseline value of the outcome. In post-hoc analyses, interaction terms were included to determine whether the impact of the program at endpoint differed by grade. Stata/MP (version 13; StataCorp LP, College Station, TX) and SAS software (V9.3; SAS Institute Inc., Cary, NC) were used.

Results

Table 1 presents baseline characteristics for study participants with complete data. In this mobile population, 212 participants (24%) left the school district before endpoint data collection. Complete data were available for: 78% of students with baseline accelerometer data; 77% of students with baseline fitness data and baseline dietary surveys; and 78% of students with baseline BMI. Students who were lost to follow-up were more likely to identify as Asian (p<0.05), compared with students who remained through endpoint data collection. There were no significant differences in age, gender, weight status, or baseline PA.

Table 1.

Demographic Characteristics of Students with Complete Data at Baseline (Fall 2011) and Endpoint (Spring 2013)

	Intervention		Control
Characteristic	Frequency	Percent	Frequency	Percent	p value^c
Accelerometer data (N=450)
Gender
Male	156	51.5	65	44.2	0.148
Female	147	48.5	82	55.8
Grade
Fourth	163	53.8	76	51.7	0.676
Fifth	140	46.2	71	48.3
Race/ethnicity
Asian	23	7.6	21	14.3	0.047
Black	16	5.3	15	10.2
Latino	156	51.5	66	44.9
Multiracial	48	15.8	19	12.9
White	17	5.6	11	7.5
Other^b	43	14.2	15	10.2
Fitness data (N=461)
Gender
Male	156	51.5	64	42.3	0.089
Female	152	48.9	86	57.3
Grade
Fourth	165	53.1	78	52.0	0.832
Fifth	146	47.0	72	48.0
Race/ethnicity
Asian	23	7.4	24	16.0	0.005
Black	18	5.8	16	10.7
Latino	165	53.1	70	46.7
Multiracial	49	15.8	16	10.7
White	16	5.1	12	8.0
Other^b	40	12.9	12	8.0
Measured height and weight sample (N=676)
Gender
Male	228	51.1	105	45.7	0.194
Female	218	48.9	125	54.4
Grade
Third	125	28.0	68	29.6	0.904
Fourth	167	37.4	83	36.1
Fifth	154	34.5	79	34.4
Race/ethnicity
Asian	33	7.4	36	15.7	0.018
Black	40	9.0	27	11.7
Latino	236	52.9	105	45.7
Multiracial	65	14.6	29	12.6
White	30	6.7	14	6.1
Other^b	42	9.4	19	8.3
Lunch tray photograph sample (N=400)
Gender
Male	129	46.9	59	47.2	1.000
Female	146	53.1	66	52.8
Grade^a
Third	85	30.9	48	38.4	0.272
Fourth	95	34.6	42	33.6
Fifth	95	34.6	35	28.0
Race/ethnicity
Asian	19	6.9	20	16.0	0.0401
Black	20	7.3	11	8.8
Latino	150	54.6	60	48.0
Multiracial	31	11.3	18	14.4
White	17	6.2	6	4.8
Other^b	38	13.8	10	8.0
Student survey sample (N=648)
Gender
Male	217	51.4	104	46.0	0.216
Female	205	48.6	122	54.0
Grade
Third	119	28.2	68	30.1	0.654
Fourth	163	38.6	79	35.0
Fifth	140	33.2	79	35.0
Race/ethnicity
Asian	32	7.6	34	15.0	0.052
Black	38	9.0	26	11.5
Latino	226	53.6	104	46.0
Multiracial	61	14.5	30	13.3
White	28	6.6	13	5.8
Other^b	37	8.8	19	8.4

Grade reported by the student at baseline.

Other variable includes students that marked American Indian/Alaskan Native, Native Hawaiian or Pacific Islander, “Other” category as well as nonresponses to the student survey.

p value based on chi-square or Fisher's exact test.

Overall, 24% of students were classified as overweight and 22% were classified as obese. EB4K with Play schools had significantly more Latino students (58% vs. 45%; p<0.01) and fewer Asian students (7% vs. 15%; p<0.01) than control schools. Additionally, there was a statistically significant, but not meaningful, difference in age between the two study groups (p<0.05).

