i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 7 8 ( 2 0 0 9 ) 788–801

journa l homepage: www. int l .e lsev ierhea l th .com/ journa ls / i jmi

Elderly persons’ perception and acceptance of usingwireless sensor networks to assist healthcare

Robert Steelea, Amanda Loa,∗, Chris Secombeb, Yuk Kuen Wongc

a Discipline of Health Informatics, Faculty of Health Sciences, The University of Sydney, Australiab Department of Computer Systems, Faculty of Information Technology, University of Technology Sydney, Australiac School of Management, Griffith University, Australia

a r t i c l e i n f o

Article history:

Received 3 March 2009

Received in revised form

26 July 2009

Accepted 2 August 2009

Keywords:

Elderly

Healthcare

Health services for the aged

a b s t r a c t

Purpose: This is an exploratory study carrying out qualitative research into the perceptions,

attitudes and concerns of elderly persons towards wireless sensor network (WSN) tech-

nologies in terms of their application to healthcare. This work aims to provide guidance

on the dimensions and items that may be included in the development of a more in-depth

questionnaire to further validate the importance of the identified factors as well as the rela-

tionships between them. This study aims to contribute to opening up a communication

channel between users and researchers, informing the research community in relation to

applications and functionalities that users deem as either desirable, inadequate or in need

of further development.

Methods: Focus groups were conducted with elderly individuals who were still living inde-

pendently. To explore elderly persons’ perceptions and thoughts on current wireless sensor

network (WSN) technologies and designs, discussion points were designed from concepts

identified from various user acceptance theories and models. Participants were given an

introduction to explain the functionality and capabilities of WSN and motes and were shown

a sample mote, the Crossbow Mica2Dot. Participants were then asked to discuss their per-

ceptions and concerns towards the likelihood of using a WSN-based healthcare system in

their home.

Findings: We have identified sixteen concepts in relation to the elderly participants’ percep-

tion, concerns and attitudes towards WSN systems. Those concepts were further classified

into six themes describing the determinants that may affect an elderly person’s acceptance

of WSNs for assisting healthcare. Some of our exploratory findings in this study indicate

for example that independence is highly valued by elderly people and hence any system or

technology that can prolong that independence tends to be highly regarded, that privacy of

WSN health data might not be as important as typically considered, and there are also indi-

cations that cost may be the most prominent determinant influencing an elderly person’s

acceptance of WSNs.

Conclusions: Our findings indicate that participants’ attitudes towards the idea of wireless

sensor networks for health monitoring are generally positive. The exploratory findings along

with the literature sugges

design and model buildin

Crown Copy

∗ Corresponding author at: University of Sydney, PO Box 170, Lidcombe,E-mail address: Amanda.lo@usyd.edu.au (A. Lo).

1386-5056/$ – see front matter. Crown Copyright © 2009 Published by Edoi:10.1016/j.ijmedinf.2009.08.001

t a number of relationships which can be used in future survey

g.

right © 2009 Published by Elsevier Ireland Ltd. All rights reserved.

NSW 1825, Australia. Tel.: +61 2 9036 7476; fax: +61 2 9351 9672.

lsevier Ireland Ltd. All rights reserved.

a l i n

1

T(tematWctcpisbfatt

ptftrifatt

ocFkteoarstabfmfcwaci

2

Tt

i n t e r n a t i o n a l j o u r n a l o f m e d i c

. Introduction

here is considerable potential for wireless sensor networkWSN)-based systems to provide an effective, scalable solu-ion for assisting elderly care. As such systems do not requirextensive hardware or infrastructure, they can be imple-ented and utilized within a home environment to prolong

n elderly person’s time at home and provide them withhe freedom to remain independent for as long as possible.

hile there is a plethora of studies addressing the techni-al improvement of wireless sensor network (WSN) healthechnologies, there is a lack of literature that explores the per-eptions of elderly persons towards such technology and theirotential acceptance or otherwise of it. For any new system, it

s important to find out the determinants of its acceptance asystems negatively perceived by the users are more likely toe rejected. This is especially important for systems designedor the elderly demographic, as most elderly people are notccustomed to modern technology. Therefore any preconcep-ions or biases they may hold towards a technology are likelyo hinder its acceptance and adoption.

This research examined the problem of whether the elderlyopulation perceives WSN technologies as a viable solutionhat can prolong their independence and if so, to identify theactors that affect an elderly person’s acceptance and expec-ations of WSN technologies. Some initial findings have beeneported in [1], and this paper reports more fully on the find-ngs and transcript responses, defines themes and conceptsrom those responses, and provides in-depth interpretationsnd discussions on the findings. This study adds insights intohe perceptions, attitudes and concerns of elderly personsowards WSN technologies for use in assisting healthcare.

A thorough literature review has found that there is a lackf empirical studies which address an elderly person’s per-eption, attitude and concerns towards WSN technologies.indings from this exploratory study aim to begin to fill thisnowledge gap. Furthermore, this study was conducted withhe goal of informing future WSN researchers and develop-rs of factors that may affect the acceptance and adoptionf WSNs. Themes and concepts affecting an elderly person’scceptance of WSNs are derived based on the participants’esponses. It is believed that future more extensive qualitativetudies can leverage the findings of this study to rigorously fur-her validate the importance of the identified factors as wells the relationships between them. Relationships identifiedetween those concepts and themes can then form a basisor future model building and provide indications on where

odel-based qualitative investigations are most likely to beruitful. This study aims to provide information on the per-eptions of elderly persons towards current WSN designs asell as to facilitate a communication channel between usersnd researchers, informing the research community on appli-ations and functionalities that users deem as either desirable,nadequate or in need of further development.

. Background

he demographic shift towards an aging population is causinghe demand for aged care to rise, putting more pressure on the

f o r m a t i c s 7 8 ( 2 0 0 9 ) 788–801 789

public and private healthcare systems. It has been predictedthat by the year 2051, Australians aged 65 and above will makeup 25.8% of the total Australian population [2], therefore theprovision of effective and efficient elderly care is increasinglycritical. Aging can lead to a steady decline in an individual’scognitive, physical and sensory functions [3–6] and these func-tional declines can result in tasks seen as simple to the youngbecoming difficult for an elderly person. This can result in theneed for them to be moved into retirement homes or otherinstitutional care facilities as they lack the ability to effec-tively and safely look after themselves. Previous studies haveshown that many elderly people prefer to live at home [7,8],but doubts upon their ability to look after themselves safelyoften prevent them from doing so.

WSN is a proposed technology to assist in this area whichhas some promise in providing an effective and scalableapproach to respond to an increasing demand for home-basedaged care and the independence of elderly persons. WSNs areusually made up of a series of sensor nodes, which are alsoknown as motes. Each mote has the capability to facilitatethe collection of a range of environmental and structural sen-sory information such as weight, blood sugar, blood oxygenlevel, ECG information, EEG information, sound, temperature,humidity, light-intensity, vibration and acceleration. Alongwith their relatively small size, as well as capability to auto-mate the collection of a wide range of sensory information,motes are suited to be used within the healthcare context todiscreetly and continuously collect an individual’s vital infor-mation. Information collected by a mote can be relayed back toa data or computing system, which can be remotely or locallylocated, for further analysis. The gathering of this aggregateddata would allow caregivers to provide healthcare servicesthat are better tailored to the specific needs of an individ-ual as well as allowing them the freedom to be away from anelderly individual while still being able to detect some healthproblems, thus potentially maximizing the time an elderlyperson may remain at home without affecting their indepen-dence or quality of life. This can potentially reduce the amountof healthcare facilities required, allow the provision of betterpreventive treatments and therefore reduce the total cost ofhealthcare.

Currently, motes are quite expensive as they are a develop-ing technology. A Crossbow Mica2Dot mote with basic sensorsaverages to around $100US each [9]. Motes are typically aroundthe size of an Australian 10c coin (as shown in Fig. 1), with alarger sized base station that they report back to, which canthen relay information onwards to a computer network or callcentre.

