Advertisement

Ambulation of People With Lower-Limb Amputations: Relationship Between Capacity and Performance Measures

      Abstract

      Parker K, Kirby RL, Adderson J, Thompson K. Ambulation of people with lower-limb amputations: relationship between capacity and performance measures.

      Objectives

      To examine the relationship between measures of ambulation capacity obtained in a clinical setting and measures of ambulation performance in the community, and to explore what demographic and clinical variables influence ambulation performance in people with lower-limb amputations.

      Design

      A cross-sectional, correlational and descriptive study.

      Setting

      Rehabilitation center and participants' homes and community environments.

      Participants

      Community-dwelling people (N=52) with lower-limb amputations at the unilateral transfemoral (n=16), unilateral transtibial (n=30), and bilateral transtibial (n=6) levels. All had been fit with prostheses for over 1 year.

      Interventions

      Not applicable.

      Main Outcome Measures

      Measures of ambulation capacity were the Locomotor Capabilities Index version 5, the 2-Minute Walk Test (2MWT), and the Timed Up and Go Test. Measures of ambulation performance included a commercially available step activity monitor (SAM; steps per day, minutes active per day, peak activity index) and self-reported performance with the Activity Restriction subscales of the Trinity Amputation and Prosthesis Experience Scales (TAPES).

      Results

      Most relationships among capacity and performance measures were in the moderate to high range (Spearman correlation coefficients, ρ=.41–.78, P<.05). The highest correlation coefficient was between the 2MWT and SAM peak activity index (ρ=.78, P=.000). A multivariate analysis found the 2MWT was significantly related to increased performance as measured by SAM mean steps per day (P=.026) and TAPES (P=.016). Depressive symptoms were also a significant predictor (P=.003) of decreased performance (TAPES).

      Conclusions

      The 2MWT, a measure of ambulation capacity, correlates well with most SAM measures of ambulation performance. Exploratory regression analysis indicated that the 2MWT is related to ambulation performance, while depression is only related to self-reported performance.

      Key Words

      List of Abbreviations:

      CES-D (Center for Epidemiological Studies Depression Scale), GLM (generalized linear model), ICF (International Classification of Functioning, Disability and Health), LCI (Locomotor Capabilities Index), LCI-5 (Locomotor Capabilities Index version 5), PEQ (Prosthesis Evaluation Questionnaire), SAM (step activity monitor), TAPES (Trinity Amputation and Prosthesis Experience Scales), 2MWT (2-Minute Walk Test), TUG (Timed Up & Go)
      MOBILITY IS A KEY component of the activity and participation domains of the ICF.
      World Health Organization
      International classification of functioning, disability and health: ICF.
      World Health Organization
      Towards a common language for functioning, disability and health: ICF.
      • Stucki G.
      • Ewert T.
      • Cieza A.
      Value and application of the ICF in rehabilitation medicine.
      The ICF provides a framework to describe and classify function and disability with respect to body functions and structure, activities, and participation. ICF defines activity as the “execution of a task or action by an individual,” and participation as the individual's “involvement in a life situation.” Qualifiers (capacity and performance) are used to describe the amount of restriction or limitation. Capacity has been defined as what can be achieved in a “standardized” or “most optimum” environment without any assistance (personal or aids), while performance refers to what people really do.
      World Health Organization
      International classification of functioning, disability and health: ICF.
      World Health Organization
      Towards a common language for functioning, disability and health: ICF.
      This construct provides the opportunity to evaluate the impact of the social and physical environment on activity through the examination of the difference between capacity and performance.
      Our interest was in exploring the relationship between what people “can do” (capacity) in a clinical environment as compared with what they “do do” (performance) in their home and community with respect to ambulation with a lower-limb prosthesis. Researchers have reviewed the various outcome measures applicable to lower-limb amputees, specifically with respect to the ICF framework.
      • Deathe A.B.
      • Wolfe D.L.
      • Devlin M.
      • Hebert J.S.
      • Miller W.C.
      • Pallaveshi L.
      Selection of outcome measures in lower extremity amputation rehabilitation: ICF activities.
      • Hebert J.S.
      • Wolfe D.L.
      • Miller W.C.
      • Deathe A.B.
      • Devlin M.
      • Pallaveshi L.
      Outcome measures in amputation rehabilitation: ICF body functions.
      Common measures of walking capacity include fixed distance or timed walk tests.
      • Deathe A.B.
      • Wolfe D.L.
      • Devlin M.
      • Hebert J.S.
      • Miller W.C.
      • Pallaveshi L.
      Selection of outcome measures in lower extremity amputation rehabilitation: ICF activities.
      • Condie E.
      • Scott H.
      • Treweek S.
      Lower limb prosthetic outcome measures: a review of the literature 1995 to 2005.
      Technology such as pedometers or activity monitors enables the measurement of steps taken by people outside the rehabilitation center, a measure of ambulation performance. Self-report measures specifically developed for people with amputations include the LCI
      • Gauthier-Gagnon C.
      • Grise M.C.
      Prosthetic profile of the amputee questionnaire: validity and reliability.
      and PEQ,
      • Legro M.W.
      • Reiber G.D.
      • Smith D.G.
      • del Aguila M.
      • Larsen J.
      • Boone D.
      Prosthesis evaluation questionnaire for persons with lower limb amputations: assessing prosthesis-related quality of life.
      which address perceived ambulatory capacity, and the Houghton Scale, which addresses perceived ambulatory performance.
      • Deathe A.B.
      • Wolfe D.L.
      • Devlin M.
      • Hebert J.S.
      • Miller W.C.
      • Pallaveshi L.
      Selection of outcome measures in lower extremity amputation rehabilitation: ICF activities.
      There are few existing studies that explore the relationship between capacity and performance for people with amputations.
      • Miller W.C.
      • Deathe A.B.
      • Speechley M.
      Lower extremity prosthetic mobility: a comparison of 3 self-report scales.
      • Bussmann J.B.
      • Grootscholten E.A.
      • Stam H.J.
      Daily physical activity and heart rate response in people with a unilateral transtibial amputation for vascular disease.
      Self-report capacity measures obtained with the LCI score were found to have no correlation to ambulation performance, as measured by the magnitude of accelerometer measures during movement.
      • Bussmann J.B.
      • Grootscholten E.A.
      • Stam H.J.
      Daily physical activity and heart rate response in people with a unilateral transtibial amputation for vascular disease.
      Moderate correlations were found between capacity measures (2MWT, r=.64; TUG test, r=−.60; LCI, r=.60; PEQ, r=.59) and a self-report measure that included aspects of ambulation performance (Houghton Scale).
      • Miller W.C.
      • Deathe A.B.
      • Speechley M.
      Lower extremity prosthetic mobility: a comparison of 3 self-report scales.
      Several researchers have measured the ambulation performance of people with lower-limb amputations in the home and community with activity monitors.
      • Bussmann J.B.
      • Grootscholten E.A.
      • Stam H.J.
      Daily physical activity and heart rate response in people with a unilateral transtibial amputation for vascular disease.
      • Holden J.
      • Fernie G.R.
      • Soto M.
      An assessment of a system to monitor the activity of patients in a rehabilitation programme.
      • Day H.J.
      The assessment and description of amputee activity.
      • Holden J.M.
      • Fernie G.R.
      Extent of artificial limb use following rehabilitation.
      • Coleman K.L.
      • Smith D.G.
      • Boone D.A.
      • Joseph A.W.
      • del Aguila M.A.
      Step activity monitor: long-term, continuous recording of ambulatory function.
      • van Dam M.S.
      • Kok G.J.
      • Munneke M.
      • Vogelaar F.J.
      • Vliet Vlieland T.P.
      • Taminiau A.H.
      Measuring physical activity in patients after surgery for a malignant tumour in the leg The reliability and validity of a continuous ambulatory activity monitor.
      • Coleman K.L.
      • Boone D.A.
      • Laing L.S.
      • Mathews D.E.
      • Smith D.G.
      Quantification of prosthetic outcomes: elastomeric gel liner with locking pin suspension versus polyethylene foam liner with neoprene sleeve suspension.
      • Berge J.S.
      • Czerniecki J.M.
      • Klute G.K.
      Efficacy of shock-absorbing versus rigid pylons for impact reduction in transtibial amputees based on laboratory, field, and outcome metrics.
      • Stepien J.M.
      • Cavenett S.
      • Taylor L.
      • Crotty M.
      Activity levels among lower-limb amputees: self-report versus step activity monitor.
      • Bussmann J.B.
      • Schrauwen H.J.
      • Stam H.J.
      Daily physical activity and heart rate response in people with a unilateral traumatic transtibial amputation.
      However, activity monitors can be expensive, and there can be logistical challenges in retrieving them. It would be convenient for clinicians and patients alike if performance in the community could be predicted by measures of capacity in the clinic. The relationship between ambulation capacity measures obtained in a clinical setting and ambulation performance, as measured by activity monitors, has not previously been systematically well explored with people with lower-limb amputations.
      Our primary objective of this study was to evaluate the relationship between measures of ambulation capacity obtained in a clinical setting and objective measures of ambulation performance in the community. Our secondary objective was to explore what demographic and clinical variables influence ambulation performance in people with lower-limb amputations. Our goal was to build upon the existing research on the objective measurement of ambulation performance of people with lower-limb amputations in their homes and communities through the use of activity monitors and to explore the relationship of this measure to ambulation capacity.
      • Bussmann J.B.
      • Grootscholten E.A.
      • Stam H.J.
      Daily physical activity and heart rate response in people with a unilateral transtibial amputation for vascular disease.
      • Holden J.
      • Fernie G.R.
      • Soto M.
      An assessment of a system to monitor the activity of patients in a rehabilitation programme.
      • Day H.J.
      The assessment and description of amputee activity.
      • Holden J.M.
      • Fernie G.R.
      Extent of artificial limb use following rehabilitation.
      • Coleman K.L.
      • Smith D.G.
      • Boone D.A.
      • Joseph A.W.
      • del Aguila M.A.
      Step activity monitor: long-term, continuous recording of ambulatory function.
      • van Dam M.S.
      • Kok G.J.
      • Munneke M.
      • Vogelaar F.J.
      • Vliet Vlieland T.P.
      • Taminiau A.H.
      Measuring physical activity in patients after surgery for a malignant tumour in the leg The reliability and validity of a continuous ambulatory activity monitor.
      • Coleman K.L.
      • Boone D.A.
      • Laing L.S.
      • Mathews D.E.
      • Smith D.G.
      Quantification of prosthetic outcomes: elastomeric gel liner with locking pin suspension versus polyethylene foam liner with neoprene sleeve suspension.
      • Berge J.S.
      • Czerniecki J.M.
      • Klute G.K.
      Efficacy of shock-absorbing versus rigid pylons for impact reduction in transtibial amputees based on laboratory, field, and outcome metrics.
      • Stepien J.M.
      • Cavenett S.
      • Taylor L.
      • Crotty M.
      Activity levels among lower-limb amputees: self-report versus step activity monitor.
      • Bussmann J.B.
      • Schrauwen H.J.
      • Stam H.J.
      Daily physical activity and heart rate response in people with a unilateral traumatic transtibial amputation.

      Methods

       Study Design

      This study was a cross-sectional, correlational and descriptive study.