Physical Activity Results

During the 5 days that students were instructed to wear their accelerometers, average wear time was 87% at baseline and 71% at follow-up, with no significant differences between intervention and control school students.

Table 2 presents change from baseline to endpoint in minutes of school-day MVPA and sedentary activity for EB4K with Play students, compared to control students. In adjusted models, there were no differences between groups in change in MVPA or sedentary time. In post-hoc analyses, students' grade significantly modified the effect of the program. As shown in stratified analyses in Table 3, students who began the intervention in fourth grade significantly decreased their sedentary activity, relative to controls (−15.2; p=0.024), with no significant change for fifth graders. Impact on MVPA diverged by grade, with fourth graders increasing by 3.1 minutes (p=0.095) and fifth graders decreasing by 3.2 minutes (0.199), but neither reached statistical significance.

Table 2.

Student-Level Data at Baseline and Endpoint, with Adjusted Change Scores

	Intervention			Control
	Baseline	Endpoint	Adjusted	Baseline	Endpoint	Adjusted	Adjusted difference
	Mean±SD	Mean±SD	Change^a±SE	Mean±SD	Mean±SD	Change^a±SE	[95% CI]
Accelerometer data (n=450)
School-day MVPA (mins)	22.1±8.8	21.3±10.0	−0.7±1.6	21.8±8.2	21.0±10.0	−1.0±2.1	−0.1 [−4.5, 4.3]
School-day sedentary (mins)	240.9±29.9	245.1±35.5	4.0±5.4	248.7±28.8	256.1±29.4	7.3±6.8	−6.7 [−20.3, 6.8]
Fitness data (n=461)
Mile run time (mins)	12.9±3.0	11.5±2.4	−1.4±0.3	12.8±2.5	11.6±2.2	−1.1±0.3	−0.2 [−0.8, 0.4]
Lunch tray photographic data (n=400)
Fruit consumed (cups)	0.4±0.5	0.3±0.2	−0.1±0.0	0.2±0.3	0.2±0.3	−0.1±0.0	0.0 [−0.1, 0.1]
Vegetable consumed (cups)	0.2±0.3	0.1±0.2	−0.1±0.0	0.3±0.4	0.2±0.3	−0.0±0.1	−0.0 [−0.2, 0.1]
Survey data (n=648)
For “yesterday,” number of times student reported:
Drinking milk	1.4±0.9	1.5±0.9	0.0±0.0	1.3±0.9	1.4±0.9	0.0±0.0	0.0 [−0.1, 0.1]
Drinking sweetened beverages	1.1±1.0	1.0±0.9	−1.0±0.0	1.1±0.9	1.1±1.0	0.0±1.0	−1.0 [−0.2, 0.1]
Eating at restaurants	0.4±0.8	0.3±0.6	−0.0±0.0	0.4±0.7	0.4±0.7	0.0±0.0	−0.0 [−0.1, 0.0]
Eating salty or sweet snacks	1.7±1.5	1.6±1.3	−0.1±0.0	1.7±1.4	1.6±1.2	−0.0±0.0	−0.0 [−0.0, 0.0]
Dietary knowledge score	2.4±1.4	4.6±1.4	1.1±0.1	2.5±1.5	2.8±1.5	0.1±0.1	0.9 [0.6, 1.2]
Anthropometric data (n=676)
BMI z-score	1.03±1.06	0.93±1.08	−0.05±0.03	0.87±1.12	0.86±1.06	0.01±0.01	−0.07 [−0.14, 0.00]

Change analyses were adjusted for baseline grade, gender, and ethnicity and account for clustering by school.

MVPA, moderate-to-vigorous actiivity; SD, standard deviation; SE, standard error; CI, confidence interval.

Bold figures are significant at p<0.05.

Table 3.

