This article was downloaded by: [University of Boras]On: 07 October 2014, At: 16:24Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Assistive Technology: The Official Journal of RESNAPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/uaty20

Investigating the International Classification ofFunctioning, Disability, and Health (ICF) Frameworkto Capture User Needs in the Concept Stage ofRehabilitation Technology DevelopmentManoj Sivan MRCSa, Justin Gallagher MSc(Eng)b, Ray Holt PhDb, Andy Weightman PhDb,Martin Levesley PhDb & Bipin Bhakta MDa

a Academic Department of Rehabilitation Medicine, University of Leeds, Leeds, UKb School of Mechanical Engineering, University of Leeds, Leeds, UKAccepted author version posted online: 17 Mar 2014.Published online: 26 Jun 2014.

To cite this article: Manoj Sivan MRCS, Justin Gallagher MSc(Eng), Ray Holt PhD, Andy Weightman PhD, Martin Levesley PhD& Bipin Bhakta MD (2014) Investigating the International Classification of Functioning, Disability, and Health (ICF) Frameworkto Capture User Needs in the Concept Stage of Rehabilitation Technology Development, Assistive Technology: The OfficialJournal of RESNA, 26:3, 164-173, DOI: 10.1080/10400435.2014.903315

To link to this article: http://dx.doi.org/10.1080/10400435.2014.903315

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor and Francis shall not be liable forany losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use ofthe Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Assistive Technology® (2014) 26, 164–173Copyright © 2014 RESNAISSN: 1040-0435 print / 1949-3614 onlineDOI: 10.1080/10400435.2014.903315

Investigating the International Classification of Functioning,Disability, and Health (ICF) Framework to Capture User Needs inthe Concept Stage of Rehabilitation Technology Development

MANOJ SIVAN, MRCS1∗, JUSTIN GALLAGHER, MSc(Eng)2, RAY HOLT, PhD2, ANDY WEIGHTMAN, PhD2,MARTIN LEVESLEY, PhD2, and BIPIN BHAKTA, MD1

1Academic Department of Rehabilitation Medicine, University of Leeds, Leeds, UK2School of Mechanical Engineering, University of Leeds, Leeds, UK

This study evaluates whether the International Classification of Functioning, Disability, and Health (ICF) framework provides a useful basisto ensure that key user needs are identified in the development of a home-based arm rehabilitation system for stroke patients. Using a quali-tative approach, nine people with residual arm weakness after stroke and six healthcare professionals with expertise in stroke rehabilitationwere enrolled in the user-centered design process. They were asked, through semi-structured interviews, to define the needs and specifica-tion for a potential home-based rehabilitation device to facilitate self-managed arm exercise. The topic list for the interviews was derived bybrainstorming ideas within the clinical and engineering multidisciplinary research team based on previous experience and existing literaturein user-centered design. Meaningful concepts were extracted from questions and responses of these interviews. These concepts obtainedwere matched to the categories within the ICF comprehensive core set for stroke using ICF linking rules. Most of the concepts extracted fromthe interviews matched to the existing ICF Core Set categories. Person factors like gender, age, interest, compliance, motivation, choice, andconvenience that might determine device usability are yet to be categorized within the ICF comprehensive core set. The results suggest thatthe categories of the comprehensive ICF Core Set for stroke provide a useful basis for structuring interviews to identify most users needs.However some personal factors (related to end users and healthcare professionals) need to be considered in addition to the ICF categories.

Keywords: robot, stroke, user involvement, user-centered design

Introduction

Stroke is the leading cause of adult onset disability with esti-mates of annual incidence of new strokes in Europe reportedto be between 200 and 300 per 100,000 populations every year(SPREAD, 2003) with around 110,000 new strokes in Englandevery year. Up to 85% of survivors experience some degree ofparesis of the upper limb at the onset (Skilbeck, Wade, Hewer,& Wood, 1983) and only 50% of survivors regain functionaluse of the affected upper limb in spite of therapeutic inter-vention (Ada, Canning, Carr, Kilbreath, & Shepherd, 1994).Motor learning principles which underpin effective arm rehabili-tation interventions include intensity of practice undertaken in anengaging/motivating environment. Novel technologies offer thepossibility of complementing conventional rehabilitation inter-ventions by augmenting intensity of practice. There are a numberof clinical environment based robotic devices (Massachusetts

∗Address correspondence to: Dr. Manoj Sivan, AcademicDepartment of Rehabilitation Medicine, Level D, Martin Wing,Great George Street, Leeds General Infirmary, Leeds, LS1 3EX,UK. Email: M.Sivan@leeds.ac.ukColor versions of one or more of the figures in the article can befound online at http://www.tandfonline.com/uaty.

Institute of Technology [MIT-Manus], Massachusetts, USA;Mirror Image Movement Enabler [MIME], Richmond, VA, USA;Gentle, University of Reading, Reading, UK) and home-userobotic devices (TheraJoy, TheraDrive) that have been developedto assist upper arm training (Johnson, Feng, Johnson, & Winters,2007; Riener, Nef, & Colombo, 2005).

