SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discusson
  7. DISCLOSURE
  8. References

A healthy lifestyle school-based obesity intervention was evaluated in a rural southern community where the rate of obesity ranks as the highest. School-age children (N = 450) ranging from 6 to 10 years of age (Mage = 8.34) participated in monthly physical activity and nutritional events during a 9-month academic year. The children's nutritional knowledge, number of different physical activities, fitness level, dietary habits, waist circumference, BMI percentile, and percentage body fat were measured pre- and postintervention. Changes on these measures were compared to students in a school employing the school system's standard health curriculum. Regression analyses with residualized change scores revealed that the intervention school showed statistically significant improvement in percentage body fat, physical activity, performance on fitness tests, and dietary habits compared to the control school. There was no evidence of differences in outcomes based on gender or ethnicity/race. With rates of obesity and overweight reaching 50% in southern rural communities, intervening early in development may offer the best outcome because of the difficulties with changing lifestyle behaviors later in adulthood. A population-based approach is recommended over a targeted approach to cultivate a culture of healthy lifestyle behaviors when children are developing their health-care habits. Evidence suggests that both boys and girls, and African-American and white children can benefit equally from such interventions.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discusson
  7. DISCLOSURE
  8. References

Obesity is linked to some of the leading causes of morbidity and early mortality (1); however, efforts to reduce this growing health risk have focused on secondary and tertiary services with limited success. The International Obesity Task Force recommends instead that we focus on preventing unhealthy weight gain before it develops (2). Theorists and clinicians recommend using the Social Learning Theory to achieve this goal because it addresses the dynamic interaction between the individual, his/her behavior, and the environment (3), rather than focusing on just the individual and his/her eating behavior for example. Intervening early in development should yield the best results because lifestyle habits are established early in childhood (4). Utilizing a population-based approach appears to offer the most promise because it is economical, reaches more people, and individualized treatment programs have proven ineffective longitudinally (5,6,7).

The most effective interventions to date address both diet and physical activity to account for the roles that caloric intake and energy expenditure play in weight gain (8). The types of treatment gains vary and depend on the outcome variables utilized. School children in the Eat Well and Keep Moving project, for example, showed a decline in television viewing time and an increase in fruit and vegetable consumption (9,10). Warren and colleagues (11) also found improvements in young children's vegetable/fruit intake as well as nutrition knowledge after they completed 20 weeks of either a nutrition education only, physical activity only, or a combined intervention. The authors concluded that it is likely that the successful outcomes for all three interventions were a function of contamination effects, as all three interventions were implemented simultaneously in the same school. They also concluded that parents should be included in school-based programs to enhance generalizations of treatment effects beyond the school setting. Edmundson et al. (12) provided empirical support for the therapeutic benefits of including parents in school-based obesity treatments.

In addition to parental involvement, the growing health disparities among ethnic/racial groups for excessive weight and weight-related diseases suggest that a representative sample of minority youth should be included in evaluations of childhood obesity interventions (8). Community-based programs have produced reductions in BMI and body fat as well as improved fitness scores among African-American parents and Mexican American youths (13,14). However, improvements in BMI may have been an anomaly as other school-based programs have not yielded significant effects on adiposity measures for minority youth (15,16). Furthermore, although clinicians espousing to an ecological framework recommend tailoring obesity interventions to the minority youth's culture to successfully effect lifestyle changes (8), few studies have examined possible ethnic/racial differences in obesity treatment outcome.

The goal of the present study was to apply the Social Learning Theory to a school-based childhood obesity intervention program in a state where the rate of obesity is the highest in the nation—Mississippi (1,17). An interdisciplinary approach was undertaken to promote healthy lifestyle behaviors early in development, which included involving teachers, the state's education department, health professionals from state academic institutions (universities), and primary caretakers—mothers; hence, the acronym—TEAM Mississippi. The program incorporated elements from established school-based programs including Pathways (7) and the CATCH project (18), and included monthly nutritional and physical activity events designed to promote healthy dietary habits and physical activity. African-American and white elementary school children were compared on outcome variables—including dietary habits, physical activity, fitness, and adiposity measures—to evaluate for possible disparities in treatment outcome.

