Effect of a Nutritiion Intervention on Fruit and Vegetable Intake in Elementary School Students

Objective: To evaluate the effectiveness of a 10-week Policy, Systems, and Environment (PSE) nutrition education curriculum to increase fruit and vegetable (FV) intake, PSE and FV knowledge, and improvement in attitudes towards FV. Design: Quasi -experimental design. Participants/Setting: N=312 (intervention=142, comparison=170) 5 grade students in low-income, urban elementary schools in Pawtucket, RI selected by established working relationship between project committee members and teachers in the Pawtucket School District. Intervention: Both groups participated in the Fresh Fruit and Vegetable Program. Intervention schools received an additional 10-week PSE education curriculum. Both groups completed pre and post-surveys. Main Outcome Measures: Self-reported dietary intake of fruits and vegetables, knowledge of PSE and FV, and attitudes towards FV. Analyses: Independent t-tests, paired sample t-tests, analysis of variance, analysis of covariance. Results: There were no significant differences in FV consumption from baseline to follow-up between or within groups. Controlling for differences at baseline, the intervention group had significantly higher PSE knowledge (p<.001), FV knowledge (p<.05), and attitudes towards FV (p<.05) at follow-up than the comparison group. Conclusions and Implications: This PSE intervention had a positive impact on knowledge and attitudes, but not behavior. PSE interventions have the potential to empower students to voice their opinions about the types of food they are served at school and home.


Importance of Fruit and Vegetable Consumption
According to recent data, children in the United States are not meeting the recommended dietary intakes for fruits and vegetables (FV) 1  Epidemiologic studies have shown a positive association between increased FV intake and a decrease in the long-term risk of obesity 2 . This is of importance given that 17.7% of children age 6 to 11 years are obese 3 . Obese children are more likely to be obese as adults putting them at risk for several obesity-related complications including diabetes, heart disease, and some cancers 4 .
The Fresh Fruit and Vegetable Program (FFVP), federally funded by the United States Department of Agriculture, is offered by more than 25% of elementary schools in the US 13,21 . The program is designed to provide an additional serving of FV to children in low-income elementary schools as a snack independent from school breakfast or lunch 13,21 . An evaluation estimating the effect of the FFVP showed that students in schools who participate in the FFVP increased FV intake by approximately one third of a cup (0.32 cups per day; p < 0.001) more than students at schools who did not participate in the FFVP 21 . Furthermore, one study of 3 low-income, urban elementary schools showed that the addition of an education component to the FFVP also increases consumption of FV 20 .
The Academy of Nutrition and Dietetics 20113 position paper, cited school-based nutrition education interventions as more effective when paired with other components such as physical activity and supplemented with FV offerings 11 . Furthermore, this paper indicated that school-based nutrition interventions in general were most successful when they included policy change, input from multiple parties (students, parents, school staff), and environmental change 11 .
School-based nutrition interventions have turned their focus to Policy, Systems and Environmental (PSE) approaches to combat childhood obesity 22 . Policy establishes rules that influence daily decisions, such as the federal government increasing the amount of FV required to be served at school lunch 23 . Systems influence how these policies are carried out, such as adding a larger variety of FV to the school lunch menu to meet federally mandated guidelines 23 . Environment is a change to the physical environment 23 . A change to the environment would be adding a garden cart to the school lunchroom to display the increased variety of FV. PSE interventions can empower children to be involved in changes that impact them directly, making them more likely to incorporate these changes into their daily routines 11,24 . A 2012 review of studies on student involvement suggests that empowering students by involving them in the decision making process has a strong positive effect on their self-esteem and ability to drive change 25  year, gender, language spoken at home (to determine ethnicity), and race.

Measurement of FV consumption. FV consumption was measured in both groups
pre and post-test using the SNAP-Ed Fruit and Vegetable Checklist (see Appendix C) to assess number of times fruit was consumed yesterday and number of times vegetables were consumed yesterday with 6 response options ranging from 0 to 5 or more times. The Fruit and Vegetable Checklist has not been validated.

