Promoting Healthy Lifestyles to Children at School: Using a Multidisciplinary Train-the-Trainer Approach Martha J. Sanders, PhD; Jesse Reynolds, MS; Nancy Bagatell, PhD; Judith A. Treu, MS; Edward O’Connor, PhD; David L. Katz, MD, MPH rrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr
Objective: The purpose of the study was to examine the efficacy of a multidisciplinary train-the-trainer model for improving fitness and food label literacy in third-grade students. Design: University student trainers taught ABC for Fitness and Nutrition Detectives, established programs to promote physical activity and nutrition knowledge, to 239 third-grade students in 2 communities over a 6-month period. A total of 110 children were in the intervention group and 129 children in the control group (2 schools each). Outcomes included the Food Label Literacy and Nutrition Knowledge test and the fitness measures of curl-ups, push-ups, 0.5-mile run, and sit and reach. Focus groups were conducted as process feedback. Setting: Four public schools in 2 different communities. Participants: A total of 200 third-grade students. Intervention: ABC for Fitness and Nutrition Detectives. Main Outcome Measures: Food Label Literacy and Nutrition Knowledge test and the fitness measures of curl-ups, push-ups, 0.5-mile run, and sit and reach. Results: Nutrition knowledge increased in the intervention group by 25.2% (P < .01). Fitness measures in the intervention schools showed greater improvement than those in the controls for curl-ups (P < .01), push-ups (P < .01), sit and reach left (P = .07), and 0.5-mile run (P = .06). Process feedback from 3 teachers and 60 students indicated satisfaction with the program. Conclusion: Adaptation of the train-the-trainer approach for Nutrition Detectives and ABC for Fitness was effective for delivering these health-related programs. KEY WORDS: childhood obesity, fitness, health,
multidisciplinary, train-the-trainer
The prevalence of obesity in US children aged 6 to 11 years has increased from 7% in 1980 to nearly 18% in 2012.1 The implications for curbing this trend are preventing chronic conditions such as type II diabetes, heart disease, and high blood pressure, and minimizing the potential for psychosocial issues such as poor body image, bullying, and decreased social participation. Modifiable factors such as diet and physical activity have been a focus of obesity-related interventions for children.2-4 School-based programs are considered ideal venues to address health issues due to close and continuous contact with students and the ability to influence health behaviors at a young age.4 However, challenges exist in implementing programs in schools within scheduling, time, cost constraints, and building internal capacity.
● School-Based Train-the-Trainer Models One approach to address these challenges is using a train-the-trainer (TTT) model to develop a team of community-based trainers to deliver a specific program, who may or may not have direct experience with the content being delivered.5 Advantages include the ability to reach a larger audience than possible through individual methods and the potential for sustaining programs. Author Affiliations: Quinnipiac University, Hamden, Connecticut (Drs Sanders, Bagatell, and O’Connor); Yale Center for Analytical Sciences, New Haven, Connecticut (Mr Reynolds); and Yale Griffin Prevention Research Center, Derby, Connecticut (Ms Treu and Dr Katz). This study was made possible by a grant from the Anthem Blue Cross and Blue Shield Foundation, LLC. The authors have no conflicts of interest to report.
J Public Health Management Practice, 2015, 21(4), E27–E35 C 2015 Wolters Kluwer Health, Inc. All rights reserved. Copyright
Correspondence: David L. Katz, MD, MPH, Yale-Griffin Prevention Research Center, 130 Division St, Derby, CT 06418 ([email protected]). DOI: 10.1097/PHH.0000000000000141
E27 Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
E28 ❘ Journal of Public Health Management and Practice Train-the-trainer models have been applied in school settings to build community knowledge and promote health and well-being. Successful TTT models include the training of high school students to teach first aid to peers in farming communities6 ; parents, teachers, and students to provide structured playground activities for children7 ; medical students and high school students to provide nutrition education to elementary school children8 ; and family child care home providers to deliver healthy eating and physical activity information to other family child care home providers.9
● TTT Program Components To ensure consistent transmittal of information and optimal delivery methods, most TTT programs include (1) a presentation by expert trainers, (2) multiple teaching methods, (3) delivery in a module format, (4) a training manual, (5) a pretest/posttest for trainers, (6) the delineation of responsibilities, and (7) the ability for trainees to practice teaching/learning activities prior to implementation.5,6,8 Their ongoing success in school settings is promoted by support for teachers and volunteers, funding to purchase equipment, adapting the program to a changing environment, and evaluation to determine program efficacy.5 Based on previously successful TTT models, this study employed a multidisciplinary process using graduate-level university students from various disciplines within a school of health sciences to directly deliver and/or train school teachers to deliver programming. These students had a broad background in the basic sciences but no specific training in health education. This model provided the opportunity to expand their skills in program planning, health promotion, and working within a school system. The model minimized the schools’ time, personnel, and costs associated with training teachers by using 2 health promotion programs that are freely available in the public domain.
● Research Purpose and Questions The purpose of the study was to examine the efficacy of a multidisciplinary TTT model at improving fitness and nutrition knowledge in third-grade students. The research questions were as follows: (1) Does a TTT-modeled delivery of an evidence-based nutrition education program improve nutrition knowledge in third-grade students? (2) Does a TTT-modeled delivery of an evidence-based daily classroom physical activity program impact physical fitness in third-grade
students? (3) What lessons learned may facilitate “reallife” school-based fitness and nutrition programs using a TTT approach?
● Methods Human participant compliance statement This study was approved by the Quinnipiac University Human Subjects Committee Institutional Review Board.
Participants Participants were third-grade students attending 4 public schools in 2 Connecticut communities, each with its own school district. Within each community, the intervention and control schools were comparable with regard to student socioeconomic status and race/ethnicity (see Table 1). Inclusion criteria were being a third-grade student, knowledge of English, and parental permission to participate. Parents were sent information letters concerning the study and returned opt-out notes only if they chose not to have their children participate.
