Take Two: The Impact of Additional Training and Resources on Functional Mobility and Balance

Home > Articles > Take Two: The Impact of Additional Training and Resources on Functional Mobility and Balance
By Phil Stevens, MEd, CPO, FAAOP

The impact of lower-limb amputation on physical activity and participation cannot be overstated. While some people who undergo lower-limb amputation have access to an initial rehabilitation program in the first few months after their amputations, for many, considerable apprehension persists, along with decreased strength, balance, endurance, and coordination. There have been a few publications in recent years about therapy programs that have been made available to individuals several years after their amputations, which provide guidance and structure as clients attempt to increase their physical abilities and activity levels. This article reviews those publications and provides insight on the potential impacts of subsequent therapy resources after initial post-amputation rehabilitation.


The Wellness Clinic Model

I first learned of the wellness clinic model several years ago when I relocated to Salt Lake City. The local physical therapy school hosted a biweekly wellness clinic for chronic stroke survivors. The rationale was no different than the one outlined above: that declines in the skills necessary for mobility can be avoided with additional physical therapy. Most of the patients that attended had been through inpatient and outpatient therapy services after their initial injuries. However, having navigated through the subacute phase of their rehabilitation, most of them had drifted away from the physical and psychological discipline required for continued gains in recovery. The University of Utah Wellness Clinic, primarily staffed by volunteer physical therapy students, provided participants a regular time and location to get out of their homes and engage in physical activity with light supervision and instruction. The model seemed to work, giving the attendees access to some unique resources during an otherwise vulnerable period.

A similar model was recently described by a team of researchers in Georgia.1 Citing the aforementioned concerns, they set out to determine if a six-week, supervised, community-based exercise program designed for individuals with lower-limb amputations would have an impact on balance, balance confidence, or gait parameters.

A convenience sample of 16 individuals with legacy lower-limb amputations was assembled. All were community-dwelling, established prosthetic users with well-fitting p rostheses who ambulated with or without assistive devices. It was a heterogeneous group ranging in age from 22 to 87 years old who had six months to 52 years of prosthetic experience.


FIGURE-OF-8 WALK TEST: As with many of the functional outcome measures that have crossed over into application with the population of individuals with amputations, the F8WT was initially described within the geriatric community.1 The test requires the participant to walk in a figure eight pattern around two cones set five feet apart, allowing for an assessment of his or her ability to walk in a straight line and negotiate a turning radius in both a clockwise and counterclockwise fashion. In its original description, the F8WT was objectively scored against three considerations: speed (i.e., the time required for completion), amplitude (i.e., the number of steps required for completion), and accuracy (i.e., the ability to navigate reasonably tight curves, defined as staying within two feet of the cones when navigating the required turns). In addition, the developers defined a subjectively rated construct of smoothness. For this consideration, subjects are assessed on whether they stop, hesitate, or change pace while negotiating the test. Each consideration is scored as 0 (any difficulty) or 1 (no difficulty) with a summed score ranging from 0 (not smooth) to 3 (smooth).

Three measures were used to track the impact of the exercise program, the Figure-of-8 Walk Test (F8WT), the Activities-specific Balance Confidence (ABC) Scale, and a set of gait parameters measured using a GAIT-Rite walkway.

The participants were placed into one of two exercise groups according to their amputation levels, transtibial or transfemoral, and met for one hour, twice per week for six weeks. The program began with a brief stretching warm-up followed by a range of strength and balance training elements. For example, stability training involved ball rolling, in which the subject stands on his or her sound limb in single-leg support while rolling a ball under the prosthesis in prescribed motions (forward/backward, side to side, clockwise/counterclockwise). Strengthening elements included using resistance bands during hip extension, abduction, and adduction. Balance elements included weight shifting in all directions in eyes-open and eyes-closed conditions, with the use of a foam standing surface to increase difficulty for some subjects, and tossing a ball with a partner. Dynamic strength and balance exercises included walking exercises where subjects stepped over or around cups set in prescribed patterns. Agility training elements included side stepping, side stepping while crossing the trailing leg in front of or behind the lead leg, and braiding. A brief cool-down period at the end of the program included stretching and yoga-based elements.

Upon completion of the six-week trial, improvements were noticed across all the performance measures. The average time to complete the F8WT decreased by 17 percent, with a decrease of two steps in the average amplitude. In addition, the average smoothness scores increased.2

Improvements in balance confidenceas measured by the ABC Scale were also noted, with the mean scores increasing from 63 to 74 on the 100-point scale. The improvements reported by those with transfemoral amputations were more pronounced than among those with transtibial amputations, with improvements on the self-report instrument of 19 percent and 14 percent respectively.1

The average self-selected walking velocity as measured with the GAITRite walkway increased by 19 percent from 75 cm/s to 89 cm/s. A more moderate increase was observed with respect to fast-as-possible gait with averages increasing 15 percent from 109 cm/s to 125 cm/s.2

Additional benefits were noted beyond the objective observations from defined performance measures. For example, 43 percent of the participants were able to reduce their reliance on upper-limb assistive devices. Further, a number of participants lost weight during the six-week exercise program, so much so that it compromised the fit of their prostheses, with 63 percent of the transtibial group and 50 percent of the transfemoral group requiring socket replacement at the conclusion of the trial.2 In short, the wellness clinic model helped these individuals realize improvements in their balance and mobility.


The Home-based Model

The second clinical trial under scrutiny was motivated by similar concerns but used a different solution. In introducing their research, the study organizers cited the inability of the majority of individuals who sustain a transfemoral amputation to meet or exceed their pre-amputation mobility levels within the first year after the amputation.3 Of greater concern, the number of individuals who are dependent on a wheelchair for mobility increases throughout the five years after amputation, from 13 percent to 39 percent.3 As with the wellness clinic model, researchers sought to determine if these trends could be reversed, facilitating improvements rather than declines in the prosthetic mobility of individuals with legacy transfemoral amputations.

