Activity and participation, quality of life and user ... - CiteSeerX

3 downloads 167 Views 402KB Size Report
Feb 11, 2011 - ing, roof windows automatically closing in case of rain and lights ...... app earance, service delivery, dev ice c o mpatibility). T2: 2.62–3.64. (cost,.
Disability and Rehabilitation: Assistive Technology, 2010; Early Online, 1–18

REVIEW

Disabil Rehabil Assist Technol Downloaded from informahealthcare.com by Cea Fontenay Auz Roses on 02/11/11 For personal use only.

Activity and participation, quality of life and user satisfaction outcomes of environmental control systems and smart home technology: a systematic review

˚ SE BRANDT1, KERSTI SAMUELSSON2, OUTI TO ¨ YTA ¨ RI3 & ANNA-LIISA SALMINEN4 A The Danish Centre for Assistive Technology, A˚rhus, Denmark, 2Department of Rehabilitation Medicine, University Hospital Linko¨ping, Linko¨ping, Sweden, 3National Institute for Health and Welfare, Helsinki, Finland, and 4Research Department, The Social Insurance Institution of Finland, Helsinki, Finland 1

Accepted October 2010

Abstract Objective. To examine activity and participation, quality of life, and user satisfaction outcomes of environmental control systems (ECSs) and smart home technology (SHT) interventions for persons with impairments. Method. A systematic review. Seventeen databases, three conference proceedings, and two journals were searched without language or study design restrictions covering the period January 1993 – June 2009. Reviewers selected studies, extracted data, and assessed the methodological quality independently. Result. Of 1739 studies identified, five effect studies and six descriptive studies were included. One study was on SHT and the remainder on ECS; functionalities were overlapping. The studies varied in most aspects, and no synthesis could be drawn. However, ECS/SHT tended to increase study participants’ independence, instrumental activities of daily living, socialising, and quality of life. Two studies showed high user satisfaction. The level of evidence was regarded as low, mainly due to small study sizes, lacking confounder control, and a majority of descriptive studies. Conclusion. Due to few and small studies and study diversity, it was not possible to determine whether ECS/SHT have positive outcomes for persons with impairment, even though the technologies seem to be promising. High quality outcomes studies such as randomised controlled trials, when feasible, and large longitudinal multi-centre studies are required.

Keywords: Assistive technology, assistive devices, occupational therapy, self-help devices, EADL

Introduction Persons with impairment have the same needs as everyone else, e.g., for housing, health care, social interaction and employment. Nevertheless, that is not always possible without personal assistance or support. Still, persons with impairment want social independence, which is about autonomy and control of their lives [1]. Environmental control systems (ECSs) have been used for many years now, aiming at enabling persons with impairment to carry out activities within the home without assistance from others and thus to live more independently, which in turn is believed to increase an individual’s quality of life [2]. By means of an input device such as a switch or a voice-controlled device, it is possible for a

person with physical impairment to control electronic equipment remotely, e.g., open/close doors and windows, and utilise entertainment centres, telephones, alarms and computers. Smart home technology (SHT) has emerged in the past decade, not only offering some of the same functionalities as ECSs, but also offering extended functionalities such as monitoring and automated functions, e.g., heating, roof windows automatically closing in case of rain and lights automatically turning on when the occupant gets out of bed. Furthermore, some ECS and SHT assist persons with cognitive impairments in terms of support, e.g., memory, security and communication. With an increasing number of people with cognitive impairment and with new technological developments, use of these technologies has been

˚ se Brandt, Department of Research & Development, Danish Centre for Assistive Technology, P.P.Ørumsgade 11, Bygning 3, DK-8000 Correspondence: A ˚ rhus C, Denmark. Tel: þ45-41-91-81-80. Fax: þ45-87-41-24-26. E-mail: [email protected] A ISSN 1748-3107 print/ISSN 1748-3115 online ª 2010 Informa UK, Ltd. DOI: 10.3109/17483107.2010.532286

2

A˚. Brandt et al.

Disabil Rehabil Assist Technol Downloaded from informahealthcare.com by Cea Fontenay Auz Roses on 02/11/11 For personal use only.

gradually increasing in the last few years [3]. In the Nordic countries, persons with severe physical impairments are entitled to receive ECS if it can greatly increase their independence in everyday activities or if it enables the person to work [4]. In other countries, this is not always the case and may depend on the person’s insurance. In any case, knowledge about the outcomes of ECS and SHT should be of great importance in many different global discussions and decisions, for example in the context of discussions on priorities.