Outcome measures Nutrition knowledge was operationalized as the ability to understand food labels and identify healthful food choices. Pre- and posttest data were collected using a 10-item Food Label Literacy for Applied Nutrition Knowledge (FLLANK) questionnaire, a validated and reliable tool designed to measure food label literacy in elementary students exposed to the Nutrition Detectives program (Cronbach α = 0.77 and intraclass correlation coefficient = 0.68 in repetitive administrations).11 The questionnaire tests students’ application of the program’s “5 clues” to make healthful food choices. Each question requires students to decide which of 2 sets of Nutrition Facts panels and ingredient lists within a category of packaged foods (breads, crackers, cereals, cereal bars, or cookies) is a better choice. Correct responses are summed for an overall score. The ABC for Fitness evaluation included tests for abdominal strength (curl-ups), upper extremity strength (push-ups), flexibility (sit and reach), and aerobic capacity (0.5-mile run). These are standardized tests used in Connecticut state fitness testing.
Procedures The study used a mixed methods design12 that utilized quantitative data to determine the efficacy of
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Promoting Healthy Lifestyles to Children
❘ E29
TABLE 1 ● School Demographic Data: 2010-2011 School Yeara
qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq Community 1 (School District 1)
Variable Socioeconomic status of students Percentage of students eligible for free or reduced price lunch Race/ethnicity of students American Indian Asian American Black Hispanic Pacific Islander White ≥2 races Total percentage of minority students in each school a From
Community 2 (School District 2)
Intervention School
Control School
Intervention School
Control School
13.5%
11.8%
18.1%
17.3%
0.0% 8.8% 14.2% 11.2% 0.0% 61.8% 4.1% 38.2%
0.6% 16.6% 11.2% 8.2% 0.0% 61.5% 1.9% 38.5%
0.0% 3.3% 2.7% 18.4% 0.0% 75.7% 0.0% 24.3%
0.0% 5.6% 5.3% 14.3% 0.0% 74.9% 0.0% 25.1%
Connecticut State Department of Education.10
the interventions and qualitative data to provide process evaluation feedback on program satisfaction. The quantitative component used a quasi-experimental, pretest/posttest design with a delayed intervention for the control group. Two schools each from 2 communities participated in the study. In each community, third-grade students in one school served as the control group and those in the other school served as the experimental group.
ence to university curriculum. The training process included key features of TTT models to ensure mastery of content, delivery methods, and ongoing trainee support (see Table 2). This approach differed from those reported in the literature in that graduate student trainees organized themselves, practiced content delivery independently, and developed additional program activities as part of their graduate level curriculum.
Program development A series of initial meetings was held with a prevention research center, university representatives, and school administrators to develop administrative support for the program and set realistic timelines for implementation and evaluation. The research team then met with principals and teachers in each intervention school to address scheduling, data collection, and expectations.
TTT approach The TTT approach was a collaborative, grant-funded academic-community partnership among a university, a prevention research center, and 2 communities with elementary schools. Prevention research center consultants provided initial training on the content and delivery of the 2 health programs (Nutrition Detectives and ABC for Fitness) to 145 university faculty and graduate students. The graduate students represented several disciplines, including occupational therapy, physical therapy, nursing, and physician assistant. Faculty in each discipline were also trained in order to provide oversight, support, and to bridge the experi-
TABLE 2 ● Components of the Multidisciplinary Train-the-Trainer Process Model qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq
Train-the-Trainer Process for School-Based Health Promotion Program University faculty recruitment Faculty attend a general information session Faculty provide suggestions to integrate Nutrition Detectives and ABC for Fitness into university curriculum Formal training session (3 hours) provided by prevention research center consultants Review issues of childhood obesity, fitness, lifestyle, and nutrition Participate in Nutrition Detectives and ABC for Fitness program training Teacher training Meet with elementary school principals and teachers to explain program Train teachers University student preparation Assign roles and specific program components to graduate students Practice presentation and hands-on activities Program implementation in elementary schools University students directly deliver Nutrition Detectives program Elementary teachers deliver ABC for Fitness program with support from University students and faculty
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
E30 ❘ Journal of Public Health Management and Practice
Program implementation Nutrition Detectives is an age-appropriate, “spy themed” program designed to help children understand the relationship between food and health and use tips (“clues”) to read food labels and make healthful food choices based on (1) recognition of marketing deceptions; (2) the first ingredient on the list; (3) avoidance of trans fat and high fructose corn syrup; (4) length of ingredient list; and (5) amount of fiber per serving in grain products.13 In this study, the 90-minute program was delivered during two 45-minute lessons within 1 week. The first lesson was a didactic presentation to help students understand nutrition and learn detective “clues.” The second lesson was a hands-on group activity to read packaged food labels and identify more healthful (“clued-in”) versus less healthful (“clue-less”) food items. The program was delivered in a total of 8 classrooms within the 2 intervention schools. ABC for Fitness is a daily physical activity program designed to interject fun, noncompetitive activities into classroom schedules around transition times when students may be restless and need refocusing. Its “activity bursts” typically last 5 to 6 minutes and include a warm-up, core aerobic activity, and a cool-down component. Activities can be imaginative moves, creative extensions of learning content, or advanced bursts delivered at regular intervals for a total of 30 minutes per day.3 In this study, classroom teachers received training, a manual, and a set of fitness cards, and then implemented the program independently with university support and consultation. In community 2, graduate students provided this support on a weekly basis as requested by the teachers.
Data collection For Nutrition Detectives, third-grade students in intervention schools completed the FLLANK questionnaire before and after program implementation. The FLLANK was administered once to students in control schools to compare baseline knowledge. Since it was assumed that control school scores would not likely change during the brief length of the intervention period (ie, two 45-minute sessions delivered within a week), a posttest was not administered. In community 1, graduate students administered the tests in control and intervention schools in November/December 2010. In community 2, graduate students administered the testing at the intervention school in February/March 2011 and at the control school in March/April 2011. For ABC for Fitness, baseline fitness tests were administered by physical education (PE) teachers during gym classes, with health science students assisting as needed to record the data. Posttesting was performed
by school PE teachers in 2 schools and by outside PE consultants at the other 2 schools. Community 1 data were collected in October 2010 (baseline) and April 2011 (posttest). Community 2 data were collected in January/February 2011 (baseline) and June 2011 (posttest). Anthropometric measures of height, weight, and body mass index (BMI) were measured in community 1 during the same testing period in the following manner: nurse practitioner students recorded the ages, height, and weights of children (clothed without shoes) using a manual scale with a height arm attachment. These data were not collected in community 2.