There are many studies discussing the algorithms involvedin maintaining the privacy and security of WSNs [10–12], butless has been done to determine to what extent users desiretheir functionalities. While it is crucial to ensure the privacyand security of the data communicated within any health-care system, a highly secured system is likely to be rejectedby users if its does not fulfill users’ expectations. Thereforethe importance of the other factors contributing to the accep-

tance of WSNs should not be neglected. It is important toinvestigate users’ needs and requirements before the designof any new systems such that true user needs can be satis-fied, allowing those systems to have a better chance in gaining

790 i n t e r n a t i o n a l j o u r n a l o f m e d i c a l

Fig. 1 – A Crossbow Mica2Dot mote in comparison with a

10c coin.

user acceptance and adoption [13]. This is especially essen-tial for systems designed for the elderly population as theyare a specialized demographic with perceived tendencies toadopt more slowly or reject technology. For this reason, feed-back from the elderly population must be sought, obtained andplanned for in WSN designs such that their implementationcan be more likely accepted.

Several models have been proposed in order to study andunderstand user acceptance behavior and factors, the mostnotable among which are Fishbein and Azjen’s Theory of Rea-soned Action (TRA) [14], Davis’ TAM model [15], and the UnifiedTheory of Acceptance and Use of Technology (UTAUT) modeldeveloped by Venkatesh et al [16]. TRA is a theory developedbased on social psychology which can be used in the predictionof a wide range of behaviors. According to TRA, individual’sbehavioral intention is dependent on that individual’s atti-tude about the behavior and the subjective norm. TAM is aspecific application of the TRA theory specifically used for thedetermination of user acceptance behaviors for introducedtechnologies. It states that actual system use is governedby two main factors, perceived ease of use (PEOU) and per-ceived usefulness (PU), and these two constructs can directlyaffect an individual’s attitude and intention towards using asystem.

Similar to TRA and TAM, UTAUT also aims to provide anexplanation on a user’s intention to use an information sys-tem. UTAUT was developed based on an empirical comparisonof eight acceptance models [16], and as such, is able to providea more unified view on user acceptance than its eight prede-cessors. UTAUT describes an individual’s behavioral intentionto use a system as well as their usage behavior is depen-dent on four main constructs: Performance Expectancy, EffortExpectancy, Social Influence and Facilitating Conditions [16].As UTAUT is a unified model aggregating constructs fromeight different acceptance models, its constructs hold signif-icant resemblance to those that have been identified in othermodels. Acceptance determinants described by the above-mentioned theories and models have been proven to affecta user’s acceptance of a technological system, therefore those

constructs are deemed to be viable starting points for the pur-pose of this study.

Several studies have attempted to determine an elderlyperson’s acceptance of healthcare-related systems, such as

i n f o r m a t i c s 7 8 ( 2 0 0 9 ) 788–801

Demiris et al.’s [17] work in determining older adults’ attitudeand perception towards the WSN infrastructure implementedat TigerPlace. TigerPlace is an “aging-in-place” initiativeundertaken as a joint collaboration between The SinclairSchool of Nursing at the University of Missouri-Columbia andThe Americare Corporation. It is an implemented “Smart-home” solution that utilizes a range of sensors to facilitatethe automation of tasks such as lighting control and the open-ing of doors and window [17]. Apart from home automation,sensors are also utilized to facilitate the continuous collectionof an elderly individual’s health status [17,18]. It was reportedthat participants from the TigerPlace project demonstrated anoverall positive attitude towards the technology, and believedit can be used to improve their everyday lives [19]. Howevera similar study with other groups of older adults may notyield similar findings as these participants were selected froma group of well-educated, upper middle class, retired elderlypeople who are more familiar and accustomed to technol-ogy [19]. Moreover, as those participants were residents atTigerPlace, therefore the findings are mostly depicting users’attitude and perception towards TigerPlace and may not berepresentative more broadly of WSN systems. Thus, to deter-mine whether an elderly person is willing to accept and adoptWSN technologies, we have conducted this exploratory studyto determine the major themes and concepts that may influ-ence an elderly person’s perception towards WSNs. These canthen provide guidance as to the dimensions to be includedin future empirical study design as input into model develop-ment.

3. Methods

As this work is an exploratory study with the aim to determinefactors affecting an elderly user’s acceptance of WSN-basedsystems. The findings provide insights into the perceptions,attitudes and concerns that an elderly person may havetowards WSN systems. The three steps of the methodology are(i) determining relevant concepts from the existing literature;(ii) conducting focus groups; and (iii) data analysis.

3.1. Literature review

This study is conducted as a result of knowledge gaps identi-fied in the existing literature. It was found that while there isan abundance of studies discussing the technical issues relat-ing to WSN technologies, users’ perceptions, especially thoseof elderly users, are usually not explored. Review of previousliterature has revealed that factors such as an elderly person’sperception of technology [20,21], financial status [21], as wellas their willingness to learn and accept technology [22] canaffect their perception and willingness to use a technology. Astudy on various acceptance models (TRA, TAM and UTAUT)has also provided an insight into the major universal factorsthat are likely to affect a user’s perception towards a technol-ogy. Therefore to determine whether an elderly person is likely

to accept WSN-based systems, we have designed our focusgroup questions based on the significant issues and concernsdescribed in previous studies as well as various acceptancetheories and models.

a l i n

qEwUiEi“vettouibo

3

IstteSfaiett

ajnsaiasapbbdp

EsgpFaaTcsc

i n t e r n a t i o n a l j o u r n a l o f m e d i c

The major factors that formed the basis of our focus groupuestions are PEOU (Effort Expectancy), PU (Performancexpectancy), Social Influence, Behavioral Usage Intention asell as users’ concerns. It should be pointed out that severalTAUT constructs hold significant similarities to constructs

dentified in the TAM model. For instance, “Performancexpectancy” is similar to PU, and “Effort Expectancy” is sim-lar to PEOU. “Social Influence” can be seen as derived fromSubjective Norm” as defined in TRA. As this study aims to pro-ide an insight into the pre-implementation concerns that anlderly person may have as well as providing information onheir expectations for a WSN-based homecare system, inves-igation on the effects of “Facilitating Conditions” has beenmitted mainly due to the fact that WSN technologies are stillnder development. We believe that the impact of “Facilitat-

ng Conditions” on the acceptance of WSN systems may note accurately determined until users have a better knowledger experience in interacting with a WSN homecare system.

.2. Conducting focus groups

n order to account for an extensive range of opinions repre-enting elderly persons’ perceptions, attitudes and concernsowards WSN technologies, participants who volunteered toake part in this study were randomly selected from severallderly community groups, golf clubs and churches withinydney, Australia. An effort was made to even the mix ofemale and male participants. All participants were 65 or oldernd resided in Sydney. To ensure that none of the views regard-ng aged care accommodations are affected by pre-existingxperiences at a nursing home or aged hostel, all elderly par-icipants in this study were living independently at the timehis study took place.

The focus groups were conducted with institutional ethicspproval and informed consent was obtained from all sub-ects. Two focus groups were conducted. The first focus groupot only acted as a forum for ideas, but also a pilot for theecond group. The purpose of the pilot group is to ensure thatll unexpected outcomes are accounted for as well as allow-ng the removal or modification of any ambiguous questions. Itlso gave an indication on how the structure of the focus grouphould be modified such that the quality of the commentsnd ideas might accurately depict how an elderly person mayerceive WSN-based technologies. It was emphasized at theeginning of each group that participants’ anonymity woulde preserved; responses and opinions made would be confi-ential and only be used for research purposes. None of thearticipants objected to the recording of the discussion.

A total of 13 elderly participants took part in this study.ach group went for approximately 2 h. The first group con-isted of three female and two male participants; the secondroup was conducted 2 weeks later, and had eight partici-ants with an equal number of male and female participants.ocus groups were chosen as it allows for the free-flow of ideasnd comments from participants, allowing the moderator tosk questions when the responses are unclear or ambiguous.

his fits the goals of an exploratory study. In order to dis-over aspects which may not have been identified in previoustudies, the elderly participants were first introduced to theoncept of WSNs for the monitoring of an elderly person’s

f o r m a t i c s 7 8 ( 2 0 0 9 ) 788–801 791

health, and were given the chance to first raise issues thatthey foresaw on their own. Particular issues identified fromthe existing literature that were not raised by the focus groupswere brought up by the facilitator and discussed.

Both focus groups were conducted by Secombe. A Cross-bow Mica2Dot mote (Fig. 1) was shown to each participantto demonstrate what a mote might look like as well as itsweight and size. Both groups were given the same introductionexplaining the concept of introducing sensors to assist elderlycare. A description of the mote’s functionality and capabilityis also given.