Methods and Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discusson
  7. DISCLOSURE
  8. References

Participants

Participants included 450 children ranging from 6 to 10 years of age (Mage = 8.34, s.d. = 1.30); 204 attended the intervention school and 246 attended the control school. The community populations were comparable and ranged from 3,500 to 4,500 residents. Median income was also comparable: $30,713 for the intervention vs. $29,904 for the control community. Approximately two thirds were African American (63% control; 58% intervention) and the remaining students were white (37% control; 42% intervention). Approximately half were male (52%) and 48% were female. The ethnic/racial composition of the sample allowed for statistical comparisons for treatment effects between African-American and white youth. The participating schools were the only two schools in their respective communities, precluding testing additional schools in the community. Tests of differences revealed that the two schools did not differ significantly on mean age, BMI percentile, percentage body fat, nutrition knowledge, physical activity level, and performance on curl-ups at baseline, F's (1, 448) = 0.01–3.19, P > 0.05. However, the intervention school's mean waist circumference was smaller (26.55 vs. 27.77 in), F(1, 448) = 5.44, P = 0.02; they were also slower in completing the shuttle run (13.99 vs. 12.48 s), F(1,448) = 81.90, P < 0.0001, but they stretched farther on the V-sit fitness test (27.63 vs. 25.11 cm), F(1,448) = 14.46, P = 0.0002.

Procedure

After obtaining institutional review and school board approval, the schools were randomly assigned to the intervention and control conditions. Interventions were not manipulated within schools because of possible contamination effects with experimental manipulations in the same school (11). Parents provided written informed consent for participation. Exclusion criteria included disabilities that precluded comprehending the questionnaires or performing the fitness tests. Only one student at the control school was excluded because s/he met this criterion. Out of the 567 children enrolled at both schools, 26 declined to participate (10 at the intervention, 16 at the control school), and 34 failed to return their consent form, leaving a total of 507 participants. Fifty-seven students had moved out of the school district by the postintervention assessment, yielding an overall attrition rate of 11% (10% for intervention school and 12% for control school). Hence, statistical analyses were conducted on the 450 students with complete data.

Baseline data, including height, weight, waist circumference, percentage body fat, performance on three fitness tests, nutrition knowledge, and self-report physical activity, were collected at both schools during a weeklong assessment. Trained professionals from the Department of Education and from local academic institutions (universities) conducted physical measurements and fitness tests individually for each child; and research assistants from academic institutions administered the nutrition knowledge and physical activity checklists with groups of 15–20 children during class time. Parent-report dietary habit questionnaires were completed by the primary caregiver at home (i.e., mother) and returned to the school. An 8-month intervention program was initiated and completed at the intervention school following completion of baseline assessments at both schools. The control school followed the state's standard health curriculum, which included didactic nutrition education, health information incorporated into academic lessons, and weekly physical education classes.

The intervention program included monthly family events that alternated between nutrition and physical activities/contests. Focus groups were held with community residents prior to initiating a program to obtain their input on treatment activities that would complement the community's activities. Examples of events included a healthy tailgating party at the high-school football game, which is typically attended by a large percentage of the community. The schoolchildren and parents prepared healthy recipes for the event. An example of a physical activity event included a parent-child softball throw contest at the beginning of the baseball season (see Table 1 for outline of intervention). Cooking and physical activity equipment/supplies were distributed as prizes to winners of the various contests. In addition to the activities, there were changes to the intervention school's food service including replacing the deep frying equipment with baking ovens. Rather than applying a prearranged program, the events were designed to coincide with community activities. At the conclusion of 8 months, a postassessment of the same variables assessed at baseline was conducted at both schools.

