We lack the data necessary to identify and characterize consequential falls experienced by people with LLA. In the absence of such data manufacturers cannot design and clinicians cannot prescribe prostheses capable of mitigating consequential falls. Research priorities concerning assessment and prevention of consequential falls also cannot be established. Current falls data, collected from small samples using ad hoc questions, does not meet the need. Funded by two prosthetic manufacturers, I lead a team of qualitative research experts working to develop, validate, and administer the first standardized amputee-specific fall circumstance and consequence survey. Our fall survey, and the data it yields, will address these gaps and impact the field by ensuring simple, accurate, meaningful, and repeatable measurement of falls by people with LLA. Guided our fall-type classification framework (Kim, 2019), we used best practices in survey development (i.e., focus groups and individual semi-structured interviews) to develop a conceptual model of falls in people with LLA, identify suitable fall-related terminology, assess fall event definitions, create candidate survey questions, and evaluate those questions for clarity, comprehension, and applicability to people with LLA. An initial validation of the survey in 210 people with LLA is ongoing. Results of this research, which have been disseminated broadly to prosthetic manufacturers and clinicans, have been submitted for peer review. Planned research includes a characterization of consequential falls in a national sample of people with LLA (n=2000) (R01, aim 1). The most consequential type(s) of falls will undergo biomechanical analyses (aim 2), which will guide the selection of motor tasks to be used in the clinical prediction of consequential falls (aim 3). Two recently funded DoD projects, the largest clinical trial of computer-controlled prosthetic knees to date, and an assessment of prosthetic foot design on balance and falls will use our survey.

There is a scarcity of valid, reliable, and practical tests to accurately predict fall risk in people with LLA. The inability to identify those at risk for falling, and thereby intervene prior to a fall, drives the persistent and elevated incidence of falls in people with LLA. To address this gap we designed, developed, and validated the Narrowing Beam Walking Test (NBWT), a novel test that progressively increases the challenge to balance control (i.e. individuals walk along a series of low beams, each narrower than the last). Funded through an NIH K-award and the DoD, our work on the NBWT has produced a testing apparatus and manual (Sawers, 2018a), as well as administration and scoring procedures that limit practice effects (Sawers, 2018a) and improve test accuracy (Sawers, 2019). The NBWT discriminates between people with LLA who do and do not have a history of falls with greater accuracy than existing tests (Sawers, 2018b), and is accompanied by key validity (Sawers, 2020a) and reliability (Sawers, 2020b) indices clinicians need for test selection, administration, and interpretation. The NBWT can therefore be used to identify people with LLA in need of intervention to offset fall risk, and help decrease the incidence of falls. Data from a multi-site prospective study is being analyzed to verify this claim. This research has begun to impact the prosthetic community. The NBWT is used by prosthetic manufacturers to evaluate foot design, and by VA centers (Seattle, Minneapolis) to assess patient outcomes. We also won a national award, and were profiled in the DoD Annual Report to Congress for this research. Guided by our research on the epidemiology of falls, we will expand this line of inquiry to develop tests that predict how and where people with LLA are most likely to experience a consequential fall, increasing the specificity with which interventions to reduce falls can be prescribed.

We have identified and begun to address major gaps in knowledge concerning muscle structure and function in people with LLA, and their implications for amputation rehabilitation (Hewson, 2020). Funded by the DoD, my inter-disciplinary team uses a multimodal approach (i.e., dynamometry, electromyography, and ultrasound imaging) to address gaps that include overlooked normalization procedures, demographic, health, and amputation contributions to muscle weakness, the identification of novel and functionally meaningful measures of strength in people with LLA, and the characterization of neural versus structural determinants of weakness in people with LLA. These data will impact the field by providing engineers with biological deficits their devices must replace, inform surgeons how current procedures affect muscle structure and function, and identify limitations therapists can assess and treat. Promising initial data (n=12) suggests that failure to scale hip extension strength to body size masks weakness in the residual and intact legs compared to controls. Evidence of residual and intact leg weakness compared to controls, but not between residual and intact legs, suggests general deconditioning rather than amputation-specific reasons for the observed strength deficits. Finally, residual limb muscle composition (i.e., % contractile vs. non-contractile tissue) is correlated (r=.72) to the rate of force development. In an ongoing clinical trial we are also testing if socket design has a modifying effect on hip muscle function in people with above-knee amputation. The prospect of improving muscle function by walking in a socket that promotes muscle activity is appealing since it could be done at home and in the community, without strength training equipment. Planned NIH and DoD proposals will challenge the device-driven paradigm that governs the field of prosthetics (Sawers, 2012) by testing the central hypothesis that deficits in muscle structure and function are modifiable, and supersede prosthetic design as determinants of balance, mobility, and endurance in people with LLA.