The Impact of an Elementary After-School Nutrition and Physical Activity Program on Children's Fruit and Vegetable Intake, Physical Activity, and Body Mass Index: Fun 5

Abstract

Objective

This study evaluated the impact of the Fun 5 program on fruit and vegetable intake, physical activity, and body mass index (BMI) percentile of overweight and obese children in Hawai'i's A+ After-School Program.

Methods

Children in 4th – 6th grades (n=119, 55% female) from six randomly selected schools participated in this longitudinal study. Time 1 (T1) measures were taken October 2007 and time 2 (T2) measures April 2008. Height and weight along with a self-report survey were used to measure fruit and vegetable intake, physical activity, and BMI.

Results

The “at risk” population, defined as fruit and vegetable intake <5 servings per day (n=30), physical activity <300 minutes per week (n=78), or Body Mass Index >85th percentile (n=48) at T1, showed a significant increase in fruit and vegetable intake, from 2.97 (±1.16) servings per day at T1 to 5.60 (±3.93) servings per day at T2 (P=<0.01) and physical activity, from 125.26 (±76.03) minutes per week of physical activity at T1 to 222.18 (±180.90) minutes per week at T2 (P=<0.01) and no change in Body Mass Index.

Conclusion

The Fun 5 program had an impact on improving fruit and vegetable intake and physical activity on the at risk population.

Introduction

From 1999 to 2004, the percent of overweight children ages 6 to11 in the United States increased from 29.8% to 37.2%,¹ representing a 25% increase over a five year period. These statistics are consistent with the increased number of overweight children in Hawai‘i. In 1984, a study found no significant under or over weight children in Hawai‘i's schools but by 2002, 33% of the children entering kindergarten were overweight.²

Fruit and vegetable (FV) intake is negatively correlated with overweight or obesity in children.^3,4 The U.S. Department of Health and Human Services recommends 5 servings of FV a day.⁵ However, the majority of Hawai‘i's school children are not consuming the recommended 5 servings of FV per day.⁶

As is the case with FV intake, participation in physical activity (PA) is also a factor in childhood obesity. As the number of over-weight and obese children has risen over the years, there has been a decrease in time spent participating in physical activity and an increase in time spent in sedentary activities.^7–9 Research shows children with higher PA levels have lower fat mass levels.¹⁰ Because energy balance is dependent on both intake and expenditure, most weight loss or weight maintenance recommendations emphasize proper nutrition and increased levels of PA.

In 2002, the Fun 5 Physical Activity & Nutrition program partnered with the Hawai‘i State Department of Education After-School Plus (A+) program to increase PA, improve nutritional status, and reduce the number of overweight children in grades K-6 throughout the State of Hawai‘i. In the pilot year (2003) and the following dissemination school year (2004/2005), the Fun 5 program was successful in increasing moderate and vigorous PA and FV intake.^11–13 Although the improvements in PA and FV intake are encouraging, the effect of the Fun 5 Program on the children's body mass index (BMI) has yet to be investigated. Further, the effect of Fun 5 on “at risk” children — those who would stand to gain most from changing their behaviors, defined as children that consume <5 servings per day of FV, participate in <300 minutes per week of PA, or have a BMI ≥85th percentile — has not been documented to date. Therefore, the purpose of this research was to replicate the overall impact of the Fun 5 Program on children's FV intake, PA, and BMI and investigate the impact of Fun 5 for at risk children.

Methods

Participants

Six O‘ahu public schools from A+ After-School program participating in Fun 5 were randomly selected from 115 schools using a random number table. It was estimated that six O‘ahu schools would provide the number of children (n=79) needed to allow for meaningful results. The study population included school age children grades 4th–6th. The University of Hawai'i, Committee on Human Subjects, approved this research project. Only students with signed parental consent forms were allowed to participate in the study.

Intervention

The Fun 5 program is designed as a train-the-implementers process and relies in part on qualified Sports, Play, and Active Recreation for Kids, Active Recreation (SPARK AR) trainers within the A+ system. Training sessions are done annually in the beginning of the school year for all participating site coordinators, group leaders, and program aides. During training sessions, A+ staff are taught the skills necessary to implement and maintain the program. Training sessions include introduction to the Fun 5 program, SPARK AR physical activities, the nutrition component, and overview of the program evaluation.

