UBC Faculty Research and Publications

A survey of stakeholder perspectives on exoskeleton technology Wolff, Jamie; Parker, Claire; Borisoff, Jaimie; Mortenson, W B; Mattie, Johanne Dec 19, 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12984_2014_Article_700.pdf [ 805.29kB ]
JSON: 52383-1.0228381.json
JSON-LD: 52383-1.0228381-ld.json
RDF/XML (Pretty): 52383-1.0228381-rdf.xml
RDF/JSON: 52383-1.0228381-rdf.json
Turtle: 52383-1.0228381-turtle.txt
N-Triples: 52383-1.0228381-rdf-ntriples.txt
Original Record: 52383-1.0228381-source.json
Full Text

Full Text

RESEARCHA survey of stakeholderofitamong many wheelchair users [3,4]. Standing and walking, fits [8,9]. Technological efforts to enable functional ambu-J N E R JOURNAL OF NEUROENGINEERINGAND REHABILITATIONWolff et al. Journal of NeuroEngineering and Rehabilitation 2014, 11:169http://www.jneuroengrehab.com/content/11/1/169robotic exoskeletons.Canada Research Chair in Rehabilitation Engineering Design, British ColumbiaInstitute of Technology, (3700 Willingdon Ave.), Burnaby (B5G 3H2), CanadaFull list of author information is available at the end of the articleeither independently or with assistance, may also improveseveral aspects of health, including blood pressure, jointrange of motion, bladder health, skin integrity, spasticity,and pain [5-7]. Simple devices such as standing frameslation (i.e. to replace the wheelchair) have been underwayfor decades. Orthotics such as long leg braces are stillprescribed, although they are rarely used by people withspinal cord injury (SCI). Newer passive orthoses, such asthe reciprocating gait orthosis [10,11] and hip guidanceorthosis [12] have been developed; however, their use isalso limited. The latest efforts concern the development of* Correspondence: Jaimie_Borisoff@bcit.ca1Occupational Science and Occupational Therapy, University of BritishColumbia, (T325 - 2211 Wesbrook Mall), Vancouver (V6T 2B5), Canada2remains for standing and walking as a means of mobilitythere are practical barriers to their everyday use as a mobility device. To further understand potential exoskeletonuse, and facilitate the development of new technologies, a study was undertaken to explore perspectives ofwheelchair users and healthcare professionals on reasons for use of exoskeleton technology, and the importance ofa variety of device characteristics.Methods: An online survey with quantitative and qualitative components was conducted with wheelchair usersand healthcare professionals working directly with individuals with mobility impairments. Respondents ratedwhether they would use or recommend an exoskeleton for four potential reasons. Seventeen design features wererated and compared in terms of their importance. An exploratory factor analysis was conducted to categorize the17 design features into meaningful groupings. Content analysis was used to identify themes for the open endedquestions regarding reasons for use of an exoskeleton.Results: 481 survey responses were analyzed, 354 from wheelchair users and 127 from healthcare professionals.The most highly rated reason for potential use or recommendation of an exoskeleton was health benefits. Of thedesign features, 4 had a median rating of very important: minimization of falls risk, comfort, putting on/taking offthe device, and purchase cost. Factor analysis identified two main categories of design features: Functional Activitiesand Technology Characteristics. Qualitative findings indicated that health and physical benefits, use for activity andaccess reasons, and psychosocial benefits were important considerations in whether to use or recommend anexoskeleton.Conclusions: This study emphasizes the importance of developing future exoskeletons that are comfortable,affordable, minimize fall risk, and enable functional activities. Findings from this study can be utilized to inform thepriorities for future development of this technology.Keywords: Exoskeleton, Powered orthoses, Spinal cord injury, Social participation, Mobility, User perspectiveBackgroundWhile wheelchairs may promote activities of daily livingand participation in the community [1,2], a strong desireoffer several of these benefits [6]. Clinical gait trainingwith body weight-support, either therapist assisted orusing robotic devices such as the Lokomat, is becomingmore widespread due to its health and rehabilitation bene-exoskeleton technologyJamie Wolff1, Claire Parker1, Jaimie Borisoff1,2,4,6*, W Ben MAbstractBackground: Exoskeleton technology has potential bene© 2014 Wolff et al.; licensee BioMed CenCommons Attribution License (http://crereproduction in any medium, provided tDedication waiver (http://creativecommounless otherwise stated.ptral. Tativeche orns.orOpen Accesserspectives onrtenson1,4,5 and Johanne Mattie3s for wheelchair users’ health and mobility. However,his is an Open Access article distributed under the terms of the Creativeommons.org/licenses/by/4.0), which permits unrestricted use, distribution, andiginal work is properly credited. The Creative Commons Public Domaing/publicdomain/zero/1.0/) applies to the data made available in this article,Wolff et al. Journal of NeuroEngineering and Rehabilitation 2014, 11:169 Page 2 of 10http://www.jneuroengrehab.com/content/11/1/169A robotic exoskeleton is a wearable, powered lowerlimb orthosis that uses a system of actuators and sensorsto achieve walking movements. Currently exoskeletonsare primarily used in supervised clinical settings forhealth and rehabilitation purposes, but are eventuallyintended for daily use as a functional mobility device[13]. The ReWalk™ exoskeleton was recently approved forhome use by the United States Food and Drug Adminis-tration, when the user is accompanied by a speciallytrained assistant [14]. Most current designs (such as theReWalk™, Ekso Bionics™, and Indego™) require the use ofarm crutches or a walker for stability. The Rex™ roboticwalking device, however, is self-supporting, requiring noother device for stability. Exoskeleton users initiate move-ment either with hand controls or using the position oftheir upper body. Primary candidates for this type oftechnology are individuals with mobility impairments,in particular those who rely on wheelchairs for mobilityand have bilateral upper