Analyses
All data were analyzed using the statistical software program, SPSS version 22.0, IBM Corp., Armonk, NY. Differences in demographic information were analyzed using independent sample t-tests.
FV variables were normally distributed and there were no differences in FV consumption at baseline. Accordingly, between group change in FV consumption was analyzed using an ANOVA.
At baseline, the intervention group scored significantly higher on the PSE knowledge assessment than the comparison group (t=3.22, p<.001). The intervention group also scored significantly higher than the comparison group on FV knowledge at baseline (t=2.71, p<.01), and attitudes towards FV at baseline (t=3.04, p<.01). Accordingly, between group change in PSE knowledge, FV knowledge and attitudes towards FV were also analyzed using ANCOVAs with the baseline value as a covariate. Within group differences for FV intake, PSE knowledge, FV knowledge, and attitudes towards FV were measured using paired t-tests.

RESULTS
There were no significant differences between groups by age (average age 10.19 ± .45), gender (55% male), or ethnicity (Hispanic or Non-Hispanic defined by students reporting whether or not they speak Spanish at home; 39% Hispanic from intervention schools, 61% Hispanic from comparison schools) ( Table 2).
There were no significant changes in FV consumption between groups from pre to post-intervention (Table 3). There were also no significant changes in FV consumption within groups from pre to post-intervention ( Table 3).
The intervention group scored significantly higher on the post PSE knowledge assessment than the comparison group when adjusted for baseline score in an ANCOVA (F=37.56, p<.001) ( Table 4). There was also a significant increase in score on the PSE knowledge assessment from pre to post-test in both the intervention group (t=-6.55, p<.001) and the comparison group (t=-2.30, p<.05) ( Table 4).
The intervention group scored significantly higher on the post FV knowledge assessment than the comparison group when adjusted for baseline score in an ANCOVA (F=3.94, p<.05) ( Table 4). Analysis of within group change showed that only the comparison group had a significant increase in FV knowledge from pre to post-test on the FV knowledge assessment (t=-2.03, p<.05) ( Table 4).
The intervention group scored significantly higher on the post FV attitude assessment than the comparison group when adjusted for baseline score in an ANCOVA (F=5.86, p<.05) ( Table 4). Analysis of within group change showed that only the comparison group had a significant increase in attitudes towards FV from pre to post-test on the FV attitude assessment (t=2.17, p<.05) ( Table 4).

