Survey Says? Findings From the Field’s Largest Study of People With Upper-limb Amputations

Home > Articles > Survey Says? Findings From the Field’s Largest Study of People With Upper-limb Amputations
By Phil Stevens, MEd, CPO, FAAOP

Under the direction of Linda Resnik, PT, PhD, FAPTA, an effort was recently undertaken to better understand the quality and outcomes of care received by veterans with upper-limb loss. The resulting survey data from over 800 individuals was the largest study of upper-limb loss the profession has had access to, with a series of insightful manuscripts expanding the collective understanding of this relatively small population.

In addition to such basic demographic considerations as age, gender, race/ethnicity, and employment, the data included the side, level, etiology, and date of amputation, as well as hand dominance.1 Among prosthesis users, data related to the prosthesis type and terminal device use were collected, along with distinctions between primary and secondary devices where appropriate.1 Utilization patterns, including frequency of use and history of prosthesis abandonment were also considered, along with any history of prosthetic training.

Satisfaction with any reported prostheses was assessed using the Trinity Amputation and Prosthetic Experience Scale (TAPES) and the OPUS Client Satisfaction with Devices scale (CSD), collectively assessing such considerations as color, shape, noise, appearance, weight, usefulness, reliability, fit, comfort, ease of donning, and cost.1

Perceived disability was assessed using the QuickDASH (the short version of the Disabilities of the Arm, Shoulder and Hand assessment), while health-related quality of life (QOL) was assessed using a veteran version of the 12-Item Short Form Health Survey (SF-12), the Veterans RAND 12-Item Health Survey (VR-12), that produces composite scores for the Physical Component Summary (PCS) and Mental Component Summary (MCS).1

Different elements and consideration of an individual's pain experience were also explored, including phantom limb pain, residual limb pain, and overuse pain affecting the contralateral limb, neck, and back. This was coupled with reports of commonly diagnosed musculoskeletal conditions affecting the sound side such as tendinitis of the wrist, elbow, fingers or thumb, and carpal tunnel syndrome.1

The impact of amputation on activities and participation was similarly assessed, with questions relating to eating, meal preparation, housework, home maintenance, computer use, lifting, and carrying. Additional content related to body image, flashbacks related to the amputation event, prosthesis embodiment, and confidence with the prosthesis were also explored.1

The result was a massive data set, collected in hundreds of individual telephone interviews of roughly 45 minutes. The data was ultimately drawn from respondents who received care in the US Department of Veterans Affairs (VA) system between 2010 and 2016 who had experienced amputations between the forequarter and wrist disarticulation levels.1 This article summarizes key findings from this historic undertaking to help facilitate the translation of these findings into the clinic setting.

The Population

The demographics of the study cohort, gathered as it was through the VA healthcare system, were more diverse in some areas than others. The average age was 63 ±14 years old, with 31 ±18 years since their initial amputations. The most prevalent dates for amputation were in the Vietnam War era, with 24 percent and 18 percent of the respondents reporting their amputations occurred in the 1960s and 70s, respectively. However, more recent amputations were also well represented, with 18 percent occurring between 2000 and 2009, and 14 percent occurring between 2010 and 2016.1

The study cohort was overwhelmingly male (97.5 percent) and white (75 percent), with a small representation of Black (11 percent) and Hispanic (9 percent) participants. The laterality of amputations was well balanced, with left amputations only slightly more prevalent than right amputations (52 percent versus 48 percent), and bilateral upper-limb amputations reported by 4 percent of the study sample. Transradial amputations were most common (36 percent), followed by transhumeral amputations (30 percent), and wrist disarticulations (16 percent). Amputations at or proximal to the shoulder were reported by 12 percent of the sample. Concurrent amputations of at least one lower limb were reported by 12 percent of the sample.1

Sixty percent of those with unilateral upper-limb amputations reported themselves as prosthesis users, while 91 percent of those with bilateral upper-limb amputations used a prosthesis on at least one side. Among unilateral amputees, 37 percent reported using two or more prostheses, while 24 percent of those with bilateral amputations reported the use of two or more devices for at least one side. Body-powered devices were the most commonly reported primary prostheses among those with unilateral (71 percent) and bilateral (78 percent) upper-limb amputations. Externally powered prostheses were primary for 21 percent and 13.5 percent of unilateral and bilateral prosthesis users, respectively. The use of multiple terminal devices was common among those with unilateral amputations (43 percent), and comparatively rare among those with bilateral amputations (14 percent). The most commonly reported primary terminal device was the body-powered hook among both unilateral (62 percent) and bilateral (73 percent) prosthesis users.1

