Comparison of Pedometer and Accelerometer Activity Levels Among Fourth and Fifth Grade Students During the School Day

Background and Purpose Pedometers are popular, relatively inexpensive tools for objectively measuring physical activity. Pedometers have been used in schools to estimate and monitor student physical activity levels. A few studies have provided mixed evidence demonstrating the validity of pedometers for measuring physical activity levels, and several have suggested pedometer step cut points corresponding with recommended levels of physical activity overall daily and during school or afterschool programs. However, there is no currently accepted step threshold for achieving the recommended 30 minutes of moderate-to-vigorous physical activity (MVPA) during the school day. Schools and investigators working with schools would benefit from having a pedometer step goal to use as a benchmark for determining whether students are achieving 30 minutes of accumulated MVPA during the school day.

Objectives This study aims to: 1) compare pedometer and accelerometer estimates of school-based physical activity levels among fourth and fifth grade students, and 2) estimate pedometer step cut points that correspond to the recommended 30 minutes of MVPA levels during the school day.

Methods Data presented here were collected as part of a larger quasi-experimental study examining changes in student physical activity levels following implementation of school-based physical activity promotion initiatives. in June 2011, 415 fourth and fifth grade students in 26 classrooms at 6 schools wore Actigraph accelerometers (models 7164, GT1M, and GT3X) for one school week as part of regular data collection. On one day students additionally wore a downloadable Omron pedometer (model HJ-720ITC) placed next to the accelerometer on a nylon belt worn on the hip. Accelerometers captured uniaxial activity levels in 1-minute epochs, and pedometers captured steps per hour. Age-specific accelerometer intensity count cut points calculated by the Freedson equation were used to identify minutes of MVPA. We calculated Pearson‘s correlations comparing total daily pedometer steps and total daily accelerometer minutes of MVPA, and calculated linear regressions to estimate pedometer steps associated with the criterion measure of accelerometer minutes MVPA. We also performed a receiver operating characteristic (ROC) curve analysis to estimate the daily pedometer step cut point corresponding to achieving 30 minutes of accumulated MVPA during the school day.

Results A total of 341 students provided valid concurrent accelerometer and pedometer data for one school day each. Fifty-five percent of these students were female, 47% were in grade 4, and most were of black (59%) or Hispanic/Latino (29%) race/ethnicity. Most students (94%) provided 6-7 hours of monitored time. Students recorded an average of 5,992 (SD 2,892) steps via pedometer during the school day, and 29.0 (SD 23.4) minutes MVPA via accelerometer. of the 341 students, 127 (37%) achieved 30 minutes of accelerometer-determined MVPA during the school day. Daily pedometer steps and daily accelerometer minutes of MVPA were moderately correlated (r=0.62, p<0.0001) overall, with similar results by sex. The correlation was higher (r=0.73) when MVPA was measured by the older model 7164 accelerometer compared with the newer models GT1M and GT3X (r=0.59). We determined daily minutes of MVPA to be estimated from daily pedometer steps according to the following equation: MVPA_school_day = steps_school_day / 76.59 – 49.20. ROC curve analysis indicated adequate sensitivity (0.63) and specificity (0.73) at 6,100 pedometer steps per school day. The total area under the curve was 0.79, demonstrating good ability to correctly classify students as achieving or not achieving 30 minutes of accumulated MVPA during the school day.

Conclusions Among fourth and fifth grade students, daily pedometer steps were moderately correlated with daily accelerometer-determined minutes of MVPA during the school day. Achieving approximately 6,100 pedometer steps during the school day may be used as a benchmark for achieving 30 minutes of MVPA. Existing evidence has suggested that among children daily step cut points between 11,000 and 14,000 correspond with 60 minutes of physical activity during the whole day. Thus, a 30-minute MVPA threshold of approximately half the 60-minute MVPA threshold seems reasonable. As the older and newer model accelerometers may have different relationships with pedometer steps, separate regression models and ROC curve analyses may be necessary. Overall, these results suggest that pedometers can be used as an inexpensive tool for accurately assessing student physical activity levels during the school day. School systems and investigators may use pedometers to monitor student physical activity levels over time.

Support/Funding Source Supported by Active Living Research (grant #68591), a national program of the Robert Wood Johnson Foundation.