Instrumented Timed Up and Go test: A reliable and valid tool for elderly with femur fracture.

Background: The Timed Up and Go (TUG) test, commonly used to predict falls, has been enhanced with an inertial sensor, creating the Instrumented TUG (iTUG). Its reliability has been evaluated in a few studies involving neurological and heterogeneous orthopaedic populations.

Aim: The aim was to assess the intra-rater, inter-rater, and test-retest reliability of the iTUG in patients with femur fractures and define an interpretative model of the iTUG and a discriminative one to detect patients at risk of fall.

Methods: 201 inpatients with femur fractures underwent the iTUG test before discharge and a telephone follow-up at 1-month post-discharge to record falls. In a subgroup of 48 patients, iTUG data were used for assessing test-retest, intra-rater, and inter-rater reliability through Intra-class Correlation Coefficients. In the total sample, iTUG reliable variables were used for exploratory factor analysis and creating logistic regression models to identify the best predictive model for falls.

Results: 27/100 iTUG variables demonstrated "excellent" reliability; of those, 15 not-redundant variables were collected in a 3-factor model explaining 74 % of total variance. 92 % of patients completed the follow-up, and 5.5 % were classified as fallers. The fall risk model created with three iTUG variables showed an R= 0.31, with an AUC= 0.91.

Conclusion: The iTUG test is a reliable tool for assessing individuals with femur fractures, with 27 highly reliable variables. The 3-factor model offers a focused framework for gait evaluation, while the fall-risk model highlights the predictive value of iTUG variables, indicating that fall risk in the elderly depends on multiple gait characteristics. This article was awarded the SIAMOC Best Paper 2024.