Daily prosthesis use was reported by 77 percent of those with unilateral limb loss, and just over half of this cohort reported eight or more hours of daily use. By contrast, 19 percent reported using their prostheses for less than two hours per day. Among those with bilateral limb loss, the usage numbers were predictably higher: 100 percent reported daily use of at least one prosthesis, and 76 percent reporting eight or more hours of daily prosthesis use.1



A separate publication was subsequently dedicated to the satisfaction reported by the 449 prosthesis users with unilateral amputations.2 Among this subcohort, transradial amputation was the predominant amputation level at 46 percent, with roughly 22 percent of transhumeral amputations and wrist disarticulations comprising much of the remaining cohort. The individual prevalence values of elbow disarticulation, shoulder disarticulation, and forequarter amputation were all less than 5 percent.

After controlling for covariates, there were no significant differences in cumulative satisfaction values by prosthesis type or terminal device type. Body-powered users tended to report greater satisfaction with respect to the reliability, durability, and weight of their devices, but were less satisfied with the wear and tear on their clothing. Myoelectric users reported lower scores with respect to weight and unintended movement.
Cosmetic prosthesis users reported lower levels of self-consciousness.2

Significant differences were observed with respect to the satisfaction associated with different amputation levels, with lower scores reported by those with transhumeral and shoulder-level amputations. Higher satisfaction scores were observed among older respondents and those who had received training with their initial prostheses. While the number of Black participants represented in the data was small (11 percent), they reported reduced satisfaction with their prostheses than their white peers, a phenomenon that suggests the need for further investigation and understanding to facilitate eventual remediation.2


Function and Quality of Life

A separate analysis was performed to better understand the impact of prostheses on the functional capacity and the physical and mental components of
QOL among individuals with unilateral upper-limb loss.3 Of the 755 individuals who met this description, users of body-powered prostheses were most common (43 percent), followed by non-users (40 percent), users of single degree-of-freedom myoelectric prostheses (8 percent), users of myoelectric prostheses with multiple degrees of freedom (5 percent) and users of cosmetic oppositional prostheses (3 percent).3

When considered by amputation level, 70 percent of the 91 individuals with amputations at the shoulder level were classified as non-users, with the most common prosthesis type among those using a prosthesis at this level being body-powered (15 percent). This trend persisted to a lesser extent at the transhumeral amputation level where 58 percent identified themselves as non-users, and 34 percent used a body-powered device as their primary prostheses. By contrast, only 23 percent of those with transradial amputations and wrist disarticulations identified as non-users, with 55 percent reporting use of body-powered prostheses, and 19 percent reporting use of some type of myoelectric prosthesis.

The analysis concluded that those individuals with unilateral upper-limb loss who choose not to use a prosthesis reported greater difficulty in performing activities, more overall disability, and reduced physical function compared to their peers who used any type of upper-limb prosthesis. Further, non-users are almost twice as likely to require help with ADLs compared to users of body-powered devices.

The use of oppositional, cosmetic prostheses, common among female prostheses users and associated with enhanced body image and psychosocial adjustment, were also associated with reduced performance on fine motor activities and general tasks. Otherwise, prosthesis configuration was not associated with any significant differences in performance outcomes.

Of interest, those who reported the use of two or more types of prosthesis reported higher scores on the physical component summary (PCS) scores of the VR-12 QOL tool, suggesting that access and proficiency with multiple devices enhanced functionality.3 Additionally, reports of access to training with an individual's current prosthesis were
associated with improved PCS scores while reports of training with an individual original prosthesis were associated with better mental component summary (MCS) scores even many years after that training.3

Time since amputation was associated with better physical function, lower disability, and a reduced likelihood of needing help with ADLs, trends that suggest that following unilateral upper-limb amputation, individuals adapt and compensate to their condition irrespective of whether they chose to use a prosthesis or not.3

Of concern, the Black veterans in this data set reported lower mental function, greater disability, and higher likelihood of needing help with ADLs than their white peers. As with discrepancies in prosthesis satisfaction scores, the reason for this disparity is unclear and requires greater understanding.