extremity function. Approxi-mately 0.6% of Canadians (210,000 people) and 0.7% ofAmericans (2.2 million people) reported using a wheel-chair, in 2006 and 2012 respectively [15,16]. Many of theseindividuals could, therefore, be potential candidates to usean exoskeleton.Exoskeletons may play a larger role in rehabilitationmoving forward [9]. A recent narrative review found thatusing exoskeletons as a method of partial assistance forrehabilitation following incomplete spinal cord injurywas an effective technique for gait retraining andstrengthening functioning muscles [17,18]. Further, asystematic review on exoskeletons in stroke rehabilita-tion found that their use in combination with physio-therapy led to an increased incidence of independentwalking [19]. Two studies examining safety training andtolerance for the ReWalk™ exoskeleton over short dis-tances demonstrated it had low safety risks, was well tol-erated, and that users improved in its use with training[20,21]. Spungen et al. [13] noted that with training,some participants were able to independently performselected home and community based skills using theexoskeleton, including walking on a slope and accessinga high shelf while standing.While there is much excitement around these new ro-botic exoskeletons, there are issues that may limit theirutility both as a therapeutic device and as a mobility de-vice. Some significant limiting factors include difficultydonning and doffing, problems transferring, slow andoften rough movement, lack of dependability, and con-cerns surrounding pressure distribution and skin integ-rity [22]. Researchers have identified four key topics forfuture development of exoskeletons: robust control, safetyand dependability, ease of wear-ability or portability, andusability/acceptance [23]. For example, if a person cannoteasily use a device, or has problems with accepting a noveltechnology, it will likely be abandoned or not used to itsfull potential [22]. For this reason, the wider acceptance ofexoskeletons for both rehabilitation and function isdependent on the end user being central to design and de-velopment of the technology [23].Despite the potential benefit of these devices, and im-portance of user acceptance, little is documented aboutstakeholder perspectives on exoskeletons. One qualita-tive study found that potential end users and mobilityspecialists were primarily concerned with the safety,cost, ease of use, and functionality of the device [24].Additional research on user perspectives and applicabil-ity of exoskeletons is needed in order to understand thefeatures that stakeholders feel are most important, inorder to guide development of safe, functional, user-friendly devices. Therefore this study was undertaken toexamine and compare stakeholder (wheelchair users andhealthcare professionals) perspectives on exoskeletontechnology, with respect to perceived importance of de-sign features and potential reasons for use.MethodsStudy designData for the study were collected using an online survey,which was developed and administered using the tai-lored design method [25]. The survey was piloted to asmall group of participants (n = 6), from both stakeholdergroups. Based on their feedback, minor adaptations weremade to wording and layout, and a final version of thesurvey was created, which was approved by the Behav-ioural Research Ethics Board of the University of BritishColumbia.The survey included 30 questions: 7 questions to collectdemographic information (age, sex, country of residence,level of education, primary diagnosis, and profession); 5questions related to past experiences and familiarity withexoskeleton technology; and 17 questions about reasonsfor use of an exoskeleton and importance of various de-sign and functionality considerations. Questions primarilyused a multiple choice response format (demographicsand reasons for use), or a 5-point Likert scale, with Likertscales ranged from 1-Very Unimportant to 5-Very Import-ant (design considerations), or 1-Strongly Disagree to5-Strongly Agree (statements about exoskeleton designcharacteristics). Participants were asked to respond Yes orNo to whether they would use an exoskeleton for: healthbenefits, rehabilitation purposes, social interactions, and/or functional day-to-day tasks. Participants were alsoasked one open-ended question. Wheelchair users wereasked: “Are there any other reasons you would use an exo-skeleton?”, whereas healthcare professionals “Are thereany other reasons you would recommend an exoskel-eton?” The full survey is available as Additional file 1.Wolff et al. Journal of NeuroEngineering and Rehabilitation 2014, 11:169 Page 3 of 10http://www.jneuroengrehab.com/content/11/1/169SampleTwo groups of stakeholders were recruited for thisstudy: wheelchair users and healthcare professionalsworking directly with wheelchair users. To be eligible forthis study, wheelchair users needed to be over 18 yearsof age, fluent in English, and use a wheelchair as a pri-mary means of mobility (self-defined). Healthcare pro-fessionals needed to be over 18 years of age, fluent inEnglish, and have experience working with individualswith mobility impairments (e.g., occupational therapists,rehabilitation assistants, nurses, physiotherapists, phys-iatrists, orthotists, assistive technology specialists, ormobility equipment vendors). Because the study aimedto evaluate perspectives on potential rather than actualuse of the devices in question, no exclusion criteria re-lated to participants’ physical abilities were set.Participants were recruited using mass emails to data-bases of research volunteers from prior studies con-ducted by the authors’ respective organizations, postingson user and healthcare professional online forums, flyersposted in rehabilitation centres, social media, and wordof mouth. Data were collected between February andJune, 2014.Data analysisThe quantitative data were analyzed using IBM SPSSStatistics for Windows, Version 22.0. Armonk, NY: IBMCorp. Descriptive statistics and graphic representationswere used to characterize the sample and to compare theimportance of different design features. Importance com-parisons were conducted using medians, and % of respon-dents rating a factor as “important” or “very