The SPARK AR component has been developed for all out-of-physical education PA programs and is designed to provide substantial opportunities for all children to actively engage in movement. The SPARK program focuses on the development of a variety of basic motor and manipulative skills, such as throwing, catching, kicking, developing positive social skills, and the ability to get along with others by reinforcing ideas such as sharing equipment and demonstrating cooperative behavior. The goal of SPARK is to increase participation in activities and personal physical skill levels, while increasing confidence in the ability to be physically active and promoting a positive attitude toward PA and health.¹⁴ The SPARK AR curriculum includes sections on management (eg, grouping, distributing and handling equipment, making teams) and instruction for inclusive activities that include four main areas: great games (eg, tag games, ball games), super sports (eg, soccer, Frisbee), dynamic dance, and other activities (eg, jump rope, relays). These activities are designed to be transferable to leisure activity in other settings (eg, home, park). SPARK AR is easy to use, as each activity is described and word-for-word instructions provided along with diagrams.

The nutrition intervention developed by Fun 5 program staff has an emphasis on FV and includes art projects that are designed to promote positive association with eating FV, for example creating posters with a rainbow of FV. Interactive nutrition booklets are provided for the children and incorporate information on FV in the form of coloring pages, cross-words, word searches, etc. During the training session, group leaders are encouraged to be healthy role models not bringing large sodas or fast food with them to the A+ program, but to be model eaters of healthy snacks during the A+ time. Positive reinforcement techniques were also encouraged at the training sessions which include giving high-fives when children bring healthy snacks.

Measures

FV intake was assessed by asking: “How many servings of fruits do you eat each day?” and the same question was asked about vegetables. An example of a serving size was included as part of the question. The single items addressing the average number of fruits and the average number of vegetables eaten each day have documented validity and reliability in adolescents¹⁵ and are positively related to the five stages of change for FV intake in children.¹⁶

PA was assessed using an adaptation of Godin & Shephard's Leisure-Time Exercise Questionnaire.^17,18 Participants indicate how many days per week they engage in strenuous, moderate, and mild PA for 0, 10, 20, 30, 40, 50, or 60+ minutes when they are not in school. Levels of PA activity were defined as part of each question. In adults, the instrument was found to be significantly related to caltrac accelerometer readings (r=0.32), metabolic equivalents (METs; r=0.36), treadmill exercise time (r=0.57), percentage of body fat (r=−0.43), and VO_2max (r=0.56).¹⁹ The instrument is also significantly related to the five stages of change for PA across populations^20,21 including children.²² Sallis and colleagues (1993) reported good test-retest reliability (r=0.81) and adequate validity (r=0.39) when compared to kilocalories expended per day in a sample of 5th, 8th, and 11th graders.²³

Weight was measured by gender-matched reseachers to the nearest .1 kg using a portable Health-o-Meter digital scale with a capacity of 330 lbs.

Height was measured by gender-matched reseachers to the nearest .1 cm using a Seca 216 Accu-Hite Stadiometer.

BMI was calculated using the CDC's online BMI calculator. The CDC BMI-for-age weight growth charts were used to establish the BMI status based on percentile ranking. BMI rakings were categorized as a BMI percentile <5th percentile underweight, 5th to <85th percentile healthy weight, 85th to <95th percentile overweight, and ≥95th percentile obese.²⁴

Treatment fidelity was assessed via yearly interviews with all A+ site coordinators who reported the incorporation of PA at least three times a week and the accessibility to nutritional materials, which were summarized as the percentage of sites attaining these goals. Relatedly, implementation quality was the extent of consistency between program execution and research protocol which was evaluated through the 15-item SPARK session checklist¹⁴ and summarized as a percentage during unscheduled visits at 10% of randomly selected sites (note: there is no protocol for a nutrition treatment fidelity site visit).

Procedures

The self-report measures were collected at the beginning and end of the 2007/2008 school year [October: Time 1 (T1) and April: Time 2 (T2)]. Data collection was repeated at T2 on those students who completed the first set of measures. This allowed for longitudinal comparison within the school year. The Statistical Package for the Social Sciences (SPSS) 16.0 and SPSS 17.0 were used for statistical analysis. A paired t-test (α = 0.05) was used to analyze the impact of Fun 5 on FV intake, PA, and BMI from T1 to T2.

Results

Demographic characteristics are presented in Table 1. One hundred and nineteen children completed both T1 and T2 measures. Fifty-five percent of the population (n=66) was female. The 4th grade represents 46% of the population (n=55), the 5th grade 40% (n=47), and 14% of the sample was 6th graders (n=17). Note: The 5th grade is the highest grade in most Hawai‘i elementary schools, which explains the smaller percentage of participants from the 6th grade. Children who completed T1 only (n=134, 54% female, 44% grade 4, 40% grade 5, 16% grade 6; data not shown) were not different demographically from children who completed T1 and T2.