       Participants

      Participants were a sample of convenience, 52 of whom were recruited through the Department of Prosthetics at a hospital-based rehabilitation center. A power calculation, testing the hypothesis that the correlation coefficient would have 80% power to detect a correlation value of 0.7 at an α level of .002 (Bonferroni adjusted for 21 comparisons), indicated that a sample size of 24 would be adequate. As many additional participants as possible were recruited within a practical period to provide a larger sample size required for exploratory analysis. Possible participants were approached by their prosthetists and, if interested in the study, were referred to the research coordinator. After consent was obtained, participants were screened by an investigator and the house staff or participant's family physician. All participants were 18 years or older, had been fit with a lower-limb prosthesis for over 1 year, were able to follow instructions, and did not have any unstable medical conditions. Participants with only partial foot amputation were excluded. Participants were collapsed into 2 diagnostic groups of amputations: transtibial (including Syme's level of amputation) and transfemoral (including knee, hip, and pelvic disarticulation amputation).

       Ethical Issues

      This study received ethical approval from the Research Ethics Board of the Capital District Health Authority. All participants provided informed consent.

       Demographic and Clinical Data

      We collected demographic and clinical data (sex, age, level of amputation, reason for amputation, years since amputation, body mass index) from each participant. The level of depressive symptomatology was obtained using the CES-D. A higher score on the CES-D indicates a greater amount of depressive symptoms; people with scores greater than 16 are considered to be depressed.
      • Radloff L.S.
      The CES-D scale: a self-report depression scale for research in general population.
      A measure of depression was included because restricted activity has been shown to be related to depression in people with lower-limb amputations.
      • Rybarczyk B.D.
      • Nyenhuis D.L.
      • Nicholas J.J.
      • Schulz R.
      • Alioto R.J.
      • Blair C.
      Social discomfort and depression in a sample of adults with leg amputations.

       Ambulation Capacity

      In this study, we defined “capacity” as a participant's ability to walk and move with his/her prosthesis and ambulation aids (canes, crutches, or walkers) in a standardized environment (rehabilitation clinic). This definition differs slightly from the ICF definition, which is the ability to walk unassisted. We obtained a subjective measure of ambulation capacity using the LCI-5. The LCI is a 4-level ordinal scale self-report measure that addresses capacity issues of transferring, walking, and climbing stairs and has been validated for people with lower-limb amputations.
      • Gauthier-Gagnon C.
      • Grise M.C.
      Prosthetic profile of the amputee questionnaire: validity and reliability.
      • Miller W.C.
      • Deathe A.B.
      • Speechley M.
      Lower extremity prosthetic mobility: a comparison of 3 self-report scales.
      • Gauthier-Gagnon C.
      • Grise M.C.
      • Lepage Y.
      The locomotor capabilities index: content validity.
      The LCI was revised by Franchignoni et al
      • Franchignoni F.
      • Orlandini D.
      • Ferriero G.
      • Moscato T.A.
      Reliability, validity, and responsiveness of the locomotor capabilities index in adults with lower-limb amputation undergoing prosthetic training.
      by splitting the highest ordinal level (score of 4) to reflect the capacity of doing tasks with (score of 4) or without (score of 5) ambulation aids to create the LCI-5. There are other measures of subjective ambulation capacity developed for people with lower-limb amputations, such as the PEQ,
      • Legro M.W.
      • Reiber G.D.
      • Smith D.G.
      • del Aguila M.
      • Larsen J.
      • Boone D.
      Prosthesis evaluation questionnaire for persons with lower limb amputations: assessing prosthesis-related quality of life.
      but for practical reasons we included only 1 subjective measure of capacity.
      The objective measures of ambulation capacity that we chose were the 2MWT and the TUG test. They are standard tests that have been shown to be reliable, valid, and sensitive to change over time for people with lower-limb amputations.
      • Brooks D.
      • Parsons J.
      • Hunter J.P.
      • Devlin M.
      • Walker J.
      The 2-minute walk test as a measure of functional improvement in persons with lower limb amputation.
      • Brooks D.
      • Hunter J.P.
      • Parsons J.
      • Livsey E.
      • Quirt J.
      • Devlin M.
      Reliability of the two-minute walk test in individuals with transtibial amputation.
      • Deathe B.
      • Miller W.C.
      • Speechley M.
      The status of outcome measurement in amputee rehabilitation in Canada.
      For the 2MWT, participants were asked to walk along a 30m level hallway as far as they could in 2 minutes. The test was timed, and the distance walked was measured to the closest 0.1m. For the TUG test, participants were seated in a standard armchair. On cue, participants got up from the chair and walked at a self-selected speed along a straight line a distance of 3m, turned and walked back to the chair to return to the seated start position. The test was timed to the nearest 0.1 seconds.

