Measurement of mobility and physical function in hospitalised trauma patients: A systematic review of instruments and their measurement properties

Introduction: It is well recognised that organised trauma systems reduce trauma patient mortality. As established systems mature, there is an increasing need to better understand the patient recovery trajectory. Mobility and physical function are key aspects of recovery, but the optimal instruments for measurement in the acute hospital setting remain unclear. Methods: A systematic review was undertaken to identify and describe mobility and physical function instruments scored by direct patient assessment, in adult trauma patients in an acute hospital setting. Databases were searched with no date restrictions. Instruments that were specific to subgroups or related to individual conditions, diseases or joints were excluded. The consensus-based standards for the selection of health measurement instruments checklist was used to assess risk of bias where relevant. Clinimetric properties were reported where possible, including reliability, validity and responsiveness. Results: Fourteen thousand one hundred and fourteen articles were identified with 37 eligible for final review, including six instruments. None had been specifically designed for use in a heterogeneous range of trauma patients. The Functional Independence Measure was the most commonly cited (n = 10 studies), with evidence of construct validity, responsiveness and minimal floor/ceiling effects (<3%). The Acute Care Index of Function (n = 1 study) was found to be valid and responsive whilst the modified Iowa Level of Assistance (n = 2 studies) was reliable and responsive, but ceiling effects ranged from 26% to 37%. Little clinimetric data were available for other measures. Conclusion: Evidence from a few studies show promise for the use of the Functional Independence Measure, Acute Care Index of Function and modified Iowa Level of Assistance to measure mobility and physical function in trauma patients, however comprehensive clinimetric data are lacking. Future research should test these scores in specifically designed clinimetric property studies in defined trauma patient populations. This would enable the identification of a gold standard measure for evaluating treatment effectiveness, enabling benchmarking between centres, allow prediction of recovery pathways and optimise trauma patient outcomes.