The success of any medical device depends on how wellit matches the purpose intended by the healthcare professionaland the needs and expectations of the individual being investi-gated or treated. This makes the perspectives of both individu-als (end-users) and healthcare professionals (professional users)paramount in the development process of medical device tech-nology. Involvement of users in design development has beenshown to be associated with higher market usability (Gould &Lewis, 1985), improved equipment safety and efficiency (Lin,1998), higher chances of successful user use (Fouladinejad &Roberts, 1996) and overall reduction in development time andcosts (Giuntini, 2000; McDonagh, Bruseberg, & Haslam, 2002).This has also led to increased regulatory requirement of userinvolvement in device development (Powers & Greenberg, 1999).King (1999) also emphasized that the adoption and success ofassistive technology depends on the aspirations, psychologicalneeds and preferences of the intended user, not simply on theirmechanical needs.

Dow

nloa

ded

by [

Uni

vers

ity o

f B

oras

] at

16:

24 0

7 O

ctob

er 2

014

Investigation of ICF Framework 165

The involvement of users can be considered at various stagesof device development from the stage of idea generation to thefinal stage of market deployment. Current literature supports theconcept that user needs should drive product development, nottechnology and commercial pressure and that users should beinvolved in all stages of device development (Bridgelal Ram,Campling, Grocott, & Weir, 2008). Models of medical devicelifecycle stages have been described by Cooper and Kleinschmidt(1986; 13 stages) Rochford and Rudelius (1997; 12 stages), andWorld Health Organization (WHO; 2003; 7 stages). A recent,concise model proposed by Shah, Robinson, and AlShawi (2009)comprised four stages in the rehabilitation assistive technologydevelopment process: concept, design, testing, and deployment.Early involvement of users in the concept and design stages canfacilitate the development of technology with improved usability,avoiding modification costs and time in comparison to involvingusers later in the device lifecycle.

International Organization for Standardization 9241-210(2010) set out the international standard for user-centered design,and specified six principles for user-centered design: that it isbased on explicit understanding of users, tasks and environments;that users are involved throughout development; that design isdriven by user-centered evaluation; that the process is iterative;that the design must address the whole user experience; and thatthe design team should include multidisciplinary skills and per-spectives. It also set out an iterative process for user-centereddesign, which involves understanding and specifying the contextof use; specifying user requirements; producing design solutions;evaluating the design against the user requirements and iteratingas necessary until the user requirements are satisfied.

There are a variety of direct and indirect methods usedto involve users throughout the device development process.Usability tests, interviews and questionnaire surveys are the mostcommonly used methods across all the stages of device devel-opment (Shah & Robinson, 2006). The methods that can beparticularly used in the concept stage are brainstorming sessions,ethnography, user meetings, interviews, focus groups, and semi-nars (Shah et al., 2009). In user-centered design literature, thereare currently no sufficiently detailed models described that canhelp researchers understand their target population’s needs in theconcept stage.

The Cambridge design exclusion calculator was developedbased on a disability follow-up survey of 1996–1997 (Grundy,Ahlburg, Ali, Breeze, & Sloggett, 1999) and is used to give anestimate of the number of people in a population who wouldbe excluded by a particular design and helps researchers todevelop more inclusive designs (Waller, Langdon, & Clarkson,2008). The calculator is based on seven impairments people withdisabilities might have: locomotion, reach, dexterity, vision, hear-ing, communication, and intellectual functions. The model mightbe used to help design technology for stroke patients but theabove impairments do not provide enough depth to understandstroke patients. This model does not take into account extendedfunctional activities, personal factors like sex, race, ethnicity,interests, motivation, and environmental factors such as therapyresources, carers and vocation. Cook and Hussey’s human activ-ity assistive technology (HAAT) model also provides a processfor structuring consideration of user’s needs in prescribing anddesigning assistive technology to place them in the context of a

Fig. 1. International Classification of Functioning Disability andHealth (ICF) model.

given activity, environment and level of ability (Cook & Polgar,2008).

One framework that is commonly used in health sciences tounderstand and capture the different aspects of any health con-dition is the WHO International Classification of Functioning,Disability, and Health (ICF; Figure 1). The ICF classifies healthcondition into domains of body structure/functions, activities,participation, personal and environmental factors related to theindividual (WHO, 2001). The ICF aims to undertake a holisticapproach to understand the different aspects of the health con-dition in an individual and providing a mechanism to ensureappropriately targeted healthcare intervention. This frameworkis internationally accepted and used extensively in research.Researchers have developed disease specific ICF Core Sets forspecific health conditions such as rheumatoid arthritis, multiplesclerosis and stroke. The comprehensive ICF Core Set categoriesfor stroke put forward by an international consensus group iswidely used in stroke related research (Geyh et al., 2004).

We have investigated the usefulness of the stroke specific com-prehensive ICF Core Set to guide researchers in understandinguser needs in the concept stage of developing restorative reha-bilitation technologies for people with stroke. As far as we areaware, such use of ICF framework in technology development,has not been described in literature so far.

Methods

The study had approval from National Research EthicsCommittee and local research and development departments.

Sample

People with stroke who had arm movement difficulties and whowere attending local NHS stroke services were recruited to thisstudy. Healthcare professionals (physiotherapists and occupa-tional therapists) involved in the care of individuals with strokewere identified from local stroke services. Written informedconsent was obtained from all participants.

The users in the context of this qualitative study wereconsidered in two groups: end-users and professional users.In the “end-user” group we included a broad range of people

Dow

nloa

ded

by [

Uni

vers

ity o

f B

oras

] at

16:

24 0

7 O

ctob

er 2

014

166 Sivan et al.

with stroke and arm weakness (including those with commu-nication difficulties, visual impairments, mobility problems andvarying degrees of arm weakness) and their carers. In the profes-sional user group, we included physiotherapists and occupationaltherapists experienced in assessing and treating people with armweakness after stroke.