Dexterity and Activity

When considering the subject of function and dexterity, it is appropriate to share the findings of a subsequent follow-up study related to the same research effort that supplemented self-report measures of function with objective performance measures. This multisite effort ultimately recruited 127 individuals with unilateral or bilateral upper-limb loss through five participating study sites where three- to four-hour physical assessments were performed.4

A diverse suite of performance measures was utilized. Dexterity was assessed using the Jebsen-Taylor Hand Function test, the Nine Hole Peg test, the Box and Block test and the Southampton Hand Assessment Procedures (SHAP). Activity performance was recorded using the Activities Measure for Upper Limb Amputation, the Brief Activities Measure for Upper Limb Amputation, and the Timed Measure of Activity Performance.4

Individuals with unilateral transradial amputations or wrist disarticulations represented the largest study cohort in the sample (59 percent) followed by individuals with transhumeral amputations or elbow disarticulations (25 percent), individuals with bilateral upper-limb loss (12 percent) and individuals with shoulder-level amputations (4 percent). Body-powered devices were used by 59 percent of the participants, with roughly half of the remaining study participants using terminal devices with multiple degrees of freedom, and half using single degree-of-freedom terminal devices. The prevalence of body-powered devices was slightly higher among individuals with bilateral limb loss (67 percent), while devices with multiple degrees of freedom were more common among transradial prosthesis users (24 percent), and single degree-of-freedom devices were more common among transhumeral prosthesis users (28 percent), and users of bilateral upper-limb prostheses (27 percent).4

Predictably, dexterity, as measured by all the aforementioned assessments, was greatest for transradial users, followed by transhumeral prosthesis users, followed by users of shoulder-level prostheses.4 The same trend was observed with respect to the activity assessments, with transradial prosthesis users reporting the best activity scores, followed by users of transhumeral and shoulder-level prostheses respectively.4 Reflecting their increased reliance on their prostheses, those with bilateral amputations demonstrated better prosthetic dexterity scores than their peers with unilateral amputations. However, they also had lower scores on measures of activities of daily living performance with a greater likelihood of requiring assistance. The study failed to find statistically significant differences in either dexterity or activity performance by prosthesis type.4


Returning to the original data sets based on patient-reported outcomes, a distinct publication was put forward to consider the relationship between gender and prosthesis use.5 This analysis was inherently limited by the lower prevalence of female veterans with upper-limb loss (21 women versus 755 men). For example, in the sample of female patients, the prevalence of transhumeral amputations uncharacteristically exceeded that of transradial amputations. However, with these limitations understood, this preliminary data suggests that following upper-limb loss, women were less likely to have ever used a prosthesis (81 percent versus 93 percent), be a current user of a prosthesis (47 percent versus 65 percent), or have received training with their original prostheses (59 percent versus 76 percent) or current prostheses (38 percent versus 66 percent).5 Compared to men, a greater proportion of women used a oppositional, cosmetic prosthesis, (38 percent versus 4 percent), and a smaller proportion used body-powered devices (38 percent versus 71 percent).5 While women reported slightly higher disability scores and slightly lower PCS scores with respect to QOL considerations, these differences failed to reach statistical significance. By contrast, the MCS scores reported by women were statistically lower than their male peers before controlling for age, amputation level, or prosthesis type.

Phantom Pain

An additional analysis of phantom limb pain in the upper-limb population has also been published.6 Participants were asked to rate the frequency and intensity of their phantom pain and residual limb pain over the previous four-week period; data was available for 776 veterans with upper-limb loss.

Phantom limb pain was a "daily to always" experience for 42 percent of the sample. However, an additional 27 percent of participants reported no phantom pain over the prior four weeks. Reports of "daily to always" phantom limb pain were higher with more proximal amputation levels, including 57 percent of those with shoulder-level amputations and 48 percent of those with transhumeral amputations compared to 35 percent of those with trans-radial amputations. Similarly, elevated phantom pain intensity levels were more likely in more proximal amputations, with "moderate" and "severe" ratings reported by 67 percent of those with shoulder-level amputations and 66 percent of transhumeral amputations compared to only 49 percent of those with transradial amputations.6

Consistent with earlier research in this area, there was a clear relationship between the presence of phantom limb pain and residual limb pain. Those participants reporting a frequency of residual limb pain of "daily to always" were almost nine times more likely to report phantom limb pain than those citing no residual limb pain.


