Validation of an accelerometer system for measuring physical activity and sedentary behavior in healthy children and adolescents

The study aims to assess the concurrent validity of the SENS -motion (R) accelerometer system for device-based measurement of physical activity and sedentary behavior in healthy children and adolescents. Thirty-six healthy children and adolescents (mean +/- standard deviation (SD) age, 10.2 +/- 2.3 years) were fitted with three SENS sensors while performing standardized activities including walking, fast walking, sitting/lying, and arm movements. Data from the sensors were compared with video observations (reference criteria). The agreement between SENS -motion((R)) and observation was analyzed using Student's t-test and illustrated in Bland-Altman plots. The concurrent validity was further evaluated using intraclass correlation coefficient (ICC) and was expressed as standard error of measurement (SEM) and minimal detectable change (MDC). Strong agreement was found between SENS and observation for walking time, sedentary time, and lying time. In contrast, moderate agreement was observed for number of steps, sitting time, and time with and without arm movement. -ICC2.1 values were overall moderate to excellent (0.5-0.94), with correspondingly low SEM% for walking time, sedentary time, lying time, and time with arm movement (2-9%). An acceptable SEM% level was reached for both steps and sitting time (11% and 12%). For fast walking time, the results showed a weak agreement between the measurement methods, and the ICC value was poor. Conclusion: SENS -motion((R)) seems valid for detecting physical activity and sedentary behavior in healthy children and adolescents with strong agreement and moderate to excellent ICC values. Furthermore, the explorative results on arm movements seem promising.

Automated summary

The study aims to assess the concurrent validity of the SENS-motion(R) accelerometer system for device-based measurement of physical activity and sedentary behavior in healthy children and adolescents. The results show strong agreement between SENS and observation for walking time, sedentary time, and lying time, while moderate agreement was observed for number of steps, sitting time, and time with and without arm movement. The ICC values were overall moderate to excellent, indicating good concurrent validity.