Table 1.

Gender, School, and Grade Distribution of Study Population (n=119)

Variable		n	%
Gender	Male	53	45
	Female	66	55
School	Site 1	29	24
	Site 2	14	12
	Site 3	23	19
	Site 4	27	23
	Site 5	8	7
	Site 6	18	15
Grade	4th	55	46
	5th	47	40
	6th	17	14

At T1 the children reported eating 7.67 (±4.68) and at T2 7.54 (±4.87) servings of FV per day (P=0.78). For PA, at T1 it was 252.35 (±220.09) and at T2 272.00 (±222.62) minutes of moderate to vigorous activity per week (P=0.37). The mean for BMI at both T1 and T2 was 71st (±25.13) percentile (P=0.97). Seventy-one out of 119 children were in the normal BMI category at both T1 and T2. The number of children in the obese category increased by 2 and the number of children in the overweight category decreased by 2 at T2 (Table 2).

Table 2.

Impact of the Fun 5 Program on Fruit and Vegetable Intake, Physical Activity, and Body Mass Index in Children Grades 4th, 5th, and 6th

Variables		n	Time 1	Time 2	t (P)
Overall Population
FV Intake (servings/day)	Mean ± S.D.	119	7.67 ± 4.68	7.54 ± 4.87	0.28 (0.78)
	Median		6.00	6.00
	% Increased				41.0
	% Same				13.7
	% Decreased				45.3
PA (min. of mod & vig. PA/week)	Mean ± S.D.	119	252.35 ± 220.09	272.00 ± 222.62	0.90 (0.37)
	Median		190.00	210.00
	% Increased				51.3
	% Same				6.1
	% Decreased				42.6
BMI Percentile	Mean ± S.D.	119	71.49 ± 25.13	71.52 ± 24.63	−0.04 (0.97)
	Median		80.00	80.00
	% Increased				46.2
	% Same				16.0
	% Decreased				37.8
BMI Category			n	_2 (p)	df
Normal		71	71
Overweight		27	25
Obese		21	23
Total		119	119	0.34 (0.84)	2
At Risk Population
1FV Intake (servings/day)	Mean ± S.D.	30	2.97 ± 1.16	5.60 ± 3.93	−3.74 (<0.01)
	Median		3.00	4.50
	% Increased				73.3
	% Same				13.3
	% Decreased				13.4
2PA (min. of mod & vig. PA/week)	Mean ± S.D.	78	125.26 ± 76.03	222.18 ± 180.90	−4.94 (<0.01)
	Median		120.00	160.00
	% Increased				65.8
	% Same				3.8
	% Decreased				30.4
3BMI Percentile	Mean ± S.D.	48	92.98 ± 4.61	92.31 ± 6.27	1.30 (0.20)
	Median		93.00	94.00
	% Increased				35.4
	% Same				35.5
	% Decreased				29.2

Note: FV = Fruit & Vegetables, PA = Physical Activity, BMI = Body Mass Index.

¹At risk defined as < 5 servings of FV per day at time 1.

²At risk defined as < 300 min. of moderate and vigerous PA per week at time 1.

³At risk defined as ≥ 85th percentile at time 1.

In addition to evaluating the overall impact of the Fun 5 program, data analysis was run to evaluate the impact of the Fun 5 program on the “at risk” populations. The at risk population was defined as children that consumed <5 servings per day of FV, participated in <300 minutes per week of PA, or had a BMI ≥85th percentile at Time 1. Thirty-one children (26%) reported eating <5 servings per day of FV, 80 children (67%) reported participating in <300 minutes per week of PA, and 48 children (40%) had a BMI ≥85th percentile at T1. The at risk population showed a significant increase in FV intake (P=<0.01) reporting an average FV intake of 2.97 (±1.16) servings per day at T1 and 5.60 (±3.93) servings per day at T2. As with FV intake, PA for the at risk population showed a significant increase (P=<0.01) in minutes per week of PA, reporting an average of 125.26 (±76.03) minutes per week at T1 and 222.18 (±180.90) minutes per week at T2. There were no significant changes in BMI percentile for the at risk population (P=0.20). The median and percent change supports the above data (Table 2).

Treatment fidelity indicators revealed a high percentage of successful implementation. The decision to execute PA at least three times per week was high (97%), accessibility to nutritional materials was high (85%), and the percentage of proper program implementation was also high (80%).