Back and Neck Pain

In addition to baseline analyses related to back and neck pain of the original 808 participants in the clinical survey effort, a follow-up survey of 585 of these participants demonstrated the fairly robust nature of back and neck pain within this community.7 More specifically, back pain was reported by just under three-quarters of both samples. Neck pain was slightly less common, reported at just over 60 percent in both cohorts. Daily back pain was reported by just over one-third of both study cohorts, while daily neck pain was only reported by approximately one-quarter of both of the populations. The average intensity values for back and neck pain was approximately five points on a ten-point pain severity scale.7

Contralateral Limb Pain

Pain in the contralateral limb has also been reported in the original baseline and follow-up survey efforts.8 As with back and neck pain, reported pain affecting the contralateral limb appeared to be robust over time. Pain in the contralateral limb was reported by roughly 72 percent of the study populations, with about one-third of these subjects describing "daily or more frequent pain." As with back and neck pain, the intensity of this pain experience averaged just under five on the ten-point pain scale. With respect to QOL considerations, the physical and mental components of the VR-12 were adversely affected by moderate and to a greater extent, severe pain intensities affecting the contralateral limb. Similarly, patient-reported disability values, as measured by the QuickDASH were higher among those reporting moderate to severe pain in the contralateral limb.8



A tremendous amount of insight has come from the ongoing analysis of this robust volume of patient-reported outcomes data from within the community of veterans who have experienced upper-limb loss. The relative prevalence of various amputation levels has been suggested, along with the general prevalence of prosthesis types and terminal devices. The overall prevalence of prosthesis use by amputation level has also been suggested along with daily prosthesis wearing patterns. The impact of amputation level on considerations of satisfaction, function, QOL, prosthetic dexterity, activity level, and proficiency with ADLs has been more firmly established, as has the value of appropriate training with both initial and current prostheses. The positive impacts of any prosthesis on dexterity and activity compared to the decision not to use a prosthesis has been established, as has the apparent value of access and proficiency with more than one prosthesis or terminal device. A more nuanced understanding of women with upper-limb loss has begun to be explored. Finally, the pain experience of those living with upper-limb loss has been reinforced with respect to the prevalence and intensity of phantom limb pain, residual limb pain, and pain in the contralateral limb. These insights should now be translated into the clinic, to help patients and clinicians alike anticipate challenges, establish expectations, and explore new solutions.


Phil Stevens, MEd, CPO, FAAOP, is a director with Hanger Clinic's Department of Clinical and Scientific Affairs. He can be contacted at


Resnik, L., S. Ekerholm, M. Borgia, and M. A. Clark. 2019. A national study of Veterans with major upper limb amputation: Survey methods, participants, and summary findings. PloS One 14(3):e0213578.

Resnik, L., M. Borgia, A. W. Heinemann, and M. A. Clark. 2020. Prosthesis satisfaction in a national sample of veterans with upper limb amputation. Prosthetics and Orthotics International 44(2):81-91.

Resnik, L., M. Borgia, and M. Clark. 2020. Function and quality of life of unilateral major upper limb amputees: Impact of prosthesis use and type. Archives of Physical Medicine and Rehabilitation 101(8):1396-1406.

Resnik, L., M. Borgia, and J. Cancio, et al. 2020. Dexterity, activity performance, disability, quality of life, and independence in upper limb veteran prosthesis users: A normative study. Disability and Rehabilitation 24:1-2.

Resnik, L. J., M. L. Borgia, and M. A. Clark. 2020. A national survey of prosthesis use in veterans with major upper limb amputation: Comparisons by gender. Physical Medicine and Rehabilitation 12(11):1086-98.

Balakhanlou, E., J. Webster, M. Borgia, and L. Resnik. 2020. Frequency and severity of phantom limb pain in veterans with major upper limb amputation: Results of a national survey. Physical Medicine and Rehabilitation.

Davey, J., S. Biester, and M. Borgia, et al. 2020. The prevalence and intensity of back and neck pain in veterans with upper limb amputation. Archives of Physical Medicine and Rehabilitation 101(11):e101.

Underwood, R., S. Biester, and M. Borgia, et al. 2020. The prevalence and impact of contralateral limb pain in veterans with upper limb amputation. Archives of Physical Medicine and Rehabilitation 101(11):e15-6.