3.3. Data analysis

The transcripts were analyzed, discussed and the findingscross-validated with a professional caregiver with over 25years experience to help identify key concepts. Any sugges-tions or indications of relationships between the concepts thatthe domain expert suggested were also noted. The transcriptswere analyzed to allow for the identification of commonconcepts and themes. This process included looking for fre-quently raised issues to identify concepts. Concepts depictingsimilar ideas were then grouped to allow for the identificationof broad themes that may play an important role in determin-ing an elderly person’s acceptance of WSN technologies. Dueto the nature as well as the size of the dataset, it was felt thatit would be more accurate to organize, analyze and track thefindings by hand rather than by software.

4. Findings

Analysis of the transcripts led to the identification of a numberof concepts that potentially affect elderly users’ acceptanceof WSNs. This section introduces the themes and conceptsidentified and the relationships between them.

4.1. Themes and concepts

Six major themes were identified from the elderly partici-pants’ comments (i) independence; (ii) perceived impact onthe quality of life; (iii) concerns associated with WSNs; (iv)user’s personal preferences; (v) design preferences; and (vi)external factors (see Table 1).

4.1.1. Independence4.1.1.1. Perception on nursing homes. Many of the elderlyparticipants had negative views on nursing homes. Most sug-gested that they would prefer a caregiver to come and visitor assist them with their household chores, commenting thatthey would only move into a nursed care facility if there wereno other choices.

. . .you can have outside people like nurses coming in to yourhome regularly. . . I don’t think anyone actually wants to go intoa nursing home unless you’re really not all there. . .they are all

forced. . .

However, several participants pointed out that nursinghomes may have positive effects on some elderly individualsdue to the increase in social contact.

792 i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 7 8 ( 2 0 0 9 ) 788–801

Table 1 – Summary of themes and concepts identified.

Theme Concept

Independence Perception on nursing homesThe significance of independence to an elderly individual

Perceived impact on the quality of life Perceived benefitsPotential changes in lifestyle

Concerns associated with WSNs CostSocial implicationAdherenceHealthConfidentiality/privacyAnxietySystem reliability

User’s personal preferences Technology self-efficacyWillingness to be trained

Design preferences Implementation typeWearable sensorsAmbient monitoringEmbedded sensors

Functionalities

External factors

. . .there are people, who have gone to nursing homes and thrivedbecause they have company. . .

4.1.1.2. Significance of independence to an elderly person. Allparticipants believed that independence is of paramountimportance; all of them prefer to remain at home for as longas possible. Each participant was asked to rate the impor-tance of continuing to live at home on a 10-point scale, froma minimum score of 1 (unimportant) to a maximum of 10(extremely important). Responses were given to the group andnot privately or in isolation. An average from each group wascalculated and one group rated independence as 9 out of 10while the other group rated it as 8 out of 10.

I would choose home, I think most people would. . .Nobody choosesto go to a nursing home. . .

4.1.2. Perceived impact on the quality of life4.1.2.1. Perceived benefits. While most participants acknowl-edge the benefits WSNs could provide, most of them onlyconsider WSNs as a “safety-net” system that provides assis-tance when emergency occurs. When asked about how theyperceive WSN can assist them after a brief description of itsfunctionalities a participant commented:

Well you might not get to the phone if you’ve fallen over andbroken a leg. . .

Other participants noted that even if an elderly individualhas a caregiver who visits them daily, they would still needto spend most of the day alone and WSN technologies canensure that if accidents happen while the caregiver is away,emergency help can be provided promptly.

Even if you had a human carer, who comes by one hour a day,there’s twenty-three hours where they are not there. The humancarer still only comes in to bathe you or wash you and then they’regone. This can provide you some level of safety. It is something

Supports available

that is always there, and it connects you to someone, if you fallover. And most elderly people do fall over. That to me, is impor-tant.

However it is evident that the functionalities and benefitsof WSN systems is a concept some elderly persons find hardto grasp. Most participants tend to relate WSNs to the con-ventional panic alarm systems such as VitalCall [23], whichrequires users to wear a necklace or wrist pendant, wherethey can be connected to an emergency response centre onthe push of a button. As one participant stated:

Well what advantage does it have over the VitalCall – that’s whatI don’t understand. Just that you don’t have to do something?That’s the only advantage – that it activates itself?

4.1.2.2. Potential changes in lifestyle. Participants generallyagreed that social wellbeing is an important aspect of theirlives which significantly contributes to their overall health. Ingeneral, the elderly participants do not perceive WSNs or sys-tems of any type to impose any changes on their social livesand stated that their social trends and habits would continuewith or without the system.

Why would that change when they’ve got this [WSN]? Theirlifestyle wouldn’t change just because it [WSN] is there. If theywere always inside before, they’ll still always be inside.

All participants strongly opposed and gave a resounding“No” to the suggestion that social aspects, such as notifyingtheir family of the lack of social activities, should be incorpo-rated into the design of WSNs. They perceive this as somethingthat should be left for them to determine and participants

expressed strong resentment at any attempt to dictate thoseterms to them.

No, I don’t think it needs that [social aspect integrated], that’s apersonal choice.

a l i n

44ecfbancichc

lao

4aorppa

4atdps

asid

4i

i n t e r n a t i o n a l j o u r n a l o f m e d i c

.1.3. Concerns associated with WSNs

.1.3.1. Cost. Cost was the most significant concern to thelderly participants; it was a frequently recurring theme andoncern in both focus groups, and is the most likely topicor participants to refer back to regardless of what issue waseing discussed. Several participants have raised the questions to whether the number of people being monitored by theetwork has an effect on its implementation and utilizationost. Most participants were also concerned about the costn maintaining a WSN system at home. Some participantsommented that there needs to be evidence demonstratingow WSNs can offer a saving in other areas before they wouldonsider getting it. Most participants view WSNs as an “extra.”

I think it would have to be relatively cheap before I’d consider it,and if it gives a saving in other areas. . . a lot of people in myage group are on fixed incomes, so they can’t afford it, and it’s anextra. It may not be included in a budget.

Several participants also stated that a WSN system is mostikely something that their children would want them to have,nd would only consider using the system if “the governmentr my children pay for it.”

I think the children would want them to have it, but I’m not sureif people would order it for themselves.

.1.3.2. Social implication. Existing panic alarm systems suchs Lifeline [24] and VitalCall [23] require users to wear a neckr wrist pendant, and will connect them to an emergencyesponse centre on the push of a button. Responses on theseanic alarm systems are generally negative, with a number ofarticipants stating either themselves, or a friend they knowre ashamed of wearing it.

I know someone with a panic alarm, and she won’t wear it out-side. If she’s out the back hanging out the washing and she falls,there’s nothing she can do until someone finds her. So now she’sjust decided that she won’t go outside, so she practically lives inher kitchen.

.1.3.3. Adherence. Most participants stated the need to wearsensor device is a potential issue as they may not adhere

o the system purely because they have forgotten to wear theevice. Some have also stated occasions when they have pur-osely refused to wear a medical device such as a hearing aidimply because they did not want to wear it that day.

I can choose to wear it or not wear it [hearing-aid] and when Iwant to rebel, I’ll just say ‘I don’t think I’m going to wear it today!’and I won’t wear it.

There were participants, who considered monitoring askind of “surveillance,” stated their reluctance to have the

ystem monitoring them all the time and wondered if themplementation would allow them to “get rid of it” whenesired.

So if you got sick of it, sick of the surveillance, could you. . . throw

it in a bucket of water?

.1.3.4. Health. There were concerns regarding whether hav-ng a WSN system at home can negatively impact a person’s

f o r m a t i c s 7 8 ( 2 0 0 9 ) 788–801 793

health. Some participants were aware of the fact that thesensors communicate with each other wirelessly and had con-cerns on whether those waves may cause cancer. Questionsabout whether embedded sensors may cause allergic reactionswere also raised. A few participants showed concerns regard-ing to the pain they may experience if a sensor is required tobe planted underneath their skin.

Could the sensor radio waves give you cancer? I think this is whatI would be worried about.

