Prospective fall risk screening in people who use a transtibial or transfemoral prosthesis: Cut-off times from an exploratory cohort study.

Abstract

Over 52% of people who use a lower-limb prosthesis fall once a year. Despite this knowledge, few studies have identified parameters that could help screen fall risk in clinical practice. We sought to determine if any clinical outcome measure could significantly distinguish 6-month prospective fallers from non-fallers. In this exploratory cohort study, participants completed seven clinical outcome measures, walked on level-ground at their self-selected pace, completed a 12-month fall history, and reported falls every two weeks over a 6-month period. Participants were grouped by level of prosthesis use (transtibial, transfemoral), and were classified as prospective fallers or non-fallers based on whether they reported a fall over the 6-month follow-up period. Mann-Whitney U tests assessed statistical significance between 6-month prospective fallers and non-fallers. For parameters that were significant, area under the curve (AUC) analyses assessed clinical applicability. Twenty-one individuals who use a unilateral prosthesis participated (age 57.6 ± 14.1 years). For transtibial (n = 14), the Four Square Step Test (FSST) and walking speed were significantly different with large effect sizes and clinical applicability (FSST p = 0.022, g = -1.122, AUC = 0.883, cut-off = 12.83 s; walking speed p = 0.022, g = -0.961, AUC = 0.883, cut-off = 1.28 m/s). For transfemoral (n = 7), the Timed Up and Go (TUG) was statistically significant with large effect size and clinical applicability (p = 0.029, g = -1.642, AUC = 1.000, cut-off = 9.44 s). Cut-off times for the FSST (≥12.83 s) and walking speed (≥1.28 m/s) in transtibial prosthesis users, as well as the TUG (≥9.44 s) for transfemoral prosthesis users, may be useful to distinguish 6-month prospective fallers from non-fallers.