Stratified Analyses for Physical Activity and Anthropometry Outcomes

	Intervention			Control
	Baseline	Endpoint	Adjusted	Baseline	Endpoint	Adjusted	Adjusted difference
	Mean±SD	Mean±SD	Change^a±SE	Mean±SD	Mean±SD	Change^a±SE	[95% CI]
Accelerometer data (n=450)
School-day MVPA (mins)
Fourth grade	21.9±8.8	23.6±10.4	2.4±1.8	22.5±9.0	19.9±7.4	−0.8±2.2	3.1 [−0.5, 6.7]
Fifth grade	22.2±8.7	18.8±8.8	−3.6±2.2	21.2±7.4	22.2±12.0	−0.4±2.6	−3.2 [−8.1, 1.7]
School-day sedentary (mins)
Fourth grade	239.8±29.5	238.9±33.6	−0.6±6.6	239.3±28.3	255.2±28.2	11.3±7.9	−15.2 [−28.3, −2.0]
Fifth grade	242.2±30.5	252.4±36.4	11.0±8.1	258.2±26.0	257.0±30.9	4.7±9.3	1.1 [−15.1, 17.2]
Anthropometric data (n=675)
BMI z-score
Third grade	1.05±1.00	0.92±1.03	−0.06±0.01	0.84±1.08	0.87±1.02	0.00±0.02	−0.06 [−0.10, −0.02]
Fourth grade	1.05±1.12	0.95±1.15	−0.01±0.02	0.92±0.98	0.83±0.98	−0.02±0.02	0.01 [−0.07, 0.08]
Fifth grade	0.98±1.05	0.92±1.06	−0.04±0.02	0.84±1.29	0.86±1.19	−0.01±0.01	−0.02 [−0.07, 0.02]

Change analyses were adjusted for baseline grade, gender, and ethnicity and account for clustering by school.

MVPA, moderate-to-vigorous actiivity; SD, standard deviation; SE, standard error; CI, confidence interval.

Bold figures are significant at p<0.05.

There was no significant change in mile run time between intervention and control school students in overall or stratified models.

Student Dietary Results

Based on lunch-tray photographic data from a subsample of 400 participants, there was no between-group difference in lunchtime FV consumption. Based on student survey data for the larger group, intervention students reported a relative 10% reduction in eating out, compared to control students (p=0.05). No other between-group differences in self-reported behaviors were significant. There was a statistically significant improvement in students' dietary knowledge at intervention schools, compared to control schools (Table 2). Students who had participated in the program had a better understanding of how to choose more nutrient-dense foods, read nutrition facts labels, and understand the MyPlate messages.

School-Level Dietary Results

Schools in the intervention group ordered significantly more cups of vegetables per student, compared to the control schools, during the second year of the intervention, as indicated by the menu production records (0.4±0.0 vs. 0.3±0.0; p<0.05). There was no significant difference observed in mean cups of fruit per student ordered (0.5±0.0 vs. 0.5±0.0; p=0.167).

Adiposity

Overall reduction in BMI z-score over the 2 years almost reached statistical significance (p=0.05). Stratified models showed that this was largely driven by a significant improvement in BMI z-scores for intervention students in third grade at baseline (−0.06±0.01 vs. 0.00±0.02; p<0.05). No group differences were observed in the fourth (−0.01±0.02 vs. −0.02±0.02; p=0.833) or fifth grade (−0.04±0.02 vs. −0.01±0.01; p=0.260) cohorts.

Discussion

To our knowledge, EB4K with Play is the only school-based intervention to date that represents a partnership between two existing national organizations that promote youth health. In contrast to programs that are developed for, and funded by, research grants, programs that leverage existing infrastructure have a heightened chance of being sustained after the research is complete. Further, synchronization of efforts in two areas known to contribute to weight status—dietary intake and PA—reflects the belief that coordinated multifaceted programs can best confront the causes of youth obesity.