Focus groups In community 2, after 7 weeks of ABC for Fitness, university faculty and graduate students conducted four 40-minute focus groups in the intervention school: 1 with all 3 teachers and 3 with students (3 classes, 20 students per classroom). Teachers responded to questions concerning their experiences in implementing the program, their perceptions of its impact on students, and recommendations for the future. The focus group was audio recorded and transcribed for analysis. Third-grade student focus groups were activitycentered.14 To encourage reflection on the experience, students first completed a worksheet independently upon which they drew pictures and answered questions regarding what they learned, liked, and disliked about the program. Students then participated in a 5-minute activity burst and shared their spontaneous reflections with peers and the researchers. The researchers probed for further detail when necessary and wrote down key phrases and quotes.
Data analysis To assess group differences at baseline between intervention and control groups, Student t test was used for continuous outcomes and χ 2 test was used for categorical outcomes. Nutrition knowledge was assessed through repeated measures analysis of variance to assess within-group differences in the intervention group on the FLLANK from pretest to posttest. Repeated measures analysis of variance was used to assess both the between- and within-group differences for anthropometric and physical activity measures. Post hoc tests were conducted to assess the influence of within-group factors such as gender, age, and school district using the General Linear Model. Data were analyzed per protocol. The statistical software package SPSS version 17.0 was used to conduct all analyses.15 All tests were conducted with a 2-tailed α level set at .05. Transcribed data from focus group discussions and third-graders’
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Promoting Healthy Lifestyles to Children
drawings were analyzed for thematic content around guiding questions.
● Results A total of 110 children comprised the intervention group and 129 children served as control with delayed intervention (see Table 3). The proportions of students from community 1 and community 2 were comparable across study groups. The average age of students and the ratio of boys to girls were also comparable between intervention and control schools. Body weight, BMI, and BMI percentile measurements were comparable between students from community 1 intervention and control schools. Intervention school students had significantly higher baseline nutrition
❘ E31
knowledge/food label literacy than those in control schools (P < .05). Furthermore, at baseline, intervention school students scored significantly higher than control school students on the following fitness measures: curl-up, push-up, and 0.5-mile run (all P < .05) (Table 3). Some inconsistencies in data collection occurred during the study (see Table 3). In community 1, some baseline fitness data were not collected from the control school because of inadequate prior communication. In community 2, the control school failed to use the correct protocol for aerobic fitness measures. Although standardized testing procedures were used for all fitness testing, in 2 schools, PE consultants provided the posttesting (due to time constraints). Furthermore, although BMI is a typical outcome measure in childhood obesity intervention
TABLE 3 ● Baseline Characteristics of Participating Third-Grade Students
qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq
Variable Community/school district Community/school district 1 Community/school district 2 Average age of participating students (years) Gender of participating students Boys Girls FLLANK (% correct) Community/school district 1 Community/school district 2 Anthropometric Measuresa Body weight (pounds) BMI (kg/m2 ) BMI percentile Fitness Test Scores: Curl-Upb (no. completed) Community/school district 1 Community/school district 2 Fitness Test Scores: Push-upb (no. completed) Community/school district 1 Community/school district 2 Fitness Test: 0.5-mile runc (minutes) Community/school district 1 Community/school district 2 Fitness Test: Sit and Reach Leftb (inches) Community/school district 1 Community/school district 2 Fitness Test: Sit and Reach Rightb (inches) Community/school district 1 Community/school district 2 a School
Intervention N = 110
Control N = 129
56 (50.9%) 54 (49.1%) 8.4 ± 0.5
73 (54.1%) 62 (45.9%) 8.3 ± 0.5
.72
44 (40.0%) 66 (60.0%) 54.3 ± 17.4 53.1 ± 19.6 55.5 ± 15.1
59 (45.7%) 70 (54.3%) 47.7 ± 21.3 45.6 ± 21.2 50.4 ± 21.4
.92 .01 .05 .15
75.0 ± 18.1 18.1 ± 3.1 66.6 ± 28.8 20.2 ± 12.1 15.9 ± 10.9 24.4 ± 11.9 10.3 ± 6.0 9.2 ± 5.2 11.4 ± 6.5 5.4 ± 1.5 5.4 ± 1.4 5.4 ± 1.6 9.4 ± 2.4 8.6 ± 2.5 10.3 ± 2.0 9.5 ± 2.3 8.8 ± 2.4 10.4 ± 2.0
73.7 ± 17.7 17.9 ± 3.2 64.1 ± 30.2 16.0 ± 8.9 ... 16.0 ± 8.9 7.2 ± 5.0 ... 7.2 ± 5.0 4.9 ± 1.4 4.9 ± 1.4 ... 9.0 ± 1.7 ... 9.0 ± 1.7 8.9 ± 1.7 ... 8.9 ± 1.7
.70 .74 .66 .01 ... .00 .01 ... .00 .02 .04 ... .19 ... .00 .06 ... .00
district 1 schools only.
b School district 1 control post only. c School district 2 intervention only.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
P
.36
E32 ❘ Journal of Public Health Management and Practice studies, the community 2 school district board of education did not approve the measure after having initially agreed.
Nutrition detectives Comparison of FLLANK scores between intervention and control schools yielded a significant difference in baseline nutrition knowledge (Table 3). From pretest to posttest, overall nutrition knowledge scores on the 10-item test increased by 25.2% in the intervention group, which was statistically significant (P < .01). At the individual item level, the increase in correct responses ranged from 6% to 53%. Each intervention school, when analyzed in isolation, had significantly improved scores on the FLLANK as well (both P < .01) (Table 4). No significant interactions were found with regard to gender, age, or school district and improved FLLANK scores (P > .05).
ABC for fitness Community 1 students in intervention and control schools significantly differed with respect to baseline anthropometric measures. With the exception of sit and
reach measures, students in intervention and control schools improved on all fitness measures at posttest. Students in intervention schools showed greater improvement on all measures than control. These differences were statistically significant on the curl-up and push-up measures (P < .01) (Table 4). The fitness outcomes for 0.5-mile run (P = .06) and sit and reach left (P = .07) approached significance. No significant interactions were found with regard to gender, age, or school district.