important”.Exploratory factor analysis (EFA) was used to examinehow responses to individual questions about differentexoskeleton design characteristics (i.e. variables) were re-lated to one another. That is, could the characteristicsbe grouped together into certain categories? EFA is amethod to extract these broad underlying categories,which are then called factors [26]. The number of factorsto be extracted was determined through examination ofa scree plot of the eigenvalues [26]. Maximum likelihoodwas the method of extraction and direct oblimin withKaiser normalization was the method of rotation. Forfactor analysis, loadings > .71 are considered excellent,>.63 are considered very good, >.55 are considered good,and > .45 are considered fair [27].The Kaiser-Meyer-Olkin (KMO) measure and Bartlett’s test of sphericitywere utilized to ensure adequacy of the sample for EFA[26]. For the KMO measure, a minimum of 0.5 is recom-mended, 0.60-0.69 is considered mediocre, 0.70-0.79 isconsidered fair and 0.80-0.89 is considered good [28].Through this method, associations and patterns amonggroups of variables were used to group potential exo-skeleton features into factor-based categories. Thesecategories were then compared for wheelchair users andhealthcare professionals using an independent samplesMann–Whitney U test. Chi-squared tests were performedto determine significant differences between stakeholdergroups for reasons to use or recommendation of anexoskeleton.Responses to open-ended questions were analyzed usingcontent analysis [29]. Analysis was based in a perspectiveof engagement in meaningful activity, and themes orga-nized using the Human Activity Assistive Technology(HAAT) model [30]. This model describes a person, an ac-tivity, and assistive technology interacting within a context[30]. The HAAT model depicts the person as possessingunderlying skills and abilities which they bring to a giventask. The assistive technology, in this case the exoskeleton,influences human performance. This occurs within a con-text, which includes the physical, social, and cultural envi-ronments. This framework was used as an analytical lensto conceptualize the multifaceted nature of the human-technology interaction within themes. Emergent coding(i.e.,exploring the content without previously formulatedassumptions about the results) was used to establishcategories from the individual responses, and inductiveanalysis, generating broader ideas based in specific details,was performed to combine categories into broader themes[29]. Relative frequencies of categories and themes wereassessed to determine the most prevalent themes withinthe responses. Responses could be coded with more thanone theme.ResultsParticipantsA total of 603 participants responded to the survey. Ofthese, 122 respondents did not meet the inclusion criteriaand/or did not fully complete the survey and were excludedfrom the data analysis. Demographic information about the481 remaining respondents is described in Table 1.Reasons to use an exoskeletonWhen participants were asked whether they would use anexoskeleton for health benefits, rehabilitation purposes,social interactions, and/or functional day-to-day tasks, thereason most frequently rated “yes” was health benefits(See Figure 1). Specific potential health benefits identifiedby respondents included pressure relief, increased circula-tion, improved bone density, improved bowel and bladderfunction, reduced risk of orthostatic hypotension and gen-eral benefits associated with standing and walking.Stakeholders were asked to agree or disagree (1-Strongly disagree to 5-Strongly agree) on three additionalstatements about exoskeleton use. Wheelchair usersagreed with two statements significantly more thanhealthcare professionals: “A powered exoskeleton is agood idea” (Chi-Square = 14.885, p = 0.005) and “ITable 1 Demographics of stakeholder groupsWheelchair Users (n = 354) Frequency PercentGender- Male 194 54.8%- Female 160 45.2%Age18-24 17 4.8%25-34 59 16.7%35-44 70 19.8%45-54 93 26.3%55-64 72 20.3%65 and above 43 12.1%Country- Canada 197 55.6%- United States 129 36.4%- Other 27 7.6%Diagnosis- SCI (paraplegia) 130 36.7%- SCI (quadriplegia) 87 24.6%- MS 30 8.5%- CP 24 6.8%- Muscular Dystrophy 19 5.4%- Post-polio 13 3.7%- Congenital SCI 12 3.4%- Stroke 10 2.8%- Other 32 9.0%Hours per day using a wheelchair0–4 hours 35 9.9%5–8 hours 40 11.3%9–12 hours 86 24.3%12+ hours 193 54.5%Previous use of an exoskeletonNo 328 95.6%Yes 15 4.4%Healthcare Professionals (n = 127) Frequency PercentGender- Male 44 34.6%- Female 83 65.4%Country- Canada 76 59.8%- United States 41 32.3%- Other 10 7.9%ProfessionOccupational Therapist 25 19.7%Physiotherapist 21 16.5%Equipment vendor 13 10.3%Wolff et al. Journal of NeuroEngineering and Rehabilitation 2014, 11:169 Page 4 of 10http://www.jneuroengrehab.com/content/11/1/169would use/recommend an exoskeleton” (Chi-Square =31.316 p = 0.001), although the majority from bothgroups had agreement with both statements. Conversely,more healthcare professionals thought users would feelself-conscious using the device in public compared towheelchair users (Chi-Square = 35.067 p = 0.001).Design featuresTable 1 Demographics of stakeholder groups (Continued)Nurse 9 7.1%Support staff 8 6.3%Rehabilitation assistant 7 5.5%Rehabilitation engineer 7 5.5%Clinic director/manager 6 4.7%Assistive technology specialist* 5 3.9%Research professional 5 3.9%Physician 3 2.4%Orthotist 2 1.6%Other** 16 12.6%Previous use of an exoskeletonNo 108 93.1%Yes 8 6.9%Breakdown of characteristics of the 481 respondents, by stakeholder group.