Stage 1: Interview Process

All participants were offered a mutually convenient time to attendfor face to face interviews undertaken by one of the authors(M. S.). The aim of interviews was to understand users’ perspec-tives, expectations of technology use (to provide arm exercises)in a home setting and the potential barriers to such technology.The research team (two clinicians [M. S. and B. B.]) and fourengineers [J. G., A. W., R. H., and M. L.]) initially identifiedthe broad interview topics by brainstorming ideas and discussionin research team meetings. The topics were based on previousexperience in user-centered design process (Holt et al., 2007;Weightman et al., 2010) and existing literature on developmentof assistive technology for arm rehabilitation (Egglestone et al.,2009; Lu et al., 2011). The nature of the interviews was semi-structured and was a combination of open-ended and close-endedquestions based on the interview checklist of topics prepared bythe research team. The nature of the interviews allowed the dis-cussion to deviate from these topics to those that were identifiedby the end user’s relevant to the technology design.

For the interview format and content of questions, patients andtheir carers were considered as the “end-user” group (AppendixI: Patient Interview Checklist; see supplemental material online).Physiotherapists and occupational therapists were considered as“healthcare professionals” (Appendix II: Healthcare ProfessionalInterview Checklist; see supplemental material online). Theywere asked questions on intensity and type of arm therapythat patients receive after stroke, home arm exercises, func-tional goals, role of technology to provide arm exercises,Information Technology (IT), computer skills, perceptions onhome-based technology, and comparison to hands-on conven-tional physiotherapy.

Stage 2: Extracting Interview Concepts

ICF linking rules have been developed to link health measuresand interventions to the ICF framework. These rules were ini-tially published in 2002 (Cieza et al., 2002) and later updatedin 2005 (Cieza et al., 2005). The authors suggested identifyingmeaningful concepts within items and responses of measures andlinking to most precise ICF category. Meaningful “concepts” arethose that describe the health condition, person, functional activ-ity or any of the environmental factors. For example, considerthe statement/item, “Pain doesn’t prevent me from walking anydistance.” Two different meaningful concepts can be identified inthis statement, “pain” and “walking.”

Meaningful concepts referring to “quality of life” are assigned“not definable-quality of life.” If a meaningful concept is not con-tained in the ICF and is clearly a personal factor, it is assigned“personal factor.” If a meaningful concept is not contained in ICFand is not a personal factor, it is assigned “not covered.” If themeaningful concept refers to a diagnosis or a health condition, itis assigned “health condition” (Cieza, et al., 2005).

Based on the above linking rules, meaningful concepts wereextracted from our interview topic questions and responses(Table 1).

Stage 3: Matching Interview Concepts to the ComprehensiveICF Core Set for Stroke

The interview concepts, which resulted from the semi-structuredinterviews were matched to the categories within the ICF com-prehensive core set for stroke. The core set has 130 categories intotal, which comprise of 46 categories from body function andstructure domain, 51 categories from activities, and participationdomain and 33 from environmental factors (Geyh et al., 2004).The personal factors domain has no categories yet.

We explain the process of extracting interview concepts andmatching them to ICF categories as proposed by Cieza et al.(2005) with the following example from our interviews.

1. Participants were asked, “Describe home arm exercises you do(end-users)/prescribe (professional-users).” The two mean-ingful concepts which can be extracted from the questionare—“home setting” and “arm exercises.”

2. A sample end-user response was, “I have private physicaltherapy at home performing squeezing exercise with affectedhand, wheel hand cycling, and working with weights.” Asample professional-user response was, “Lack of motivationand cognitive problems could contribute to poor compli-ance with home exercises.” The meaningful concepts thatcan be extracted from above two responses are—“privatetherapy,” “fine hand skills,” “stretching exercises,” “strength-ening exercises,” “motivation,” “cognitive problems,” and“compliance.”

3. The extracted concepts are linked to one or more of the com-prehensive ICF Core Set categories, which convey the samemeaning:• Home setting—e210 “Physical geography.”• Private therapy—e580 “Health services, systems, and

policies.”• Cognitive problems—b110 “Consciousness functions,”

b114 “Orientation functions and others.”

Results

Nine people with stroke and residual arm weakness, and six expe-rienced neuro-rehabilitation physiotherapists and occupationaltherapists were enrolled in the user-centered design process.We will refer to the individuals with stroke as end-users in thisarticle.

Participants (End-users)

• The time since stroke varied from 1 year to 3 years• Five participants had left side weakness and four had right side

weakness.• Five participants had weakness in their non-dominant arm and

four in their dominant arm.• Five participants had problems with speech and language, four

of them had word finding difficulty (expressive dysphasia) andone had problems with articulation (dysarthria).

Dow

nloa

ded

by [

Uni

vers

ity o

f B

oras

] at

16:

24 0

7 O

ctob

er 2

014

Investigation of ICF Framework 167

• All participants had problems with weakness in the affectedarm, five of them had stiffness and three of them had problemswith sensation in the affected arm.

• Four participants had experienced problems with vision (fielddefects) after stroke, which had improved by the time ofrecruitment. One participant had visual inattention on theaffected side.