Process evaluation qualitative findings Teacher perceptions Process evaluation themes centered around the benefits of program structure, classroom management, students’ awareness of their bodies, and support from graduate students. Although some teachers had used movement informally in the classroom prior to the program, they appreciated the consistency and structure of ABC for Fitness (warm-up, core activity, cooldown) as it provided a routine for the classroom. They enjoyed the opportunity to customize the frequency and timing of the activity bursts in their classrooms during the school day.
TABLE 4 ● Change From Baseline Characteristics of Participating Third-Grade Students
qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq Control
P
25.2 ± 22.2 27.3 ± 22.2 23.1 ± 22.3
N/A N/A N/A
.00 .00 .00
2.6 ± 28.8 1.2 ± 5.8 6.3 ± 44.2 9.1 ± 15.9 16.2 ± 14.6 1.2 ± 13.5 2.4 ± 5.6 3.0 ± 5.1 1.7 ± 6.2 − 0.5 ± 1.9 − 0.4 ± 1.5 − 0.6 ± 2.3 0.3 ± 1.6 0.4 ± 1.6 0.1 ± 1.7 0.0 ± 1.6 0.2 ± 1.5 − 0.2 ± 1.7
2.6 ± 8.6 0.6 ± 1.1 3.1 ± 11.4 0.7 ± 4.2 ... 0.7 ± 4.2 0.3 ± 1.3 ... 0.3 ± 1.3 0.0 ± 1.3 0.0 ± 1.3 ... −0.1 ± 0.6 ... −0.1 ± 0.6 −0.0 ± 0.6 ... −0.0 ± 0.6
.99 .53 .63 .00 ... .80 .00 ... .13 .06 .15 ... .07 ... .61 .82 ... .41
Variable
Intervention
FLLANK (%) (within group change in scores) Community/school district 1 Community/school district 2 Anthropometric Measuresa Body weight (pounds) BMI (kg/m2 ) BMI percentile Fitness Test Scores: Curl-upb (no. completed) Community/school district 1 Community/school district 2 Fitness Test Scores: Push-upb (no. completed) Community/school district 1 Community/school district 2 Fitness Test: 0.5-mile walk/runc (minutes) Community/school district 1 Community/school district 2 Fitness Test: Sit and Reach Leftb (cm) Community/school district 1 Community/school district 2 Fitness Test: Sit and Reach Rightb (inches) Community/school district 1 Community/school district 2 Abbreviation: N/A, not applicable. a School district 1 schools only. b School district 1 control post only. c School district 2 intervention only.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Promoting Healthy Lifestyles to Children
The teachers found the support from the graduate students useful for embedding curricular goals into activity bursts, clarifying intensity levels for core activities, and addressing any behavioral and student selfregulation issues that arose during the activity bursts. One teacher noted that “the graduate students helped me realize that activity bursts do not have to be all high-impact jumping. They also helped me establish strategies for a few students who needed help calming down.” Teachers indicated that third graders responded well to the “ABC Rules” they developed to promote classroom management. Teachers also developed their own strategies to help students understand the structure of the activity bursts, including the “ABC places” in the classroom. The most important outcomes that teachers reported were students’ positive attitudes toward fitness, awareness of their bodies, and improved fitness levels. One teacher stated, “I believe most of my students have developed a different mental idea of their own personal stamina with regard to exercise.” Another teacher said that “most [students] would be able to describe cool down and warm up versus core activity and the differences that happen in their bodies when they are doing each type of exercise.” The teachers did not report any changes in academic or behavioral outcomes as a result of the program.
Student perceptions Student themes centered around their preferences for the type of activity break and learning about the relationship between exercise and their bodies. Students enjoyed the game-like quality of the bursts and the opportunity “get up and move.” They most liked the highintensity bursts such as jumping jacks, running, and jumping. They reported learning about the importance of exercise and how to monitor their heart rate, stretch, and cool down. One third grader stated, “When you get active, your heart pumps more blood.” Another student who learned how to calm down explained, “You breathe in deeply through your nose and out through your mouth when you want to calm down.” They were enthusiastic about continuing the activity bursts in the classroom.
● Discussion Elementary school students’ food label literacy improved significantly after exposure to the Nutrition Detectives program when taught using the TTT approach. Findings demonstrate that children were able to perform on par with (if not slightly better than) children from a previously published Nutrition Detec-
❘ E33
tives pilot study.13 In our study, third-grade students improved their FLLANK scores by greater than 25%. This outcome exceeds that found in the pilot study,13 in which children in grades 2, 3, and 4 improved their FLLANK scores by 18% across grade levels, with a 23% improvement among third-grade students. These findings indicate that the TTT approach adapted for this study (ie, training university students to deliver the program in partnership with elementary school teachers) was as effective with regard to student performance. The findings also indicate that this gain in nutrition knowledge and skills in identifying healthful food choices, if sustained over time, could lead to improved health benefits among children who apply their knowledge and skills to improve the nutritional quality of their food choices. Body mass index findings were also similar to those from an ABC for Fitness pilot study in which intervention group students did not significantly change with regard to BMI at follow-up.3 Students in the current study displayed similar characteristics at followup. However, these results are tenuous at best because the entire sample of participants was not measured, as 1 school district decided not to measure BMI once program implementation began. Changes in physical fitness data from baseline in the intervention group showed significant improvements when compared with control on certain fitness measures and approached trend-level significance (P < .10) on others: 0.5-mile run and the sit and reach left measures. Trend-level significance suggests that including a greater number of students in the analysis could have led to statistical significant differences between groups. It is important to note that these data were not consistently collected at all time points across study groups and, therefore, some comparisons were “underpowered.” Students in an ABC for Fitness pilot study3 also demonstrated similar improvements in physical activity measures, therefore, suggesting that the program delivery of ABC for Fitness in this current study demonstrates fidelity to the original program design.
