| Sumario: | INTRODUCTION Cadence variability, or the distribution of steps taken per minute throughout the day, can be determined by analyzing data obtained from accelerometers (Arch, 2018). Based on Medicare Functional Classification Level (K-Level) guidelines, which dictate the type of prosthetic technology that can be prescribed for individuals with lower-limb loss (LLL), the ability to vary cadence is a required skill to justify the prescription of higher-level (i.e., K3- and K4-level) prosthetic componentry (HCFA, 2001). While it is crucial that K-Level assignment reflect actual mobility level, clinical judgment of cadence variability can be subjective (Borrenpohl, 2016). Use of performance-based outcome measures may improve the objectivity of K-Level classification; however, associations between outcome measures conducted in a clinical setting and real-world cadence variability are unknown. The purpose of this study was to determine if better performance on clinical measures of functional mobility (i.e., gait speed, the L-Test, Figure-of-8 Walk Test) is predictive of increased cadence variability during community ambulation, assessed via FitBit® OneTM monitors among individuals with LLL. METHODS Participants: Fifty prosthetic users, aged 18–85 years, with a unilateral transtibial amputation were included in this study funded by the Orthotics and Prosthetics Education and Research Foundation and approved by the University of Delaware Institutional Review Board for Human Subjects. Individuals with bilateral amputations or weight-bearing restrictions of the residual limb were excluded. Procedures: Participants provided demographic and amputationrelated information and completed physical performance measures. Self-selected and fast gait speeds were determined using the 10-Meter Walk Test (10MWT), a test used to calculate walking speed among individuals with LLL (Roffman, 2016). Participants completed the L-Test and the Figure-of-8 Walk Test (F8WT), two valid assessments of walking ability among individuals with LLL (Deathe, 2005; Hess, 2010). Participants then wore a FitBit® monitor around the prosthetic ankle for seven days following the on-site examination. Cadence variability was determined as the scale parameter of a Weibull probability density function fit to cadence data calculated from the FitBit®; a lower scale parameter indicates a narrower distribution of cadence variability (Arch, 2017). Individuals with ≥ 5 days of FitBit® data were included in analyses (n = 42). Data Analysis: After checking assumptions, four linear regression models were used to explore relationships between each outcome measure and cadence variability, while controlling for gender, age, and time elapsed since the initial amputation, (p < .0125). RESULTS Sixty-four percent of the sample was male (n = 27). Mean age was 59 ± 12 years, and median time elapsed since the initial amputation was eight (2, 18) years. Mean cadence variability was 30.4 ± 8.1. Covariates explained a significant amount of the variance (p < .001) for cadence variability in all models (self-selected and fast gait speed: 46.8% and 42.1%, respectively; L-Test: 42.1%; F8WT: 42.1%). After controlling for covariates, self-selected gait speed explained the greatest amount of additional variance in cadence variability (18.4%, p < .001), while all other tests explained a smaller but significant amount of the variance (fast gait speed: 15.6%, p = .001; L-Test: 10.7%, p = .008; F8WT: 10.6%; p = .008). For every 1 m/s increase in self-selected and fast gait speeds, cadence variability increased by 17.1 and 9.8 respectively. For every 1-second increase in time to complete the L-Test and F8WT, cadence variability decreased by .4 and .8, respectively. DISCUSSION Objective determination of K-Level is crucial for appropriate prosthetic prescription; however, the routine use of accelerometers to quantify cadence variability is not feasible. Thus, it is important to identify clinical measures that provide insight into real-world cadence variability among adults with LLL. Our findings suggest that faster gait speeds and better walking performance predict increased cadence variability during community ambulation above and beyond age, gender, and time since amputation. CONCLUSION Gait speed, the L-Test, and the F8WT are predictive of real-world cadence variability among adults with unilateral transtibial LLL.
|
|---|