Definitions of environmental control systems and smart home technology In Europe, the term ECS is mostly used, and when denoting specific devices the term may be environmental control unit. The term assistive products for controlling from a distance is used for the ISO 9999 classification of assistive devices, defined as ‘devices for enabling remote control and operation of electronic and electrical equipment within the living environment to enable independent living’ [5, p. 51]. This definition is used in the present review. In North America, the term electronic aids to daily living has in many cases replaced the term ECS [6,7]. There is no consensus about the definition of SHT, which is mostly characterised by an internal network, intelligent control, and/or home automation consisting of products within the home and links to services and systems outside the home. One definition is for instance ‘A smart house is a house that has highly advanced automatic systems for lighting, temperature control, multi-media, security, window and door operations and many other functions’ [8, p. 1]. Other terms such as domotics or integrated control are also used to refer to SHT. Contrary to SHT, ECSs are not integrated with other systems in the home.

Activity and participation, quality of life and user satisfaction ECS and SHT are used to enable activities in the home for persons with impairments and are in some cases also used to facilitate social participation, which, in turn, is expected to increase quality of life [9]. Activity and participation are defined and delimited in a number of ways; for the purpose of this review, we adopted the definition used by the International Classification of Functioning, Disability and Health (ICF) [10] because of its broad support and understanding of activity and participation, ranging from simple activities (tasks) to social inclusion, which seems to match the kinds of

outcomes that ECS and SHT provide. In the ICF, activity is defined as ‘the execution of a task or action by an individual’, and participation as ‘involvement in a life situation’ [10, p. 10]; these appear to be separate terms, but when it comes to their concrete classification in the ICF, activity and participation are listed jointly, indicating an overlapping of the two terms. Activity and participation may be executed with or without personal assistance, i.e. independently [10], which is often the aim of ECS and SHT [5]. Activity and participation can be considered an outcome dimension denoting effectiveness [11], which should be the aim of ECS interventions and of at least some kinds of SHT. Effectiveness is contrasted with efficacy and is assessed in natural rather than artificial environments, i.e. in the homes of the users [12,13]. However, effectiveness does not necessarily correlate with a user’s satisfaction with an assistive device or application or with the extent to which it contributes to a user’s quality of life. Consequently, it is important to focus not only on effectiveness but also on users’ perspectives in terms of their perceptions of quality of life and satisfaction outcomes of assistive technology (AT) [11,14]. Quality of life can be considered an important outcome of ECS and SHT interventions. The concept is comprehended in different ways, and in the present review, we adopt a broad understanding of quality of life in order not to exclude any relevant study; for instance, we include psychosocial impact as a quality of life outcome dimension, since we consider this concept as closely related to the concept of quality of life. We have applied the definition agreed upon by the World Health Organization in development of the WHOQOL-BREF: ‘Quality of life is defined as individuals’ perceptions of their position in life in the context of the culture and value systems in which they live and in relation to their goals, expectations, standards and concerns’ [15, p. 7]. User satisfaction with assistive devices is one of the outcomes of AT interventions that has been frequently investigated, in most cases by means of the Quebec User Evaluation of Satisfaction with AT (QUEST) instrument [16,17], which defines user satisfaction as ‘a person’s positive or negative evaluation of those distinct dimensions of the assistive device that are influenced by one’s expectations, perceptions, attitudes and personal values’ [17, p. 6]. This definition is used in this review. When initiating this study, a number of literature reviews had been conducted on the opportunities of ECT and/or SHT interventions to improve quality of life, independent living, well-being for persons with physical and cognitive impairments and to make ageing-in-place possible [18–22], but no systematic review on outcomes of ECT or SHT interventions for persons with impairment in their everyday life in

Environmental control and smart home technology terms of activity and participation (i.e. effectiveness), quality of life and user satisfaction was available. Knowledge about such outcomes is essential for professionals in order to be able to make an informed decision on ECS and SHT interventions.

Disabil Rehabil Assist Technol Downloaded from informahealthcare.com by Cea Fontenay Auz Roses on 02/11/11 For personal use only.

Objective To examine whether persons with impairment benefit from ECSs and SHT in their everyday life in terms of activity and participation, quality of life and user satisfaction, and to appraise the studies’ quality.

