<!-- FEATURE 4 --> Apparently unaffected by the 2,300 miles that separate their physical locations, physiatrist A. Barry Deathe, MD, FRCP(C), of London, Ontario, Canada, and researcher William C. Miller, MSc, PhD, OT, of Vancouver, British Columbia, Canada, have collaborated for more than a decade to contribute to the development and validation of clinically relevant outcome measures in rehabilitation for people with amputations. While an exhaustive review of their academic contributions is beyond the scope of this article, their work with the L Test of Functional Mobility (L Test) and the Activities-specific Balance Confidence (ABC) Scale warrant detailed review. <h3>L Test: Initial Development</h3> One of the goals of rehabilitation following amputation is to enable patients to return to mobility, or regain the ability to walk and change body positions when needed. When mobility is restored, activities such as transfers, level-ground walking, turns, and household ambulation can be routinely performed. In the modern climate of healthcare accountability, a mechanism of quantifying prosthetic mobility is increasingly in demand. In a 2005 publication, Deathe and Miller describe the evolution of just such a measure within a clinical setting.<sup>1</sup> In their Regional Amputee Program, they were using the Timed Up and Go (TUG) test and two-minute walk test (2MWT) to assess prosthetic mobility. However, they were finding neither test to be fully conducive to their needs. The TUG test was simply too easy for younger individuals and many of the fit, elderly patients, leading to an observed ceiling effect. The longer 2MWT was difficult to administer due to the requirement of a 20 meter long hallway free of distractions. In the authors' words, "[O]ur need was to find or develop a test that could be easily and quickly administered concurrent with each patient visit that assisted with determining ability to walk with prosthetic devices."<sup>1</sup> Recognizing the value of including transfer skills in addition to ambulation, the authors wanted to preserve that element of the TUG test in a more demanding test. The authors describe the evolution of their solution as follows: "Observations of a patient's gait during clinics showed that we usually asked the patient to get up and walk out of the room, turn and go down the hall, then return to the room and sit down. This walking path, representing an "L" configuration, required turns to both the right and left. Standardizing the distance (3m x 7m) led to the development of a potentially more demanding, yet practical, modification of the TUG that we have titled the L Test of Functional Mobility (L Test)."<sup>1</sup> To validate the new instrument, a convenience sample of 102 subjects who were attending regular clinic appointments during a six-month treatment window was studied. This included the performance of a number of walking tests-the L Test, 2MWT, TUG test, and ten-meter walk test (10MWT)-during two testing sessions, conducted by two separate raters and broken up by the completion of several self-report questionnaires. The L Test was performed three times during each session to assess any potential learning affect, and willing subjects returned two weeks after that initial study protocol to perform the L Test series a second time. [caption id="" align="alignnone" width="398"]<img class="" style="margin: 4px; border: 1px solid #000000;" src="https://opedge.com/Content/OldArticles/images/2016-05/2016-05_04-01.jpg" alt="table 1" width="398" height="471" /> Table 1. Mean performance values for the L Test according to amputation levels and etiologies, the need for a walking aid, and age. Adapted from Deathe and Miller.[/caption] Ninety-three of the subjects who began the study protocols completed all of the tests, defining the original study cohort. Of these, 27 returned for the retest protocols, comprising a secondary cohort. The primary cohort was predominantly male (78 percent) with transtibial amputations (74 percent) and traumatic amputation etiologies (60 percent). However, transfemoral amputations (26 percent) and vascular amputation etiologies (40 percent) were also well represented.<sup>2</sup> Within the primary cohort, the average completion time for the initial L Test was 32.6 seconds, a time that did not change substantially with the second performance (32.9 seconds). The normative data according to amputation etiologies and levels, along with age and walking aid considerations, are shown in Table 1. The most intuitive clinometric consideration to most readers is reliability. Performance on this test appears to be fairly consistent across multiple iterations and different test administrators. However, other validation values were also observed. For example, as expected, performance on the L Test correlated very strongly with performance on the TUG test and during the 10MWT. Similarly, a strong negative correlation was observed with the 2MWT, meaning patients with lower L Test performance values covered greater distances during the 2MWT.