Discussion

The Fun 5 program had a positive effect on FV intake (p=<0.01) and minutes of moderate and vigorous PA (p=<0.01) in the “at risk” population. Although the sample size of the children reporting <5 servings per day of FV was small (n=30), these children reported an average increase of 2.63 serving per day from T1 to T2. At T2, this population of children was meeting their recommended FV intake of 5 servings per day reporting an average FV intake of 5.60 servings per day. Seventy-eight children were participating in <300 minutes per week of PA at T1 and showed an average of 96.92 minutes per week or a 77% increase of PA at T2. These results are encouraging and indicate that the Fun 5 Program has a positive impact on FV intake and PA in the at risk population.

The overall effectiveness of the Fun 5 program on children's FV intake and PA is difficult to determine. At both T1 and T2 the children reported approximately of 2.5 servings above the recommended five servings of FV per day. Children are currently either eating the recommended number of FV servings per day, or possibly over reporting their FV intake. Research in the area of FV intake measures has shown a tendency to over report FV intake.²⁵ Over reporting may be the result of social desirability, an indication that these children are aware of the emphasis on FV intake, or that the Fun 5 program is successful in improving the children's awareness of the importance of eating more FV.

Although not significant, participants in the Fun 5 program reported an average increase of about 20 minutes per week of moderate and vigorous PA from T1 to T2. These results are promising as any improvement in PA is beneficial.²⁶ In addition, using the same self-report measures as this study, the Fun 5 program has consistently shown improvements in time spent in PA¹³. Similar school based intervention programs resulted in little or no change in physical activity. For example, no intervention effects were observed for activity or inactivity among the Baylor GEMS, Eat Well and Keep Moving, and HIP-HOP to Health Jr. interventions.²⁷

The children's BMI percentile averaged in the 71st percentile at both T1 and T2 and no significant difference in the number of children in the overweight and obese categories was seen. Because children are growing, the observation of no change in BMI percentile and BMI categories is encouraging. A review of the effectiveness of school-based obesity intervention programs by Baranowski, et al, reveals that the BMI results are comparable to similar intervention programs.²⁸

Of concern is the children's BMI average in the 71^nd percentile and 48 children in the overweight and obese categories, which represented 40% of this population. Currently, the prevalence of overweight and obese US children is 31.9%.²⁹ Because this is a relatively small sample size the number of overweight and obese children in this study may not necessarily represent the A+ After School program population. Still these statistics warrant the need for ongoing research to combat the large number of obese and overweight children in Hawai‘i's public after-school programs.

The measures employed, although practical, are not without limitations. BMI measurements may not accurately measure the degree of adiposity in children.³⁰ A three-page survey was developed to measure strenuous, moderate, and mild activity; sedentary behaviors, self efficacy, fruit intake and vegetable intake; knowledge of healthy eating; and parent modeling. The three-page survey proved to be challenging because of the children's relatively low attention spans. Also, collecting data in the field has the limitation of not being able to control the environment in which the measurements are taken. In order to collect the information, brief instruments were used to measure FV intake and PA. Objective observation, 24-hour recall for FV and PA, food frequency questionnaire, or accelerometers may produce more accurate data. Another limitation of this study was the lack of a control group. Although the treatment fidelity results indicate proper implementation and current results replicated the behavioral effects of Fun 5, a comparison with a control group would have added further validity to the evaluation of the effects of the Fun 5 program.

Although the Fun 5 program was designed to increase PA and FV intake, which would theoretically reduce the prevalence of obesity among Hawai‘i's youth, there may be other positive outcomes to consider. Of interest would be improvements in academic performance and long-term reduction in development of chronic diseases such as diabetes and heart disease. Significant findings in any of the above areas would further validate the importance and benefits of incorporating obesity prevention programs such as the Fun 5 program into other programs in Hawai‘i's public schools and the community.

We seem to be a long way from seeing a reversal in the trend of increasing overweight and obesity among U.S. children. Better understanding of the role of diet and PA in the development and prevention of obesity should be complemented with better understanding of impacting behavioral changes. How much influence the media, food availability, and sedentary lifestyles have on childhood obesity are also empirical questions. Ongoing research in the field of behavioral change theories and understanding the role of society and the environment on childhood obesity would facilitate the development and dissemination of effective obesity prevention programs such as Fun 5.

Disclosure Statement

Funded by the Hawai‘i Medical Service Association, an Independent Licensee of the Blue Cross and Blue Shield Association.

Cara Sandoval Iversen was responsible for initial design, data gathering, data analysis, and writing the article. Dr. Nigg is the PI of the Fun 5 program, mentored the primary author, reviewed and approved, and/or edited all aspects of this paper. Dr. Titchenal consulted on the project, and approved and/or edited all aspects of this paper. None of the authors identify any conflict of interest.