• Three participants had mild to moderate pain in the affectedarm; one of them had previously received an injection inthe shoulder area in the past for pain and all three were onanalgesic medication for the pain. Pain limited the range ofmovements in their arm.

• One participant had cognitive problems in areas of short-termmemory since stroke.

• Two participants reported on-going problems with their moodsince the stroke.

• None of the participants were in employment.

Healthcare Professionals

• The six healthcare professionals enrolled in the user-centereddesign process all had at least five years experience of workingwith individuals with stroke.

• Four of them were physiotherapists and two were occupationaltherapists.

• Four of them provided therapy in both inpatient and out-patient settings. Two therapists provide therapy in a communitysetting.

• All of them were involved in providing and advising home-based arm exercise programs for their patients with stroke.

Stage 1: Interview results

The interview topic results were grouped as comments made byend-user group and healthcare professional group. Responses toclosed questions and similar comments made by more than oneindividual are reported in third person in this paper. Individualcomments in response to open questions are reported as directquotes.

Therapy Received After Stroke

End-user group:

• All nine participants felt that there was need for continuingtherapy at home once inpatient and outpatient therapy finished.

• Participants on-average received 3–4 sessions 45-minphysiotherapy sessions per week while they were inpatientsafter stroke. Following discharge from hospital outpatientor community based rehabilitation therapy varied from onesession per week to four times a week and continued up to12 weeks.

• All participants felt they would benefit from additional therapyfor their arm movement difficulties:

“I felt I needed more therapy for arm but did not receive itlong term.”

“There was more emphasis on walking (lower limb function)than upper limb exercises in initial therapy after stroke.”

“National Health Service (NHS) resources are limited andwaiting time for out-patient therapy is long.”

“I wish to perform more arm exercises and want to improvearm function for better performance in daily activities andbetter quality of life.”

Professional user group:

• All six healthcare professionals agreed that there was a gapbetween acute/sub-acute therapy and long-term therapy afterstroke in current NHS.

“Waiting list for out-patient therapy could be up to 6 monthsfrom the time of accepting a referral.”

• Three months to 2 years after stroke was reported as the timeperiod when maximal recovery of arm function was likely tooccur, although all indicated that there is continued recoveryeven beyond this period depending on intensity and type oftherapy received by patients.

“There is scope for change up to 5 years post-stroke.”

“Additional therapy would result in improved recovery offunctional arm movement.”

Type of Arm Exercise

Professional user group:

• All healthcare professionals commented their therapy planis personalized to individual patient goals. The interventionsare targeted at maintaining pain free passive range of armmovement and improving arm weakness.

“Intensity of community therapy is based on the initial assess-ment, they can be considered in three categories: high need(therapy thrice in a week), moderate need (twice in a week)and low (once a week), the sessions are one hour each.”

“Distraction and stimulation techniques are used for visualsensory inattention or neglect.”

• Patients are advised to exercise within their pain free range ofmovement.

Home Exercises

End-user group:

• All participants had been given home-based arm exercises bytheir therapists, which involved passive stretching exercisesand active involvement in daily activities of living.

“I use hand beads and baking for fine hand control exercises.”

• Seven participants were no longer practicing home exerciseson a regular basis

“I did not continue doing home exercises after few months aslost motivation.”

Dow

nloa

ded

by [

Uni

vers

ity o

f B

oras

] at

16:

24 0

7 O

ctob

er 2

014

168 Sivan et al.

“I have private physical therapy at home performing squeezingexercise with affected hand, wheel hand cycling and workingwith weights.”

• Two participants occasionally went to gym and worked withcross-trainer exercise machines and lifting weights.

Professional user group:

• Prescribed home exercises were based on functional dailyactivities goals in discussion with patients.

“The exercise program includes stretching and sensitisationexercises as well as trunk stabilization and balance exercisesto optimize arm movement.”

“Digital photos of exercises are given to patients to remindthem how to do the exercises.”

• Patients varied in their compliance with home exercises.

“A compliance rate of one-third of recommended amount ofexercises is acceptable.”

“Lack of motivation and cognitive problems after stroke couldcontribute to poor compliance.”

“I recommend patients to do arm exercises (on top of dailyactivities) for at least 15 min everyday.”

Functional Activity Goals

End-user group:

• All participants did not use the affected arm in daily activitiesas much as before stroke. They had developed compensatorystrategies of performing daily tasks.

“I lift kettle with my unaffected arm now.”

“I use the affected arm only to hold things in place while theunaffected arm does most of the activity.”

• The participants wanted to improve their ability to use the armin daily activities such as combing hair, washing, dressing,cooking, and eating with knife and fork.

“I want to improve my writing.”

“I want my affected arm to be less tight while doing activities.”

“I want to get back to swimming and drive a manual car[currently driving an automatic car].”

Professional user group:

• The healthcare professionals directed therapy based on indi-vidual functional activity goals and encouraged patients to usetheir affected arm as much as possible in daily activities and asearly as possible after stroke.

“Therapy is tailored to patients needs.”

Use of Technology at Home for Arm Therapy

End-user group:

• Two participants have used the Wii video game console forentertainment and therapy at home.

“I use my unaffected arm to operate the Wii remote device asthe affected arm does not have sufficient strength to operatethe device. I play golf, tennis and bowling with my son.”

“I like the competitive element of Wii games.”

Professional user group:

• The healthcare professionals sometimes recommend use of Wiifor arm therapy in selected patients.