4.1.3.5. Confidentiality/privacy. A majority of the participantsshowed no concern about having their health informationtransmitted wirelessly. One participant even stated, “My Life isan open book” and do not perceive any concerns regarding theconfidentiality of his health information as long as a camerais not used to monitor him. Participants place their preferenceon ensuring that help is readily available over privacy or theconfidentiality of their health information.

When you get old you don’t really care about privacy, you justwant to know someone will come to help you.

4.1.3.6. Anxiety. Several participants have stated that theyare apprehensive towards technology and demonstrated clearconcern about the possibility of having to interact with a WSNsystem. The majority of the participants emphasized on thefact that computer systems are hard to understand due totheir old age. Many have even stated that they are “terrifiedof a computer.”

Technology has no brains. And that’s the scary thing about it.Because you have to be so precise using it and it can’t make up. . .

if you make the wrong choice, the wrong button. . .

4.1.3.7. System reliability. Few participants were interested onwhether the system is reliable and accurate. Several of themwere concerned about what would happen when the batteryruns out or how they will be notified when the system mal-functions.

What is the expected life of the machine? I know one chap whowas on a pacemaker and he’s having trouble at present, and heknows he’s faced some symptoms, and that his battery may runflat. . . if those batteries last 1 year, 2 years, whatever it is, it mightbe running down at the vital time. So how would you know it isable to run down?

Others were concerned about the accuracy of the system,questioning whether the system can diagnose illnesses oraccidents and accurately determine the elderly person’s needsto the same level as doctors or caregivers.

I think machines can’t diagnose to the same level as people, I meansometimes you might only need the ambulance to come and getyou back up on your feet and then call your next of kin, you mightnot need to go to hospital or anything. . .

Another participant also stated similar concern:

Is there some way of cancelling it [alert sent from WSN]? Forexample if you’re in the shower and you bend over to pick up yoursoap and it thought you’d fallen– there could be false alarms. . .

i c a l

794 i n t e r n a t i o n a l j o u r n a l o f m e d

and I don’t want it sending for the ambulance if I’ve only bumpedmy knee.

4.1.4. User’s personal preferences4.1.4.1. Technology self-efficacy. Similar to computer self-efficacy, technology Self-efficacy refers to an individual’sperception of his or her ability to interact and use a computersystem or technology and is comprised of four core compo-nents: previous experiences with technology, observation ofother’s experience, verbal persuasion and affective arousal[25].

Our findings show that the majority of the participantshave had at least one negative experience with technology,and several of them cited previous incidents when they ortheir friends have unsuccessfully interacted with technology.

I had a similar problem with my DVD player recently, when some-one came around to put it in for us they showed us how to use it,but we didn’t get anything written down and the book of instruc-tions we can’t understand. We’ve finally figured out how to usemost of it. It can be learnt, but it does strike fear.

Others have also stated their desire to have a simple tounderstand system, commenting:

I think the system should be as simple as possible. After all themore functions this thing does. . . the harder it will be to learn.

4.1.4.2. Willingness to be trained. It is apparent that the capa-bility to interact with the system is a significant concern tothe participants; with many of them showing a strong desireto be able to interact successfully with the system. In order tofind out how to improve the usability of WSN systems for anelderly person, the participants were ask to make several sug-gestions on what can be provided to make it easier for themto interact with the WSN system. Many have stated the desirefor a demonstration or training to be given so that they would“understand how to use the thing [WSN system].” Others also com-mented on the fact that a manual should be provided so thatthey can refer back to it when they cannot remember whatwas taught during the training.

Training, I think it is essential to have some sort of training. Itshould be part of the installation process. They should teach youwhile they install it.

It was also pointed out that the manual should be free ofjargon and easy to understand.

I think the manual should be written by someone who was edu-cated in Australia – not one of those foreign manuals that has allthe strange ways of putting things.

4.1.5. Design preferences4.1.5.1. Implementation type. Currently there are three mainapproaches to implementing sensor networks; wearable sen-sors, ambient monitoring and embedded sensors. A briefexplanation was given to the participants describing how the

sensors could be placed differently under different imple-mentation approaches, the participants were then asked tocomment on which implementation approach they prefer andpoint out the perceived flaws.

i n f o r m a t i c s 7 8 ( 2 0 0 9 ) 788–801

4.1.5.1.1. Wearable sensors. General attitudes towardsthe idea of wearable sensors were quite apprehensive withparticipants showing more enthusiasm towards the idea ofembedding sensors into clothing accessories such as a ring ora watch. Participants generally related wearable sensors to ahearing aid and indicated a willingness to wear a sensor deviceif it proved to be a practical solution, although some were waryof the fact that they may not be able to remember to wearthe sensors on them all the time. Most participants initiallyresponded that they do not mind the concept of wearing sen-sors on their body; however several questioned how they canstill be monitored when they are not wearing any clothes oraccessories. Participants generally perceive wearable sensorsto have similar capabilities with current panic alarm systemssuggesting that panic alarms can be better accepted if they aremade smaller or less obtrusive.

. . .they [panic alarm] just hang around your neck, so why wouldsomeone want to wear something else? So why don’t they justmake the panic alarms smaller?

There were also suggestions that wearable sensors shouldbe discreet and disguised into a “watch or a ring” as mostelderly persons do not want to be seen wearing a “healthmonitoring device.”

4.1.5.1.2. Ambient monitoring. A majority of the par-ticipants were concerned as to the fact that an ambientmonitoring implementation would be expensive as most ofthem assumed that more sensors are required to equip awhole house. Several participants showed concerns over whatwould happen if accidents occur beyond the vicinity of thesensors, pointing out potential flaws on existing “smart home”solutions such as Tigerplace [26].

. . . they wouldn’t work if you weren’t in here [home], would they?It’s no good having a sensor in the lounge room if you’ve fallenover in the garden.

Questions were raised on whether a pure ambient imple-mentation is adequate to ensure that an elderly person ismonitored at all times. There were also suggestions that ahybrid solution consisting of both ambient and embeddedsensors may be better equip in solving a number of perceivedproblems.

4.1.5.1.3. Embedded sensors. Embedded sensors forelderly care are conceptually similar to RFID implant moni-toring, which has already been deployed with limited successin some environments, such as border control and warehousemonitoring [27].

Participants demonstrated a strong interest in embeddedsensors and expressed a strong acceptance of the concept ofembedding sensors under the skin. Participants pointed outthat not only would it be invisible to others, the need to worryor remembering to wear the sensors would also be eliminated.

Surprisingly, attitudes towards embedded sensors wereextremely positive despite the fact that most participantsassociated this concept with the “micro-chipping” of dogs.

However there were concerns about the pain that may becaused by the insertion of the sensors, some participants alsoquestioned whether the sensors need to be taken out in orderto replace the battery.

a l i n

m

4otewwpvftl

am

si

ssselobkhtetaip

n

i n t e r n a t i o n a l j o u r n a l o f m e d i c

So what happens when the battery runs out? Would you have togo to the doctor and get him to cut you open again?

Quite a number of participants stated that they would notind to be chipped as long as the pain is not significant.

If it’s quick and painless and the benefits are there, I don’t see anyreason not to (be chipped)

.1.5.2. Functionalities. It is difficult to obtain an agreementn the type of interface, functionalities and how the sys-em handles error, as each one of the elderly participantsxpressed different desires and has different perception onhat functionalities and how they should be implementedithin a health monitoring system. For example, while somearticipants suggested that they would like to have access toiew the system’s status on a screen, others said that they pre-er to have a beeping sound that alerts them of errors. Most ofhe participants indicated that they would like to have someevel of control or interaction with the system.

You’ve got to be able to have control of it. I think you should havea screen somewhere, that maybe you can check if you think youmay have set it off, well you can go see if you have or not. Itshould also tell you that the battery is there. . .

The participants have stated that the ability to detect fallsnd monitor basic bodily functionalities should be imple-ented into a WSN health monitoring system.

Well I think that as long as it can measure temperature, heart-beat and pulse, that would accommodate for 90% of problems.It should detect falls I guess, some sensor to measure falls. . .itshouldn’t monitor anything it doesn’t need to.

Participants stated that while they do not mind having theystem monitoring them 24/7, the incorporation of a cameranto such system would be “too intrusive.”