3

Subsequently, studies that only addressed input/ control devices or were laboratory studies were excluded. Furthermore, smart home applications controlling water, energy management, metering, security such as alarms and motion detectors and health, e.g., telecare, were excluded because these applications do not require the active control of the user. In case of control intervention, any relevant control, e.g., human assistance, other kinds of AT or no intervention was accepted. Outcomes. Studies on the following outcomes were considered: activity and participation, user satisfaction and quality of life.

Method Search strategy Study design A systematic review.

Inclusion criteria Study design. We primarily included effect studies, i.e. controlled studies, pre post studies and secondarily descriptive studies, i.e. cohort studies without baseline data, cross-sectional and qualitative studies. Descriptive studies represent a lower level of evidence [23,24]. In spite of that, these types of studies were included in this review, since they may provide valuable information about ECS and SHT interventions, and furthermore, by excluding such studies, studies on satisfaction and psychosocial impact would be excluded [25,26]. Studies that compared expected or imagined outcomes with actual outcomes were not considered to be effect studies. From these studies, only actual experiences were considered, and the studies were classified as descriptive. Study population. Inclusion criterion was persons with impairment. Interventions and control interventions. Studies on EHS or SHT applications in the user’ home, controlled by the user or assisting the user in being active or in participation in everyday activities, were included. SHT applications were delimited to systems or parts of systems for: . . . .

Home entertainment (audio visual, Internet) Domestic appliances (cooking, cleaning, maintenance alerts) Information and communication (phone, Internet) Environment (programmable lighting, heating)

Based on the study objective, the authors who are all experienced within the field of AT and a research librarian developed key words and synonyms based on the inclusion criteria. In addition, an expert on SHT was consulted. The Ovid MEDLINE search strategy that we used is presented in Appendix 1. Since databases use slightly different keywords, for instance in the OVID MEDLINE ‘self-help device’ is used for ‘assistive device’, while in CINAHL the term ‘assistive technology device’ is used, the search strategy was adapted for use in the other database searches. We searched for studies on ECS and SHT interventions without language restrictions from the following databases: CCTR, CLCMR, DARE, CLHTA, CLEED, CINAHL, Ovid MEDLINE(R), Ageline, Assia, Sociological Abstracts, PsycInfo, Ebsco ASE, Science Direct, Social Care Online, SocIndex FT and StakesLib. Original studies as well as reviews were searched in order to find possible secondary sources. Beyond this, we hand-searched conference proceedings from RESNA, AAATE and ICADI conferences as well as the journals Assistive Technology and Technology and Disability Journal, which were the major AT journals covering the search period. The searches were delimited to January 1993–June 2009, because ECS and SHT older than that cannot be expected have the same functionalities as newer technologies.

Selection of studies Two authors (AAB, ALS) first screened the titles and abstracts of the findings and excluded duplicate articles and articles that were obviously not within the inclusion criteria. In cases of disagreement or when eligibility could not be decided on the basis of title and abstract, or when articles were considered

4

A˚. Brandt et al.

eligible, copies of full text articles were obtained and reviewed by the two authors. Disagreements were solved by discussion resulting in a final list of eligible articles.

studies regarding study population, interventions, types of outcomes and assessment instruments used.

Disabil Rehabil Assist Technol Downloaded from informahealthcare.com by Cea Fontenay Auz Roses on 02/11/11 For personal use only.

Results Data extraction and quality assessment

Search

Data from the full text articles were extracted and inserted into a study specific descriptive table by two authors (AAB, ALS) and controlled by a third author (KS or OT). The table was designed by the authors in order to capture critical information about the studies and included the following: study design and length of follow-up; study population (number of participants, age, diagnosis and functioning); intervention/control intervention (technology and implementation process); outcomes (activity and participation, quality of life and user satisfaction) and support from AT industry. The internal quality of the studies was appraised by three authors by means of a criterion list modified from Borghouts et al. [27]: whether the study described the population selection sufficiently well; whether the study size was sufficiently large (over 10 patient year); whether the follow-up time was sufficiently long (4 mo or more); whether the proportion of drop-outs was acceptable (less than 20%) and if their characteristics were described; whether outcome measures and data presentation were congruent with the study aims; whether confounder control was performed; and finally, whether the psychometric properties of the instruments were reported. In our modification, the two latter criteria substituted the original criteria ‘appropriate analysis techniques’ and ‘prognostic factors’, because we considered the two chosen criteria to be more relevant in our review. In addition, the identified study with a randomised controlled trial (RCT) design was especially appraised by means of a list of criteria developed by van Tulder et al. [28] regarding selection bias, performance bias, attrition bias and detection bias. External validity and clinical applicability were appraised by three authors using four questions adapted from Shekelle et al. [29]: whether the description of the participants, interventions and setting were sufficiently detailed; whether all clinically relevant outcomes were measured and reported and whether the effect size was clinically important. From the original Shekelle criteria, we deleted one question that dealt with treatment benefits in relation to adverse outcomes since they were not expected. A 10% gain was considered clinically important, derived from research related to back pain and function [30]. The results of the studies were only described and not synthesised due to considerable variation among