Relationship to previous school-based healthy lifestyle programs This study supports findings from a recent review of more than 25 school-based physical activity and nutrition education programs focusing on childhood obesity prevention.4 The review indicates that most physical activity/nutrition programs demonstrate measurable changes in nutrition knowledge on the basis of valid pretest/posttest comparisons. However, most studies of short duration (
Objective: The purpose of the study was to examine the efficacy of a multidisciplinary train-the-trainer model for improving fitness and food label literacy in third-grade students. Design: University student trainers taught ABC for Fitness and Nutrition Detectives, established programs to promote physical activity and nutrition knowledge, to 239 third-grade students in 2 communities over a 6-month period. A total of 110 children were in the intervention group and 129 children in the control group (2 schools each). Outcomes included the Food Label Literacy and Nutrition Knowledge test and the fitness measures of curl-ups, push-ups, 0.5-mile run, and sit and reach. Focus groups were conducted as process feedback. Setting: Four public schools in 2 different communities. Participants: A total of 200 third-grade students. Intervention: ABC for Fitness and Nutrition Detectives. Main Outcome Measures: Food Label Literacy and Nutrition Knowledge test and the fitness measures of curl-ups, push-ups, 0.5-mile run, and sit and reach. Results: Nutrition knowledge increased in the intervention group by 25.2% (P < .01). Fitness measures in the intervention schools showed greater improvement than those in the controls for curl-ups (P < .01), push-ups (P < .01), sit and reach left (P = .07), and 0.5-mile run (P = .06). Process feedback from 3 teachers and 60 students indicated satisfaction with the program. Conclusion: Adaptation of the train-the-trainer approach for Nutrition Detectives and ABC for Fitness was effective for delivering these health-related programs. KEY WORDS: childhood obesity, fitness, health,
multidisciplinary, train-the-trainer
The prevalence of obesity in US children aged 6 to 11 years has increased from 7% in 1980 to nearly 18% in 2012.1 The implications for curbing this trend are preventing chronic conditions such as type II diabetes, heart disease, and high blood pressure, and minimizing the potential for psychosocial issues such as poor body image, bullying, and decreased social participation. Modifiable factors such as diet and physical activity have been a focus of obesity-related interventions for children.2-4 School-based programs are considered ideal venues to address health issues due to close and continuous contact with students and the ability to influence health behaviors at a young age.4 However, challenges exist in implementing programs in schools within scheduling, time, cost constraints, and building internal capacity.
● School-Based Train-the-Trainer Models One approach to address these challenges is using a train-the-trainer (TTT) model to develop a team of community-based trainers to deliver a specific program, who may or may not have direct experience with the content being delivered.5 Advantages include the ability to reach a larger audience than possible through individual methods and the potential for sustaining programs. Author Affiliations: Quinnipiac University, Hamden, Connecticut (Drs Sanders, Bagatell, and O’Connor); Yale Center for Analytical Sciences, New Haven, Connecticut (Mr Reynolds); and Yale Griffin Prevention Research Center, Derby, Connecticut (Ms Treu and Dr Katz). This study was made possible by a grant from the Anthem Blue Cross and Blue Shield Foundation, LLC. The authors have no conflicts of interest to report.
J Public Health Management Practice, 2015, 21(4), E27–E35 C 2015 Wolters Kluwer Health, Inc. All rights reserved. Copyright
Correspondence: David L. Katz, MD, MPH, Yale-Griffin Prevention Research Center, 130 Division St, Derby, CT 06418 ([email protected]). DOI: 10.1097/PHH.0000000000000141
E27 Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
E28 ❘ Journal of Public Health Management and Practice Train-the-trainer models have been applied in school settings to build community knowledge and promote health and well-being. Successful TTT models include the training of high school students to teach first aid to peers in farming communities6 ; parents, teachers, and students to provide structured playground activities for children7 ; medical students and high school students to provide nutrition education to elementary school children8 ; and family child care home providers to deliver healthy eating and physical activity information to other family child care home providers.9
● TTT Program Components To ensure consistent transmittal of information and optimal delivery methods, most TTT programs include (1) a presentation by expert trainers, (2) multiple teaching methods, (3) delivery in a module format, (4) a training manual, (5) a pretest/posttest for trainers, (6) the delineation of responsibilities, and (7) the ability for trainees to practice teaching/learning activities prior to implementation.5,6,8 Their ongoing success in school settings is promoted by support for teachers and volunteers, funding to purchase equipment, adapting the program to a changing environment, and evaluation to determine program efficacy.5 Based on previously successful TTT models, this study employed a multidisciplinary process using graduate-level university students from various disciplines within a school of health sciences to directly deliver and/or train school teachers to deliver programming. These students had a broad background in the basic sciences but no specific training in health education. This model provided the opportunity to expand their skills in program planning, health promotion, and working within a school system. The model minimized the schools’ time, personnel, and costs associated with training teachers by using 2 health promotion programs that are freely available in the public domain.
● Research Purpose and Questions The purpose of the study was to examine the efficacy of a multidisciplinary TTT model at improving fitness and nutrition knowledge in third-grade students. The research questions were as follows: (1) Does a TTT-modeled delivery of an evidence-based nutrition education program improve nutrition knowledge in third-grade students? (2) Does a TTT-modeled delivery of an evidence-based daily classroom physical activity program impact physical fitness in third-grade
students? (3) What lessons learned may facilitate “reallife” school-based fitness and nutrition programs using a TTT approach?
● Methods Human participant compliance statement This study was approved by the Quinnipiac University Human Subjects Committee Institutional Review Board.
Participants Participants were third-grade students attending 4 public schools in 2 Connecticut communities, each with its own school district. Within each community, the intervention and control schools were comparable with regard to student socioeconomic status and race/ethnicity (see Table 1). Inclusion criteria were being a third-grade student, knowledge of English, and parental permission to participate. Parents were sent information letters concerning the study and returned opt-out notes only if they chose not to have their children participate.
