Advancing balance assessment in stroke rehabilitation: a comparative exploration of sensor-based and vonventional balance tests

Balance impairments in stroke rehabilitation are commonly assessed using the Trunk Control Test (TCT), Berg Balance Scale (BBS), and Mini Balance Evaluation System Test (Mini-BESTest). However, these conventional tests are subjective, susceptible to floor and ceiling effects, and time-intensive. Inertial measurement units (IMUs) may address these limitations by providing objective, impairment-level metrics, not captured by conventional tests. This observational study explored the measurement properties of an IMU-based balance assessment of postural sway, and compared them with conventional tests in routine stroke rehabilitation. Stroke survivors from five Dutch rehabilitation centers were assessed at admission and discharge using conventional and IMU-based balance tests during sitting and standing tasks. Floor and ceiling effects were evaluated, and relationships between measures were examined using correlation analysis. At admission, 105 participants were measured, and 90 at discharge. IMU measures showed no floor or ceiling effects despite skewed distributions. IMU stance-tasks correlated moderately with the BBS and Mini-BESTest (18–29% variance explained), whereas IMU sitting-tasks showed weak to no relationship with the TCT. IMU-based balance assessment of postural sway captures balance-related information that is partially different from conventional tests. Although IMUs offer practical advantages, further research is needed to establish the clinical relevance of postural sway measurements alongside conventional tests.