<sup>1</sup> The end result of this initial effort was the introduction of a more demanding version of the TUG test with high clinical utility (that is, easily performed in most clinical environments). This utility is also consistent with the concept of face validity, meaning that the instrument measures prosthetic mobility in a fashion that captures the real needs of most patients, such as getting up from a chair in the living room, exiting the room into the hallway to retrieve something from the kitchen, and returning to a sitting position in the living room. Compared to the TUG test, not only was the distance traveled more than tripled, from 6 meters to 20 meters, but the test added 90 degree turns to both the left and right-an activity that can be challenging to some patients with transfemoral amputations. The net result was a measure with a reduced ceiling effect relative to its predecessor, the TUG test.<sup>1</sup> <h3>L Test: Sensitivity to Change</h3> The utility, reliability, and validity of the L Test are evidenced in its initial publication. The answer to the more nuanced question, how sensitive the L Test is to change, would take another decade to reach publication.<sup>2</sup> This trial began with a carefully selected cohort of patients who were attending the South Western Ontario Amputee Program, all of whom were identified by their physiatrists as "requiring a major intervention" such as a new socket or referral for additional therapy. These patients performed the L Test at the time of those determinations, received the indicated interventions, and returned to the clinic where a second L Test was performed. Of the 33 patients included in the study, the most commonly identified problems were a poor prosthetic fit (33 percent), residual limb problems (30 percent), and residual limb shrinkage (15 percent). Of these subjects, 64 percent required a new socket. Other interventions included modification to an existing prosthesis (9 percent) and the prescription of medication (9 percent). Upon their return to the clinic, in addition to retaking the L Test, subjects were asked, "Since we last saw you, have you experienced any change in your ability to get up and walk with your prosthesis?" Not surprisingly, the vast majority (82 percent) said they had. They were then asked if this ability had improved or gotten worse, and rated the amount of change on a seven-point Likert scale (0 = almost the same and 6 = a very great deal better/worse). The authors then set out to determine the relationship between changes in reported abilities and L Test performance. Encouragingly, the L Test appeared sensitive enough to capture the changes experienced by most of the patients who had required and received major prosthetic interventions. The mean performance score decreased by an average of 6 seconds following the intervention (46 to 40 seconds). Even among the four patients who reported a decrease in prosthetic ability, an average improvement of 4.5 seconds was observed. As hypothesized, those who reported greater amounts of improvement demonstrated greater average reductions on the L Test of 13.5 seconds than those reporting more measured improvements, 3.5 seconds.<sup>2</sup> In addition, the authors ultimately assert that the minimal clinically important difference for the L Test was 4.5 seconds, meaning that an improvement of 4.5 seconds best distinguishes those patients who experience greater or more measured levels of improvements through their prosthetic interventions.<sup>2</sup> <h3>Balance Confidence</h3> Just as Deathe and Miller were concerned about measuring limitations in prosthetic mobility, they were interested in better understanding potential barriers to prosthetic mobility. To do so, they examined the concept of fear of falling in terms of balance confidence. Based on Albert Bandura's theory of self-efficacy, that confidence is at least as important to predicting and changing behavior as the presence of the appropriate skill level, balance confidence can be summarized as an individual's belief that he or she can perform a given activity or action without losing balance or sustaining a fall. [caption id="" align="alignnone" width="515"]<img style="margin: 4px; border: 1px solid #000000;" src="https://opedge.com/Content/OldArticles/images/2016-05/2016-05_04-02.jpg" alt="table 2" width="515" height="1309" /> Table 2. Average ABC scores according to a number of sociodemographic, amputation-related, health-related, and psychologic factors. Extracted from Miller et al.[/caption] In contrast to attempts to quantify fear of falling, assessing balance confidence is measured over a continuum of activities from easy to more difficult, allowing a more refined assessment of where limitations might be present. Additionally, inquiries regarding a person's degree of confidence in performing a given activity are less threatening than asking about his or her fears.<sup>3</sup> In one of their early studies on balance confidence, Miller and Deathe administered the ABC Scale to a convenience sample of 435 individuals with lower-limb amputations who were attending one of two outpatient limb-loss clinics in Ontario. While Miller and Deathe did not develop the ABC Scale, which was originally created for use in the able-bodied geriatric community, they were the first to publish research about using the self-report instrument for those with limb loss. The ABC Scale is a 16-item survey that asks participants to rate their confidence in their ability to perform tasks in a number of different commonly encountered situations and environments without losing their balance or becoming unsteady. A score of 100 represents full balance confidence. Analysis of their observations confirms several expected relationships along with some surprises (Table 2). Older patients tend to report lower balance confidence, as do women. Perhaps less intuitive, advanced education and increased income were associated with higher balance confidence scores. Higher scores, predictably, were associated with increased prosthetic experience, nonvascular amputation etiologies, and those who did not use assistive devices during indoor ambulation. The presence of a fall in the past 12 months did not appear to affect ABC scores and those who sustained injuries in a fall reported only modestly reduced ABC values compared to those who had not. By contrast, those who admitted a fear of falling reported substantially lower ABC scores. Perceived general health values were also closely related with reported ABC scores.