The search of the electronic literature databases identified 1726 documents, and 13 were identified from other sources (websites, reference lists, personal archives and experts). Of these, 41 documents were duplicates, leaving 1698 documents for screening. One thousand six hundred fifteen of these were deemed not to be within the inclusion criteria, and thus, 83 documents were retrieved in full text for examination. On the basis of failure to meet the inclusion criteria, 70 documents were excluded, e.g., because the documents merely described technological opportunities, they were not about the targeted categories of technology, they were reviews or they were not about the outcome dimensions of interest. This left the 13 articles that are included in this review. Two studies were each reported by two articles, so that 11 studies were identified in total [7,31–42]. Figure 1 describes the selection process.

Study design and size All studies were published within the last 10 year (Table I). Of the five effect studies, two had a controlled design [36,40,41], one of which was a RCT reported in two different articles [40,41], which had the longest follow-up time and the largest number of study participants of all included studies. In the three pre-post studies, study sizes ranged from eight to 10 participants, and follow-up time varied from 2 wk to 6 mo [31,32,42]. Of the six descriptive studies, one study compared data with a former study, which we judged as a descriptive cross-sectional study rather than an effect study, because it was not able to make comparisons at the level of individual persons [37]. Another study had an interrupted time series design, but since baseline data were about expectations for the interventions rather than real experiences, the design was regarded as a cohort study without baseline [38], i.e. a descriptive study. Likewise, a third study was a cohort study without baseline data, reported in two articles [7,39]. The number of participants in the descriptive studies varied from 7 to 47. None of the studies that reported on funding had received any support from the AT industry. One study did not report whether support had been received [40].

Disabil Rehabil Assist Technol Downloaded from informahealthcare.com by Cea Fontenay Auz Roses on 02/11/11 For personal use only.

Environmental control and smart home technology

5

Figure 1. Flowchart showing the process of identifying and selecting the included studies.

Study populations Most study participants had physical impairments, but some also had other impairments such as poor vision or speech difficulties, and in one study, the target group was persons with cognitive impairments [31]. Impairments were reported at different levels and in different ways, making it impossible to get an overview. Likewise, the study participants had a broad range of diagnoses: frequently spinal cord injury, multiple sclerosis and muscular dystrophy. Participant age varied substantially ranging from six to 89 year of age with a mean age of 54 year, and most study participants were adults of 30–60 year of age. However, in the identified RCT, the mean age of the participants was over 70 year [40,41].

Interventions and control interventions Based on the description in the articles, SHT was only implemented in the RCT study [40,41] and ECS in the remaining studies (Table I). However, the functionalities of the implemented technologies seemed to overlap. The implemented applications varied among the studies and were administered in different combinations. The most frequent ECS or SHT applications were control of lighting, remote

opening and closing of doors, operation of telephones, control of home entertainment equipment and presence of security alarms. The implementation process was described to differing extents in eight of the studies. In six studies, the system was adapted to the needs of the user [31–33,41,43,44], and in five [31,33,35,41,44], the users were trained in the use of the system. In the two controlled studies, the control intervention was no intervention [36,40,41].