*This category included job titles such as seating specialist, AT provider**This category consists of specific job titles of which there were two or fewerincidences which could not be grouped into the other categories, e.g. socialworker, disability services provider.Participants ranked 17 design features on a Likert scalefrom 1 (Very Unimportant) to 5 (Very Important). De-scriptive statistics were used to illustrate the differences inimportance between the features (see Table 2). Whenconsidering all participants as one group, four of the 17potential design features were rated with a Median of 5(i.e. “very important”): minimizes risk of falling, purchasecost, comfort, and putting on/taking off the device. Appear-ance and length of training time were overall rated lowestwith a Median of 3 “neither important nor unimportant”.To help compare which design features were most im-portant across both groups, the percentage of partici-pants rating each feature as important or very importanton the Likert scale was calculated. Comfort, minimizesrisk of falling, repair and maintenance cost, and pur-chase cost was rated as important or very important bybetween 75 and 80% of all participants. Six other fea-tures were rated important by between 70 and 74% ofparticipants: range of battery life, ease of putting on andtaking off, ability to walk on uneven surfaces, portabilityof the device, amount of energy needed for use and abil-ity to carry out daily tasks while standing.When the stakeholder groups were examined separ-ately, a similar trend to the overall data was evident ineach group. However, some features showed a discrepancyin opinion between stakeholder groups. One discrepancywas when asked to identify an appropriate price rangefor a powered exoskeleton, the median reported priceby healthcare professionals was $10,000-$20,000USD,Figure 1 Reasons to use of recommend an exoskeleton. Participants wrecommend an exoskeleton for health benefits, rehabilitation purposes, soccommonly supported reason by both stakeholder groups. Error bars denotTable 2 Importance of exoskeleton design featuresExoskeletondesign featuresMeanimportanceStandarddeviationMedianimportanceMinimizes risk of falling 4.54 0.828 5Purchase cost 4.39 0.912 5Comfort 4.38 0.838 5Repair and maintenance cost 4.34 0.844 4Wolff et al. Journal of NeuroEngineering and Rehabilitation 2014, 11:169 Page 5 of 10http://www.jneuroengrehab.com/content/11/1/169Ease of putting on andtaking off the device4.25 1.033 5Range of battery life 4.23 0.859 4Ability to walk onuneven surfaces4.22 0.922 4Amount of energy 4.15 1.015 4needed for useAbility to carry out dailytasks while standing4.13 0.946 4Portability of the device 4.09 0.942 4Ability to toilet 4.05 1.071 4Ability to use to getin and out of a car3.97 1.033 4Ability to climb stairs 3.91 1.029 4Ability to use withoutarm crutches3.71 1.006 4Walking speed 3.64 0.985 4Length of training tobecome proficient3.34 1.082 3Overall appearance 3.23 1.177 3Valid N = 405Descriptive statistics used to illustrate the difference in importance betweenratings of 17 potential design features. These features were ranked byrespondents on a Likert scale from 1 – Very Unimportant, to 5 – Very Important.compared to the median reported by wheelchair users ofunder $10,000USD. An overall trend when comparingdesign features was that healthcare professionals ratedevery feature as more important than did wheelchairusers, with the exception of the ability to walk withoutarm crutches. Additionally, variance was larger for wheel-chair users than health care professionals for all import-ance questions. Figure 2 shows the relative importance ofall 17 features to both stakeholder groups.Exploratory factor analysisExploratory factor analysis found two underlying factorsthat encapsulated the 17 question items regarding theimportance of different potential features of the technol-ogy. Of the 17 items, 8 items loaded onto factor 1, and 9items loaded onto factor 2. All but two items loaded as fairor above, defined as > .45 (see Table 3). Cross-loading, de-fined as a variable which loads as > .30 on both factors[26], was evident for two items: portability and batterylife/range. However, both items loaded more strongly ontoere asked to respond “Yes” or “No” to whether they would use orial interactions, and functional tasks. Health benefits was the moste 95% confidence intervals.factor 2. Overall appearance did not load well onto eitherfactor, although it loaded more strongly onto factor 1.Factor 1 included items generally related to device char-acteristics (Factor 1 was labeled Technology Characteris-tics), whereas Factor 2 included items related to activitiesand tasks (Factor 2 was labeled Functional Activities).Overall, Technology Characteristics were rated as slightlymore important than Functional Activities (Mean = 4.078,SD = 0.689 and Mean = 3.995, SD = 0.706 respectively).Sampling adequacy was good as determined by theKMO measure (KMO = 0.903) [26]. Bartlett’s test ofsphericity indicated that correlations between items weresufficiently large for factor analysis, Chi-Square = 3577.059,p < 0.001.Independent samples Mann–Whitney U tests wereused to determine differences in perceived importanceof each factor across stakeholder groups. Importance ofTechnology Characteristics (Factor 1) and importance ofkedorrtsWolff et al. Journal of NeuroEngineering and Rehabilitation 2014, 11:169 Page 6 of 10http://www.jneuroengrehab.com/content/11/1/169Functional Activities (Factor 2) both varied significantlybetween wheelchair users and healthcare professionalsFigure 2 Importance of design features. 17 design features were ranThe percentage of respondents who identified features as either ‘4 - Imprate all features as more important than their wheelchair user counterpa(U = −4.651, p = 0.001 and U = −2.288, p = 0.022, respect-ively). In both cases the healthcare professionals ratedthese factor as more important than wheelchair users.Qualitative analysisContent analysis of the open-ended questions regardingfurther reasons to use or recommend an exoskeletonshowed consistent underlying themes both within andbetween groups (See Table 4). Response rate to the open-ended questions was 47.7% of total wheelchair user surveyrespondents and 33.9% of total healthcare professionalsurvey respondents.The themes identified among both stakeholder groupsare illustrated in Figure 3. Four major categories fromthe HAAT model were represented in the themes: Person,Activity, Context, and Assistive Technology [30]. Threecommon themes were found in both wheelchair user andhealthcare professional populations. Psychosocial Benefits(Person) was the most common theme identified byhealthcare professionals and the second most commontheme identified by wheelchair users. One participantresponded, “I’d like to stand and kiss my husband, I’d liketo meet people eye to eye again, I’d like to breathe the airup there”. The Health