       Ambulation Performance

      We defined “performance” as a participant's mobility with his/her prosthesis and ambulation aids (if any) in the home and community environment. We obtained a subjective measure of ambulation performance in the community with the TAPES Activity Restriction subscales. TAPES is a self-report measure that assesses a person's adjustment to the prosthesis with respect to psychosocial factors, activity restrictions, satisfaction, and pain.
      • Gallagher P.
      • Maclachlan M.
      The development and psychometric evaluation of the Trinity Amputation and Prosthesis Experience scales (TAPES).
      This measure has been shown to have adequate to high internal consistency and face, content, construct, and predictive validity for people with lower-limb amputations.
      • Gallagher P.
      • Maclachlan M.
      The development and psychometric evaluation of the Trinity Amputation and Prosthesis Experience scales (TAPES).
      • Gallagher P.
      • Maclachlan M.
      The Trinity Amputation and Prosthesis Experience scales and quality of life in people with lower-limb amputation.
      Because the focus of our study was ambulation, only the Activity Restriction subscales were used. A limitation of this measure is that the use of the individual subscales has not been validated, and overall, only preliminary validation has been reported.
      • Condie E.
      • Scott H.
      • Treweek S.
      Lower limb prosthetic outcome measures: a review of the literature 1995 to 2005.
      • Deans S.A.
      • McFadyen A.K.
      • Rowe P.J.
      Physical activity and quality of life: a study of a lower-limb amputee population.
      The TAPES Activity Restriction subscales ask about how limited people are in performing various activities during a typical day because of their prostheses. The subscales address the functional, social, and athletic limitations encountered during a typical day. A higher TAPES Activity Restriction subscales score is indicative of more restrictions encountered. We considered this measure to be reflective of performance because the questions are directed towards activities done during a typical day, and many of them are associated with home and community life. Although there are limitations with respect to this outcome measure, we believed it addressed a more broad range of walking activities than the Houghton Scale, and it has also been developed specifically for the population of interest.
      We obtained an objective measure of ambulation performance in the community over a 1-week period using a SAM attached to the prosthesis at approximately ankle level on 1 side. The SAM we used was an accelerometer- and microprocessor-based activity monitor
      StepWatch 3 Activity Monitor; Orthocare Innovations, 6405 218th St SW, Ste 100, Mountlake Terrace, WA 98043-2180.
      capable of detecting steps of people with varying walking styles and mobility limitations.
      • Coleman K.L.
      • Smith D.G.
      • Boone D.A.
      • Joseph A.W.
      • del Aguila M.A.
      Step activity monitor: long-term, continuous recording of ambulatory function.
      It has been shown to be accurate for people with lower-limb amputations.
      • Coleman K.L.
      • Smith D.G.
      • Boone D.A.
      • Joseph A.W.
      • del Aguila M.A.
      Step activity monitor: long-term, continuous recording of ambulatory function.
      The output of the SAM is the number of steps taken with 1 leg (or strides) in 1 minute. We used the number of steps by both legs each minute (ie, the SAM output doubled). The main variables of interest were the total number of steps per day and the number of minutes active each day. We also included the percent time spent in low activity (1–30 steps/min), medium activity (>30–60 steps/min), and high activity (>60 steps/min) and the peak activity index, which is a measure of the average of the highest 30 minutes of step activity (not necessarily continuous) during a 24-hour day. All measures were averaged over the period worn. Because of a limited number of available SAMs and the logistical difficulty of retrieving the SAMs after the period of data collection, objective measures of ambulation performance were obtained on a subset of the participants (n=27), those who lived within 100km of the rehabilitation center.

       Procedure

      After informed consent, screening, and recording of demographic and clinical data, we obtained subjective measures first, followed by objective measures of ambulation capacity. Participants were fit with the SAM and sent home to wear the SAM for a 7-day period.

       Statistical Analysis

      Statistical analyses were performed using SAS
      Version 9.1; SAS Institute Inc, 100 SAS Campus Dr, Cary, NC 27513-2414.
      statistical software, version 9.1. We performed descriptive statistics (means, medians, SDs, ranges) on all measures. We performed exploratory correlation analyses investigating the relationships among capacity measures and among performance measures.
      Spearman correlation coefficients (ρ) were used to determine the strength of the relationships among each measure of capacity (LCI-5, 2MWT, TUG) and each measure of performance (TAPES, SAM output measures). The level of statistical significance was defined as α equal to .05. Because of the exploratory focus of this study, we did not Bonferroni-adjust the level of statistical significance for multiple comparisons. A correlation coefficient value below .29 was interpreted as having little to no correlation. We considered a value of .30 to .49 as low, .50 to .69 as moderate, and values greater than .70 as high correlations.
      • McDowell I.
      Measuring health: a guide to rating scales and questionnaires.
      To explore what demographic and clinical data influence ambulation performance, TAPES and the SAM mean steps per day as the dependent measures were modeled with a multivariate GLM with a gamma distribution (constructed with the Proc Genmod in SAS).
      • McCullagh P.
      • Nelder J.A.
      Generalized linear models.
      Independent variables included all the demographic and clinical data and the 2MWT. The 2MWT was chosen as the best measure to represent ambulation capacity because it is an objective measure of ambulation, while the TUG also includes sit-to-stand transfers. All independent data were fit first to univariate models. Variables at the P less than 0.1 level were entered into the exploratory multivariate model. The scaled deviance was used to test the goodness of fit of the model.

      Results

       Demographic and Clinical Data

      Initially, 54 participants enrolled in the study. Two participants dropped out because of scheduling difficulties. Demographic and clinical data for the remaining participants are provided in table 1. Most participants, 43 (82.7%), did not indicate any depressive symptomatology as measured by the CES-D.
      Table 1Demographic and Clinical Data of the Participants (N=52)
      CharacteristicTotal
      Unilateral amputation46 (88.5)
      Sex
       Male41 (78.8)
       Female11 (21.2)
      Reason for amputation
       Vascular20 (38.5)
       Trauma26 (50.0)
       Other6 (11.5)
      Level of amputation
       Unilateral transfemoral16 (30.8)
       Unilateral transtibial30 (57.7)
       Bilateral transtibial6 (11.5)
      Age (y)55.2±15.8
       Median54.7
       Range20.1–88.7
      Years since amputation13.6±11.1
       Median10.9
       Range1.0–41.6
      Body mass index (n=49)29.2±4.5
       Median29.0
       Range18.1–39.0
      CES-D (0–60)10.2±7.9
       Median9.0
       Range0.0–29.0
      NOTE. Values are n (%), mean ± SD, or as otherwise indicated.