<sup>3</sup> While such values are helpful in describing the relationships between a number of presentation variables and balance confidence, additional work was required to affirm the reliability and validity of the measure among people with amputations and was subsequently published.<sup>4</sup> Reliability was verified by asking a sample of patients to complete the ABC Scale on two separate occasions, four weeks apart. Convergent validity was verified by comparing ABC scores to physical performance measures mentioned earlier. A positive correlation was found, as expected, between higher ABC scores and 2MWT distances. Similarly, a negative correlation was found between higher ABC scores and reduced times to complete the TUG test.<sup>4</sup> Discriminative validity was confirmed by the ability of the ABC scores to discriminate with regard to average prosthetic wearing times, stair climbing ability, and reported walking distances (Table 3). [caption id="" align="alignnone" width="538"]<img style="margin: 4px; border: 1px solid #000000;" src="https://opedge.com/Content/OldArticles/images/2016-05/2016-05_04-03.jpg" alt="table 3" width="538" height="420" /> Table 3. ABC scores were able to discriminate with regard to average prosthetic wearing times, stair climbing ability, and reported walking distances. Extracted from Miller et al.[/caption] <h3>Combination</h3> With prosthetic mobility measured and characterized through the L Test and balance confidence recorded through the ABC Scale, the last Deathe and Miller contribution this article addresses is an effort to determine to what extent these two variables might predict the actual utilization of a lower-limb prosthesis. As the authors state, "Acknowledging that learning a new skill, such as prosthetic ambulation, largely relies on having the appropriate physical attributes, whether the individual uses the new skill or not also involves on having the self-efficacy or confidence to perform targeted activities."<sup>5</sup> To test this relationship, one more convenience sample from Ontario was assembled. The researchers recorded the L Test and ABC scores of those individuals at the time of their discharge from the inpatient rehabilitation that followed their amputations, and then again at one and three months post-discharge. At the three-month follow-up appointment, prosthetic utilization was measured.<sup>5</sup> Using regression analyses, they determined that patient performance values on those two outcome measures at the time of discharge predicted 64 percent of the observed variance in prosthetic utilization during social activity. These two variables were better predictors of such activity than were age, gender, amputation level, or reliance on walking aids.<sup>5</sup> <h3>Conclusion</h3> As the requirements for the objective assessment of patient outcomes continue to gain traction in the healthcare community, the O&P profession is indebted to the efforts of this Canadian duo. Their efforts to produce an accurate assessment of prosthetic mobility and validate an existing measure of balance confidence among patients with lower-limb amputations have yielded two straightforward outcome measures-the L Test and the ABC Scale-that can be easily included in prosthetic rehabilitation to document both mobility and self-efficacy while also predicting prosthetic utilization. <em>Phil Stevens, MEd, CPO, FAAOP, is in clinical practice with Hanger Clinic, Salt Lake City.</em> <h4 style="color: #000000;">References</h4> <ol style="font-size: 85%;"> <li>Deathe, A. B., and W. C. Miller. 2005. The L Test of functional mobility: Measurement properties of a modified version of the timed "up & go" test designed for people with lower-limb amputations. <em>Physical Therapy</em> 85 (7):626-35.</li> <li>Rushton, P. W., W. C. Miller, and A. B. Deathe. 2015. Minimal clinically important difference of the L Test for individuals with lower limb amputations: A pilot study. <em>Prosthetics and Orthotics International</em> 39 (6):470-6.</li> <li>Miller, W. C., M. Speechley, and A. B. Deathe. 2002. Balance confidence among people with lower-limb amputations. <em>Physical Therapy</em> 82 (9):856-65.</li> <li>Miller, W. C., A. B. Deathe, and M. Speechley. 2003. Psychometric properties of the Activities-specific Balance Confidence Scale among individuals with a lower-limb amputation. <em>Archives of Physical Medicine and Rehabilitation</em> 84:656-61.</li> <li>Miller, W. C., and A. B. Deathe. 2011. The influence of balance confidence on social activity after discharge from prosthetic rehabilitation for first lower limb amputation. <em>Prosthetics and Orthotics International</em> 35 (4):379-85.</li> </ol>