“It depends on the type of arm impairment and availability ofequipment. It can be engaging and can augment the intensityof arm therapy.”

“Computer gaming exercise may be an adjunct to conventionaltreatments if chosen appropriately and based on individualclinical assessments.”

Information Technology (IT) Skills and Computer Games

End-user group:

• All participants had either used computers in past or currentlyuse them on a regular basis. They described their IT skills asbasic and used the computer for Web browsing, shopping, andemails.

• Seven participants had laptop computers, one had a desktopcomputer, two had both (a laptop and a desktop computer).One participant did not have a computer at home.

• Seven participants’ family members or carers had basic com-puter skills and could use computers for internet web browsing,shopping, and email purposes.

• Three participants have played computer games, such as cardgames, street games, formula one car racing games, darts, andChinese checkers with their family at some point.

“I would like to play computer games based on golf, bowling,and tennis.”

“Games based on Sudoku or shopping would be interesting.”

“Games based on space (asteroids) or word building(Scrabble) will be good.”

Professional user group:

• The healthcare professionals felt computer games might keepsome patients interested.

“Not all would find using a computer easy and interest-ing, especially the elderly patients and those with cognitiveproblems.”

“The nature of games that will engage patients will varyamong individuals depending on sex, vision problems, inter-ests, and previous experience.”

Dow

nloa

ded

by [

Uni

vers

ity o

f B

oras

] at

16:

24 0

7 O

ctob

er 2

014

Investigation of ICF Framework 169

Individual Perceptions on Role of Arm RehabilitationTechnology in Home Setting

End-user group:

• All participants believed that using a home-based technologyaimed at arm exercises would help them perform more armexercises. They felt it would give them more independence intheir rehabilitation program and motivate them to engage morein the exercise program.

“It can improve one’s concentration and thinking ability.”

“It can improve hand-eye coordination and fine skills.”

“It would have been ideal if it were available straightaway afterstroke, when I was discharged from hospital.”

“It might give me a purpose to get out of bed and use thedevice every day.”

• Seven of the participants preferred to use the technology on anindividual basis at home.

“I prefer using it on my own as feel I would become con-scious of my problems when using such devices in front ofother people.”

“Competition is good but not with able-bodied people.”

“I do not like stroke clubs and resource centers.”

• Two participants preferred to also use it in a multi-user set-ting with other individuals. They were interested in the ideaof a collaborative approach where one could play with anotherindividual remotely via an Internet connection.

“Using such technology in hospital setting would benefitpatients early after stroke.”

Professional user group:

• All healthcare professionals supported the idea of having tech-nology that could help individuals perform arm exercises intheir homes.

“Such technology could improve thinking ability andcognition.”

• Five healthcare professionals felt patients would prefer usingsuch technology both on an individual basis and in multi-userapproach in community centers and stroke clubs.

“Patients are generally motivated in groups but some patientscould become self-conscious and threatened in a group.”

“Younger people like competition.”

“ Elderly patients might not be keen on technology.”

“There might be some potential difficulties installing thedevice in community centers like maintenance of device,

transport facilities for patients to get to the centers andinability to access the device when facilities were closed.”

Comparison of Arm Rehabilitation Technology to ConventionalHands-on Physiotherapy

End-user group:

• All participants stated that technology would provide themwith additional therapy to their physiotherapy and would ben-efit the eventual recovery of their arm function. They statedthat home-based technology could increase their ability toindependently perform arm exercises in their free time.

“Home technology might save time going to hospitals andwould help me undertake therapy when my child (2 years old)is asleep.”

“Using computer games could increase my concentration andthinking ability that has been affected by stroke.”

“I am unsure if playing computer games could be used toprovide useful exercises for arm recovery.”

“I would not know whether the correct exercises were beingundertaken using the technology as they would not be super-vision by professionals.”

Professional user group:

• All healthcare professionals felt technology can augmenthands-on physiotherapy after stroke.

“It might provide intensity to the exercise program andencourage patients to perform exercises when not under super-vision.”

“It might empower the patient with increasing independence.”

“It can increase engagement and have positive impact oncognitive function.”

“It might be cheaper, convenient and flexible (in terms oftaking breaks within sessions when tired).”

• Four healthcare professionals stated a major drawback wouldbe the lack of ability to monitor patient performance andquality of movements undertaken.

“It might be difficult to correlate device movements to thefunctional relevant daily activity movements.”

“One concern is the huge costs involved in developing suchtechnology and whether it is cost-effective in the long run.”

Expectations From the Home-based Arm Rehabilitation Device

End-user group:

• Seven participants wanted the device to be simple and easy touse with its footprint suitable for installation in a home setting.

Dow

nloa

ded

by [

Uni

vers

ity o

f B

oras

] at

16:

24 0

7 O

ctob

er 2

014

170 Sivan et al.

• Participants wanted the device to be deployable in a livingroom (3 participants), kitchen (2 participants), or bedroom(4 participants). All 4 participants’ bedrooms were first-floorrooms with a staircase as the sole route to get to the rooms.

“It should be a tidy piece of equipment.”

“It has to be easily movable and portable.”

“The device must be safe especially electrical faults.”

“Hope there will be access to engineers to fix technical issuesas they arise.”

Professional user group:

• All healthcare professionals felt the device should be safe andeasy to set up and use. They suggested the device should becompact and easily installable in home setting.