       Capacity and Performance Data

      Capacity and performance data are presented in table 2. For the LCI-5 score there was a noticeable ceiling effect, with almost half of the participants obtaining the highest score possible. Some participants removed the SAM early, resulting in shorter recording times of 4 days for 2 participants and 6 days for 3 participants. For these participants, the SAM results were averaged over the number of full 24-hour days of device wear.
      Table 2Capacity and Performance Data
      ParameternMean ± SDMedianRange
      Capacity
       LCI-5 (0–56)5250.2±9.153.511–56
       2MWT (m)52111.4±49.4110.02.4–210.7
       TUG (s)5213.6±8.611.20.0–47
      Performance
       TAPES mobility subscore (0–24)5210.8±6.110.00–23
       SAM (steps/d)274217±30273218249–12,714
       SAM (min/d)27227±12022446–631
       SAM low activity time (% of 24h day)2712.8±6.313.13.0–33.8
       SAM medium activity time (% of 24h day)272.2±1.81.70.0–8.1
       SAM high activity time (% of 24h day)270.8±0.90.40.0–3.3
       SAM peak activity index (steps/min)2748.6±22.247.35.6–83.0
      Spearman correlation coefficients among capacity and performance measures are provided in table 3. Significant correlations were found for all but 1 of the relationships. Figure 1 shows the relationship between the 2MWT and SAM steps per day.
      Table 3Relationship Among Capacity and Performance Measures
      Performance MeasuresCapacity Measures
      LCI-52MWTTUG
      TAPES activity subscore n=52−.663 (.000)−.599 (.000).424 (.002)
      SAM (steps/d) n=27.640 (.000).747 (.000)−.554 (.003)
      SAM (min/d) n=27.651 (.000).607 (.001)−.413 (.032)
      SAM % low activity (% of 24h day) n=27.571 (.002).446 (.020)−.241 (.227)
      SAM % medium activity (% of 24h day) n=27.583 (.001).700 (.000)−.538 (.004)
      SAM % high activity (% of 24h day) n=27.588 (.001).762 (.000)−.569 (.002)
      SAM peak activity index (steps/min) n=27.570 (.002).776 (.000)−.575 (.002)
      NOTE. Values are Spearman correlation coefficients; P values are inside parentheses.
      Figure thumbnail gr1
      Fig 1Relationship between the 2MWT and mean SAM steps per day for each participant.
      Exploratory analysis revealed high correlation coefficients (ρ values >0.7) among all capacity measures (P<.001). The TUG was negatively correlated to the LCI-5 (ρ=−.715) and the 2MWT (ρ=−.841). The 2MWT was positively correlated to the LCI-5 (ρ=.819). Similarly, high correlations were found between mean SAM steps per day and mean SAM minutes active per day and other SAM output measures, with coefficient values ranging from .718 to .966 (P<.000). The peak activity index was highly correlated to the percent time spent in medium and high activity, with coefficient values of .920 and .957 (P=.000), respectively. The percent time spent in medium and high activity were highly correlated, with a coefficient value of .867 (P=.000). The percent time spent in low activity was moderately correlated to peak activity index and percent time spent in medium and high activity, with coefficient values ranging from .537 to .666 (P range, .000–.004). TAPES had low to moderate negative correlations with the SAM output measures, with coefficient values ranging from −.451 to −.695 (P range, .000–.018).

       Regression Model

      For the univariate models with the TAPES Activity Restriction subscales as the dependent measure, 5 independent variables were significant. The TAPES score increased with increased age (P=.022) and increased CES-D score (P=.006). The TAPES score decreased with an increase in years since amputation (P=.012), subjects with amputations due to trauma or other reasons (P=.012), and an increased 2MWT score (P=.000). These 5 variables were entered into the multivariate GLM. Only the CES-D and the 2MWT remained significantly related to the TAPES Activity Restriction subscales (table 4).
      Table 4Parameter Estimates of Multivariate Regression Model Analysis
      Dependent VariablesPredictor VariablesEstimateSEP
      TAPES Activity2MWT (m)−.004.002.016
       Restriction subscalesCES-D.024.008.003
      Age (y).002.005.708
      Trauma or other amputation−.127.166.444
      Years since amputation−.008.006.203
      Intercept2.643.397.000
      SAM mean steps per day2MWT (m)31.14313.096.026
      Age (y)−17.16345.523.690
      Transfemoral amputation−1433.2561009.471.169
      Intercept2239.5093715.390.553
      For the univariate models with SAM mean steps per day as the dependent measure, only 2 independent variables were significantly related. The SAM mean steps per day decreased with an increased age (P=.033) and increased with the 2MWT (P=.000). These variables along with the demographic variable indicating level of amputation (P=.098) were entered into the multivariate GLM. The 2MWT was the only measure significantly related to the SAM mean steps per day (see table 4).

      Discussion

      The primary purpose of this study was to evaluate the relationship between measures of ambulation capacity obtained in a clinical setting and measures of ambulation performance in the community in persons with lower-limb amputations. Although performance, as measured by activity monitors, and capacity ambulation measures have been used as outcome measures in people with lower-limb amputations, the relationship between these measures has not been well explored.
      • Coleman K.L.
      • Boone D.A.
      • Laing L.S.
      • Mathews D.E.
      • Smith D.G.
      Quantification of prosthetic outcomes: elastomeric gel liner with locking pin suspension versus polyethylene foam liner with neoprene sleeve suspension.
      • Stepien J.M.
      • Cavenett S.
      • Taylor L.
      • Crotty M.
      Activity levels among lower-limb amputees: self-report versus step activity monitor.
      Most capacity measures were moderately correlated to performance measures except between the 2MWT and steps per day, peak activity index, and percent time spent in medium and high activity, which had high correlation coefficient values (ρ range, .700–.776). Similarly, low to moderate correlations have been found in other populations in evaluating the relationship between self-report capacity measures and daily step activity, while moderate to high correlations were found between capacity walking tests such as the 6-minute and 10-m walk tests and daily step activity.
      • Mudge S.
      • Stott N.S.
      • Walt S.E.
      Criterion validity of the StepWatch activity monitor as a measure of walking activity in patients after stroke.
      • Smith D.G.
      • Domholdt E.
      • Coleman K.L.
      • del Aguila M.A.
      • Boone D.A.
      Ambulatory activity in men with diabetes: relationship between self-reported and real-world performance-based measures.
      Miller et al
      • Miller W.C.
      • Deathe A.B.
      • Speechley M.
      Lower extremity prosthetic mobility: a comparison of 3 self-report scales.
      also found moderate correlations between the 2MWT, TUG, LCI, and a self-report performance measure in subjects with lower-limb amputations. Because other factors, such as the physical environment and social support, influence ambulation performance, it is not surprising that some moderate relationships between capacity and performance were obtained. Ours is the first study to systematically measure a relationship between both objective and subjective measures of ambulation capacity and objective measures of ambulation performance as recorded by a SAM in people with lower-limb amputations.
      A secondary objective was to explore what demographic and clinical data influenced ambulation performance in people with lower-limb amputations. The multivariate analysis indicated that the CES-D score and the 2MWT were significant predictors of performance of ambulation in the community as measured by the TAPES Activity Restriction subscales. A recent study
      • Asano M.
      • Rushton P.
      • Miller W.C.
      • Deathe B.A.
      Predictors of quality of life among individuals who have a lower limb amputation.
      found depression, as measured by the CES-D, to be a significant predictor of quality of life of people with lower-limb amputation, followed by self-reported physical mobility. Increased age and having an above-knee amputation individually decreased the mean steps per day but not significantly in the multivariate model. A significant decrease in amount walked with increased age or having an above-knee amputation was seen by Holden and Fernie
      • Holden J.M.
      • Fernie G.R.
      Extent of artificial limb use following rehabilitation.
      in people with unilateral amputations during the early rehabilitation process and 2 years after discharge. In our study, the 2MWT was the only significant predictor of mean steps per day.

