THE EFFECT OF A MENTOR BASED EXERCISE INTERVENTION ON BODY MASS INDEX Z-SCORES, DUAL-ENERGY X-RAY ABSORPTIOMETRY : TOTAL % BODY FAT LEVELS, DEPRESSION, AND FITNESS LEVELS IN OBESE ADOLESCENTS

Abstract

Project MENTOR was a mentor-based exercise intervention for overweight children, ages 12-18 years. Undergraduate exercise physiology students (n=10) served as mentors to facilitate healthy lifestyle modifications through exercise based on C.A.R.E. (competence, autonomy, relatedness, and enjoyment from self-determination theory) and lifestyle coaching using cognitive-behavioral strategies. Mentors led exercise training three times a week and implemented a weekly lesson focused on self-regulatory skills and goal-setting related to healthy eating and lifestyle activity. The current study examined MENOR's [sic] impact on BMI-Z scores, DXA-total percent body fat (% BF), depression, and physical fitness levels assessed by VO₂ max and total treadmill time percent change. Obese female (n=11) and male (n=12) adolescents, including Caucasian (n=13) and African American (n=10) participants were randomly assigned to two conditions: a 14 weeks intervention or wait-list control. Five additional participants completed a 22-week intervention. Participants from all conditions were assessed at baseline, after 14 weeks, and 3 months follow-up. The 28 participants had an initial BMI Z-score of 2.17 (SD=0.40) and DXA-total % BF score of 43.66% (SD=6.28), VO₂ max values of 27.15 (SD=6.87) ml/kg/min along with a depression t-score of 53.53 (SD=1.79). At baseline, there were no significant differences between groups on any of the outcome variables. Repeated measures ANOVA's revealed nonsignificant time by group interactions for BMI Z-scores (p=.12) and DXA-total % BF (p=.18). Although the interaction effects were not significant, effect sizes were evaluated using Cohen d given the low statistical power associated with small sample sizes. There was minimal change in the control group's BMI Z-score and DXA-total % BF at 14-weeks and follow-up compared to baseline with Cohen d ranging from .005-.09. In comparison, both intervention groups made small to moderate improvements in BMI Z-scores and % BF at post-assessment (BMI Z-scores: d=.31 for the 22-week intervention and d=.23 for the 14-weeks; DXA-% BF: d=.63 for the 22-weeks and d=.23 for the 14 weeks). The 14-week intervention group did not maintain changes in BMI Z-scores or % BF levels at follow-up (d=.02 for BMI Z-scores and d=.04 for % BF). The 22-week intervention group, however, was able to improve their BMI Z-scores (d=.62) and % BF (d=.57) at follow-up. There was a significant interaction for VO₂ max values (p=.01). Based on effect sizes, the 22-week intervention (Cohen d=.003 and .12) and control group (Cohen d=.04 and .21) showed minimal change in VO₂ across the three time points; whereas the 14-weeks intervention had a large increase from pre to post-assessment (d=.89) and then decreased toward baseline at follow-up (d=.29). For depression, there was a nonsignificant group X time interaction (p= .21). Inspection of effect sizes revealed that all three groups showed a moderate decrease in depression scores from pre to post test. Whereas both the control and prolonged intervention maintained those improvements, the 14 weeks intervention did not. In summary, although both intervention groups showed some improvement across all outcome variables post intervention, only the prolonged 22-week intervention maintained improvements after a no treatment follow-up. Overall, these findings suggest that a longer treatment may yield sustained improvement or maintenance in weight related outcomes.