Psychometrics of Wearable Devices Measuring Physical Activity in Ambulant Children With Gait Abnormalities: A Systematic Review and Meta-analysis.

OBJECTIVE

To evaluate psychometrics of wearable devices measuring physical activity (PA) in ambulant children with gait abnormalities due to neuromuscular conditions.

DATA SOURCES

We searched PubMed, Embase, PsycINFO, CINAHL, and SPORTDiscus in March 2023.

STUDY SELECTION

We included studies if (1) participants were ambulatory children (2-19y) with gait abnormalities, (2) reliability and validity were analyzed, and (3) peer-reviewed studies in the English language and full-text were available. We excluded studies of children with primarily visual conditions, behavioral diagnoses, or primarily cognitive disability. We performed independent screening and inclusion, data extraction, assessment of the data, and grading of results with 2 researchers.

DATA EXTRACTION

Our report follows Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. We assessed methodological quality with Consensus-based Standards for the selection of health measurement instruments. We extracted data on reported reliability, measurement error, and validity. We performed meta-analyses for reliability and validity coefficient values.

DATA SYNTHESIS

Of 6911 studies, we included 26 with 1064 participants for meta-analysis. Results showed that wearables measuring PA in children with abnormal gait have high to very high reliability (intraclass correlation coefficient [ICC]=0.81; 95% confidence interval [CI], 0.74-0.89; =88.57%; ICC=0.99; 95% CI, 0.98-0.99; =71.01%) and moderate to high validity in a standardized setting ( =0.63; 95% CI, 0.36-0.89; =99.97%; =0.68; 95% CI, 0.57-0.79; =98.70%; =0.69; 95% CI, 0.58-0.80; =87.02%). The methodological quality of all studies included in the meta-analysis was moderate.

CONCLUSIONS

There was high to very high reliability and moderate to high validity for wearables measuring PA in children with abnormal gait, primarily due to neurological conditions. Clinicians should be aware that several moderating factors can influence an assessment.