and Physical Benefits (Person)theme was the most prevalent theme represented in theresponses of wheelchair users and second for healthcareprofessionals. One wheelchair user noted, “The healthbenefits alone would be worth it”. A healthcare profes-sional replied that they perceived the device’s benefits toon a Likert scale from 1 – Very Unimportant, to 5 – Very Important.tant’ or ‘Very Important’ is shown. Healthcare professionals tended to. Error bars denote 95% confidence intervals.be “mostly for health and rehab”. The third theme foundin both groups was Use in Daily Life (Activity, Context)and included functional and accessibility considerations.One wheelchair user respondent noted, “More independ-ence in getting around a community not structured forwheelchair users”, and another, “Try doing the dishes,cook delicious meals for my family… walking up anddown the stairs in my own beautiful home”. One themeunique to wheelchair users was Larger Impacts (Context),i.e., using exoskeleton technology as a means of contribut-ing to development, or as a method of advocacy and visi-bility for individuals with disabilities. A theme unique tohealthcare professionals was “Client-driven” (Person) andincluded exoskeleton use because of client interest, or as amethod of motivating clients in the rehabilitation process.One healthcare professional respondent noted, “Motiv-ation during the rehab process. It would be more excitingfor a patient to use an exoskeleton during therapy to walksomewhere instead of on a treadmill, like the Lokomat orother similar devices”.Two final themes related to potential problems usingthe device. Some respondents felt that the device wouldnot work (Technology) for reasons such as impractical-ity, inefficiency, a potential for harm, and an inherentlyhigh cost that would prohibit many individuals from use.Respondents posed questions such as “I’m really curiousif you fall, what can and will go wrong?” and “Have youquadriplegia, hemiplegia, joint contractures, and lowbone density), although some users were simply not in-terested in walking in such a device. For a complete re-port of all participant responses to the open-endedquestions, please see Additional file 2.DiscussionPerspectives on exoskeletonsThis is one of the first studies to examine the perspec-tives of healthcare professionals and potential end-userson exoskeleton technology. Previous research on adop-tion of assistive technology devices in general has alsoTable 3 Exploratory factor analysisExoskeleton design features Factor 1(Technologycharacteristics)Factor 2(Functionalactivities)Purchase cost 0.778Repair and maintenance cost 0.758Comfort 0.701 −0.128Ease of putting on and takingoff the device0.694Minimizes risk of falling 0.659 −0.107Amount of energy needed for use 0.659Length of training to become proficient 0.509Wolff et al. Journal of NeuroEngineering and Rehabilitation 2014, 11:169 Page 7 of 10http://www.jneuroengrehab.com/content/11/1/169Overall appearance 0.375 −0.107Ability to climb stairs −0.212 −0.855Ability to carry out daily taskswhile standing−0.757Ability to use to get in and out of a car −0.731Ability to walk on uneven surfaces 0.110 −0.672considered the pressure sore issues that could result?”. Asecond group of respondents felt while the device mayhave merit, they personally would not use it (Person,Technology). In both groups, this was predominantlydue to the inability to use the device given their (or theirclients’) impairment (examples given included: highWalking speed −0.573Ability to toilet 0.142 −0.495Portability of the device 0.395 −0.471Range of battery life 0.310 −0.459Ability to use without arm crutches 0.171 −0.419Associations (i.e. loadings) of individual design features and the two factors (i.e.categories) are revealed through exploratory factor analysis. Higher numbersindicate a stronger association between the design feature (variable) and thefactor, where > .71 are considered excellent, >.63 are considered very good, >.55are considered good, and > .45 are considered fair. Values between −0.100 and0.100 have been excluded from this table. These loadings allow the designfeatures to be grouped into two major categories, where Factor 1 representsTechnology Characteristics and Factor 2 represents Functional Activities. Italicizedloading values indicate the factor which the design feature was grouped into.Table 4 Qualitative themesTheme*Psychosocial Benefits Roles & relationships,curiosity/interest, “cooHealth and Physical Benefits Health, pressure manUses in Daily Life Leisure, employmentLarger Impacts Research & developmClient-driven Client goals, motivatiDevice will not work Potentially harmful, inNot compatible with my impairment Hemiplegia, quadripleamputee, obesity, muThemes derived from responses to the open-ended question “Are there any other rfull copy of responses is included as an additional file.*Themes are ordered by prevalence within the qualitative responses.identified safety and cost as priorities for users [31]. Ourstudy expands these findings to conclude that thesesame features are important to both users and health-care professionals when considering exoskeletons specif-ically. These features are in line with interactionsbetween the person and their assistive technology withintheir context of use, as described by the HAAT model[30]. These considerations are important to the use andadoption of the technology [32].Two of the most considered factors in recent researchregarding exoskeletons are comfort and safety. Contem-porary studies have focused on falls risk as well as othersafety concerns of the device, such as proper fit in orderto maximize comfort and minimize pressure areas[20,21,33,34]. Safety was also identified by users in a studyby Matthews et al. [31] as a primary concern for any as-sistive technology. Within our study, concerns were raisedby respondents within the open-ended questions that thetechnology had potential for harm, both in terms of pres-sure issues and falls risk, and that wheelchairs remained asafer, more effective option. While current trials of exo-skeletons show low safety risks [20,21], these are in super-vised clinical settings with a trained therapist guarding theuser from falls. If exoskeletons are to be used