Outcomes Activity and participation. In the five effect studies, activity and participation outcomes were measured with different instruments making comparisons impossible (Table I). The RCT carried out by Sundar et al. [40] and Tomita et al. [41], in which the participants were older people and the intervention SHT, only showed limited effects: instrumental activities of daily living (IADL) remained stable for the intervention group, but declined for the control group. In addition, more study participants from the intervention group still lived at home after 3 year when compared to the control group. The other controlled study by Rigby et al. [36] also showed that ECS users were more autonomous in IADL than non-users and that they were able to perform most of

Pre-post study 4 mo (n ¼ 4) 6 mo (n ¼ 4)

A-B-A design (single case pre-endpost study) 2 wk

Crosssectional study

Boman et al. [31]

Croser et al. [32]

Harmer & Bakheit [33]

Study

Study design and length of follow-up Technology: Basic ECS: central control of, e.g., burglar motion detector, lighting and heating; electronic key; home network (laptop computer) used for alarms, bookings of common facilities, calendar, etc.; video camera at the entrance door. Advanced ECS: remote control of, e.g., light, coffee machine; photo controlled water taps; photo telephone, stove guard; lamps activated at night; timer. Implementation process: Parts of the technology were adapted to the user. Before moving in the users got at 2–3 days introduction period to learn to use the equipment, after moving in 1–2 hours’ training 4–5 timers weekly during the first 3 wk.

Technology: Voice activated ECS, scanning radio remote, multi CD, adapted telephone, adapted and standard touch lamp, modified bed control. Individually selected solutions. Implementation process: When relevant the equipment or input devices were adapted to the user.

Technology: Six different ECS brands (described in detail in the article) to control lighting, ventilation and heating, open/close curtains, identify caller at the door, lock/ unlock the door, leave/enter property, summon help, operate a telephone, converse and correspond, control radio/TV/video. Four different types of input control devices were used. The ECS had been used for a mean of 62.3 mo (range: 7 mo–14 years). Implementation process: The equipment and input devices were adapted to the user. All users and 85% of their care-givers had received training on how to use the various functions of the systems.

n¼8 Mean years of age 26.6 (range 6–58 years) Diagnosis: Tetraplegia (n ¼ 3); CP (n ¼ 5) Functioning: NR except ‘low independence’

n ¼ 16 Mean years of age: Male 44.6 years (range 30–62 years); female: 58.1 years (range 34–89 years) Diagnosis: MS (n ¼ 11); MD dystrophy (n ¼ 2); SCI (n ¼ 1); Friedrich’s ataxia (n ¼ 1); CP (n ¼ 1) Functioning: Impaired motor function: quadriplegia with head movements only (n ¼ 6), weakness all limbs (n ¼ 1), paraplegia (n ¼ 7), lower limbs weakness (n ¼ 2).

Intervention/control intervention

n¼8 Mean years of age 40.1 (range 24–53 years) Diagnosis: Traumatic brain injury (n ¼ 6); stroke (n ¼ 2) Functioning: Self-care: independent (n ¼ 7); moderate assistance (n ¼ 1). Memory function: moderate (worst memory function) (n ¼ 3); poor (n ¼ 5).

Study population

Table I. Description of the identified studies.

Outcomes

(continued)

Activity/participation: Statistically significant gains in 7 of the 17 tasks: Control lighting (p ¼ 0.01), identify caller at the door (p ¼ 0.002), lock/unlock door (p ¼ 0.003), summon help (p ¼ 0.003), operate a telephone (p ¼ 0.04), work from home (p ¼ 0.03), control radio/TV/video (p ¼ 0.02). Measured by a modified version of the LOMEC. The amount of care needed decreased for 13 users, 1 felt there had been no difference, 2 responded ‘don’t know’. How much time NR. Measured by a SSQ. Qualitative data showed that ECS had increased independence (n ¼ 8). Quality of life and user satisfaction: not investigated.

Activity/participation: All rated their independence higher when using the ECS for the activities of highest priority: 6 of 8 participants rated their independence 5 or more points on a 10-point scale (higher score more independent). For the activity of next highest priority: 3 of 7 participants rated their independence 5 or more points on a 10-point scale. No data presented, no statistical analysis was performed. Measured by a SSQ. Quality of life and user satisfaction: not investigated.

Activity/participation: Accomplishment of most important activities (MIA): Improvement of 2 points in 19 of 34 MIA (p 5 0.0.5) for 5 of 8 participants. Group level: Z ¼ 2.10 (p 5 0.05). Satisfaction with occupational performance: Improvement of 2 points in 24 of 34 MIA (p 5 0.0.5) for 6 of 8 participants. Group level: Z ¼ 2.38 (p 5 0.05). Measured by the COPM: higher score, better performance and higher satisfaction with performance. NS change in: sleep and rest, eating, home management, recreation and work. Measured by the SIP. Quality of life: Self-perceived quality of life improved: Z ¼ 2.19 (p 5 0.05). Measured by a VAS. Psychosocial function increased Z ¼ 2.38 (p 5 0.05). Measured by the SIP. User satisfaction: not investigated.