Table 1.  Standard control health curriculum and intervention program
inline image

Measures

Adiposity measures. Each child's height was measured without shoes using a scale-mounted stadiometer. Weight was measured in light clothing without shoes using a physician's balance scale. BMI was calculated for each child (weightkg/heightm2) and BMI percentile was determined using the most recent Centers for Disease Control's BMI for age- and sex-specific growth charts (19). Obesity is defined as BMI ≥95th percentile for age and sex. Bioelectrical impedance analysis was performed on each child to estimate percentage body fat using a BF-350A body composition analyzer (Tanita, Arlington Heights, IL). Bioelectrical impedance analysis has been found to correlate significantly with dual energy X-ray estimates of fatness (0.86), skinfold thickness (0.83), and BMI (0.82) (20). Intraclass correlations based on replicate measures are high (>0.994). The children's percentage body fat was measured at the same time of the day at pre- and postassessment. Each child's waist circumference was also measured in inches by placing a tape measure around the abdomen at the level of the iliac crest.

Nutrition knowledge. Students completed the Know Your Body Questionnaire, a 12-item multiple-choice measure of nutrition knowledge for children (21). Response items included two possible choices that were presented pictorially for questions about healthy food choices. For example, one item asks respondents to select which is better for you to eat, fried chicken or chicken baked in an oven? The reading level for the measure is at the 1st grade level. Correct responses were summed to yield a total score.

Fitness. The children completed three fitness tests from the President's Challenge Physical Activity and Fitness Awards Program including the shuttle run, curl-ups, and V-sit. The outcome variables included the number of seconds in which the child completed the shuttle run, the number of curl-ups completed in 60 s, and the distance in centimeters stretched for the V-sit.

Physical activity. Students completed a 21-item checklist of age-appropriate physical activities (e.g., swimming, jumping rope). Similar checklists have been used in health promotion studies and correlate with an interview format (r = 0.76, P < 0.001), a measure of heart rate (r = 0.57, P < 0.001), and accelerometer scores (r = 0.30, P < 0.001) (22,23,24). The children checked each activity that they engaged in the previous day for a minimum of 15 min. Responses were summed for a total score.

Dietary habits. The children's primary caregivers completed the 17-item Child Dietary Fat Questionnaire about the child's dietary habits. Caregivers indicated on a 5-point scale (1 = never, 5 = always) how frequently they served high fat foods and prepared fried foods for the child. For example, respondents are asked, “In the past month, for breakfast, how often did you serve breakfast meats (bacon or sausage)?” A psychometric study of the measure revealed that corresponding items correlated with total fat intake (r = 0.68, P < 0.0001), saturated fatty acids intake (r = 0.75, P < 0.0001), and dietary cholesterol intake (r = 0.57, P < 0.0001) (25). Test-retest reliabilities for predicting children's dietary intake of fat, saturated fatty acids, and cholesterol are 0.41, 0.66, and 0.64, respectively. Criterion-related validity evaluated against 4 days of dietary records have yielded significant correlations for total fat (r = 0.54, P < 0.0001), for saturated fatty acids (r = 0.36, P < 0.01), and dietary cholesterol (r = 0.55, P < 0.0001). A sum score was used in analyses after reversing items representing unhealthy dietary behavior. Higher sum scores reflect healthy dietary habits. The caregivers also completed an identical form in regard to their own intake of high fat foods. The questionnaire includes the same questions from similar measures, e.g., Food Frequency Questionnaire (26).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discusson
  7. DISCLOSURE
  8. References