DISCUSSION
While many studies have evaluated the efficacy of PSE interventions on dietary outcomes, nutrition-related knowledge, and attitudes, few have looked at urban fifth grade students 9,12,13,[28][29][30][31][32][33][34][35] . Successful studies used several strategies to employ change including involvement of parents and Wellness Committees and a structured nutrition education program 8,27,33,35,37,38 . What sets this study of urban fifth grade students apart from other interventions is the use of student empowerment to drive change.
Other elementary school-based PSE interventions have found significant decreases in unhealthy food items being brought to school form the outside, increases in nutrition knowledge, and improved attitudes towards healthy foods 8,27,28 . However, this study found no change in intake of FV. Nevertheless, this study found significant improvements in knowledge and attitudes.
Although there were no statistically significant demographic differences between groups, 61% of students in the comparison group identified their language spoken at home as Spanish, compared to only 39% in the intervention group. National data indicates Hispanic students, on average, score lower than Non-Hispanic students at reading, which may explain why the intervention group scored significantly higher on the knowledge assessment at baseline than the comparison group 39 . Comprehension of the knowledge assessment may have been further complicated by the fact that the assessment was not validated and the assessments were provided in English only, which may not have been appropriate for Spanish speaking students.
Studies aimed at increasing FV intake typically see an increase around 1/3 cup per day, compared to this study which found no significant change in FV intake 11 . Most other studies use a multi-item food frequency questionnaire to measure dietary intake, such as The Block Food Frequency Questionnaire used in Shaping Healthy Choices, and these food frequency questionnaires may be more accurate than the two-item instrument used in this study 8,11,26 . During administration of the assessment, educators read each question out loud for the students, clearly explaining the directions, but students often worked ahead disregarding the directions. Additionally, in some classrooms, the FV assessment was administered on a Monday, so students were reporting FV consumption for a weekend day which can vary greatly from FV consumption during the week.
The significant increase in PSE and FV knowledge is consistent with other studies who have assessed changes in knowledge following interventions which incorporate an education component 5,8,11 . To our knowledge, this is the first study of its kind to evaluate a change in PSE knowledge. These results are encouraging, but further research is necessary to determine whether students understand the concept of PSE. The significant improvement in attitudes towards FV is consistent with Campbell, et al., who found significant improvement in attitudes towards healthy eating using a validated survey 28 . The improvement in attitudes associated with this study were supported by focus groups conducted with students post-intervention, which revealed that they felt empowered by being involved in the decision-making process 40 . The majority of students in both intervention schools voted "yes", they would eat the recipe again, and the recipes (apple cucumber salad, and roasted carrots) were added to the cycle lunch menu. Empowering students has been shown to improve self-esteem and increase the likelihood that they will participate in a meaningful way where they share their voice 25 . Future research, using a validated attitude survey is needed.

Strengths and Limitations
A strength of this study is that, to the author's knowledge, it is the first PSE nutrition intervention with elementary school students that focuses on empowering students, and uses that empowerment to drive change. There were several limitations though.
Dietary assessments with children are challenging and the accuracy is questionable 41 .
The Fruit and Vegetable Checklist that was used in this study may not be a sensitive instrument to assess change. Although significant differences in baseline score were statistically controlled for using ANCOVA, future research should focus on schools being more evenly matched by student demographics. Additional demographics, such as race, would have been useful, but the instrument used to collect this information was poorly understood by students. Schools in this study were not randomly selected, and including more schools would have provided a larger sample with the potential for more robust results.

IMPLICATIONS FOR RESEARCH AND PRACTICE
This study was associated with improved knowledge and attitudes, however, there was no difference in FV intake. Focusing on student empowerment may lead to longer term changes. Future research needs more sensitive tools, but this intervention focusing on empowerment is promising. This study served as a pilot study to be replicated and built upon. SNAP-Ed has revised the program and is delivering it in the Providence, RI school district as, "Students Take Charge", but has embraced the same focus on empowering students to drive change through PSE related activities.    This literature review will discuss the importance of nutrition education interventions for increasing FV intake among elementary school students. Additionally, it will discuss the significance of PSE interventions.

Childhood Obesity
According to data provided by the 2011-2012 National Health And Nutrition Examination Survey (NHANES), 32% of children age 2-19 years are overweight or obese with the percentage of these children measuring as obese at nearly 17% 2,17 . For children within this age bracket, obesity is defined as being at or above the 95 th percentile on the BMI-for-age growth chart provided by the Centers for Disease Control (CDC) 18,19 . Examining the data by smaller age groups of 2-5 years, 6-11 years, and 12-19 years, a steady increase in obesity from 8.9% to 17.5% to 20.5% respectively, can be observed 17 .

23
Childhood obesity has been linked to a high prevalence of metabolic syndrome among children, that rises with increased obesity 20 . Additionally, childhood obesity increases the risk for obesity in adulthood which increases the likelihood of acquiring obesity-related diseases such as type 2 diabetes or heart disease later in life 21,22 .
According to the CDC, 25% of deaths in the United States (US) are represented by heart disease, making it the leading cause of death 23

Impact of the Nutrition Environment
A child's nutrition environment plays a significant role in the types of foods they choose to eat, and therefore can significantly influence the amount of FV they consume 29