Outcome measures Nutrition knowledge was operationalized as the ability to understand food labels and identify healthful food choices. Pre- and posttest data were collected using a 10-item Food Label Literacy for Applied Nutrition Knowledge (FLLANK) questionnaire, a validated and reliable tool designed to measure food label literacy in elementary students exposed to the Nutrition Detectives program (Cronbach α = 0.77 and intraclass correlation coefficient = 0.68 in repetitive administrations).11 The questionnaire tests students’ application of the program’s “5 clues” to make healthful food choices. Each question requires students to decide which of 2 sets of Nutrition Facts panels and ingredient lists within a category of packaged foods (breads, crackers, cereals, cereal bars, or cookies) is a better choice. Correct responses are summed for an overall score. The ABC for Fitness evaluation included tests for abdominal strength (curl-ups), upper extremity strength (push-ups), flexibility (sit and reach), and aerobic capacity (0.5-mile run). These are standardized tests used in Connecticut state fitness testing.
Procedures The study used a mixed methods design12 that utilized quantitative data to determine the efficacy of
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Promoting Healthy Lifestyles to Children
❘ E29
TABLE 1 ● School Demographic Data: 2010-2011 School Yeara
qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq Community 1 (School District 1)
Variable Socioeconomic status of students Percentage of students eligible for free or reduced price lunch Race/ethnicity of students American Indian Asian American Black Hispanic Pacific Islander White ≥2 races Total percentage of minority students in each school a From
Community 2 (School District 2)
Intervention School
Control School
Intervention School
Control School
13.5%
11.8%
18.1%
17.3%
0.0% 8.8% 14.2% 11.2% 0.0% 61.8% 4.1% 38.2%
0.6% 16.6% 11.2% 8.2% 0.0% 61.5% 1.9% 38.5%
0.0% 3.3% 2.7% 18.4% 0.0% 75.7% 0.0% 24.3%
0.0% 5.6% 5.3% 14.3% 0.0% 74.9% 0.0% 25.1%
Connecticut State Department of Education.10
the interventions and qualitative data to provide process evaluation feedback on program satisfaction. The quantitative component used a quasi-experimental, pretest/posttest design with a delayed intervention for the control group. Two schools each from 2 communities participated in the study. In each community, third-grade students in one school served as the control group and those in the other school served as the experimental group.
ence to university curriculum. The training process included key features of TTT models to ensure mastery of content, delivery methods, and ongoing trainee support (see Table 2). This approach differed from those reported in the literature in that graduate student trainees organized themselves, practiced content delivery independently, and developed additional program activities as part of their graduate level curriculum.
Program development A series of initial meetings was held with a prevention research center, university representatives, and school administrators to develop administrative support for the program and set realistic timelines for implementation and evaluation. The research team then met with principals and teachers in each intervention school to address scheduling, data collection, and expectations.
TTT approach The TTT approach was a collaborative, grant-funded academic-community partnership among a university, a prevention research center, and 2 communities with elementary schools. Prevention research center consultants provided initial training on the content and delivery of the 2 health programs (Nutrition Detectives and ABC for Fitness) to 145 university faculty and graduate students. The graduate students represented several disciplines, including occupational therapy, physical therapy, nursing, and physician assistant. Faculty in each discipline were also trained in order to provide oversight, support, and to bridge the experi-
TABLE 2 ● Components of the Multidisciplinary Train-the-Trainer Process Model qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq
Train-the-Trainer Process for School-Based Health Promotion Program University faculty recruitment Faculty attend a general information session Faculty provide suggestions to integrate Nutrition Detectives and ABC for Fitness into university curriculum Formal training session (3 hours) provided by prevention research center consultants Review issues of childhood obesity, fitness, lifestyle, and nutrition Participate in Nutrition Detectives and ABC for Fitness program training Teacher training Meet with elementary school principals and teachers to explain program Train teachers University student preparation Assign roles and specific program components to graduate students Practice presentation and hands-on activities Program implementation in elementary schools University students directly deliver Nutrition Detectives program Elementary teachers deliver ABC for Fitness program with support from University students and faculty
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
E30 ❘ Journal of Public Health Management and Practice
Program implementation Nutrition Detectives is an age-appropriate, “spy themed” program designed to help children understand the relationship between food and health and use tips (“clues”) to read food labels and make healthful food choices based on (1) recognition of marketing deceptions; (2) the first ingredient on the list; (3) avoidance of trans fat and high fructose corn syrup; (4) length of ingredient list; and (5) amount of fiber per serving in grain products.13 In this study, the 90-minute program was delivered during two 45-minute lessons within 1 week. The first lesson was a didactic presentation to help students understand nutrition and learn detective “clues.” The second lesson was a hands-on group activity to read packaged food labels and identify more healthful (“clued-in”) versus less healthful (“clue-less”) food items. The program was delivered in a total of 8 classrooms within the 2 intervention schools. ABC for Fitness is a daily physical activity program designed to interject fun, noncompetitive activities into classroom schedules around transition times when students may be restless and need refocusing. Its “activity bursts” typically last 5 to 6 minutes and include a warm-up, core aerobic activity, and a cool-down component. Activities can be imaginative moves, creative extensions of learning content, or advanced bursts delivered at regular intervals for a total of 30 minutes per day.3 In this study, classroom teachers received training, a manual, and a set of fitness cards, and then implemented the program independently with university support and consultation. In community 2, graduate students provided this support on a weekly basis as requested by the teachers.
Data collection For Nutrition Detectives, third-grade students in intervention schools completed the FLLANK questionnaire before and after program implementation. The FLLANK was administered once to students in control schools to compare baseline knowledge. Since it was assumed that control school scores would not likely change during the brief length of the intervention period (ie, two 45-minute sessions delivered within a week), a posttest was not administered. In community 1, graduate students administered the tests in control and intervention schools in November/December 2010. In community 2, graduate students administered the testing at the intervention school in February/March 2011 and at the control school in March/April 2011. For ABC for Fitness, baseline fitness tests were administered by physical education (PE) teachers during gym classes, with health science students assisting as needed to record the data. Posttesting was performed
by school PE teachers in 2 schools and by outside PE consultants at the other 2 schools. Community 1 data were collected in October 2010 (baseline) and April 2011 (posttest). Community 2 data were collected in January/February 2011 (baseline) and June 2011 (posttest). Anthropometric measures of height, weight, and body mass index (BMI) were measured in community 1 during the same testing period in the following manner: nurse practitioner students recorded the ages, height, and weights of children (clothed without shoes) using a manual scale with a height arm attachment. These data were not collected in community 2.