As shown in Table 2, 32 and 33% of the children in the intervention and control schools, respectively, fell at or above the 95th percentile for BMI at baseline, and they had an average of 26.17 and 27.15% body fat, respectively. We regressed the post-test scores for the various adiposity measures on the respective pretest scores (27). The resulting standardized residuals (residualized change score) represent change in the adiposity measure adjusted for baseline variances. Using linear regression analyses with the residualized change score for the respective adiposity measure as the criterion variable, a treatment effect was only observed for percentage body fat. More specifically, the intervention school showed a statistically significant decline in percentage body fat, whereas the control school children's percentage body fat remained fairly stable, F(1,449) = 5.56, P = 0.02 (see Figure 1). The intervention school reported engaging in significantly more physical activities from baseline to postintervention whereas the control school reported a decline in physical activities, F(1,449) = 4.56, P = 0.04 (see Figure 2). The intervention school also showed improvement in their dietary fat intake compared to the control school, F(1,449) = 12.30, P < 0.0005 (see Figure 3). Interestingly, the caregivers from the intervention school did not show significant changes in their own dietary fat intake. Nevertheless, their dietary fat intake remained stable whereas the caregivers from the control school reported more dietary fat intake at postintervention compared to baseline level, F(1,449) = 4.32, P < 0.04. The children from the intervention school also showed statistically significant improvement in their performance on two of the three fitness tests from the President's Challenge, including curl-ups, F(1,449) = 30.69, P < 0.0001, and the shuttle run, F(1,449) = 52.24, P < 0.0001. Tests for gender and ethnic/racial differences yielded no significant moderating effects on outcome variables (ts = −0.06 — 1.28, P > 0.10).

Table 2.  Tests of comparisons for intervention and control schools on outcome variables
inline image
image

Figure 1. Treatment effect on the children's percentage body fat.

Download figure to PowerPoint

image

Figure 2. Treatment effect on the number of physical activities the children engaged in the previous day.

Download figure to PowerPoint

image

Figure 3. Treatment effect on the children's healthy dietary behavior.

Download figure to PowerPoint

Discusson

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discusson
  7. DISCLOSURE
  8. References

Rural southern children are especially vulnerable to obesity with rates reaching as high as 33% in the present study. This rate is twice the national average (1), which is likely a function of differences in methodology. Most prevalence rates for obesity, for example, derive from random phone interviews with community residents (1,17), whereas the rates of obesity and overweight in the present study were obtained from measuring the children directly. The urgency for effective childhood obesity interventions is apparent from the growing health, social, and economical costs associated with the rising rates of obesity. Individual treatments have yielded short-term weight loss (2), but barriers to maintaining healthy lifestyle behaviors including the lack of environmental support appear to preclude maintaining initial treatment gains. School-based interventions designed to instill healthy lifestyle behaviors by engaging at-risk children and their parents in nutritional and physical activities might be a viable solution for preventing childhood obesity, as children in the present study showed a statistically significant reduction in their percentage body fat compared to students in the school system's standard health curriculum. The health benefits of the program also extend to improved performance on fitness tests and reductions in dietary fat intake. Although statistically significant, the intervention school showed a <1% change in mean percentage body fat. This magnitude of change hardly meets criteria for clinical significance if applying a change of 1–2 s.d. (28). However, when applying percentage body fat cutoff scores for obesity, the boys nearly fell in the overweight status (cutoff = 25%) at post-test. The cutoff percentage body fat score for obesity is higher for girls (32%). Hence, they already scored below the range for obesity in the present study. Yet, it is noted that achieving successful weight management is an ongoing process that involves managing multiple parameters including caloric intake, energy expenditure, social and cultural dietary influences, and other factors. Hence, weight loss and management requires incremental changes. Research further suggests that incremental improvements in body weight/fat and maintenance over time is healthier than large changes in short periods of time or frequent multiple changes (29).

Interestingly, while children in the intervention program showed improvements in their dietary fat intake, their parents did not. It is important to note, however, that the parent's dietary fat intake remained stable, whereas the parents from the control condition reported an increase in their dietary fat intake at postintervention. Perhaps changes in parents' dietary behaviors require more time because of their intransigent health-care behaviors. It is also noted that improvements in the children's dietary behaviors occurred in spite of the lack of any statistically significant changes in their nutrition knowledge; thus supporting assertions that changing children's eating behaviors is not necessarily dependent on their knowledge about nutrition.