Impact of Increased Nutrition Knowledge
Nutrition education can increase a child's nutrition knowledge, and an increase in nutrition knowledge can lead to healthier food choices 15,39,40  There were over 500 participants including both intervention and control schools 15 .
The program included interactive assemblies, videos, lessons and games incorporating the OrganWise Guys TM , characters designed to make children aware of the different parts of their bodies and how to keep them healthy 15 . The program was designed for teachers to incorporate into their existing lessons and were supported by their local university cooperative extension 15 . Changes in nutrition knowledge and self-efficacy were measured pre-and post-intervention using an assessment designed specifically for this program which was validated during the pilot study 15 . At the end of the 12- week intervention, participants in the intervention group showed a greater increase in nutrition knowledge compared to the control group (F(1,553) = 6.99, p = 0.00) 15 .
Participants in the intervention group also showed a greater increase in self-efficacy than the control group (F(1, 404) = 8.38, p = 0.00), but no significant difference in FV preferences 15 .

Impact of Nutrition Education on Attitudes Towards FV
A recent study examining children's attitudes towards healthy eating concluded children do not view choosing healthy foods as an important factor in their dietary intake, posing a risk to an increase in childhood obesity rates 42 . Nutrition education interventions, however, may have a positive influence on children's attitudes towards FV 8,9 .
Healthy Buddies, a health promotion program conducted in British Columbia elementary schools, partnered younger students (grades kindergarten to 3) with older students (grades 4 to 7) as healthy-living mentors 43 . Each week older students were responsible for teaching 30-minute health lessons to the younger students after receiving the same lessons from intervention teachers for a total of 21 lessons 43  intervention group habit scores for Kindergarten -Grade 3 increased by 2.3 points (p<0.001) and Grades 4 -7 increased by 1.7 points (p<0.001); attitudes improved significantly in intervention schools (p<0.002) 43 . These results indicate that nutrition knowledge has a positive impact on behaviors, habits and attitudes towards healthy foods 43 .

Impact of Nutrition Education on FV Consumption
Increasing nutrition knowledge is an important first step in helping children make healthier choices. Many studies go beyond looking at just an increase in knowledge or the ability to make a healthier choice, and measure changes in consumption of healthy foods 5,7,9,16,39,[44][45][46][47][48][49][50][51][52] . Often these studies look specifically at changes in FV consumption. A review of schools who participate in the Fresh Fruit and Vegetable Program (FFVP) indicated that these schools offer more FV and intake of FV is higher among students in these schools than schools who do not participate in the program 53 .
The Overweight and Obesity, school-based nutrition education interventions were found to be more effective when paired with other components such as physical activity and supplemented with FV offerings 9 . Furthermore, the paper indicates that school-based nutrition interventions in general are most successful when they include policy change, input from multiple parties (students, parents, school staff), and environmental change 9 .

PSE Interventions
Policy, Systems and Environmental (PSE) interventions use these three avenues (PSE) to drive change. This model is commonly used in community interventions with a fair amount of success 55,56 . In 2010, 50 communities across the U.S. received Communities Putting Prevention to Work grants from the CDC to implement PSE interventions 55 . Community teams developed strategies to improve health over a 2- year period in the following categories: nutrition, physical activity, and tobacco use 55 .
Progress was assessed at 12 months at which point community teams on average reached 43% of the targeted population 55 .