Overall, students in the present study did not increase their MVPA over 2 school years of exposure to EB4K with Play. Though disappointing, these results are not surprising given that EB4K with Play focused on engaging students in active play during existing PA times rather than increasing the amount of time dedicated to PA during the school day. Previous studies that have focused on increasing the quality of existing physical activities without increasing quantity have not successfully increased PA during the school day.^31,32 In contrast, the Kinder-und Jugendsportstudie trial, which provided first- and fifth-grade students with two additional 45-minute PA sessions per week, succeeded in increasing school-day MVPA by 13 minutes over 1 school year.³³ These findings suggest that it may be difficult to significantly increase MVPA without increasing time allotted for it.³⁴

Post-hoc analyses in the present study revealed a differential impact of EB4K with Play by grade on student activity levels. Exposed students who moved from fourth to fifth grade during the study significantly decreased their minutes of sedentary activity, and tended to increase their MVPA, wheraes older students showed no improvements. Differences by grade may represent the effect of developmental changes that coincide with onset of adolescence, specifically onset of self-consciousness, which has been shown to influence youths' activity choices.³⁵ As a result, older students may balk at imposed structured activities, which could explain the largely null results to date of school-based PA interventions among middle school youth.^31,36

To the extent that the post-hoc analyses among fourth-grade students represent the intervention's impact among children in younger grades, the 15-minute decrease in school-day sedentary time could represent a clinically meaningful impact. Greater time spent in sedentary activities is associated with higher prevalence of obesity, metabolic syndrome, and hypertension among youth.^6,37

Similar to PA results, researchers found no overall improvement in students' adiposity levels as measured by BMI z-scores, but stratified analyses revealed a significant improvement among the youngest intervention group students. A number of teachers reported that the nutrition education and motivational activities were better suited for younger students, which may explain the differential impact by grade. This finding is notable given that the intervention lasted only 20 months and BMI z-score change is often not achieved in that time frame. It is also consistent with other school-based obesity prevention programs that have found positive results in younger elementary age students.^14,15 Studies of school-based interventions that are adequately powered to examine impact across grade strata would help to identify the most developmentally appropriate program curriculum.

Although intervention schools ordered significantly more vegetables at endpoint than control schools, changes in FV consumption during lunch did not differ between intervention and control students. This finding may stem from changes in Federal National School Lunch Program standards that occurred during the intervention. At endpoint, 100% fruit juice could only be used at only one meal to meet the daily fruit requirement. Therefore, the two intervention schools that, at baseline, had substituted juice for whole fruit offerings when they ran out of fruit during lunchtime could no longer use juice for this purpose at endpoint. This regulatory change inhibited our ability to match menus pre- and post-test. Further, in some instances, different or additional vegetables were served at the two data collection points owing to foods being carried over from previous school meals, also reducing our ability to accurately identify consumption changes over time.

There were no significant changes found in students' self-reported dietary data. It is likely that the student survey was not sensitive enough to detect student-level dietary changes. Multiday diet records might have been more sensitive; however, the cost of tool administration, data entry, and data analysis were prohibitive in this study.

The evaluation did find that more vegetables were available to students at intervention schools than control schools during the second school year. Although new packaging equipment purchased for use by the district enabled all schools to increase FV offerings, the increase was only observed in the intervention schools. This indicates that equipment and kitchen modernization can facilitate change, but may not be enough on their own to alter behavior. RD coaches worked with each intervention school's energy balance team to facilitate the increase in offerings.

This study had limitations that should be considered. The study took place in a low-income, urban school district in Northern California and may not generalize to other settings. Though the infrastructure for the intervention exists, schools would have to identify funding to pay for the program. Loss to follow-up was close to 25% for some outcomes, which could introduce bias into our results. Whereas grade significantly modified the effect of the intervention, observing differences by grade was not an a priori hypothesis; this increases the likelihood of a type I error. Because only fourth- and fifth-grade students wore accelerometers during this evaluation, there are no additional data to support the conclusion that EB4K with Play reduces sedentary time more effectively in younger students. Program implementation may have benefitted from greater coordination between the RD and Playworks coaches. Whereas Playworks coaches are extensively trained in behavioral management, teachers reported during year 1 that RD coaches needed to improve their classroom management and student engagement skills. Having Playworks train the RD coaches more intensively in these skills could enhance future EB4K with Play program implementation.