No, definitely no photos. It’s like big brother, and I’m not havingthat

Participants have suggested that a WSN health monitoringystem should behave differently under different circum-tances. While several participants suggested that usershould be allowed to turn the system off when desired oth-rs were more wary of the idea and suggested putting a timeimit on how long it can be turned off before it turns itself backn again. There were also suggestions that if the system is toe turned off, then it should notify the elderly person’s next ofin, such that a family member is aware that the monitoringas stopped. Some mentioned that as the system’s purpose iso ensure that medical assistance will arrive promptly whenmergencies happen, therefore it should not allow users tourn it off. Instead it should have some sort of a “situationware” mechanism as there may be situations where explicitnterruptions from the system may be undesirable or inappro-riate.

Is it possible for it to have two modes? Can be interrupted mode,

and can’t be interrupted mode? Because if you went to sleep andyou had a heart attack, you would want it to interrupt you.

As mentioned above, all of the participants had at least oneegative experience with technology. They expressed their

f o r m a t i c s 7 8 ( 2 0 0 9 ) 788–801 795

frustration in not being able to understand what happenedwhen error occurs and hence expressed a desire for the sys-tem to have “common sense” and give out meaningful andeasy to understand error messages.

Well I think if it picks you up in two places at once, that’s its fault.I don’t think it should interrupt you asking something like “Areyou both alright?” It should get its act together. I think it shouldhave a signal that says “I’m faulty, get someone to fix me.” I thinkan error signal is important.

4.1.6. External factorsResponses from the participants indicate that user acceptanceof WSNs can be affected by external factors including, but notlimited to, types of supports available, the type of care that theparticular elderly individual is receiving, the elderly person’scurrent housing situation as well as who is willing to subsidizethe WSN system. Most participants pointed out that the avail-ability and quality of user support is crucial in determiningwhether users can interact successfully with the technology.Our findings also indicated that elderly individuals with goodsocial bonds are less likely to think that they have a need forWSNs. It has been pointed out that a WSN system, while it hasthe potential to extend an elderly person’s time at home, mayonly be useful before they are needed to enter into a nursingfacility.

If you had this at home, you could still have meals on wheelscoming to feed you if you needed it. But in between this time,where you’re still at home but not quite in a hostel yet, I thinkthat’s where this really fits.

4.2. Relationships between identified concepts andexisting intention to use models

There are indications that elderly people do not solely basetheir intention to use or acceptance of a system on PEOUand PU. Other factors, such as implementation design andconcerns associated with WSNs are needed to be addressedbefore an elderly person would use a WSN-based healthsystem. In this section, we will provide support and expla-nation based on the transcript-based findings and relatedliterature for the probable relationships between the con-cepts and themes discovered and the factors present inexisting models representing ‘intention to use’ such as TRA,TAM and UTAUT. These relationships along with the con-cepts and themes are indications of the dimensions thatshould be further explored in future larger studies. Due tothe infancy of the WSN technology, we were not able todetermine whether Facilitating Conditions such as knowl-edge or experience with WSN systems has an effect on theiracceptance.

4.2.1. Factors affecting perceived ease of useAn elderly person’s preference on the system design, per-sonal preferences and several external factors it appears canaffect a user’s PEOU of the system. An embedded solution

with an easy to understand and good usability is the mostaccepted implementation among the participants, suggestingthat interaction with the system may be “an automatic thing”or “as simple as ‘push the button’.” Suggestions of an ambi-

i c a l

796 i n t e r n a t i o n a l j o u r n a l o f m e d

ent solution were met with indifference, with participantsquestioning its reliability as well as its ease of use. It hasalso been suggested that simplicity in the design of the inter-face would have a positive effect on the system’s PEOU withparticipants stressing the fact that the system is for “emer-gency only” and therefore the system should be kept as simpleas possible to allow users to quickly learn how to interactwith it.

We have also found that user’s personal preference, whichis contributed to by factors such as the user’s “technologyself-efficacy” as well as their “willingness to be trained,” canaffect a WSN system’s PEOU. While several participants havequestioned their ability to successfully interact and utilize aWSN system, they have also, at the same time, indicated anintention to use WSN. There is however, an indication thatusers’ concern with their ability to interact with the system isalso a concern about whether they can remember how to uti-lize all functionalities the system offers. Therefore it can beseen that a low level of “technology self-efficacy” may neg-atively impact a system’s PEOU. It has been observed thatlow levels of “self-efficacy” may be an important determinanton one’s “willingness to be trained”; participants who weremore concerned on whether they can successfully interactwith a WSN system are more likely to hold the perceptionthat training sessions or demonstrations are necessary inallowing them to successfully interact with or use the WSNsystem.

External factors, such as the availability of training, sup-port or a well written manual not only can affect a system’sPEOU but also “User’s Personal Preferences.” The quality of thesupports available can affect an elderly person’s willingnessto be trained; where good quality supports can minimize their“apprehensiveness” and “fear” towards WSNs, hence makingit easier to interact with the system.

4.2.2. Factors affecting perceived usefulnessBoth “independence” and “perceived impact on the quality oflife” have been found in our exploratory results to have anaffect on PU. Participants have commented that they wouldconsider using the system if it is deemed useful, reliable andprovides obvious benefits to their current lifestyle.

If the thing is good, and it works, then we go for it. However, ifwe see something that is useless, and obtrusive, and is changefor change’s sake, then no. Not Interested

Determination in sustaining one’s independence can affecta WSN system’s PU and any indication on WSN’s ability toimprove an elderly individual’s quality of life would appear tohave a positive impact on the system’s PU. It has been observedthat negative perception on nursing homes can increase anelderly person’s desire to remain at home. Participants witha relatively negative attitude towards nursing homes not onlyare more likely to perceive WSNs as useful, but also are moreinclined to accept technologies that show potential in prolong-ing their time at home or independence.

4.2.3. Factors affecting behavioral intention to useAmong all concerns identified, cost is the most prevalent con-cern of our participants and has been observed to be one ofthe main determinants in defining WSN acceptance. While

i n f o r m a t i c s 7 8 ( 2 0 0 9 ) 788–801

the participants are aware of the benefits that a WSN sys-tem can provide, they were not willing to utilize the systemif its cost is substantial, as most of them hold the percep-tion that it is an “emergency response system.” Participantshave suggested that the system should cost no more than“a dollar a day” for them to consider whether or not theywill use the system.“Social implication” similar to “socialinfluence” as described in UTUAT, is also a significant deter-minant of whether a user would intend to use WSNs as fearof being stigmatized as frail can result in an elderly person’srefusal to wear a wearable sensor device. Participants havestressed that the system should be unobtrusive so that “every-body doesn’t say ‘Look at that woman, she’s wearing one of thosethings’.”

Existing literature has suggested that users are tradition-ally concerned about privacy in wireless communications[13,28]; responses from our elderly participants show other-wise. Apart from a strong rejection in the use of a camera,participants showed no concerns regarding the possibility ofsomeone “listening in” to their data. Participants do not per-ceive their medical data, such as heartbeat, pulse or bloodpressure to hold any significant value to an outsider, withone participant dismissing the “Privacy” issue with the fol-lowing comment: “What’s in it that’s private?.” It should bestressed that while the importance of privacy should notbe diminished, a highly secured system is unlikely to beaccepted unless it also fulfills and overcomes the require-ments and concerns that elderly people have towards WSNsystems.

Apart from “Concerns,” “Design Preference” has also beenfound to have contribution towards user’s “Behavioral Inten-tion to Use.” For example, responses from the participantshave suggested that the system should be designed suchthat it addresses the problems of remembrance, rebellion andobtrusiveness; otherwise users will resent the technology andwill not fully utilize it even if it is deemed useful.

5. Interpretation

Overall, findings from this exploratory study suggest that anelderly person’s acceptance of WSN-based systems is affectedby numerous factors, but two factors had particular priority –cost and whether the user is able to interact with such system.Cost is the most frequently discussed issue and it suggests thatunless the system is affordable or has its cost offset, it may notbe widely accepted and adopted even if it is deemed useful.This is contributed to by the fact that most of the elderly par-ticipants involved in this study are either on fixed and limitedincomes, or on the pension, and as they perceived the use of aWSN health monitoring as an elective option, it does not ratenecessarily high on their financial agenda.