Focus groups In community 2, after 7 weeks of ABC for Fitness, university faculty and graduate students conducted four 40-minute focus groups in the intervention school: 1 with all 3 teachers and 3 with students (3 classes, 20 students per classroom). Teachers responded to questions concerning their experiences in implementing the program, their perceptions of its impact on students, and recommendations for the future. The focus group was audio recorded and transcribed for analysis. Third-grade student focus groups were activitycentered.14 To encourage reflection on the experience, students first completed a worksheet independently upon which they drew pictures and answered questions regarding what they learned, liked, and disliked about the program. Students then participated in a 5-minute activity burst and shared their spontaneous reflections with peers and the researchers. The researchers probed for further detail when necessary and wrote down key phrases and quotes.
Data analysis To assess group differences at baseline between intervention and control groups, Student t test was used for continuous outcomes and χ 2 test was used for categorical outcomes. Nutrition knowledge was assessed through repeated measures analysis of variance to assess within-group differences in the intervention group on the FLLANK from pretest to posttest. Repeated measures analysis of variance was used to assess both the between- and within-group differences for anthropometric and physical activity measures. Post hoc tests were conducted to assess the influence of within-group factors such as gender, age, and school district using the General Linear Model. Data were analyzed per protocol. The statistical software package SPSS version 17.0 was used to conduct all analyses.15 All tests were conducted with a 2-tailed α level set at .05. Transcribed data from focus group discussions and third-graders’
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Promoting Healthy Lifestyles to Children
drawings were analyzed for thematic content around guiding questions.
● Results A total of 110 children comprised the intervention group and 129 children served as control with delayed intervention (see Table 3). The proportions of students from community 1 and community 2 were comparable across study groups. The average age of students and the ratio of boys to girls were also comparable between intervention and control schools. Body weight, BMI, and BMI percentile measurements were comparable between students from community 1 intervention and control schools. Intervention school students had significantly higher baseline nutrition
❘ E31
knowledge/food label literacy than those in control schools (P < .05). Furthermore, at baseline, intervention school students scored significantly higher than control school students on the following fitness measures: curl-up, push-up, and 0.5-mile run (all P < .05) (Table 3). Some inconsistencies in data collection occurred during the study (see Table 3). In community 1, some baseline fitness data were not collected from the control school because of inadequate prior communication. In community 2, the control school failed to use the correct protocol for aerobic fitness measures. Although standardized testing procedures were used for all fitness testing, in 2 schools, PE consultants provided the posttesting (due to time constraints). Furthermore, although BMI is a typical outcome measure in childhood obesity intervention
TABLE 3 ● Baseline Characteristics of Participating Third-Grade Students
qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq
Variable Community/school district Community/school district 1 Community/school district 2 Average age of participating students (years) Gender of participating students Boys Girls FLLANK (% correct) Community/school district 1 Community/school district 2 Anthropometric Measuresa Body weight (pounds) BMI (kg/m2 ) BMI percentile Fitness Test Scores: Curl-Upb (no. completed) Community/school district 1 Community/school district 2 Fitness Test Scores: Push-upb (no. completed) Community/school district 1 Community/school district 2 Fitness Test: 0.5-mile runc (minutes) Community/school district 1 Community/school district 2 Fitness Test: Sit and Reach Leftb (inches) Community/school district 1 Community/school district 2 Fitness Test: Sit and Reach Rightb (inches) Community/school district 1 Community/school district 2 a School
Intervention N = 110
Control N = 129
56 (50.9%) 54 (49.1%) 8.4 ± 0.5
73 (54.1%) 62 (45.9%) 8.3 ± 0.5
.72
44 (40.0%) 66 (60.0%) 54.3 ± 17.4 53.1 ± 19.6 55.5 ± 15.1
59 (45.7%) 70 (54.3%) 47.7 ± 21.3 45.6 ± 21.2 50.4 ± 21.4
.92 .01 .05 .15
75.0 ± 18.1 18.1 ± 3.1 66.6 ± 28.8 20.2 ± 12.1 15.9 ± 10.9 24.4 ± 11.9 10.3 ± 6.0 9.2 ± 5.2 11.4 ± 6.5 5.4 ± 1.5 5.4 ± 1.4 5.4 ± 1.6 9.4 ± 2.4 8.6 ± 2.5 10.3 ± 2.0 9.5 ± 2.3 8.8 ± 2.4 10.4 ± 2.0
73.7 ± 17.7 17.9 ± 3.2 64.1 ± 30.2 16.0 ± 8.9 ... 16.0 ± 8.9 7.2 ± 5.0 ... 7.2 ± 5.0 4.9 ± 1.4 4.9 ± 1.4 ... 9.0 ± 1.7 ... 9.0 ± 1.7 8.9 ± 1.7 ... 8.9 ± 1.7
.70 .74 .66 .01 ... .00 .01 ... .00 .02 .04 ... .19 ... .00 .06 ... .00
district 1 schools only.
b School district 1 control post only. c School district 2 intervention only.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
P
.36
E32 ❘ Journal of Public Health Management and Practice studies, the community 2 school district board of education did not approve the measure after having initially agreed.
Nutrition detectives Comparison of FLLANK scores between intervention and control schools yielded a significant difference in baseline nutrition knowledge (Table 3). From pretest to posttest, overall nutrition knowledge scores on the 10-item test increased by 25.2% in the intervention group, which was statistically significant (P < .01). At the individual item level, the increase in correct responses ranged from 6% to 53%. Each intervention school, when analyzed in isolation, had significantly improved scores on the FLLANK as well (both P < .01) (Table 4). No significant interactions were found with regard to gender, age, or school district and improved FLLANK scores (P > .05).
ABC for fitness Community 1 students in intervention and control schools significantly differed with respect to baseline anthropometric measures. With the exception of sit and
reach measures, students in intervention and control schools improved on all fitness measures at posttest. Students in intervention schools showed greater improvement on all measures than control. These differences were statistically significant on the curl-up and push-up measures (P < .01) (Table 4). The fitness outcomes for 0.5-mile run (P = .06) and sit and reach left (P = .07) approached significance. No significant interactions were found with regard to gender, age, or school district.
