Test-Retest reliability and preliminary reliable change estimates for Sway Balance tests administered remotely in community-dwelling adults

Objective: Impaired balance and postural stability can occur with advanced age, following traumatic brain injury, in association with neurological disorders and diseases, and as the result of acute or chronic orthopedic problems. The remote assessment of balance and postural stability could be of value in clinical practice and research. We examined the test-retest reliability and reliable change estimates for Sway Balance Mobile Application tests (Sway Medical, Tulsa OK, USA) administered remotely from the participant's home.

Method: Primarily young, healthy community-dwelling adults completed Sway Balance Mobile Application tests remotely on their personal mobile devices once per week for three consecutive weeks while being supervised with a video-based virtual connection. Sway Balance tests include five stances (i.e., feet together, tandem right foot forward, tandem left foot forward, single leg right foot, single leg left foot), which are averaged to compute a Sway Balance composite score from 0 to 100, with higher scores indicating better postural stability. We examined test-retest reliability (measured with intraclass correlation coefficients, ICCs) and preliminary reliable change estimates for 70%, 80%, and 90% confidence intervals.

Results: Participants included 55 healthy adults (ages = 26.7 ± 9.9 years, interquartile range = 20–30, range = 18–58; 38 [69%] women). Test-retest reliability for the Sway Balance composite score across three weeks was.88. Test-retest reliability for individual stances ranged from 62 to 83 (all ps < 0.001). At the 80% confidence interval, preliminary reliable changes estimates were 9 points for the Sway Balance composite score.

Conclusions: For a remote administration, test–retest reliability was moderate-to-good for all Sway Balance stances, as well as for the Sway Balance composite score. Reliable change estimates may allow clinicians to determine whether an improvement or decline in performance is greater than the expected improvement or decline due to measurement error in young adults.