Disabil Rehabil Assist Technol Downloaded from informahealthcare.com by Cea Fontenay Auz Roses on 02/11/11 For personal use only.

6 A˚. Brandt et al.

Cohort with no baseline T1: time after acquisition NR T2: 6–9 mo after T1 Comparison group with no experience of ECS use: not included in this review

Same as Jutai et al.

Cross sectional study

Jutai [7] (Same study as Stickel et al. [39])

Stickel [39] (Same study as Jutai et al. [7] )

KantoRonkanen [34]

Study

Study design and length of follow-up

Table I. (Continued).

Technology: Eleven different of ECS brands and types, ranging from direct access systems to integrated voice access systems for call system, phone, audio-visual equipment, door opener, lighting, computer, heating and ventilation, powered bed. Implementation process: NR.

Same as Jutai et al

Technology: Different individually selected ECS for controlling door opening (100%), lights (91%), window opening (74%), domestic appliance (74%), phone (70%), door intercom (39%), helper-alarm (35%). Implementation process: NR.

At T1 n ¼ 20; at T2 n ¼ 15 Mean years of age: 31.1 years (range 18–55 years)/32.9 years (range 21–55 years). T1/T2: Diagnosis: DMD: (n ¼ 14)/(n ¼ 10); SMA: (n ¼ 5)/(n ¼ 4); MD: (n ¼ 1)/ (n ¼ 1) Functioning: 57.2/57.0 (FIM: higher score better performance).

Users n ¼ 23, carers n ¼ 24 (assisting 17 users, whether formal or informal not reported) Mean years of age of users NR, range 20–67 years Mean years of age of carers NR, range 21–78 years Diagnosis: NR Functioning: NR

Intervention/control intervention

At T1 n ¼ 20; at T2 n ¼ 16 Mean years of age: T1 31 + 10.1/T2 32.4 + 10.6. T1/T2: Diagnosis: DMD: (n ¼ 14)/(n ¼ 11); SMA: (n ¼ 5)/(n ¼ 4); Other MD: (n ¼ 1)/(n ¼ 1) Functioning: Total score 57.1/57.1; motor function* 23.3/23.0; cognitive skills{ 33.8/34.1. (FIM: higher score better performance).

Speech: normal (n ¼ 3), slurred but understood (n ¼ 12), weak voice (n ¼ 1). Vision: poor (n ¼ 12), good (n ¼ 4)

Study population

Outcomes

(continued)

Activity/participation: Qualitative data: neighbours can visit; can open door to friends; can meet with friends. Increased independence in going in/out and opening window. Quality of life and user satisfaction: not investigated.

User satisfaction: Mean global satisfaction: T1: 4.16; T2: 3.89. Highest mean satisfaction: T1: 4.40–4.00 (Simplicity of use; maintenance; weight; safety; dimensions). T2: 4.39– 4.14 (safety, installations, simplicity of use, multipurposefulness, training) Lowest mean satisfaction: T1: 2.82–3.58 (cost, follow-up services, appearance, service delivery, device compatibility) T2: 2.62–3.64 (cost, repairs/servicing, transportation, follow-up services, device compatibility). Measured by the QUEST 1.0: higher score better satisfaction. Quality of life: See Jutai et al.

Quality of life: Psychosocial impact stable over time: competence: 1.83/1.79. Adaptability: 1.42/1.39; selfesteem: 1.46/1.55 (NS). Competence scores were higher than adaptability (p ¼ 0.002) and self-esteem (p ¼ 0.016). Measured by the PIADS: higher score more positive psychosocial impact. Activity and participation: not investigated. User satisfaction: See Stickel et al. [39]

Disabil Rehabil Assist Technol Downloaded from informahealthcare.com by Cea Fontenay Auz Roses on 02/11/11 For personal use only.

Environmental control and smart home technology 7

Crosssectional study

Controlled study

Crosssectional study including a

Maguire et al. [35]

Rigby et al. [36]

Ripat [37]

Study

Study design and length of follow-up

Table I. (Continued).

Technology: ECS activated by a single switch most common system. Most systems were in operation in the living room and bedroom and had 6–10 functions. Implementation process: 20% had a practical trial of ECS before it was installed in their home.

