Reproducibility in measuring physical activity in children and adolescents with an acquired brain injury

Abstract

Aim: To examine the reproducibility in measurement of physical activity performance using the ActiGraph^® GT3X+ accelerometer in children aged 8–16 years with Acquired Brain Injury (ABI).

Methods: Reproducibility of standardized tasks: Thirty-two children with ABI (12 years 1 month, SD = 2 years 4 months; 20 males; Gross Motor Function Classification System I = 17, II = 15) performed the following activities on 2 consecutive days while wearing an accelerometer and a heart rate monitor: quiet sitting, slow walking (SW), moderate walking (MW), fast walking (FW) and rapid stepping on/off a block (STEP). Intra-class correlation coefficients (ICC) were calculated. Performance variability: Fifty-one participants (12 years 1 month, SD = 2 years 5 months; 27 males; GMFCS I = 26, II = 25) wore an accelerometer for 4 days in the community and reliability coefficients were calculated using standardized 12-hour time spent in moderate-to-vigorous physical activity (MVPA).

Results: Test–re-test reproducibility was excellent for all activities (SW, ICC = 0.90; MW, ICC = 0.83; FW, ICC = 0.91; STEP, ICC = 0.89). Three days of monitoring produced excellent variability estimates of MVPA (R = 0.78).

Conclusion: Therapists can confidently use accelerometry as a reproducible measure of physical activity under standardized walking and stepping conditions, as well as in the community for children with ABI.

Keywords:

• Physical activity
• children
• adolescents
• acquired brain injury
• accelerometer

Acknowledgements

All authors were responsible for the design, ethics applications, reporting and project management. RNB was responsible for funding the research study. EB took on a lead role for data collection. EB and LB were responsible for data analysis. All authors contributed to the writing of the manuscript. The authors would also like to thank Naomi Westwood (Allied Health Assistant) and Melinda Lewis (Study co-ordinator) for their contributions to recruitment and overall management of the study.

Declaration of interest

The project is supported by a Foundation for Children Grant and Smart Futures Co-Investment Program Grant. EB is supported by an Australian Postgraduate Award. LB is supported by an Australian National Health and Medical Research Council (NHMRC) Early Career Fellowship 1070623. LS is supported by a NHMRC Early Career Fellowship 1090828. RB is supported by a NHMRC Fellowship 1105038. The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.