Conclusions

Findings from this study contribute to the literature of school-based nutrition and PA interventions by evaluating a partnership of existing and complementary organizations as a means for delivering sustainable programming in schools. The present evaluation suggests that EB4K with Play improves dietary knowledge and may have a positive impact on PA and adiposity, particularly in younger children. Future interventions should consider the differential impact that PA and dietary interventions may have on students by age, with a focus on programming that is effective for students approaching adolescence. Additionally, future interventions should strive to improve the nutrition environment and increase time available for youth to participate in PA during the school day, given that changing children's environments is likely to have the greatest impact on their behaviors.

Footnotes

Acknowledgments

This study was funded by the Academy of Nutrition and Dietetics Foundation (Chicago, IL). The authors acknowledge the RD Nutrition Coaches: Sandra Brown, Anca Cunningham, Elena Huey, Susie Liles, Ruth Schrager, Karaliese Brown, Jill West, Lisa Medrow, and Kuda Chimanya; the Playworks Recess Coaches: Peter Best, Laura Kocourek, Paul Pasternak, and Grace Reinhalter; the Playworks staff: Alison Towney, Jackie Hemann, Evan Kennedy, and Lisa Clemson; and Academy of Nutrition and Dietetics staff: Jan Foster.

Author Disclosure Statement

No competing financial interests exist.

References

Ogden

, Carroll

, Kit

, et al. Prevalence of childhood and adult obesity in the United States, 2011–2012. JAMA, 2014; 311:806–814.

May

, Kuklina

, Yoon

. Prevalence of cardiovascular disease risk factors among US adolescents, 1999–2008. Pediatrics, 2012; 129:1035–1041.

Freedman

, Mei

, Srinivasan

, et al. Cardiovascular risk factors and excess adiposity among overweight children and adolescents: The Bogalusa Heart Study. J Pediatr, 2007; 150:12–17.e2.

Singh

, Siahpush

, Kogan

. Rising social inequalities in US childhood obesity, 2003–2007. Ann Epidemiol, 2010; 20:40–52.

Ebbeling

, Leidig

, Sinclair

, et al. A reduced-glycemic load diet in the treatment of adolescent obesity. Arch Pediatr Adolesc Med, 2003; 157:773–779.

Must

, Tybor

. Physical activity and sedentary behavior: A review of longitudinal studies of weight and adiposity in youth. Int J Obes (Lond), 2005; 29(Suppl 2):S84–S96.

Ekelund

, Luan

, Sherar

, et al. Moderate to vigorous physical activity and sedentary time and cardiometabolic risk factors in children and adolescents. JAMA, 2012; 307:704–712.

Hess

, Slavin

. Snacking for a cause: Nutritional insufficiencies and excesses of U.S. children, a critical review of food consumption patterns and macronutrient and micronutrient intake of U.S. children. Nutrients, 2014; 6:4750–4759.

National Research Council. Accelerating Progress in Obesity Prevention: Solving the Weight of the Nation. The National Academies Press: Washington, DC, 2012.

10.

CDC. School health guidelines to promote healthy eating and physical activity. MMWR Morb Wkly Rep, 2011; 60:1–76.

11.

White House Child Nutrition Fact Sheet. Child Nutrition Reauthorization: Healthy, hunger-free kids act of 2010. 2010. Available at www.whitehouse.gov/sites/default/files/Child_Nutrition_Fact_Sheet_12_10_10.pdf Last accessed July 25, 2014.

12.

Williams

, Henley

, Williams

, et al. Systematic review and meta-analysis of the association between childhood overweight and obesity and primary school diet and physical activity policies. Int J Behav Nutr Phys Act, 2013; 10:101.

13.

Lavelle

, Mackay

, Pell

. Systematic review and meta-analysis of school-based interventions to reduce body mass index. J Public Health, 2012; 34:360–369.

14.

Manger

, Manger

, Minno

, et al. Obesity prevention in young schoolchildren: Results of a pilot study. J Sch Health, 2012; 82:462–468.

15.

Lytle

, Murray

, Perry

, et al. School-based approaches to affect adolescents' diets: Results from the TEENS study. Health Educ Behav, 2004; 31:270–287.

16.