Other studies have also reported improvement in dietary behavior following school-based interventions that included family participation (12) but this study is unique because it is one of only a handful of studies that report significant effects on an adiposity measure. The present program's relatively short intervention may have precluded more extensive treatment effects on other measures of adiposity including changes in BMI percentile and waist circumference. Although others have also failed to find treatment effects on children's BMI, they have observed effects on other anthropometric measures (e.g., skinfold thickness; 11,25,30,31,32). These findings suggest that more sensitive measures of changes in body composition may be warranted when examining treatment effects for childhood obesity interventions (31). Although we did not observe a significant treatment effect on BMI percentile, the intervention school children's mean BMI percentile was stable at postintervention; whereas the control school children's BMI percentile increased 5% points. Perhaps these findings might lead to statistically significant differences with longer interventions. On the other hand, there was an increase in the percentage of children in the intervention school that fell in the 85th–95th BMI percentile; thus raising questions about the validity of the BMI percentile as a measure of adiposity. Others have also criticized using BMI percentile as a reliable measure of adiposity (33). Hence, a multimethod approach is recommended to assessing adiposity in future research.

Although school-based obesity intervention programs have been tested with minority groups, this is the first study to date that examined ethnic/racial differences in treatment outcome within a rural southern community where the rate of obesity is among the highest in the nation. The present treatment effects support approaching childhood obesity from a community-based approach rather than from a targeted approach. Furthermore, designing a program with consideration given to the community's culture is recommended over applying a generic intervention program. However, the lack of interaction effects observed for gender and ethnicity/race suggests that the present program is equally effective for girls and boys and for African American and white youth alike. Perhaps it is the social contagion that influences lifestyle behaviors and not ethnic/racial factors per se. Nevertheless, further research is recommended to ascertain the role of culturally sensitive variables in treatment outcome for minority vs. nonminority youth (8).

Methodological limitations

Methodological limitations include the use of self- and/or parent-report measures for physical activity and dietary fat intake. Nevertheless, objective adiposity and fitness measures were included to overcome possible reporting biases. Although significant effects were observed on these measures, objective measures of dietary behavior and physical activity are strongly recommended for further research. Significant group differences on waist circumference and performance on the V-sit and shuttle run fitness tests at baseline further limit conclusions about changes for these variables at postintervention. However for one variable, the shuttle run test, the intervention school took longer to complete the test compared to the control school at baseline and yet improved significantly at postintervention, whereas the control school's performance worsened. Although the intervention group made treatment gains on some variables, both the intervention and control groups performed about the same on the V-sit test and had somewhat larger waist circumferences at postintervention. Developmental factors such as physical growth and the nature of the fitness tests may have contributed to these findings.

Finally, research suggests that different adiposity measures may not be used interchangeably and should be evaluated separately (20). Furthermore, some adiposity measures are under scrutiny. BMI, for example, is criticized for not being a precise indicator of underlying proportion of fat and lean tissue (33). At any particular BMI, body composition varies greatly in children depending on gender, age, maturity, race, height, and body fat distribution. However, BMI has been found to be predictive of elevated fat mass and percentage body fat when above the 85th percentile in children. Likewise, Freedman and Sherry (34) reported recently that a BMI for age ≥95th percentile of the Centers for Disease Control reference population is a moderately sensitive and specific indicator of excess adiposity among children. As mentioned, a multimethod assessment of adiposity including waist circumference, BMI percentile, and percentage body fat was used to overcome the limitation of using just one method. Future research is recommended to ascertain the most appropriate fitness and biological markers for measuring childhood obesity.

The present findings are consistent with a community-based intervention recently conducted in France that reported a decline in the prevalence of overweight children (35). An entire French community including the mayor and restaurant owners was involved in promoting physical activity and healthy eating. The children in the present intervention also showed lifestyle changes that likely contributed to changes on an adiposity measure. Interestingly, the children showed positive changes in their dietary fat intake in spite of the fact that their caregivers did not. It would seem that both the caregiver and child would need to exhibit changes to promote enduring lifestyle changes. However, it may be that caregivers are heeding lifestyle changes for their children, whereas they require more time to change their own habitual behaviors. Other interventions that address more specific issues that can interfere with healthy eating and exercise behaviors, such as stress management, may be necessary to effect changes in the caregivers' behaviors. In the meantime, as recommended by Katan (36), “obesity may be a problem that cannot be solved by individual persons but requires community action.” The earlier this action is undertaken in children's development, the greater the likelihood of sustaining treatment gains. The simplicity of the TEAM Mississippi program and its efficacy suggest that this program could be one option as it can be administered easily by school health champions (e.g., school nurse, counselor) with minimal financial expenditure. Local businesses, vendors, the Department of Education, and local universities helped to offset expenses by donating services and products (e.g., door prizes, food ingredients), which resulted in total expenditures of $900.