Technology: ECS that controlled at least two devices such as emergency alert device, specialised telephone, door openers, lights, or stereo. Implementation process: NR. Control intervention: No intervention

Technology: ECS that can operate any of the typically controlled devices in a home environment. Implementation process: Systems customised to the individual user’s needs.

n ¼ 32 ECS users (n ¼ 16)/ECS nonusers (n ¼ 16) Mean years of age 39.7/39.8 (range 21–67/23–60) Diagnosis: SCI: C3/4 and above (n ¼ 4/2); C4/5 (n ¼ 9/6); C5/6 (n ¼ 2/7); C6/7 (n ¼ 1/1) Functioning: NR

n ¼ 15 experienced users (former study new users n ¼ 7) Mean years of age study participants 50 + 15 (former study 50 + 9.6)

Intervention/control intervention

n ¼ 46 Mean years of age 42 years (range 17–68 years) Diagnosis: MS (n ¼ 15); tetraplegia (n ¼ 12); CP (n ¼ 6); motor neuron disease (n ¼ 3); Friedreich’s ataxia (n ¼ 3); MD (n ¼ 3); RA (n ¼ 1) Syringomyelia (n ¼ 1); traumatic brain injury (n ¼ 1); Arthrogryposis multiplex congenital (n ¼ 1) Functioning: All severely disabled: mean 2.6 (range 0–9) (Barthel-20: higher score better performance).

Study population

Outcomes

(continued)

Activity/participation: (Only reported for experienced users): Mean performance 9.0 ( + 1.13); mean satisfaction 9.0 (+1.14). The two scores highly correlated (r ¼ 0.82, p ¼ 0.0002). Measured by the

Activity/participation: ECS users had statistically significantly greater functional abilities for 75% of the tasks compared with nonusers (p 5 0.05): identify caller and control admission, summon help, converse using the phone, correspond in writing, have ability to position self, control lighting, control AV equipment, engage in hobby or recreation activity, perform voluntary or paid work at home. No difference in: leave or re-enter home; operate phone; control heating and air condition. Measured by the LOMEC. ADL, mobility and communication: no difference between ECS users and non-users. The users were more autonomous in IADL than non-users (p ¼ 0.02). Measured by the SMAF. More users than non-users lived alone. Quality of life: (only ECS users, non-users excluded): Competence: mean 2.02; adaptability: mean 1.66; selfesteem: mean 2.03. Measured by the PIADS: higher score more positive psychosocial impact. User satisfaction: not investigated.

Quality of life: 93% of users felt that the ECS improved their quality of life. Measured by a SSQ. (Secondary aim of study. Primary aim was to evaluate provision of ECS: is not reported in the present review.) Activity and participation and user satisfaction: not investigated.

Disabil Rehabil Assist Technol Downloaded from informahealthcare.com by Cea Fontenay Auz Roses on 02/11/11 For personal use only.

8 A˚. Brandt et al.

Mixed methods: interrupted time series study{ and qualitative data. One month and 1 wk prior to ECS and at month 1, 3 and 6 after acquisition

Randomised controlled study

Sundar et al. [40] &

comparison with a former study (Ripat and Strock)

Ripat & Strock [38]

Study

Study design and length of follow-up

Table I. (Continued).

Former study see Ripat and Strock [38]

Technology: Customised ECS. All used same ECS and different types of input devices. Number of devices controlled varied from 3 to 8. Implementation process: Devices and switches were selected by occupational therapist who trained participants in use and assisted with troubleshooting.

Technology: SHT (X-10 Active Home kit and other modules): computer (100%); ActiveHome software (100%); lighting (95%); wall switch (31%); remote control (100%); chime for security or medication (87%); motion detector (49%); coffee maker (59%); power flash for window/door security (74%).

n¼7 Mean years of age 50.6 years (range 39–62 years) Diagnosis: MS (n ¼ 3); SCI (n ¼ 3); MD (n ¼ 1) Functioning: NR

At baseline n ¼ 113, at follow-up n ¼ 78.

Intervention/control intervention

Years of experience with ECS: study participants 10 + 7.8 (range 1–25) (former study 3 months) Diagnosis: NR (former study see Ripat and Strock [38]) Functioning: NR except 73% had non progressive disability (former study 43%)

Study population

Outcomes

(continued)

Activity/participation: IADL: Baseline/3 years follow-up: IG 11.9/11.8, CG 12.3/11.9 (difference p50.05). Measured by the OARS: higher score more impairment.