Myers

, Gerstein

, Foster

, et al. Energy balance for kids with play: Design and implementation of a multi-component school-based obesity prevention program. Child Obes, 2014; 10:251–259.

17.

USDA, Food and Nutrition Service. HealthierUS School Challenge: Application criteria and guidance. 2014. Available at www.fns.usda.gov/hussc/application-criteria-and-guidance Last accessed November 21, 2014.

18.

Trost

, Ward

, Moorehead

, et al. Validity of the computer science and applications (CSA) activity monitor in children. Med Sci Sports Exerc, 1998; 30:629–633.

19.

Sasaki

, John

, Freedson

. Validation and comparison of ActiGraph activity monitors. J Sci Med Sport, 2011; 14:411–416.

20.

Alhassan

, Sirard

, Spencer

, et al. Estimating physical activity from incomplete accelerometer data in field studies. J Phys Act Health, 2008; 5(Suppl 1):S112–S125.

21.

Evenson

, Catellier

, Gill

, et al. Calibration of two objective measures of physical activity for children. J Sports Sci, 2008; 26:1557–1565.

22.

Castro-Pinero

, Mora

, Gonzalez-Montesinos

, et al. Criterion-related validity of the one-mile run/walk test in children aged 8–17 years. J Sports Sci, 2009; 27:405–413.

23.

Swanson

Digital photography as a tool to measure school cafeteria consumption. J Sch Health, 2008; 78:432–437.

24.

Hanks

, Wansink

, Just

. Reliability and accuracy of real-time visualization techniques for measuring school cafeteria tray waste: Validating the quarter-waste method. J Acad Nutr Diet, 2014; 114:470–474.

25.

Thiagarajah

, Fly

, Hoelscher

, et al. Validating the food behavior questions from the elementary school SPAN questionnaire. J Nutr Educ Behav, 2008; 40:305–310.

26.

Fernandes

MM.

The effect of soft drink availability in elementary schools on consumption. J Am Diet Assoc, 2008; 108:1445–1452.

27.

Hoelscher

, Day

, Kelder

, et al. Reproducibility and validity of the secondary level School-Based Nutrition Monitoring student questionnaire. J Am Diet Assoc, 2003; 103:186–194.

28.

USDA. Choose MyPlate food groups. 2011. Available at www.choosemyplate.gov/food-groups Last accessed July 28, 2014.

29.

CDC. National Health and Nutrition Examination Survey (NHANES): Anthropometry Procedures Manual. CDC: Atlanta, GA, 2007.

30.

Kuczmarski

, Ogden

, Guo

SSea

. CDC growth charts for the United States: Methods and development. Vital Health Stat, 11 2002;(246):1–190. Available at www.cdc.gov/growthcharts Last accessed August 16, 2004.

31.

Grydeland

, Bergh

, Bjelland

, et al. Intervention effects on physical activity: The HEIA study—A cluster randomized controlled trial. Int J Behav Nutr Phys Act, 2013; 10:17.

32.

Magnusson

, Sigurgeirsson

, Sveinsson

, et al. Assessment of a two-year school-based physical activity intervention among 7–9-year-old children. Int J Behav Nutr Phys Act, 2011; 8:138.

33.

Kriemler

, Zahner

, Schindler

, et al. Effect of school based physical activity programme (KISS) on fitness and adiposity in primary schoolchildren: Cluster randomised controlled trial. BMJ, 2010; 340:c785.

34.

Herrick

, Thompson

, Kinder

, et al. Use of SPARK to promote after-school physical activity. J Sch Health, 2012; 82:457–461.

35.

Elkind

, Bowen

. Imaginary audience behavior in children and adolescents. Dev Psychol, 1979; 15:38–44.

36.

Jago

, McMurray

, Drews

, et al. HEALTHY intervention: Fitness, physical activity, and metabolic syndrome results. Med Sci Sports Exerc, 2011; 43:1513–1522.

37.

Tremblay

, Colley

, Saunders

, et al. Physiological and health implications of a sedentary lifestyle. Appl Physiol Nutr Metab, 2010; 35:725–740.