• Two healthcare professionals stated the system should haveoptions to individualize exercise programs depending onpatient deficits and needs.

“The technology must be simple, appealing and motivating forthe patients.”

“The computer tasks must be meaningful and functionally rel-evant based on the principles of motor relearning. The deviceshould help maintain the range of movement and improvestrength in weak arm.”

“I like the idea of ability of device to provide assistance whenuser is unable to complete task with affected arm.”

“It would be difficult to engage some elderly people who arenot used to computers and games.”

“Hygiene issues must be considered while designing anddelivering the device to people’s homes.”

“Patients must have access to engineers and healthcare profes-sionals who have knowledge about the technology.”

Stage 2: Extracting Interview Concepts

Meaningful concepts extracted from each interview topic ques-tions and responses/discussions with end-users and healthcareprofessionals are listed in Table 1. In case of duplication ofconcepts, they are listed only once in the table.

Stage 3: Matching Interview Themes to ICF Core Set forStroke

Concepts linked to the most relevant ICF Core Set category orcategories is shown in Table 2. Two concepts were assigned asrelated to “health condition,” 14 concepts were assigned to “per-sonal factor,” and 1 concept was assigned “not covered.”

Discussion

The process of matching the concept stage interview conceptswith the Comprehensive ICF Core Set categories indicated thatmost concepts were covered within the ICF Core Set categories.Some interview concepts relate to the personal factors domainin ICF, but there are no “personal factor” categories described inthe Comprehensive ICF Core Set. The overlap of topics can berepresented in a schematic diagram (Figure 2).

Table 1. Meaningful concepts extracted from interviews.

Interview topic Concepts emerging from discussions

Demographic characteristics Age, gender, type/time of stroke, side of weakness, dominant side, speech and languageskills, vision, stiffness, weakness, sensation, pain, cognitive problems, mood, andemployment

Therapy and exercises Duration of exercises, intensity of exercises, walking, upper limb specific exercises, timesince stroke, National Health Services (NHS) resources, daily activities, quality of life,and personal goals

Home exercises Home setting, private therapy, fine hand skills, stretching exercises, strengthening exercises,endurance exercises, sensitization exercises, balance, compliance, motivation, andcognitive problems

Functional activity goals Use of affected arm, writing, daily activities, combing hair, washing, dressing, cooking,eating, swimming, and driving

Home technology Playing games, engagement, intensity of therapy, personal choice, and leisure interestsIT skills and computer games Computer use, owning a computer, playing games, cognitive skills, gender, vision, interest,

and experience,Individual perception Concentration, thinking, coordination, fine skills, time since stroke, purpose in life,

competition, motivation, social life, age, and community resourcesComparison between technology and

hands-on physiotherapyTherapy principles, cognitive skills, independence, supervision by professionals, cost, and

maintenanceExpectations from home-based device External look of device, expectations of users/professionals/friends/family carers,

motivation, assistance, safety, hygiene, engagement, and meaningful exercises

Dow

nloa

ded

by [

Uni

vers

ity o

f B

oras

] at

16:

24 0

7 O

ctob

er 2

014

Investigation of ICF Framework 171

Table 2. Interview concepts and matched ICF Core Set categories.

Interview topic ICF code

Age and gender Personal factorType/time of stroke Health conditionSide of weakness Health conditionDominant side Personal factorSpeech and language problems b167, b330Weakness b730Stiffness b735Involuntary movements b755Sensory function problems b260, b265, b270Pain b280Vision problems b210Inattention b140Memory problems b144Cognition problems b164Mood problems b152Employment d850, d855Therapy/exercises d210, d220, d230, e580Duration of exercises b455Walking d450Upper limb exercise d440, d445Daily activities d230, d430, d510, d520,

d530, d540, d550National Health Service (NHS)

servicesd580

Quality of life Not definableIntensity of exercises b740Fine hand skills b440Home setting e210Private therapy e580Strengthening exercise b730Stretching exercise b710Sensitization exercise b270Endurance b740Balance b770Compliance Personal factorMotivation Personal factorCognitive problems b110, b114, b117, b140,

b144, b152, b156,b164, b167, b172, b176

Usage of affected arm Not coveredWashing d510, d530Dressing d540Cooking d630Eating d550Combing hair d520Writing d170Swimming d920Driving d475Playing games using technology d210, d220Engagement b140Intensity of therapy b740Personal choice Personal factorLeisure interests d920Computer usage e115Owning a computer e165Gender Personal factorInterest Personal factor

(Continued)

Table 2. (Continued)

Interview topic ICF code

Experience Personal factorConcentration b140Thinking ability b164Coordination b755Purpose in life Personal factorCompetition Personal factorMotivation Personal factorSocial life e325Age Personal factorCommunity health resources e575, e580Independence Personal factorSupervision by professionals e355, e360Cost e165Convenience Personal factorExternal look of device e150, e155, personal

factorExpectations of friends, family, staff e410, e420, e425, e440,

e450, e455Engagement Personal factorAssistance e115Hygiene e150, e155Safety e150, e155

Fig. 2. Concepts in user-centered design process.

This implies that the ICF framework or Core Set for strokecan be used by researchers as a tool to understand the criticalproblems or needs of stroke survivors. This will help in the devel-opment of an inclusive technology that meets these needs andcaters to the target population. Some examples from each ICFdomain are discussed below.