for func-tional activities, this will be in less controlled environ-ments and may require some trade-off between the safetyand overall function of the devices. Additionally, deviceAssociated categoriespsychological, quality of life, independence, eye-level social interaction,l”, social, experienceagement, pain control, walking, standing, exercise, transfers, rehabilitation, functional day-to-day tasks, access, outdoor useent, visibility, advocacyon, use of available resourcesefficient, impractical, too expensive, dislike aestheticgia, low bone density, contractures, lack of arm/hand use, poor balance,scular dystrophy, uneven lower extremitieseasons you would use/recommend an exoskeleton?” using content analysis. AionithWolff et al. Journal of NeuroEngineering and Rehabilitation 2014, 11:169 Page 8 of 10http://www.jneuroengrehab.com/content/11/1/169developers may look to proactively design technology tomitigate falls in unsupervised settings.Cost was identified by users as a potential concern inprevious research examining reasons for choosing toadopt assistive technology [31]. Currently, purchasing aReWalk exoskeleton for personal use costs just under$70,000USD [35], substantially higher than the reportedacceptable cost in this study of under $20,000USD.Some survey respondents reported that they would notuse an exoskeleton due to the fact that it may cost toomuch to purchase and maintain as a personal device.Similar contextual and economic barriers were identifiedby a recent study assessing the adoption of robotics inrehabilitation; cost was one of the largest concernsraised in this study, due in part to the unknown cost-effectiveness of robotic devices [32]. Our findings alsoFigure 3 Qualitative themes. Themes derived from open-ended questWC users and 43 HCPs. Respondents could cite more than one theme wshow that stakeholders have similar concerns with boththe purchase and maintenance costs of exoskeletons.Features of the technology were grouped by explora-tory factor analysis into two separate categories of designfeatures which resonated with two components of theHAAT model, Activity and Assistive Technology [30].These factors were named Functional Activities andTechnology Characteristics. In the HAAT model, the twocomponents interact with the person and their context,providing a comprehensive understanding of how multi-faceted the user-assistive technology relationship can be.Our study results show a similar relationship was per-ceived by potential stakeholders of exoskeletons.There were some small but significant differences be-tween the importance of the two categories when com-pared by stakeholder group. Technology Characteristicswere slightly more important to health care professionals,which may be related to the current use of exoskeletonsmainly for health benefits and rehabilitation where the clin-ical setting necessitates significant involvement from thehealthcare professional. Therefore, technology characteris-tics which support rehabilitation would be necessary whenattempting to integrate exoskeletons into their practice[32]. In the current clinical context, it may be more import-ant to consider the perspectives of healthcare professionals,as they mediate most present use of exoskeletons. In future,wheelchair users’ perspectives may become more salient asthe devices move towards individual, functional use. It isalso worthwhile to acknowledge that both factors fell withinthe range of “important” to both stakeholder groups. Thiswould indicate that a multifaceted perspective on develop-ment of exoskeletons is key; stakeholders are invested bothin the design of the technology, as well as what the technol-ogy enables users to accomplish.Health benefits of standing and walking are frequentlyidentified in current literature, a perception which ap-responses using content analysis. Total n for this question was 169in an answer. Error bars denote 95% confidence intervals.pears to be mirrored in the perceptions of stakeholdersin this study [5,6]. This may be reflective of health bene-fits being the most studied benefit of exoskeletons intheir current form. It could also reflect the priorities ofusers and healthcare professionals in seeking to optimizephysical health for better long-term outcomes.Psychosocial benefits, though not well documented inthe literature, were also noted as a perceived benefit to theuse of exoskeletons. Opportunities for eye-level social inter-action, and the joy, hope, and confidence that users felt thatstanding and walking could bring them were identified byseveral wheelchair users. This shows that the potentialbenefits of standing and walking, especially outside of theclinical rehabilitation setting, can include psychosocial aswell as physical benefits. Healthcare professionals ratedrecommending an exoskeleton for psychosocial reasonsmore highly than wheelchair users, which could potentiallyrelate to the health care professional Client-Driven themeidentified here, specifically, targeting motivation and psy-chosocial benefits to accomplish physical goals.Wolff et al. Journal of NeuroEngineering and Rehabilitation 2014, 11:169 Page 9 of 10http://www.jneuroengrehab.com/content/11/1/169The potential for use as a rehabilitative device wasidentified as a further reason for use of an exoskeleton.Use of exoskeletons in rehabilitation settings for SCI andstroke populations has been portrayed positively in theliterature, however, this reason was not as frequentlystated in this study’s quantitative and qualitative resultswhen compared to health benefits and social interactions[17,19]. This may be a reflection of the narrow potentialuser population that would meet the physical require-ments to both use an exoskeleton and have the potentialto benefit from its rehabilitative effects (e.g., incompleteparaplegia).Though use of an exoskeleton for functional dailytasks was identified as a potential reason for use, it wasrated lower than others. This may be due to the currentlimitations of the technology, which includes a relativelyslow walking speed. However, some wheelchair users ap-peared to have higher expectations than are feasible withcurrent technology. This perception creates a potentialdiscord with the realistic functional benefits of using