JMIR Publications

ABSTRACT

Background:

Existing modes of collecting self-reported 24-hour movement information from children, including digital assessments, have not been demonstrated to be of acceptable validity when compared to objective measurements. My E-Diary for Activities and Lifestyle (MEDAL) is an interactive web-based diary developed to collect time-use information from children aged 10 years and older.

Objective:

This study compared self-reported and accelerometer-measured time spent in movement behaviour among children in Singapore aged 10–11 years.

Methods:

Participants recorded their daily activities using MEDAL over two specified weekdays and two weekend days, and wore an Actigraph accelerometer on their non-dominant wrist throughout the study to objectively assess movement behaviours. Spearman correlation coefficient and intraclass correlation coefficient were used to compare the accelerometer measurements and self-reports for each movement behaviour. Bland-Altman plots were generated to investigate trends of bias in the self-reports.

Results:

Among the participants aged 10-11 years (n=49, 59% boys), we observed that children reported lower light physical activity (LPA), and higher moderate-to-vigorous physical activity (MVPA), inactivity and night sleep than accelerometer-measured. There was moderate to strong correlation between self-reported and accelerometer-measured MVPA (r=0.37, 95% CI 0.20–0.54), inactivity (r=0.36, 95% CI 0.18–0.54) and night sleep (r=0.58, 95% CI 0.43–0.74); the correlation for LPA was poor (r=0.19, 95% CI 0.02–0.36). Agreement was poor for all behaviours (MVPA ICC 0.24, 95% CI 0.07–0.40; LPA ICC=0.19, 95% CI 0.01–0.36; inactivity ICC=0.29, 95% CI 0.11–0.44; night sleep ICC=0.45, 95% CI 0.29–0.58). There was stronger correlation and agreement on weekdays for inactivity and night sleep; conversely, there was stronger correlation and agreement for MVPA and LPA on weekend days. Finally, we observed that with increasing MVPA, children tended to report higher MVPA than accelerometer measurements. There were no clear trends for the other behaviours.

Conclusions:

MEDAL may be used to assess movement behaviours of children. Based on self-reports, the children are able to estimate their time spent in MVPA, inactivity and night sleep, although actual time spent in these behaviours may differ from accelerometer-derived estimates; self-reported LPA warrant cautious interpretation. Observable differences in reporting accuracy exist between weekdays and weekend days.