       Study Limitations

      There were several study limitations. The sample size of this study was small and was one of convenience, recruited from the outpatient prosthetic department. As many of these participants came to the department for self-referred reasons, such as adjustments and other fit or prosthetic component issues, we were already biased in measuring a population that used their prosthetic limbs routinely. This sample may be healthier and more active than the overall population of people with amputations. Because of the small sample size, all data were included, and outliers may have been present that would have decreased our correlation results. The small sample size, specifically with the group fit with the SAM (n=27), limited the knowledge gained from the exploratory prediction modeling. Typically, for every variable modeled in a regression, 10 to 12 cases would be required. In this study, 8 clinical and demographic variables were included in the model, suggesting that a sample size of 80 to 96 would have been more appropriate. A larger-scale study will be required to confirm any exploratory regression findings.
      Some variation in activity can be due to seasonal effects, specifically in climates that experience 4 distinct seasons, such as Michigan, where activity increased by as much as 20% between April and September in healthy adult populations.
      • Pivarnik J.M.
      • Reeves M.J.
      • Rafferty A.P.
      Seasonal variation in adult leisure-time physical activity.
      Most of our data (82.7%) were collected during this time frame; the remainder were collected during the fall in a climate similar to that of Michigan. Because of the small sample size, we did not adjust for any seasonal variation, and this may have been a study limitation. Because the participants knew their walking performance was being measured by the SAM, this may have caused them to alter the amount and type of activity performed and skew the results. There was also a wide range of variability in the mean steps per day, from 249 to 12,714 steps/d. The SD in mean steps per day (±3027) was within the range of previously published studies, while the total mean steps per day of 4217 was approximately 1500 steps/d lower than what has been obtained in other studies investigating daily activity in the population of people with lower-limb amputations.
      • Stepien J.M.
      • Cavenett S.
      • Taylor L.
      • Crotty M.
      Activity levels among lower-limb amputees: self-report versus step activity monitor.
      • Klute G.K.
      • Berge J.S.
      • Orendurff M.S.
      • Williams R.M.
      • Czerniecki J.M.
      Prosthetic intervention effects on activity of lower-extremity amputees.
      Percent time spent in low, medium, and high activity was within the range of other published data using similar definitions of activity levels.
      • Coleman K.L.
      • Boone D.A.
      • Laing L.S.
      • Mathews D.E.
      • Smith D.G.
      Quantification of prosthetic outcomes: elastomeric gel liner with locking pin suspension versus polyethylene foam liner with neoprene sleeve suspension.
      The differences in our obtained measures as compared with other studies may be due to differences in the population studied and our inclusion of people with bilateral amputations, or could be due to differences in the environment and the influence of rural as compared with an urban community environment. We focused on a subset of SAM outcome measures, but other variables such as bouts of activity per day may also have been relevant.
      The strength of the relationship between capacity, as measured by the TUG test, and perfomance was lower than expected. This may have occurred because of a ceiling effect with the TUG that has been observed in people with amputations who are fit and younger, a bias that may have been present in our studied population.
      • Deathe A.B.
      • Miller W.C.
      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.
      As well, the TUG test reflects other dimensions of mobility such as sit-to-stand transfers along with ambulation capacity, which may help explain the moderate relationship to performance. The ceiling effect seen in the LCI-5 was not anticipated because the measure was developed from the original LCI to help avoid this effect.
      • Franchignoni F.
      • Orlandini D.
      • Ferriero G.
      • Moscato T.A.
      Reliability, validity, and responsiveness of the locomotor capabilities index in adults with lower-limb amputation undergoing prosthetic training.
      To date, no published studies have indicated any ceiling effect of the LCI-5.
      • Deathe A.B.
      • Wolfe D.L.
      • Devlin M.
      • Hebert J.S.
      • Miller W.C.
      • Pallaveshi L.
      Selection of outcome measures in lower extremity amputation rehabilitation: ICF activities.
      This may also be a result of a bias present in our studied population.
      All of the relationships found between capacity measures and the TAPES Activity Restriction subscales score were at the moderate level, suggesting that different aspects of ambulation were assessed between capacity and self-report performance measures. The TAPES Activity Restriction subscales measure a broad range of activities performed during a typical day and involve aspects of both activity and participation. Because of the broad range of activities measured, low to moderate relationships between the TAPES Activity Restriction subscales score and the TUG test and 2MWT were anticipated.
      Many of the measures used reflect different aspects of ambulation and mobility. The LCI-5 includes activities such as walking up stairs or carrying objects that would not be reflected in the 2MWT or TUG test. The TAPES Activity Restriction subscales include social activity restriction, which may or may not be reflected in the measured steps per day. Some aspects of the social activities, specifically work and hobbies, may not have been applicable to many of the participants. Moderate relationships between and among some of the performance and capacity measures are understandable because of the different aspects of ambulation and mobility being measured. As seen in recent literature, this discrepancy could also be explained by the construct capability, which would measure the opportunity for work and hobbies in their home and environment.
      • Holsbeeke L.
      • Ketelaar M.
      • Schoemaker M.M.
      • Gorter J.W.
      Capacity, capability, and performance: different constructs or three of a kind?.
      • Morris C.
      Measuring participation in childhood disability: how does the capability approach improve our understanding?.
      Although there appear to be limitations in the use of the TAPES Activity Restriction subscales, because so few measures address prosthetic use and the performance of activity, we believe this is still a useful clinical measure and hope future studies will address the validity of the subscore. Typically, observational, functional tests or self-report measures are used in a clinical setting to determine the ambulatory function of people with lower-limb amputations. In practice, these measures are used to reflect a person's actual ambulation performance in daily life because few practical alternatives are readily available. Overall, the 2MWT appears to be a good measure that is simple to obtain in the clinic and reflects the community ambulation performance of people with lower-limb amputations.
      Although this study focused on the relationship between ambulation capacity and performance, exploratory analysis of what clinical and demographic variables help predict ambulation performance and recent literature suggest that other aspects such as depression help explain the variance in ambulation performance. In our study, 17.3% of the participants indicated depressive symptomatology as measured by the CES-D score as compared with levels ranging from 24.8% to 28.7% in similar community-dwelling populations.
      • Asano M.
      • Rushton P.
      • Miller W.C.
      • Deathe B.A.
      Predictors of quality of life among individuals who have a lower limb amputation.
      • Darnall B.D.
      • Ephraim P.
      • Wegener S.T.
      • et al.
      Depressive symptoms and mental health service utilization among persons with limb loss: results of a national survey.
      The discrepancy may be due to our smaller sample size and other differences in our population such as marital status, poverty, pain, existing comorbid conditions, and education that are known to be related to depression but were not measured.
      • Darnall B.D.
      • Ephraim P.
      • Wegener S.T.
      • et al.
      Depressive symptoms and mental health service utilization among persons with limb loss: results of a national survey.
      Future large-scale studies should incorporate measures looking at social support, depression, and other sociodemographic, health, and environmental variables to explore what helps explain the variance in community ambulation performance.