The concept of having to interact with a WSN systemwas not positively met by most participants, despite mostof them have stated a desire to have some means of sys-tem control made available to them. This can be explained

by the fact that most participants were unfamiliar with theconcept of having a WSN system for elderly care as well asthe fact that most participants demonstrated a low level of“technology self-efficacy.” However we have found that par-

a l i n

ttttssantWss

fifwsfwHfstts

ombsoism

itatstndso

pcrstrrdcna

th

i n t e r n a t i o n a l j o u r n a l o f m e d i c

icipants who are “apprehensive about technology” or have “fear”owards technologies are more likely to indicate a willingnesso undertake training as they generally hold the perceptionhat more knowledge on the usage of WSN systems can makeystem interactions seem easier; allowing them to interactuccessfully with the system. This also indicates that thevailability of support can directly affect an individual’s “tech-ology self-efficacy” as well as PEOU. Fear of not being ableo interact with the system and the lack of confidence in

SN systems can cause the lack of interest in using WSNystems despite participants’ positive attitudes towards WSNystems.

The willingness to undergo training was a surprisingnding. However during our session with the healthcare pro-essional, she strongly denied that elderly people are typicallyilling to undertake training, no matter how basic, to use

omething deemed as elective. She stressed that, in her pro-essional opinion, while an elderly person may indicate aillingness to undertake training, reality proves otherwise.er experience reveals that there is little or no willingness

rom an elderly person to learn how to interact with a newystem, and even if they are shown how to use the system,hey will quickly forget, indicating that systems designed forhe elderly population should have a simple and easy to under-tand interface.

Most of the elderly participants were unable to fully rec-gnize the potentials a WSN system may provide. As such,ost of them did not perceive WSN systems to have the capa-

ilities in offering benefits that are not provided by existingystems, claiming that the only advantage WSN systems havever existing panic alarm systems is its capability to activatetself. Participants also suggested that if panic alarm systemsuch as VitalCall can be made smaller and less obtrusive, thereay not be a need for WSN systems at all.The embedded implementation was the preferred WSN

mplementation when compared to others. This may be due tohe fact that most participants have a preference for a simplend easy to understand interface and hold the perception thathe amount of interactions required can be reduced if the sen-or is embedded under the skin. While there were suggestionshat an embedded sensor implementation has negative con-otations and similarities to the concept of micro-chippingogs, it was generally agreed that it is the most acceptableolution as it can overcome problems of remembrance andbtrusiveness.

Regardless of the implementation methodology, all of thearticipants held the perception that a WSN system for elderlyare should be for emergencies only. Suggestions to incorpo-ate social aspects, such as reminders for social events, weretrongly rejected. This may have a relation to the fact that all ofhe participants in this study are elderly people who possess aeasonable awareness and intelligence in regards to their cur-ent situation in life, their surroundings as well as the conceptsiscussed during the focus group sessions, therefore are moreapable in controlling the social aspects in their live. Less cog-itively able elderly persons may hold different opinions or

ppreciation towards such functionality.

The elderly participants did not perceive that a WSN sys-em can change the breadth or quality of their social lives;owever they were concerned as to how they may be perceived

f o r m a t i c s 7 8 ( 2 0 0 9 ) 788–801 797

if they were seen to be wearing a device associated with a sys-tem that is designed for elderly care. There are indications thatissues regarding “social implication” tend to concern femaleindividuals more than their male counterparts. It has beenobserved that female participants demonstrated more con-cerns in being stigmatized as frail if they were seen wearing a“device,” commenting that they would not want to wear anysensor devices unless it is unobtrusive. While the male partic-ipants were all aware of issues relating to “social implication,”they did not demonstrate the same level of concern as seenin female participants. It should also be noted that all sce-narios and examples given to describe reasons for refusing towear a panic alarm device were all related to a female elderlyperson. This also provided a better insight into why an embed-ded implementation was the most accepted methodology, asit can be fully hidden from others. Therefore it is likely thata WSN system will experience rejection from the elderly pop-ulation if its utilization and implementation is obtrusive orassociated with the impression that it is specially designedfor a frail individual.

Participants were least concerned about the privacy of theirhealth data, stating that the security of their medical data isnot as important as ensuring the timely arrival of assistancesduring emergencies. The healthcare professional also holdsa similar opinion, confirming that in her professional experi-ence, while some elderly people may ask questions relatingto the privacy of their medical data, most are not overly con-cerned as knowing that help will arrive promptly far outweighsthe drawbacks of jeopardizing the privacy of their medicaldata. Nevertheless, the elderly participants’ lack of concernover the privacy of their health data is likely to be causedby a lack of awareness over the risks involved. A lack of pri-vacy and security precautions may likely negatively affect atechnologically savvy user’s intention to use a system.

While the elderly participants perceive WSNs as a workablesolution for healthcare and agree that WSNs have the poten-tial to extend an elderly person’s independence, they have alsopointed out that it is unlikely to fully replace the role of nursingcare. However it has been suggested that a WSN system canbe a serious alternative to aged hostels and provide significantbenefits to those who can effectively look after themselves andwho wish to remain residing at their own home for as long aspossible.

6. Discussion

This exploratory study aimed to identify and explore factorscontributing to an elderly person’s acceptance and perceptionof WSN-based health monitoring systems. As such it pro-vides insights on the thoughts, concerns and perceptions ofan elderly person on current and emerging WSN technolo-gies. This work takes advantage of the fact that the elderlyparticipants had no prior knowledge or experience with WSNsystems and as such, participants were able to freely discussany perceived issues on WSN systems without preconcep-

tions. Moreover, as the selection criteria did not includerestrictions on factors such as the participants’ ethnic, educa-tion and financial background, therefore findings obtained inthis study may provide a more generalized view that portrays

i c a l

798 i n t e r n a t i o n a l j o u r n a l o f m e d

the perceptions and thoughts of the current elderly popula-tion, complementing and differing from other studies such asDemiris et al.’s study on a group of elderly people’s opinionstowards an implemented “smart home” solution – TigerPlace[19]. Our findings are based on an elderly person’s concep-tual perception of WSN systems rather than actual usage.This is because our aim is to discover issues which needto be addressed such that WSN researchers and designerscan develop and implement health-related WSN systems thatmore fully meet an elderly user’s needs and expectations.

In the context of developing a WSN healthcare systemfor the elderly population, expectancies in performance andeffort may not be the most important determinants of WSNacceptance. While existing studies state that even if users dis-like a certain system they will adopt it as long as they perceivethe system useful [15,29], our findings found otherwise. It isacknowledged that results obtained via different methods, forexample, surveys and focus groups may not be directly com-parable. Also, perhaps due to participants’ assumptions thatthey will be asked to cover the cost to install and utilize a WSNsystem in their home, cost was found to be the most criticaldeterminant in determining an elderly person’s acceptance ofWSN-based systems. Additionally, there were also strong indi-cations that regardless of how users may deem a WSN systemuseful, it will still experience rejection and will not be acceptedif it is not affordable. This is contributed to by the fact thatmost participants consider a WSN system as an “extra” andnot a necessity, despite it relating to their health and wellbe-ing. There are also suggestions that WSN systems are morelikely to be accepted by the elderly population if the cost isoffset by a third party, such as the elderly user’s children orthe government.

While other studies hold the opinion that the developmentof training that reduces “computer anxiety” and improves“computer self-efficacy” can improve user’s acceptance of asystem [29], our findings suggest that systems with a simpleinterface that require the least amount of interaction are morelikely to be accepted by an elderly person. This is mainly dueto the fact that elderly users may find it challenging to mem-orize what functionalities different buttons may serve, theyalso have different design requirements to the average usersas their abilities are diverse; for example, some may have hear-ing or visual impairments. It has been found that appropriatetraining can improve a system’s PEOU, functionalities, inter-face and implementation designs that cater for the needs ofan elderly user may have a bigger effect on a system’s PEOU.

Our findings also point out that perception of how WSNsystems may change an individual’s lifestyle can affect per-ceptions of their usefulness. Contrary to existing views thatthe introduction of a mobile system may reduce an individ-ual’s social contact [13,30], participants generally hold theopinion that it is unlikely for a WSN-based system to changetheir social trends and habits, dismissing suggestions that theintroduction of a WSN system may have negative impacts ontheir lifestyles. It can be seen that a WSN system’s potential toprolong an elderly person’s time at home, which also has the

potential to enhance their quality of life, contributes positivelyto its PU.