thedevice. Examples of this included a number of responsesfrom users who felt that an exoskeleton would enhanceindependence in daily life. Many stakeholders also iden-tified ease of putting on and taking off the device as animportant consideration. While these functions may belimited with the current technology, they can, neverthe-less, provide direction towards the design of desirablefeatures or functions of future exoskeletons.Implications for future developmentsIf exoskeletons are to be adopted as mainstream mobilitydevices, additional research and development is requiredto enhance the affordability, comfort, safety, and ease ofuse of exoskeletons to achieve stakeholder goals. Otherareas of attention are also surely important to stake-holders. However, to reduce participant burden in thisstudy, some, more detailed questions, were not included.Further study into areas such as specific falls preventionand/or recovery strategies, specifics of hardware and con-trol designs, and directly addressing how the device couldcontrol for issues related to spasticity, contractures, orother individual needs is indicated going forward. Manywheelchair users expressed interest in using the devices toincrease visibility and advocacy. This shows the readinessand willingness of the wheelchair user community toengage in and support development of new technology,which is invaluable for developers and researchers. Asexoskeleton development continues, it will be importantto re-evaluate and expand on stakeholder perspectives tomaximize their utility and adoption [32].Study limitationsThe study had four main limitations. Firstly, the format ofan online survey limited the sample to those individualswho had access to a computer and who were fluent inEnglish. Secondly, participants were primarily from NorthAmerica. These may have resulted in issues with howrepresentative the sample is of the broader population. Avolunteer bias may have impacted the types of responses.The voluntary nature of participation in an online surveymeans that it is likely that participants already had someinterest or opinion on exoskeleton technology. It is alsopossible that there was a social desirability bias to respondpositively towards questions about exoskeletons [36].ConclusionsAn online survey was conducted to determine stakeholderperspectives on exoskeleton technology. Wheelchair usersand health-care professionals reported that there could bepotential health, psychosocial, and functional benefits tothe use of exoskeletons. They also identified safety, pur-chase cost, maintenance costs, ease of use, and comfort asvery important when considering whether or not theywould use or recommend this type of device. Several otherfeatures were also identified as important. Features relat-ing to functional activities and characteristics of the tech-nology were both identified as important by healthcareprofessionals and wheelchair users, indicating the need toaddress both in exoskeleton research and development.Findings from this study lay groundwork for future researchinto stakeholder perspectives on exoskeleton technologies,aiming to inform the ongoing development of these devicesin a user-centred direction.Additional fileAdditional file 1: Survey questions.Additional file 2: Responses to open-ended questions.AbbreviationsBCIT: British Columbia Institute of Technology; CP: Cerebral palsy; EFA: Exploratoryfactor analysis; HAAT: Human Activity Assistive Technology; ICORD: InternationalCollaboration on Repair Discoveries; KMO: Kaiser-Meyer-Olkin measure;MD: Muscular dystrophy; MS: Multiple sclerosis; RA: Rehabilitation Assistant;RESNA: Rehabilitation Engineering and Assistive Technology Society of NorthAmerica; SCI: Spinal cord injury; UBC: University of British Columbia.Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsAll authors participated in the design of the study, development of the survey,active recruitment of participants, and revised the manuscript for content. JW &CP collected the data, performed analysis of quantitative and qualitative data, anddrafted the manuscript. All authors read and approved the final manuscript.AcknowledgmentsWe wish to thank all the participants in the study, as well as the staff at theUniversity of British Columbia Department of Occupational Science andOccupational Therapy and the British Columbia Institute of Technology fortheir support with this project. This work was supported by funding from theNatural Sciences and Engineering Research Council of Canada and the RickHansen Institute. We would also like the thank ICORD for the use of facilityspace for meetings.Paraplegia 1995, 33:409–415.17. del-Ama AJ, Koutsou AD, Moreno JC, de-los-Reyes A, Gil-Agudo A, Pons JL:Review of hybrid exoskeletons to restore gait following spinal cordinjury. J Rehabil Res Dev 2012, 49(4):497–514.18. Mehrholz J, Kugler J, Pohl M: Locomotor training for walking after spinalcord injury. Cochrane Database Syst Rev 2008, 2, CD006676.19. Mehrholz J, Elsner B, Werner C, Kugler J, Pohl M: Electromechanical-assistedtraining for walking after stroke. Cochrane Database Syst Rev 2013, 7:Wolff et al. Journal of NeuroEngineering and Rehabilitation 2014, 11:169 Page 10 of 10http://www.jneuroengrehab.com/content/11/1/16911. Ijzerman MJ: The influence of the reciprocal cable linkage in theadvanced reciprocating gait orthosis on paraplegic gait performance.Prosthetics Orthot Int 1997, 2:52–61.12. Butler PB, Major RE, Patrick JH: The technique of reciprocal walking using thehip guidance orthosis (HGO) with crutches. Prosthetics Orthot Int 1984, 8:33–38.13. Spungen AM, Asselin P, Fineberg D, Knezevic S, Pisano T, Harel NY, Agranova-Breyter I, Kornfeld S: Identification of skills and level of assistance for home/community use of an exoskeleton for persons with paraplegia: Apreliminary report. J Spinal Cord Med 2013, 36(5):506.14. Food and Drug Administration: FDA allows marketing of first wearable,motorized device that helps people with certain spinal cord injuries to walk2014. http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm402970.htm.15. National Institutes of Health: How many people use assistive devices2012. http://www.nichd.nih.gov/health/topics/rehabtech/conditioninfo/pages/people.aspx.Author details1Occupational Science and Occupational Therapy, University of BritishColumbia, (T325 - 2211 Wesbrook Mall), Vancouver (V6T 2B5), Canada.2Canada Research Chair in Rehabilitation Engineering Design, British ColumbiaInstitute of Technology, (3700 Willingdon Ave.), Burnaby (B5G 3H2), Canada.3MAKE+, British Columbia Institute of Technology, (3700 Willingdon Ave.),Burnaby (B5G 3H2), Canada. 