      Conclusions

      The 2MWT, a measure of ambulation capacity, correlates well with most SAM measures of ambulation performance. Exploratory regression analysis indicated that the 2MWT is related to ambulation performance, while depression is only related to self-reported performance.
      Suppliers
      aStepWatch 3 Activity Monitor; Orthocare Innovations, 6405 218th St SW, Ste 100, Mountlake Terrace, WA 98043-2180.
      bVersion 9.1; SAS Institute Inc, 100 SAS Campus Dr, Cary, NC 27513-2414.

      Acknowledgments

      We thank Catherine McPhail, CP, of the Queen Elizabeth II Health Sciences Centre for her contribution in participant recruitment.

      References

        • World Health Organization
        International classification of functioning, disability and health: ICF.
        World Health Organization, Geneva2001
        • World Health Organization
        Towards a common language for functioning, disability and health: ICF.
        World Health Organization, Geneva2002
        • Stucki G.
        • Ewert T.
        • Cieza A.
        Value and application of the ICF in rehabilitation medicine.
        Disabil Rehabil. 2002; 24: 932-938
        • Deathe A.B.
        • Wolfe D.L.
        • Devlin M.
        • Hebert J.S.
        • Miller W.C.
        • Pallaveshi L.
        Selection of outcome measures in lower extremity amputation rehabilitation: ICF activities.
        Disabil Rehabil. 2009; 31: 1455-1473
        • Hebert J.S.
        • Wolfe D.L.
        • Miller W.C.
        • Deathe A.B.
        • Devlin M.
        • Pallaveshi L.
        Outcome measures in amputation rehabilitation: ICF body functions.
        Disabil Rehabil. 2009; 31: 1541-1554
        • Condie E.
        • Scott H.
        • Treweek S.
        Lower limb prosthetic outcome measures: a review of the literature 1995 to 2005.
        J Prosthet Orthot. 2006; 18: 13-45
        • Gauthier-Gagnon C.
        • Grise M.C.
        Prosthetic profile of the amputee questionnaire: validity and reliability.
        Arch Phys Med Rehabil. 1994; 75: 1309-1314
        • Legro M.W.
        • Reiber G.D.
        • Smith D.G.
        • del Aguila M.
        • Larsen J.
        • Boone D.
        Prosthesis evaluation questionnaire for persons with lower limb amputations: assessing prosthesis-related quality of life.
        Arch Phys Med Rehabil. 1998; 79: 931-938
        • Miller W.C.
        • Deathe A.B.
        • Speechley M.
        Lower extremity prosthetic mobility: a comparison of 3 self-report scales.
        Arch Phys Med Rehabil. 2001; 82: 1432-1440
        • Bussmann J.B.
        • Grootscholten E.A.
        • Stam H.J.
        Daily physical activity and heart rate response in people with a unilateral transtibial amputation for vascular disease.
        Arch Phys Med Rehabil. 2004; 85: 240-244
        • Holden J.
        • Fernie G.R.
        • Soto M.
        An assessment of a system to monitor the activity of patients in a rehabilitation programme.
        Prosthet Orthot Int. 1979; 3: 99-102
        • Day H.J.
        The assessment and description of amputee activity.
        Prosthet Orthot Int. 1981; 5: 23-28
        • Holden J.M.
        • Fernie G.R.
        Extent of artificial limb use following rehabilitation.
        J Orthop Res. 1987; 5: 562-568
        • Coleman K.L.
        • Smith D.G.
        • Boone D.A.
        • Joseph A.W.
        • del Aguila M.A.
        Step activity monitor: long-term, continuous recording of ambulatory function.
        J Rehabil Res Dev. 1999; 36: 8-18
        • van Dam M.S.
        • Kok G.J.
        • Munneke M.
        • Vogelaar F.J.
        • Vliet Vlieland T.P.
        • Taminiau A.H.
        Measuring physical activity in patients after surgery for a malignant tumour in the leg.
        J Bone Joint Surg Br. 2001; 83: 1015-1019
        • Coleman K.L.
        • Boone D.A.
        • Laing L.S.
        • Mathews D.E.
        • Smith D.G.
        Quantification of prosthetic outcomes: elastomeric gel liner with locking pin suspension versus polyethylene foam liner with neoprene sleeve suspension.
        J Rehabil Res Dev. 2004; 41: 591-602
        • Berge J.S.
        • Czerniecki J.M.
        • Klute G.K.
        Efficacy of shock-absorbing versus rigid pylons for impact reduction in transtibial amputees based on laboratory, field, and outcome metrics.
        J Rehabil Res Dev. 2005; 42: 795-808
        • Stepien J.M.
        • Cavenett S.
        • Taylor L.
        • Crotty M.
        Activity levels among lower-limb amputees: self-report versus step activity monitor.
        Arch Phys Med Rehabil. 2007; 88: 896-900
        • Bussmann J.B.
        • Schrauwen H.J.