DISCLOSURE

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discusson
  7. DISCLOSURE
  8. References

The authors received funding from the University of California, San Francisco National Center of Excellence in Women's Health and the Johnson & Johnson Company to conduct this study. The authors designed the study, performed all the analyses, and the sponsors had no editorial control over the article.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discusson
  7. DISCLOSURE
  8. References
  • 1
    Centers for Disease Control and Prevention. Overweight and obesity: U.S. http:www.cdc.govobesitydatatrends.html. 2009.
  • 2
    Basdevant A, Boute D, Borys JM. Who should be educated? Education strategies: could children educate their parents? Int J Obes Relat Metab Disord 1999;23 Suppl 4:S102; discussion S12.
  • 3
    Gittelsohn J, Merkle S, Story M et al. School climate and implementation of the Pathways study. Prev Med 2003;37:S97106.
  • 4
    Greening L, Stoppelbein L, Konishi C, Jordan SS, Moll G. Child routines and youths' adherence to treatment for type 1 diabetes. J Pediatr Psychol 2007;32:437447.
  • 5
    Harrell JS, Gansky SA, McMurray RG et al. School-based interventions improve heart health in children with multiple cardiovascular disease risk factors. Pediatrics 1998;102:371380.
  • 6
    Edmunds L, Walters E. Childhood obesity. In: Moyer VA et al. (eds.) Evidence Based Paediatrics and Child Health. BMJ Books: London, 2001, pp 141153.
  • 7
    Davis SM, Clay T, Smyth M et al. Pathways curriculum and family interventions to promote healthful eating and physical activity in American Indian schoolchildren. Prev Med 2003;37:S24S34.
  • 8
    Wilson DK. New perspectives on health disparities and obesity interventions in youth. J Pediatr Psychol 2009;34:231244.
  • 9
    Gortmaker SL, Peterson K, Wiecha J et al. Reducing obesity via a school-based interdisciplinary intervention among youth: Planet Health. Arch Pediatr Adolesc Med 1999;153:409418.
  • 10
    Gortmaker SL, Cheung LW, Peterson KE et al. Impact of a school-based interdisciplinary intervention on diet and physical activity among urban primary school children: eat well and keep moving. Arch Pediatr Adolesc Med 1999;153:975983.
  • 11
    Warren JM, Henry CJ, Lightowler HJ, Bradshaw SM, Perwaiz S. Evaluation of a pilot school programme aimed at the prevention of obesity in children. Health Promot Int 2003;18:287296.
  • 12
    Edmundson E, Parcel GS, Perry CL et al. The effects of the child and adolescent trial for cardiovascular health intervention on psychosocial determinants of cardiovascular disease risk behavior among third-grade students. Am J Health Promot 1996;10:217225.
  • 13
    Engels HJ, Gretebeck RJ, Gretebeck KA, Jiménez L. Promoting healthful diets and exercise: efficacy of a 12-week after-school program in urban African Americans. J Am Diet Assoc 2005;105:455459.
  • 14
    Johnston CA, Tyler C, McFarlin BK et al. Weight loss in overweight Mexican American children: a randomized, controlled trial. Pediatrics 2007;120:e1450e1457.
  • 15
    Himes JH, Ring K, Gittelsohn J et al. Impact of the Pathways intervention on dietary intakes of American Indian schoolchildren. Prev Med 2003;37:S55S61.
  • 16
    Stone EJ, Norman JE, Davis SM et al. Design, implementation, and quality control in the Pathways American-Indian multicenter trial. Prev Med 2003;37:S13S23.
  • 17
    Lower Mississippi Delta Nutrition Intervention Research Consortium. Self-reported health of residents of the Mississippi delta. J Health Care Poor Underserved 2004;15:645662.