4ICORD (International Collaboration on RepairDiscoveries), (818 West 10th Avenue), Vancouver (V5Z 1M9), Canada. 5PrincipalInvestigator, Rehabilitation Research Program, GF Strong Rehabilitation Centre,(4255 Laurel Street), Vancouver (V5Z 2G9), Canada. 6Biomedical EngineeringProgram, University of British Columbia, (2010-2332 Main Mall), Vancouver (V6T1Z4), Canada.Received: 13 September 2014 Accepted: 12 December 2014Published: 19 December 2014References1. Cooper RA, Boninger ML, Spaeth DM, Ding D, Guo S, Koontz AM, Fitzgerald SG,Cooper R, Kelleher A, Collins DM: Engineering better wheelchairs to enhancecommunity participation. IEEE Trans Neural Syst Rehab Eng 2006, 14:438–455.2. Wee J, Lysaght R: Factors affecting measures of activities andparticipation in persons with mobility impairment. Disabil Rehabil 2009,31:1633–1642.3. Hammel J, Southall K, Jutai J, Finlayson M, Kashindi G, Fok D: Evaluatinguse and outcomes of mobility technology: A multiple stakeholderanalysis. Disab Rehab Assist Technol 2013, 8(4):294–300.4. Kittel A, Di Marco A, Stewart H: Factors influencing the decision toabandon manual wheelchairs for three individuals with a spinal cordinjury. Disabil Rehabil 2002, 24(1):106–114.5. Adams MM, Hicks AL: Comparison of the effects of body-weight-supportedtreadmill training and tilt-table standing on spasticity in individuals withchronic spinal cord injury. J Spinal Cord Med 2011, 34(5):488"494.6. Arva J, Paleg G, Lange M, Lieberman J, Schmeler M, Dicianno B, Rosen L:RESNA position on the application of wheelchair standing devices.Assist Technol: Off J RESNA 2009, 21(3):161–168.7. Kressler J, Thomas CK, Sanchez J, Gant K, Ginnety K, Gonzalez H, Anderson KD,Nash MS, Field-Fote EC, m-Noga EW, Cilien DC: Understanding therapeuticbenefits of overground bionic ambulation: Exploratory case series inpersons with chronic, complete spinal cord injury. Arch Phys Med Rehabil2014, Article in press:1–14.8. Lam T, Wolfe DL, Domingo A, Eng JJ, Sproule S: Lower Limb RehabilitationFollowing Spinal Cord Injury. In Spinal Cord Injury Rehabilitation Evidence.Version 5.0. Edited by Eng JJ, Teasell RW, Miller WC, Wolfe DL, Townson AF,Hsieh JTC, Connolly SJ, Noonan VK, Loh E, McIntyre A. Vancouver: SCIRE;2014:1–74.9. Sale P, Franceschini M, Waldner A, Hesse S: Use of the robot assisted gaittherapy in rehabilitation of patients with stroke and spinal cord injury.Eur J Phys Rehab Med 2012, 48:111–121.10. Bernardi M, Canale I, Castellano LDF, Felici F, Marchetti M: The efficiency ofwalking of paraplegic patients using a reciprocating gait orthosis.16. Statistics Canada: Participation and Activity Limitation Survey 2006.http://www.statcan.gc.ca/pub/89-628-x/89-628-x2007001-eng.htm.CD006185.20. Esquenazi A, Talaty M, Packel A, Saulino M: The ReWalk powered exoskeletonto restore ambulatory function to individuals with thoracic-level motor-complete spinal cord injury. Am J Phys Med Rehab 2012, 91(11):911–920.21. Zeilig G, Weingarden H, Zwecker M, Dudkiewicz I, Bloch A, Esquenazi A: Safetyand tolerance of the ReWalk™ exoskeleton suit for ambulation by people withcomplete spinal cord injury: A pilot study. J Spinal Cord Med 2012, 35(2):96–101.22. Cowan RE, Fregly BJ, Boninger ML, Chan L, Rodgers MM, Reinkensmeyer DJ:Recent trends in assistive technology for mobility. J NeuroEng Rehab 2012, 9:20.23. Pons JL: Rehabilitation exoskeletal robotics. The promise of an emergingfield. IEEE Eng Med Biol Mag 2010, 29:57–63.24. Stone J, Fortin C, Borisoff J, Rushton PW, Mattie J, Miller W: Enhancingmobility by combining wheelchair and exoskeleton technology:Stakeholder’s perspectives. In Proceedings of the Canadian Association ofOccupational Therapists Conference. 30 May -1 June 2013. Victoria, Canada;2013:46.25. Dillman DA, Smyth JD, Christian LM: Internet, Mail, and Mixed-Mode Surveys:The Tailored Design Method. 3rd edition. Hoboken, N.J: Wiley; 2008.26. Field AP: Discovering statistics using SPSS: And sex and drugs and rock ‘n’ roll.SAGE: Los Angeles; 2009.27. Comrey AL, Lee HB: A first course in factor analysis. 2nd edition. Hillsdale:Lawrence Erlbaum Associates; 1992:488.28. Kaiser HF: An index of factorial simplicity. Psychometrika 1974, 39(1):31–36.29. Berg BL: Chapter 11: An introduction to content analysis. In Qualitativeresearch methods for the social sciences. 6th edition. Boston, MA: Allyn andBacon; 2007:238–267.30. Cook A, Hussey S: Assistive Technologies: Principles and Practice. St. Louis, MO:Mosby; 1995.31. Matthews JT, Beach SR, Downs J, Bruine de Bruin W, Mecca LP, Schulz R:Preferences and concerns for quality of life technology among older adultsand persons with disabilities: National survey results. Technol Disabil 2010,22(1):5–15.32. Turchetti G, Vitiello N, Trieste L, Romiti S, Geisler E, Micera S: Whyeffectiveness of robot-mediated neurorehabilitation does not necessarilyinfluence its adoption. IEEE Rev Biomed Eng 2014, 7:143–153.33. Arazpour M, Bani MA, Hutchins SW, Jones RK: The physiological cost indexof walking with mechanical and powered gait orthosis in patients withspinal cord injury. Spinal Cord 2013, 51(5):356–364.34. Lee H, Song WK, Sohn R, Kim R: User and expert feedback for lower-extremity robotic exercise prototype. In Proceedings of the IEEE/SICEInternational Symposium on System Integration (SII): 16–18 December 2012;Fukuoka, Japan. 2012:7–12.35. CBS News: First patient takes ReWalk robotic exoskeleton home.http://www.cbsnews.com/news/firstpatient-takes-rewalk-robotic-exoskeleton-home/.36. Krumpal I: Determinants of social desirability bias in sensitive surveys: aliterature review. Qual Quant: Int J Methodol 2013, 47(4):2025–2047.doi:10.1186/1743-0003-11-169Cite this article as: Wolff et al.: A survey of stakeholder perspectives onexoskeleton technology. Journal of NeuroEngineering and Rehabilitation2014 11:169.


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items