Some trivial but important “body function” factors like side ofstroke can be crucial in informing the design of the system. Onehas to consider this aspect in designing a home based exercisetechnology so that it can be adjusted to right arm or left arm use

Dow

nloa

ded

by [

Uni

vers

ity o

f B

oras

] at

16:

24 0

7 O

ctob

er 2

014

172 Sivan et al.

and make it efficient in terms of the utility of the technology.Understanding visual inattention and field defects is importantto design an adjustable monitor screen to suit individuals withspecific deficits. Keeping the display and user login instructionssimple is important to enable those with language and cogni-tion problems use the technology. The importance of pain andassessing the impact it has on technology usage is importantfor researchers to be able to advise the individual on deviceusage time and enable tailor-made therapy with range of pain-freemovement.

Understanding the ICF “activity and participation” factorslike functional daily activities will help researchers designgames/tasks which replicate those desired activities and makethe technology more meaningful and functionally relevant to theuser. ICF “environmental factors,” such as home environment,physical space available for device, appearance of device, carersupport, perceptions, and attitudes of people (including profes-sionals) around the individual, have a major influence on theindividual’s progress and usage of technology. Catering to theseneeds is likely to increase acceptability and usage.

Finally ICF “personal factors” are arguably the most impor-tant factors that can determine the success of any technology.These include individual’s perception of the technology, self-efficacy, and belief in therapy, computer skills, motivation, inter-est, experience and liking for technology. These factors coulddetermine how much the technology will appeal to/motivatethe individual and determine the extent of engagement of theindividual.

There are some limitations to this study. First, a relativelysmall sample of end-users and healthcare professionals wereincluded for the interviews. There is no established literature inuser-centered design as to what the minimum sample size shouldbe in such a process. The sample in this study was adequateto give researchers a flavor of user needs and expectations andto understand the recommendations of skilled healthcare pro-fessionals. Secondly, meaningful concepts are ideally linked tothe most precise third-level ICF category as per the ICF linkingrules (Cieza et al., 2005) but we have linked them only to thesecond-level category available in the comprehensive ICF coreset. The aim of this study was only to explore whether ICF pro-vides a useful template for user-feedback process and not to testthe accuracy of the linking process. Hence, the linking to theavailable categories is justified.

In summary, the ICF is a useful framework for researchers inthis area to explore the needs of users and understand their expec-tations from a rehabilitation technology. The comprehensive coreset provides categories within the main ICF domains that couldbe used as a useful template when planning the concept stageuser involvement process. Personal factors however are yet tobe categorized in the core set and will need to be included inthe concepts to be explored. We recommend the comprehensiveICF Core Set be used by researchers for thorough assessment andunderstanding of user needs. Involving users early in technologydevelopment process will lead to improved design of technologywhich will lead to greater likelihood of usability, adherence totreatment protocols both within clinical trials and routine use andsubsequent adoption of technology by health services.

This approach can be explored further in future research. It isalready established that users need to be involved in all stages of

device development: concept, design, testing, and deployment.It should be possible to use the ICF framework in a similar fash-ion to other stages of device development as well and could bethe subject for future studies in this area. Also, we have studiedthe ICF disease-specific framework for stroke in this study, thiscan be extended to any other chronic condition and technologyused to help individuals with that condition. This approach canbe extended to situations where chronic conditions coexist forexample brain and spinal cord injury where technology has to beinformed by needs in both health conditions.

Acknowledgements

The authors would like to thank all the stroke survivors and ther-apists who participated in this study. We are also grateful to Dr.Rory O’Connor who reviewed the manuscript.

Supplemental Material

Supplemental data for this article can be accessed on the pub-lisher’s website.

References

Ada, L., Canning, C. G., Carr, J. H., Kilbreath, S. L., & Shepherd,R. B. (1994). Task specific training of reaching and manipulation.Amsterdam: Elsevier Science.

Bridgelal Ram, M., Campling (nee Browne), N., Grocott, P., & Weir,H. (2008). A methodology for a structured survey of the healthcareliterature related to medical device users. Evaluation, 14(1), 49–73.

Cieza, A., Brockow, T., Ewert, T., Amman, E., Kollerits, B., Chatterji, S.,. . . Stucki, G. (2002). Linking health-status measurements to the inter-national classification of functioning, disability and health. Journalof Rehabilitation Medicine: Official Journal of the UEMS EuropeanBoard of Physical and Rehabilitation Medicine, 34(5), 205–210.

Cieza, A., Geyh, S., Chatterji, S., Kostanjsek, N., Ustun, B., & Stucki,G. (2005). ICF linking rules: An update based on lessons learned.Journal of Rehabilitation Medicine: Official Journal of the UEMSEuropean Board of Physical and Rehabilitation Medicine, 37(4),212–218. doi:10.1080/16501970510040263

Cook, A. M., & Polgar, J. (2008). Cook and Hussey’s assistive technology:Principles and practices (3rd ed.). St. Louis, MO: Mosby.

Cooper, R. G., & Kleinschmidt, E. J. (1986). An investigation into the newproduct process: steps, deficiencies, and impact. Journal of ProductInnovation Management, 3(2), 71–85.

Egglestone, S. R., Axelrod, L., Nind, T., Turk, R., Wilkinson, A., Burridge,J., . . . Rodden, T. (2009, April). A design framework for a home-basedstroke rehabilitation system: Identifying the key components. Paper pre-sented at the 3rd International Conference on Pervasive ComputingTechnologies for Healthcare, London, England.