Although it has been stressed that the preservation of pri-vacy within the healthcare domain is important and any loss

i n f o r m a t i c s 7 8 ( 2 0 0 9 ) 788–801

of privacy can have an adverse impact on the stakeholders[21], most participants disagreed and were of the opinion thattheir medical data hold no value to any malicious third party.However this may be due to a lack of understanding in mobileand wireless technologies as well as the dangers resulted fromallowing a malicious third party to gain access to one’s healthdata. Opinions and perceptions regarding the privacy of one’shealth data are likely to change when users have been edu-cated over the risks engendered by a failure in ensuring theprivacy of their health data. It should be stressed that the lackof emphasis on privacy is due to the perceptions of the elderlyparticipants, this study does not suggest for privacy issues tobe overlooked. In fact, due to the lack of awareness in theimportance of ensuring the privacy of health data among theelderly, WSN researchers and developers may have an obli-gation to further ensure the privacy and security of the datatransmitted from motes.

Despite the fact that several participants demonstratedanxiety towards the need to interact with a WSN system, thereare no indications that issues regarding usability can causerejection or unwillingness to use WSN systems. While thismay not agree with results from previous studies which statethat user anxiety can have a negative impact on their “behav-ioral intention to use” a system [31], we cannot conclude thatanxiety has no impact on usability or willingness to use. Fur-ther investigations are needed. As this is a study based onfindings from two focus groups, therefore opinions may beinfluenced by a more dominant participant who might be lead-ing the discussion. It may be beneficial to determine whether ahigh level of “computer anxiety” can discourage elderly users’intention to use WSN systems for elderly care through the useof questionnaires.

6.1. Guidance for future studies

Findings from this study can be leveraged for future stud-ies undertaking a more in-depth investigation on the factorsaffecting whether an elderly person will accept or reject aWSN-based system. Future empirical studies in this areashould include a larger group of participants involving adiverse group of elderly people from various geographical,education and cultural backgrounds. It is also believed thatfindings from this study can act as a guidance on the dimen-sions and items that should be included in a more in-depthquestionnaire aiming to investigate the determinants for theelderly users’ acceptance for WSN-based systems. Moreover,the relationships between the concepts, as described in ourfindings, can act as a basis for the development of a poten-tial acceptance model for elderly persons’ perception of WSNhomecare systems. Investigating how middle-aged personsperceive WSN is also likely to provide benefit for the designand development of WSN systems, as insights from themiddle-aged group is likely to provide indications on how WSNsystems might be perceived in the future.

Due to the lack of awareness as well as the fact that motetechnology is still under development, most of our elderly

participants have been observed to have difficulties in fullyunderstanding the benefits that motes can provide. Investi-gations on whether an elderly person is likely to accept andutilize a WSN-based homecare system are therefore neces-

a l i n

sivwsuobicabwams

aehhdtoadbdutrs

6

Apdcibdtfawlimc

etsshWiat

i n t e r n a t i o n a l j o u r n a l o f m e d i c

ary when a complete WSN-based homecare system has beenmplemented. It is believed that future studies should pro-ide detailed descriptions of the implementation, interface asell as functionalities of the evaluated WSN-based homecare

ystem and any attitudes, perceptions and concerns from itssers should be documented in detail. Detailed descriptionsn how users may interact and utilize such system should alsoe provided. As elderly users are not the only stakeholders to

nteract with the WSN-based system, it may also be benefi-ial to investigate a healthcare worker’s perceptions, concernsnd attitudes towards such systems. The utilization of a WSN-ased homecare system is also likely to impose changes on theorkflow of a clinician. Therefore future investigations shouldlso investigate the impact on the workflow of clinicians thatay be caused by the utilization of a WSN-based homecare

ystem.There were also indications from our findings that current

cceptance models such as TAM or UTAUT may need to bextended or modified due to the uniqueness of a WSN-basedealthcare system designed for elderly users. Our findingsave revealed factors such as the importance of indepen-ence, perceived impact on quality of life as well as attitudeowards nursing homes can affect an elderly user’s perceptionn whether a system is useful. These factors are believed toffect whether a WSN system is deemed useful, and should beetermined by a future, larger study as to whether they shoulde included as an extension for the TAM or UTAUT model toetermine whether the elderly people are likely to accept (andtilize) a WSN homecare system. For example, Future ques-ionnaire design can then include representative questionselating to each of the concepts and themes identified in thistudy.

.2. Limitations

lthough this study does not encompass a spectrum of elderlyeople from various cultural and geographical backgrounds, itoes illustrate elderly persons’ perceptions, attitudes and con-erns towards current and emerging WSN systems. This studys limited to Sydney, Australia, and as such, findings may note generically applied to all elderly people across the globeue to factors such as cultural differences and monetary fac-ors. However it is expected that similar results may be foundrom other developed countries such as America or Canada,s those countries have similar socio-economic backgroundsith Australia. Furthermore, the state of the Australian pub-

ic healthcare system is likely to affect the findings; an elderlyndividual with limited access to public healthcare services

ay hold different perception on the value of a WSN health-are system.

While participants were randomly selected from variouslderly community groups, we acknowledge a deliberate selec-ion bias that may over represent elderly individuals who weretill living independently. As such, our findings may repre-ent the opinion of those who are prejudiced against nursingomes and other forms of institutional healthcare facilities.

hile this bias provides an important insight into the signif-

cance of independence to an independent elderly individuals well as the effort they are prepared to undertake to extendhat independence, we do not know how the opinions of those

f o r m a t i c s 7 8 ( 2 0 0 9 ) 788–801 799

elderly individuals may compare to those who are alreadyresiding in an institutional healthcare facility. It should bepointed out that one of the purpose to install a WSN sys-tem in a residential environment is to ensure the wellbeingof those elderly individuals who resides on their own, andto ensure the prompt arrival of healthcare assistance whenemergency occurs, therefore the benefits provided by a WSNsystem may not be as prominent for those who are alreadyresiding in a healthcare facility. In spite of this, it may provefruitful to also explore the perceptions, attitudes and concernstowards WSNs of those who are already residing within a caredenvironment. Furthermore, as the utilization of a WSN systemwithin a care facility is also likely to impact the healthcareworkers, it may also be necessary to explore the healthcareworker’s perceptions towards the utilization of a WSN system.

7. Conclusions

This research was an exploratory qualitative study conductedto investigate the elderly population’s perception of WSN forelderly care. It identifies a set of possible concepts and themesaffecting WSN acceptance and its usage. It is believed that amore detailed questionnaire-based study involving a largerand more diverse group of elderly people can leverage thefindings of this study to enable the derivation of a set ofdeterminants that can provide a more generalized perspec-tive on factors affecting an elderly person’s perception andacceptance of WSN systems. As WSN systems involves mobileor wireless technologies that are not utilized in a traditionalinformation system, it is likely that a new model is requiredto accurately determine the acceptance of a WSN system, butlikely deriving from existing intention to use models such asTAM and UTAUT to require extension or modification beforethey can. Similarly, as the needs of an elderly individual woulddiffer to an average person, it is likely for acceptance mod-els describing an elderly person’s acceptance for a systemto require new/different constructs. Our findings can provideguidance on the development of an acceptance model specif-ically for determining an elderly person’s acceptance of WSNsystems as this study clearly indicates a range of thoughts,concerns and perceptions of the elderly people on currentWSN designs.

Findings from this study suggest that cost is the most pre-dominant factor determining whether an elderly person willaccept WSN technologies. It has also been found that differ-ent elderly people desire a different level of control over thesystem. It appears from the study that the ability to controlthe system must be readily available; otherwise it is likely tobe rejected by an elderly user as the perceived lack of con-trol may possibly be regarded as an affront to their decisionmaking ability.

It is a primary finding of this exploratory study that elderlypeople perceive sensor networks as a viable elderly care solu-tion, which can prolong their time living at home, on conditionthat factors regarding cost and the ability to control are suit-

ably addressed. Our findings also show that while there areindications that the elderly population would accept WSNtechnologies, they will not accept it if they cannot recognizeor do not agree with the benefits it promises to provide.

800 i n t e r n a t i o n a l j o u r n a l o f m e d i c a l

Summary pointsWhat is already known on the topic?