        • Stam H.J.
        Daily physical activity and heart rate response in people with a unilateral traumatic transtibial amputation.
        Arch Phys Med Rehabil. 2008; 89: 430-434
        • Radloff L.S.
        The CES-D scale: a self-report depression scale for research in general population.
        Appl Psychol Meas. 1977; 1: 385-401
        • Rybarczyk B.D.
        • Nyenhuis D.L.
        • Nicholas J.J.
        • Schulz R.
        • Alioto R.J.
        • Blair C.
        Social discomfort and depression in a sample of adults with leg amputations.
        Arch Phys Med Rehabil. 1992; 73: 1169-1173
        • Gauthier-Gagnon C.
        • Grise M.C.
        • Lepage Y.
        The locomotor capabilities index: content validity.
        J Rehabil Outcome Meas. 1998; 2: 40-46
        • Franchignoni F.
        • Orlandini D.
        • Ferriero G.
        • Moscato T.A.
        Reliability, validity, and responsiveness of the locomotor capabilities index in adults with lower-limb amputation undergoing prosthetic training.
        Arch Phys Med Rehabil. 2004; 85: 743-748
        • Brooks D.
        • Parsons J.
        • Hunter J.P.
        • Devlin M.
        • Walker J.
        The 2-minute walk test as a measure of functional improvement in persons with lower limb amputation.
        Arch Phys Med Rehabil. 2001; 82: 1478-1483
        • Brooks D.
        • Hunter J.P.
        • Parsons J.
        • Livsey E.
        • Quirt J.
        • Devlin M.
        Reliability of the two-minute walk test in individuals with transtibial amputation.
        Arch Phys Med Rehabil. 2002; 83: 1562-1565
        • Deathe B.
        • Miller W.C.
        • Speechley M.
        The status of outcome measurement in amputee rehabilitation in Canada.
        Arch Phys Med Rehabil. 2002; 83: 912-918
        • Gallagher P.
        • Maclachlan M.
        The development and psychometric evaluation of the Trinity Amputation and Prosthesis Experience scales (TAPES).
        Rehabil Psychol. 2000; 45: 130-155
        • Gallagher P.
        • Maclachlan M.
        The Trinity Amputation and Prosthesis Experience scales and quality of life in people with lower-limb amputation.
        Arch Phys Med Rehabil. 2004; 85: 730-736
        • Deans S.A.
        • McFadyen A.K.
        • Rowe P.J.
        Physical activity and quality of life: a study of a lower-limb amputee population.
        Prosthet Orthot Int. 2008; 32: 186-200
        • McDowell I.
        Measuring health: a guide to rating scales and questionnaires.
        3rd ed. Oxford University Press, New York2006
        • McCullagh P.
        • Nelder J.A.
        Generalized linear models.
        2nd ed. Chapman & Hall, London1989
        • Mudge S.
        • Stott N.S.
        • Walt S.E.
        Criterion validity of the StepWatch activity monitor as a measure of walking activity in patients after stroke.
        Arch Phys Med Rehabil. 2007; 88: 1710-1715
        • Smith D.G.
        • Domholdt E.
        • Coleman K.L.
        • del Aguila M.A.
        • Boone D.A.
        Ambulatory activity in men with diabetes: relationship between self-reported and real-world performance-based measures.
        J Rehabil Res Dev. 2004; 41: 571-580
        • Asano M.
        • Rushton P.
        • Miller W.C.
        • Deathe B.A.
        Predictors of quality of life among individuals who have a lower limb amputation.
        Prosthet Orthot Int. 2008; 32: 231-243
        • Pivarnik J.M.
        • Reeves M.J.
        • Rafferty A.P.
        Seasonal variation in adult leisure-time physical activity.
        Med Sci Sports Exerc. 2003; 35: 1004-1008
        • Klute G.K.
        • Berge J.S.
        • Orendurff M.S.
        • Williams R.M.
        • Czerniecki J.M.
        Prosthetic intervention effects on activity of lower-extremity amputees.
        Arch Phys Med Rehabil. 2006; 87: 717-722
        • Deathe A.B.
        • Miller W.C.
        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.
        Phys Ther. 2005; 85: 626-635
        • Holsbeeke L.
        • Ketelaar M.
        • Schoemaker M.M.
        • Gorter J.W.
        Capacity, capability, and performance: different constructs or three of a kind?.
        Arch Phys Med Rehabil. 2009; 90: 849-855
        • Morris C.
        Measuring participation in childhood disability: how does the capability approach improve our understanding?.
        Dev Med Child Neurol. 2009; 51: 92-94
        • Darnall B.D.
        • Ephraim P.
        • Wegener S.T.
        • et al.
        Depressive symptoms and mental health service utilization among persons with limb loss: results of a national survey.
        Arch Phys Med Rehabil. 2005; 86: 650-658