A cohort of eight individuals who had undergone transfemoral amputations between three and 31 years ago, with an average period post-amputation of just over 15 years, volunteered to participate in the trial.3 Additionally, study participants were required to have a self-selected walking speed between 0.67 to 1.12 m/s. The lower threshold ensured that the individuals could physically tolerate the exercise program that would follow, while the upper threshold ensured adequate room for improvement throughout the trial.

The training protocol was home-based participation in a structured treadmill usage program. Participants were instructed to engage in a training session that entailed a series of two-minute intervals at progressively increased speeds three times per week for eight weeks. For most subjects, this consisted of bouts at .89, 1.12, and 1.34 m/s. However, this varied slightly with the capacity of the individual. This series of progressively increased walking speeds was repeated five times within each session. Participation in the exercise program was tracked with externally mounted activity monitors that confirmed a 90 percent compliance rate among the volunteers.3

Metabolic energy consumption variables and gait speed were among the variables tracked by the researchers at baseline, four weeks, and eight weeks. Within the former, both metabolic efficiency and metabolic energy costs were monitored. With respect to metabolic efficiency, an initial mean improvement of 5 percent was observed at the four-week mark, with an additional 4.5 percent improvement recorded over the next four weeks. Thus, the net benefit was a decreased energy expenditure of almost 10 percent. Similarly, average energy cost, which takes into account distance traveled, decreased by 6 percent through the first four weeks, and another 3 percent over the next four weeks for a total decrease of just over 9 percent.

Gait speeds were monitored according to self-selected walking velocities, maximum walking velocities, and performance on a two-minute walk test (2MWT). The average self-selected walking speeds increased by 13.5 percent through the first four weeks, and an additional 3.7 percent through the next four weeks, for a total improvement of 18 percent (0.96 m/s, 1.09 m/s, and 1.13 m/s respectively). Similarly, a 17 percent increase in average maximum walking velocities was realized (1.53 m/s to 1.79 m/s). In the 2MWT, the average distance traveled increased from 151m to 168m at four weeks and 181m at eight weeks, for a net improvement of about 20 percent.3

Thus, by simply adhering to a prescribed home-based exercise program, people with legacy transfemoral amputations were able to realize significant increases in the self-selected and maximal walking speeds with substantial reductions in associated energy costs.3


The Recall Outpatient Clinic Model

The final model under consideration is the recall outpatient model in which patients with extensive prosthetic experience are brought back to the gym for additional physical therapy. This model was described in a recent paper from the Columbia University Medical Center.4 A convenience sample of five community-dwelling adults with amputations were re-cruited to participate in four weekly therapy sessions with two assessment sessions occurring prior to and after the program. These subjects averaged 54 years old with an average of three years of prosthesis use. The training sessions comprised manual therapy, including hip mobilization of the affected limb, targeted stretching and core strengthening exercises, and prosthetic mobility activities including sit-to-stand transfers and those activities deemed to be the most challenging among the tasks of the Berg Balance Scale (BBS).3

Among the measures conducted before and after the four-week program were the Houghton Scale, ABC, BBS, and 2MWT. Reported values on the Houghton Scale were unchanged, indicating that average prosthesis wear and indoor/outdoor prosthesis use was unchanged. Improvements in the ABC fell shy of statistical significance, but largely because of the low number of study subjects. The average reported score on the ABC rose from 68 to 81, a clinically meaningful improvement as scores greater than 80 on the ABC are consistent with the balance confidence of highly functional elderly adults, while scores between 50 and 80 are consistent with those reported by individuals in retirement homes or with chronic illness.4

Unsurprisingly, the additional practice on the BBS led to significant improvements on the measure, with the average score increasing from 45/56 to 52/56. Finally, 2MWT values increased by 30 percent from 100m to 130m.5 While caution is always indicated in studies of only five subjects, the observed improvements were encouraging.



This series of recent publications underscores the ongoing need for engagement and activity after the subacute recovery following amputation. The chronic phase of post-amputation rehabilitation appears to represent a vulnerable time for this population, where initial progress can be eroded by the adoption of a lifestyle characterized by reduced activity and participation. Intriguingly, the amputees in the studies responded across a range of resources, including formal hands-on physical therapy, group exercise programs, and a simple home-based treadmill program. Collectively, the studies underscore the ability of this population to realize improvements beyond those captured in early rehabilitation when additional resources are invested in individuals' activities and function.


Phil Stevens, MEd, CPO, FAAOP, is in clinical practice with Hanger Clinic, Salt Lake City. He can be contacted at philmstevens@hotmail.com.



1.  Hess, R. J., J. S. Brach, S. R. Piva, and J. M. VanSwearingen. 2010. Walking skill can be assessed in older adults: Validity of the Figure-of-8 Walk Test. Physical Therapy 90(1):89.

2.  Miller, C. A., et al. 2017. The effect of a supervised community-based exercise program on balance, balance confidence, and gait in individuals with lower limb amputation. Prosthetics and Orthotics International 1-9.

3.  Darter, B. J., D. H. Nielsen, H. J. Yack, and K. F. Janz. 2013. Home-based treadmill training to improve gait performance in persons with a chronic transfemoral amputation. Archives of Physical Medicine and Rehabilitation 94(12)2440-7.

4.  Wong, C. K., et al. 2016. Impact of a four-session physical therapy program emphasizing manual therapy and exercises on balance and prosthetic walking ability of people with lower-limb amputation: A pilot study. Journal of Prosthetics and Orthotics 28:95-100.