• Previous work has focused on developing WSN tech-nologies instead of addressing what an elderly personmay prefer.

• WSN technologies can allow an elderly person to “age-in-place.” But little is known on an elderly person’sperception of WSNs and how it may impact or changetheir quality of life.

What this study added to our knowledge?

• An exploratory study indicating 6 themes and relatedconcepts that affect the perception and potential useof WSNs for the purpose of supporting healthcare forthe elderly population.

• Specific findings suggested:◦ Cost is the most predominant factor in determin-

ing an elderly person’s intention to use WSNs. Theelderly participants show a strong desire for the costof acquiring a WSN system to be offset by either thegovernment or their children.

◦ There are clear indications that an elderly per-son will readily accept technology but discriminatebased on its perceived value.

◦ Privacy is not perceived as a highly significant con-cern. It does not see to have a significant affect onan elderly person’s perception or acceptance of WSNsystems.

◦ Contrary to prior belief, the elderly participants donot perceive WSNs or any other technologies to haveany impact on their social live.

◦ Embedded sensors were the most preferred imple-mentation method. It has the greatest potential tofunction under various situations and is the onlyimplementation that was perceived to be able toovercome issues regarding remembrance, rebellionand obtrusiveness.

r

[18] M. Rantz, M. Skubic, S. Miller, J. Krampe, Using Technology

Conflict of interest statement

This work was conducted independently by the authors. It wassolely supported by the university and did not involve fundingfrom an external or commercial source.

Acknowledgements

The authors thank the domain expert for her critical feed-back. They would also like to thank the reviewers for theirconstructive input.

Contributions. RS led the analysis and study. CS conducted

the focus groups and preliminary analysis on the data. Allauthors contributed to data analysis and drafting and review-ing of the manuscript.

i n f o r m a t i c s 7 8 ( 2 0 0 9 ) 788–801

e f e r e n c e s

[1] R. Steele, C. Secombe, W. Brookes, Using Wireless SensorNetworks for Aged Care: The Patient’s Perspective. PervasiveHealth Conference and Workshops, 2006.

[2] Australian Bureau of Statistics (ABS), National and StateSummary Tables, 2008.

[3] V.Z. Ogozalek, A comparison of the use of text andmultimedia interfaces to provide information to the elderly,in: Proceedings of the SIGCHI Conference on Human Factorsin Computing Systems: Celebrating Interdependence,Boston, Massachusetts, United States, 1994.

[4] J.L. Fozard, Distribution of age and technology, Interactions 4(2) (1997) 34–46.

[5] S.J. Czaja, Using Technologies to Aid the Performance ofHome Tasks. Handbook of Human Factors and the OlderAdult, 1997.

[6] R. Heller, J. Jorge, R. Guedj, EC/NSF workshop on universalaccessibility of ubiquitous computing: providing for theelderly event report, in: Proceedings of the 2001 EC/NSFWorkshop on Universal Accessibility of UbiquitousComputing: Providing for the Elderly, Alcácer do Sal,Portugal, 2001.

[7] S.L. Willis, Everyday problem solving, in: J.E. Birren, K.W.Schaie (Eds.), Handbook of the Psychology of Aging, fourthedition, Academic Press, San Diego, 1996, pp. 287–307.

[8] C. Russell, A certain age: women growing older, in: I.M.P.S.Feldman (Ed.), Meanings of Home in the Lives of OlderWomen (and Men), Allen & Unwin, Sydney, 1999,pp. 36–55.

[9] Crossbow Technology. Crossbow Technology: WirelessSensor Networks, accessed at http://www.xbow.com/Products/productdetails.aspx?sid=164#pricebox on 22.05.09.

[10] A. Perrig, R. Szewczyk, J.D. Tygar, V. Wen, D.E. Culler, SPINS:security protocol for sensor networks, Wireless Networks 8(2002) 521–534.

[11] C. Karlof, N. Sastry, D. Wagner, TinySec: a link layer securityarchitecture for wireless sensor networks, in: Proceedings ofthe 2nd International Conference on Embedded NetworkedSensor Systems, Baltimore, MD, USA, ACM, 2004.

[12] R. Di Pietro, L.V. Mancini, A. Mei, Random key-assignmentfor secure wireless sensor networks, in: Proceedings of the1st ACM workshop on Security of ad hoc and SensorNetworks, Fairfax, Virginia, ACM, 2003.

[13] J. Abascal, A. Civit, Universal access to mobile telephony as away to enhance the autonomy of elderly people, in:Proceedings of the 2001 EC/NSF Workshop on UniversalAccessibility of Ubiquitous Computing: Providing for theElderly, Alcácer do Sal, Portugal, 2001.

[14] M. Fishbein, I. Ajzen, Belief, Attitude, Intention andBehavior: An Introduction to Theory and Research,Addison-Wesley, Reading, MA, USA, 1975.

[15] F.D. Davis, Perceived usefulness, perceived ease of use, anduser acceptance of information technology, MIS Quarterly 13(3) (1989) 319–340.

[16] V. Venkatesh, M.G. Morris, B.D. Gordon, F.D. Davis, Useracceptance of information technology: toward a unifiedview, MIS Quarterly 27 (3) (2003) 425–478.

[17] M.J. Rantz, K.D. Marek, M. Aud, H.W. Tyrer, M. Skubic, G.Demiris, A. Hussam, A technology and nursing collaborationto help older adults age in place, Nursing Outlook 53 (1)(2005) 40–45.

to Enhance Aging in Place. Smart Homes and HealthTelematics, Springer, Berlin/Heidelberg, 2008, pp. 169–176.

[19] G. Demiris, M. Rantz, M. Aud, K. Marek, H. Tyrer, M. Skubic,A. Hussam, Older adults’ attitudes towards and perceptions

a l i n

CHI ‘01 Extended Abstracts on Human Factors in ComputingSystems, ACM, Seattle, Washington, 2001.

i n t e r n a t i o n a l j o u r n a l o f m e d i c

of smart home technologies: a pilot study, MedicalInformatics and the Internet in Medicine 29 (2004) 87–94.

[20] R.L. Thompson, C.A. Higgins, J.M. Howell, Personalcomputing: toward a conceptual model of utilization, MISQuarterly 15 (1) (1991) 125–143.

[21] R. Gururajan, M. Toleman, J. Soar (Eds.), Necessity for a NewTechnology Acceptance Model to Predict Adoption ofWireless Technology in Healthcare. HIC, 2004.

[22] H.J. Wisnicki, Wireless networking transforms health care:physicians’ practices better able to handle work flow,increase productivity (The human connection),Ophthalmology Times 27 (21) (2002) 38–41.

[23] VitalCall, VitalCall (Chubb Security Australia). VitalCall, 2008(accessed at http://www.vitalcall.com.au/ on 22.05.09).

[24] Lifeline, Lifeline—The Most Trusted Medical Alert ServiceProvider, 2008 (accessed at http://www.lifelinesys.com/ on

22.05.09).

[25] F.G. Barbeite, E.M. Weiss, Computer self-efficacy and anxietyscales for an Internet sample: testing measurementequivalence of existing measures and development of newscales, Computers in Human Behavior 20 (2004) 1–15.

f o r m a t i c s 7 8 ( 2 0 0 9 ) 788–801 801

[26] Tigerplace (accessed at http://www.tigerplace.net/ on22.05.09).

[27] J. McHugh, A Chip in your Shoulder: Should I Get An RFIDImplant? Slate, 2004 (accessed athttp://www.slate.com/id/2109477 on 22.05.09).

[28] J.R. Moehr, J. Schaafsma, C. Anglin, S.V. Pantazi, N.A. Grimm,S. Anglin, Success factors for telehealth—a case study,International Journal of Medical Informatics 75 (10–11) (2006)755–763.

[29] M. Rahimpour, H.N. Lovell, B.G. Celler, J. McCormick,Patients’ perceptions of a home telecare system,International Journal of Medical Informatics 77 (7) (2008)486–498.

[30] J.L. Gregg, Tearing Down Walls for the Homebound Elderly.

[31] F.C. Tung, S.C. Chang, Nursing students’ behavioral intentionto use online courses: a questionnaire survey, InternationalJournal of Nursing Studies 45 (9) (2008) 1299–1309.