Process evaluation qualitative findings Teacher perceptions Process evaluation themes centered around the benefits of program structure, classroom management, students’ awareness of their bodies, and support from graduate students. Although some teachers had used movement informally in the classroom prior to the program, they appreciated the consistency and structure of ABC for Fitness (warm-up, core activity, cooldown) as it provided a routine for the classroom. They enjoyed the opportunity to customize the frequency and timing of the activity bursts in their classrooms during the school day.
TABLE 4 ● Change From Baseline Characteristics of Participating Third-Grade Students
qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq Control
P
25.2 ± 22.2 27.3 ± 22.2 23.1 ± 22.3
N/A N/A N/A
.00 .00 .00
2.6 ± 28.8 1.2 ± 5.8 6.3 ± 44.2 9.1 ± 15.9 16.2 ± 14.6 1.2 ± 13.5 2.4 ± 5.6 3.0 ± 5.1 1.7 ± 6.2 − 0.5 ± 1.9 − 0.4 ± 1.5 − 0.6 ± 2.3 0.3 ± 1.6 0.4 ± 1.6 0.1 ± 1.7 0.0 ± 1.6 0.2 ± 1.5 − 0.2 ± 1.7
2.6 ± 8.6 0.6 ± 1.1 3.1 ± 11.4 0.7 ± 4.2 ... 0.7 ± 4.2 0.3 ± 1.3 ... 0.3 ± 1.3 0.0 ± 1.3 0.0 ± 1.3 ... −0.1 ± 0.6 ... −0.1 ± 0.6 −0.0 ± 0.6 ... −0.0 ± 0.6
.99 .53 .63 .00 ... .80 .00 ... .13 .06 .15 ... .07 ... .61 .82 ... .41
Variable
Intervention
FLLANK (%) (within group change in scores) Community/school district 1 Community/school district 2 Anthropometric Measuresa Body weight (pounds) BMI (kg/m2 ) BMI percentile Fitness Test Scores: Curl-upb (no. completed) Community/school district 1 Community/school district 2 Fitness Test Scores: Push-upb (no. completed) Community/school district 1 Community/school district 2 Fitness Test: 0.5-mile walk/runc (minutes) Community/school district 1 Community/school district 2 Fitness Test: Sit and Reach Leftb (cm) Community/school district 1 Community/school district 2 Fitness Test: Sit and Reach Rightb (inches) Community/school district 1 Community/school district 2 Abbreviation: N/A, not applicable. a School district 1 schools only. b School district 1 control post only. c School district 2 intervention only.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Promoting Healthy Lifestyles to Children
The teachers found the support from the graduate students useful for embedding curricular goals into activity bursts, clarifying intensity levels for core activities, and addressing any behavioral and student selfregulation issues that arose during the activity bursts. One teacher noted that “the graduate students helped me realize that activity bursts do not have to be all high-impact jumping. They also helped me establish strategies for a few students who needed help calming down.” Teachers indicated that third graders responded well to the “ABC Rules” they developed to promote classroom management. Teachers also developed their own strategies to help students understand the structure of the activity bursts, including the “ABC places” in the classroom. The most important outcomes that teachers reported were students’ positive attitudes toward fitness, awareness of their bodies, and improved fitness levels. One teacher stated, “I believe most of my students have developed a different mental idea of their own personal stamina with regard to exercise.” Another teacher said that “most [students] would be able to describe cool down and warm up versus core activity and the differences that happen in their bodies when they are doing each type of exercise.” The teachers did not report any changes in academic or behavioral outcomes as a result of the program.
Student perceptions Student themes centered around their preferences for the type of activity break and learning about the relationship between exercise and their bodies. Students enjoyed the game-like quality of the bursts and the opportunity “get up and move.” They most liked the highintensity bursts such as jumping jacks, running, and jumping. They reported learning about the importance of exercise and how to monitor their heart rate, stretch, and cool down. One third grader stated, “When you get active, your heart pumps more blood.” Another student who learned how to calm down explained, “You breathe in deeply through your nose and out through your mouth when you want to calm down.” They were enthusiastic about continuing the activity bursts in the classroom.
● Discussion Elementary school students’ food label literacy improved significantly after exposure to the Nutrition Detectives program when taught using the TTT approach. Findings demonstrate that children were able to perform on par with (if not slightly better than) children from a previously published Nutrition Detec-
❘ E33
tives pilot study.13 In our study, third-grade students improved their FLLANK scores by greater than 25%. This outcome exceeds that found in the pilot study,13 in which children in grades 2, 3, and 4 improved their FLLANK scores by 18% across grade levels, with a 23% improvement among third-grade students. These findings indicate that the TTT approach adapted for this study (ie, training university students to deliver the program in partnership with elementary school teachers) was as effective with regard to student performance. The findings also indicate that this gain in nutrition knowledge and skills in identifying healthful food choices, if sustained over time, could lead to improved health benefits among children who apply their knowledge and skills to improve the nutritional quality of their food choices. Body mass index findings were also similar to those from an ABC for Fitness pilot study in which intervention group students did not significantly change with regard to BMI at follow-up.3 Students in the current study displayed similar characteristics at followup. However, these results are tenuous at best because the entire sample of participants was not measured, as 1 school district decided not to measure BMI once program implementation began. Changes in physical fitness data from baseline in the intervention group showed significant improvements when compared with control on certain fitness measures and approached trend-level significance (P < .10) on others: 0.5-mile run and the sit and reach left measures. Trend-level significance suggests that including a greater number of students in the analysis could have led to statistical significant differences between groups. It is important to note that these data were not consistently collected at all time points across study groups and, therefore, some comparisons were “underpowered.” Students in an ABC for Fitness pilot study3 also demonstrated similar improvements in physical activity measures, therefore, suggesting that the program delivery of ABC for Fitness in this current study demonstrates fidelity to the original program design.
Relationship to previous school-based healthy lifestyle programs This study supports findings from a recent review of more than 25 school-based physical activity and nutrition education programs focusing on childhood obesity prevention.4 The review indicates that most physical activity/nutrition programs demonstrate measurable changes in nutrition knowledge on the basis of valid pretest/posttest comparisons. However, most studies of short duration (