31
One PSE intervention associated with the grant in low-income school districts of King County, Washington focused on obesity prevention among high school students 14 .
This intervention included student-led promotions of healthy eating campaigns, farm to school initiatives, nutrition training for school cafeteria staff, and participation in community health coalitions from 2010-2012 14 . Students in King County school districts who received the intervention were compared with students in low-income school districts in other counties who did not receive any intervention using logistic regression analysis 14 . There was a significant decrease (p=.02) in obesity in King County students compared to students in other school districts 14 .
School-based nutrition interventions have started to incorporate Policy, Systems and Environmental (PSE) approaches at the elementary school level, which present a different dynamic than middle school or high school, and will be the focus of this review. PSE interventions can empower children to be involved in changes that impact them directly, making them more likely to incorporate these changes into their daily routines 9,12 . A 2012 review of studies on student involvement suggests that empowering students by involving them in the decision making process has a strong positive effect on their self-esteem and ability to drive change 57  p<.001) 7 . There was also a significant increase in nutrition knowledge when compared to the control schools (p<.001), but there were no significant changes in FV intake or vegetable preference 7 .
The Healthy Options for Nutrition Environments in Schools (Healthy ONES) study is a randomized group trial that looked to involve school staff, parents and children in policy and environmental changes that would make the school cafeteria a healthier eating environment and would encourage children to make healthier choices 4 .
Children were observed monthly using an observation system designed and tested specifically for this study during school meals over a period of 2 years to determine the amount of unhealthy food and beverage items that were brought in from the outside 4 . Additionally, to determine change in obesity rates, student's BMI were measured at three time points 4 . Results indicated significant change in that intervention schools showed a decrease in outside unhealthy food items over time (p=.005), while control schools showed an increase in outside unhealthy food items over time (p=.04), but no significant changes in BMI were observed 4 .
Sallis, et al. conducted a randomized control trial of a school-based nutrition intervention 10 . The objective was to decrease the amount of fat students were consuming while increasing physical activity through policy and environmental change 10 . School staff, parents and students were all involved in implementing policy and environmental changes which included working with food services to offer more low-fat options in the school lunchroom, as well as making changes to the physical education curriculum 10 . Changes in physical activity were observed using two separate validated methods, Systems for Observing Fitness Instruction Time and Systems for Observing Play and Leisure Activity of Youth 10 . Amount of fat consumed at school lunch was measured by documenting meals provided by food service and analyzing fat content using Nutrition Data Systems, and also collecting nutrition facts labels from students who brought a bag lunch from home 10 . While there was no significant decrease in the amount of fat consumed, there was a significant increase (p<0.009) in physical activity 10 .

Assessment with Children
It is important to note the challenges associated with conducting dietary assessments with students. Some common instruments used to assess dietary intake are 24-hour food recalls, dietary records, dietary histories, Food Frequency Questionnaires, and observations of children's diets 59 . All of these methods can be complicated by a child's cognitive ability, surrogate reporting, or social desirability 59 . Children under the age of 12 have a limited attention span, and possess a limited ability to recall dietary intake unless it is in the immediate past 59,60 . Additionally, they often have a difficult time comprehending assessment questions and require a great deal of guidance, especially considering their limited knowledge of food preparation, measurement, and ability to estimate portion sizes 59,60 . Some types of reporting, such as dietary records, can be completed by parents or caregivers providing information for the child 59 . However, child self-reports are less accurate because children also have a tendency to respond with the most socially desirable answer, which results in under-or over-reporting 59 . Studies will often cite these challenges as limitations to achieving desired outcomes 5,7 . Furthermore, when conducting assessments in predominantly low-income school districts, it is important to consider that children of low socioeconomic status generally perform at a lower academic level than children of higher socioeconomic status 61 . According to national test scores, an additional academic disparity exists by ethnicity with Whites averaging higher test scores than Blacks and Hispanics in math and reading 62 .

Conclusion
Childhood obesity is a large-scale problem with nearly one-third of children in the US being classified as overweight or obese 2,17 . Children who are obese are more likely to be obese as adults and therefore have a higher risk for acquiring obesity related diseases later in life 21,22 . Increasing FV intake can help combat this obesity epidemic 28 .

35
Nutrition interventions have proved to be an effective means of increasing FV intake 5,9,16,45 . This review found that the most effective nutrition interventions have multiple components, combining nutrition education with physical activity, the opportunity to try FV, policy change, community involvement, and empowering children to be involved in the decision making process 4,7,9,10,57,58 .