Psychometric Properties of Brief-Balance Evaluation Systems Test Among Multiple Populations: A Systematic Review and Meta-analysis

Objectives: To synthesize evidence regarding the psychometric properties of the Brief-Balance Evaluation Systems test (BESTest) in assessing postural controls across various populations.

Data sources: Articles were searched in nine databases from inception to March 2020.

Study selection: Two reviewers independently screened titles, abstracts, and full-text articles to include studies that reported at least one psychometric property of the Brief-BESTest. There were no language restrictions.

Data extraction: The two independent reviewers extracted data (including psychometric properties of Brief-BESTest) from the included studies. The methodological quality of the included studies was appraised by the Consensus-based Standards for the selection of health status Measurement Instruments checklist, while the quality of statistical outcomes was assessed by the Terwee et al method. A best evidence synthesis for each measurement property of the Brief-BESTest in each population was conducted.

Data synthesis: Twenty-four studies encompassing 13 populations were included. There was moderate to strong positive evidence to support the internal consistency (Cronbach's alpha > 0.82), criterion validity (ρ ≥ 0.73, r ≥ 0.71) and construct validity (ρ ≥ 0.66, r ≥ 0.50, area under curve > 0.72) of the Brief-BESTest in different populations. Moderate to strong positive evidence supported the responsiveness of the Brief-BESTest in detecting changes in postural controls of patients 4 weeks after total knee arthroplasty or patients with subacute stroke after 4-week rehabilitation. However, there was strong negative evidence for the structural validity of this scale in various neurologic patients. The evidence for the reliability of individual items and measurement errors remains unknown.

Conclusions: The Brief-BESTest is a valid (criterion- and construct-related) tool to assess postural control in multiple populations. However, further studies on the reliability of individual items and minimal clinically important difference of the Brief-BESTest are warranted before recommending it as an alternative to the BESTest and Mini-BESTest in clinical research/practice.