Fouladinejad, F., & Roberts, J. R. (1996, October). A computer aided learn-ing package for training users and maintainers of medical equipment.Paper presented at the Proceedings of the 18th Annual InternationalConference of the IEEE, “Bridging Disciplines for Biomedicine,”Amsterdam, the Netherlands.

Geyh, S., Cieza, A., Schouten, J., Dickson, H., Frommelt, P., Omar,Z., . . . Stucki, G. (2004). ICF core sets for stroke. Journal ofRehabilitation Medicine: Official Journal of the UEMS EuropeanBoard of Physical and Rehabilitation Medicine 36(44), 135–141.doi:10.1080/16501960410016776

Giuntini, R. E. (2000). Developing safe, reliable medical devices. MedicalDevice and Diagnostic Industry, 22, 60–67.

Dow

nloa

ded

by [

Uni

vers

ity o

f B

oras

] at

16:

24 0

7 O

ctob

er 2

014

Investigation of ICF Framework 173

Gould, J. D., & Lewis, C. (1985). Designing for usability: key principles andwhat designers think. Communications of the ACM , 28, 300–331.

Grundy, E., Ahlburg, D., Ali, M., Breeze, E., & Sloggett, A. (1999).Research report 94: Disability in Great Britain. London, England:Corporate Document Services.

Holt, R., Makower, S., Jackson, A., Culmer, P., Levesley, M., Richardson,R., . . . Bhakta, B. (2007, June). User involvement in developingRehabilitation Robotic devices: An essential requirement. Paper pre-sented at the 10th International Conference on Rehabilitation Robotics,Noordwijk, the Netherlands.

International Organization for Standardization. (2010). Ergonomics ofhuman-system Interaction Part 210: Human-centred design for inter-active systems (ISO 9241-210). Geneva, Switzerland: Author.

Johnson, M. J., Feng, X., Johnson, L. M., & Winters, J. M. (2007). Potentialof a suite of robot/computer-assisted motivating systems for personal-ized, home-based, stroke rehabilitation. Journal of NeuroEngineeringand Rehabilitation, 4, 6. Retrieved from http://www.jneuroengrehab.com/content/4/1/6.

King, T. W. (1999). Assistive technology: Essential human factors. Boston:Allyn & Bacon.

Lin, L. (1998, October). Human error in patient-controlled analgesia:Incident reports and experimental evaluation. Paper presented at the42nd Annual Meeting of the Human Factors and Ergonomics Society,Chicago, IL.

Lu, E. C., Wang, R. H., Hebert, D., Boger, J., Galea, M. P., & Mihailidis, A.(2011). The development of an upper limb stroke rehabilitation robot:Identification of clinical practices and design requirements through asurvey of therapists. [Comparative Study Research Support, Non-U.S.Gov’t]. Disability and rehabilitation. Assistive Technology, 6, 420–431.doi:10.3109/17483107.2010.544370

McDonagh, D., Bruseberg, A., & Haslam, C. (2002). Visual product evalua-tion: exploring users “emotional relationships with products.” AppliedErgonomics, 33, 231–240.

Powers, D. M., & Greenberg, N. (1999). Development and use of ana-lytical quality specifications in the in vitro diagnostics medical

device industry. Scandinavian Journal of Clinical and LaboratoryInvestigation, 59, 539–543.

Riener, R., Nef, T., & Colombo, G. (2005). Robot-aided neurorehabilita-tion of the upper extremities. Medical & Biological Engineering andComputing, 43(1), 2–10.

Rochford, L., & Rudelius, W. (1997). New product development process:Stages and successes in the medical products industry. IndustrialMarketing Management, 26(1), 67–84.

Shah, S. G., & Robinson, I. (2006). User involvement in healthcaretechnology development and assessment: Structured litera-ture review. International Journal of Health Care QualityAssurance Incorporating Leadership in Health Services, 19,500–515.

Shah, S. G., Robinson, I., & AlShawi, S. (2009). Developing medicaldevice technologies from users’ perspectives: A theoretical frame-work for involving users in the development process. InternationalJournal of Technology Assessment in Health Care, 25, 514–521.doi:10.1017/S0266462309990328

Skilbeck, C. E., Wade, D. T., Hewer, R. L., & Wood, V. A. (1983). Recoveryafter stroke. Journal of Neurology, Neurosurgery & Psychiatry, 46(1),5–8.

Stroke Prevention and Educational Awareness Diffusion (SPREAD). (2003).The Italian guidelines for stroke prevention and treatment. Milano,Italy: Hyperphar Group.

Waller, S., Langdon, P. M., & Clarkson, P. J. (2008). Converting disabilitydata into a format suitable for estimating design exclusion. London,England: Springer-Verlag.

Weightman, A. P. H., Preston, N., Holt, R., Allsop, M., Levesley, M., &Bhakta, B. (2010). Engaging children in healthcare technology design:developing rehabilitation technology for children with cerebral palsy.Journal of Engineering Design, 21, 579–600.

World Health Organization. (2001). The international classification of func-tioning, disability and health—ICF.

World Health Organization. (2003). Medical device regulations: Globaloverview and guiding principles.

Dow

nloa

ded

by [

Uni

vers

ity o

f B

oras

] at

16:

24 0

7 O

ctob

er 2

014