|Abstract||Obesity and inactivity in children are at all-time highs and have been steadily increasing over the last thirty years. The school environment provides the ideal setting to reach a large number of children across diverse populations to try to reverse these trends. However, there are many inconsistent results in school-based physical activity interventions due to implementation length, time for activity, and the use of structured physical activities. The LiiNK project is a whole child intervention addressing those gaps by providing children 60-minutes of outdoor, unstructured play each day that they are in school while control children receive 30-minutes. Previous physical activity data on LiiNK 1st and 2nd grade children has shown they will average more than 100-minutes of moderate to vigorous physical activity (MVPA) during the school day. Even with this increase in activity, prior data on body mass index (BMI) in LiiNK children has also shown inconsistent results. This raised the question whether BMI was the most accurate assessment in populations with high amounts of MVPA daily since it does not directly measure changes in body composition. Our first study examined obesity classification differences between BMI and bio-electrical impedance analysis (BIA). We chose to use BIA over other body composition techniques since it directly measures body fat and has high specificity and sensitivity in categorizing those who are obese. We found that there was about a 30% difference in how BMI and BIA categorized children into different obesity categories, with the biggest difference in overfat. Similar discrepancies were also seen in Studies 2 and 3. These results confirmed our prior suspicions that BMI may not be telling the whole story on the effects of the LiiNK project on childhood obesity. This new measure was then used to examine changes in body composition among LiiNK intervention and control children in Studies 2 and 3. In Study 2, body composition was measured in 2nd-5th grade children finding 2nd graders decreased obesity prevalence at a higher rate than other sub groups. This may have been a result of a daily increase in MVPA as demonstrated by prior LiiNK physical activity data, so in Study 3 we measured physical activity and body composition in 3rd and 4th grade children. We found that intervention 3rd grade children with 60-minutes of recess had the highest MVPA minutes, which led to the greatest decrease in obesity prevalence. In addition, we found that physical activity decreased with age and that this may be due to differences in play preferences as children get older and by sex and race. Across both studies, we found lower obesity prevalence and body fat percentage in females and older children. Among older children, these results suggest there may be accumulative longitudinal effects from increased time for recess. In conclusion, BMI should be used with caution when assessing body composition in populations with increased MVPA. More reliable measures such as BIA should be used to obtain more accurate results across all body fat categories. Additionally, outdoor, unstructured play is a reliable tool to increase physical activity and decrease obesity rates in children. Finally, 60-minutes of outdoor, unstructured play will allow children to experience the greatest improvements in physical activity and body composition.