An environmental intervention to prevent excess weight gain in African-American students: a pilot study

Abstract

Purpose: To examine the influence of an environmental intervention to prevent excess weight gain in African-American children.

Design: Single-group repeated measures.

Setting: The intervention was delivered to a school composed of African-American children.

Participants: Approximately 45% (N = 77) of enrolled second through sixth grade students.

Intervention: The 18-month intervention was designed to alter the school environment to prevent excess weight gain by making healthier eating choices and physical activity opportunities more available.

Measures: Body mass index percentile was the primary outcome variable. Body mass index z score was also calculated, and percent body fat, using bioelectrical impedance, was also measured. Total caloric intake (kilocalories) and percent kilocalories from fat, carbohydrate, and protein were measured by digital photography. Minutes of physical activity and sedentary behavior were selfreported.

Analysis: Mixed-models analysis was used with covarying baseline values.

Results: Boys maintained, whereas girls increased, percent body fat over 18 months (p = .027). All children decreased percent of kilocalories consumed from total and saturated fat and increased carbohydrate intake and self-reported physical activity during the intervention (p < .025). Body mass index z score, sedentary behavior, and total caloric intake were unchanged.

Conclusion: The program may have resulted in maintenance of percent body fat in boys. The percent body fat in girls steadily increased, despite similar behavioral changes as boys. School-based interventions targeting African-American children should investigate strategies that can be effective across gender.

Figure 1. Change in percent body fat for boys and girls by body mass index percentile category

There was a three way interaction for percent body fat (p = .027). At the end of treatment, girls had significantly increased their percent body fat, whereas there was no significant change in boys. Error bars refer to standard error. For each measurement period, the sample sizes were 77, 59, and 55, respectively.