  • 18
    Luepker RV, Perry CL, McKinlay SM et al. Outcomes of a field trial to improve children's dietary patterns and physical activity. The Child and Adolescent Trial for Cardiovascular Health. CATCH collaborative group. JAMA 1996;275:768776.
  • 19
    Kuczmarski RJ, Ogden CL, Grummer-Strawn LM et al. CDC growth charts: United States. Adv Data 2000;127.
  • 20
    Gutin B, Litaker M, Islam S et al. Body-composition measurement in 9–11-y-old children by dual-energy X-ray absorptiometry, skinfold-thickness measurements, and bioimpedance analysis. Am J Clin Nutr 1996;63:287292.
  • 21
    Williams CL, Carter BJ, Eng A. The “Know Your Body” program: a developmental approach to health education and disease prevention. Prev Med 1980;9:371383.
  • 22
    Marcoux MF, Sallis JF, McKenzie TL, Marshall S, Armstrong CA, Goggin KJ. Process evaluation of a physical self-management program for children: SPARK. Psychol Health 1999;14:659677.
  • 23
    Sallis JF, McKenzie TL, Alcaraz JE et al. The effects of a 2-year physical education program (SPARK) on physical activity and fitness in elementary school students. Sports, Play and Active Recreation for Kids. Am J Public Health 1997;87:13281334.
  • 24
    Sallis JF, Strikmiller PK, Harsha DW et al. Validation of interviewer- and self-administered physical activity checklists for fifth grade students. Med Sci Sports Exerc 1996;28:840851.
  • 25
    Allison DB, Faith MS, Heo M, Kotler DP. Hypothesis concerning the U-shaped relation between body mass index and mortality. Am J Epidemiol 1997;146:339349.
  • 26
    National Cancer Institute. Risk Factor Monitoring and Methods: U.S. <http:riskfactor.cancer.govdietusualintakesffq.html>
  • 27
    Cohen J, Cohen P. Applied Multiple Regression/Correlation Analyses for the Behavioral Sciences. 2nd edn. Lawrence Erlbaum: New York, 1983.
  • 28
    Jacobson NS, Truax P. Clinical significance: a statistical approach to defining meaningful change in psychotherapy research. J Consult Clin Psychol 1991;59:1219.
  • 29
    Barlow SE, Dietz WH. Obesity evaluation and treatment: Expert Committee recommendations. The Maternal and Child Health Bureau, Health Resources and Services Administration and the Department of Health and Human Services. Pediatrics 1998;102:E29.
  • 30
    Mcmurray RG, Harrell JS, Bangdiwala SI et al. A school-based intervention can reduce body fat and blood pressure in young adolescents. J Adolesc Health 2002;31:125132.
  • 31
    Neumark-Sztainer D, Story M, Hannan PJ, Rex J. New Moves: a school-based obesity prevention program for adolescent girls. Prev Med 2003;37:4151.
  • 32
    Singh AS, Chin A, Paw MJ, Brug J, van Mechelen W. Dutch obesity intervention in teenagers: effectiveness of a school-based program on body composition and behavior. Arch Pediatr Adolesc Med 2009;163:309317.
  • 33
    Demerath EW, Schubert CM, Maynard LM et al. Do changes in body mass index percentile reflect changes in body composition in children? Data from the Fels Longitudinal Study. Pediatrics 2006;117:e487e495.
  • 34
    Freedman DS, Sherry B. The validity of BMI as an indicator of body fatness and risk among children. Pediatrics 2009;124 Suppl 1:S23S34.
  • 35
    Romon M, Lommez A, Tafflet M et al. Downward trends in the prevalence of childhood overweight in the setting of 12-year school- and community-based programmes. Public Health Nutr 2009;12:17351742.
  • 36
    Katan MB. Weight-loss diets for the prevention and treatment of obesity. N Engl J Med 2009;360:923925.