Quality of life: Total score, the adaptability, competence, and self-esteem subscale scores were positive at all five measurement occasions (range 0.08–2.83) (differences NS). A post hoc analysis showed lowest level of adaptability and self-esteem at T3 (p 5 0.05), and a NS increase at T4 and T5. Competence NR. Measured by the PIADS (higher score more positive psychosocial impact). Activity and participation and user satisfaction: not investigated.

COPM: higher score, better performance and higher satisfaction with performance. Important tasks facilitated by functions facilitated by ECS use: 28% increase/maintain independence; 25% entertainment; 20% communication of basic needs; 10% socialization; 9% to meet safety needs; 6% productivity functions. Quality of life: experienced users compared with new users from former study NS: total: 1.85/1.59, adaptability: 1.61/1.52, competence: 2.00/1.70, self esteem: 1.85/1.47. Measured by the PIADS: higher score more positive psychosocial impact. User satisfaction: not investigated.

Disabil Rehabil Assist Technol Downloaded from informahealthcare.com by Cea Fontenay Auz Roses on 02/11/11 For personal use only.

Environmental control and smart home technology 9

Pre-post study 3 months

Vincent et al. (42)

Implementation process: Technology was selected and adapted; users were trained in use of the technology. In case of problems with the products or the PC, a specialised nurse solved the problem. In average 5 visits the first year and 3 the second. Control intervention: No intervention

Technology: Prototype ECS. Applications: Remote control functions, e.g., door lock release, light switch control, hands-free phone, electric bed control; verbal reminders, e.g., stove will be turned off, take your pills, time and date; automatic functions, e.g., lights, temperature, stove turns off, alarm system. Implementation process: Needs assessment by health care professional.

Users n ¼ 5; Informal caregivers n ¼ 5 Mean years of age 59.6 (range 20–82) Diagnosis: Fragile health (n ¼ 1); Spina bifida (n ¼ 1); Parkinson’s disease (n ¼ 1); Severe arthritis and poor sight (n ¼ 1); Quadriplegia (n ¼ 1) Functioning: Different levels of walking difficulties (n ¼ 5); slight memory disturbances (n ¼ 2); Cannot use hands (n ¼ 2)

Intervention/control intervention

Selected/baseline/follow-up: Sundar et al: IG n ¼ NR/46/33; CG n ¼ NR/ 67/43 Tomita et al: IG n ¼ 53/46/34; CG n ¼ 71/67/44 Mean years of age: IG 72.0; CG 75.6 (p50.05) Diagnosis: Arthritis, hypertension, cataracts, cardiovascular disease etc. Functioning IG/CG: Motor function (FIM: higher score better performance) 78.21/79.95. Movement function (SIP: lower score better performance) 104.85/ 80.89. Cognitive function (MMSE: higher score better performance) 29.74/29.43. Cognitive function (FIM) 34.03/ 33.73.

Study population

Activity/participation: Results for each participant and each item were presented, but not statistics at group level. A tendency was increased satisfaction with activity performance (meal preparation, sleep, excretory hygiene, health care, household; telecommunication not clear result). Measured by 13 items of the LIFE-H. User satisfaction: Mean satisfaction 4.13 (SD + 0.55), range 3.00–5.00. Measured by the QUEST 1.0: higher score better satisfaction. Quality of life: not investigated.

Living at home after 3 years: IG 80.4%, CG 65.7% (p ¼ 0.043). Quality of life and user satisfaction: not investigated.

Outcomes

ADL: activities of daily living; CG: control group; COPM: Canadian occupational performance measure; CP: cerebral palsy; DMD: Duchenne muscular dystrophy; ECS: environmental control system; FIM: functional independence measure; LIFE-H: life habits; IADL: instrumental activity of daily living; IG: intervention group; LOMEC: The Lincoln Outcome Measures for Environmental Controls and Audit of Installation Quality; MD: muscular dystrophy; MMSE: mini mental state examination; MS: multiple sclerosis; NR: not reported; OARS: The Duke Older Americans Resources and Services Procedures; PIADS: The Psychosocial Impact of Assistive Devices Scale; QUEST: The Quebec User Evaluation of Satisfaction with assistive Technology; RA: rheumatoid arthritis; RBMT: Rivermead Behavioural Memory Test; SCI: spinal cord injury; SHT: smart home technology; SIP: Sickness Impact Profile; SMA: spinal muscular atrophy; SMAF: The Functional Autonomy Measuring Scale; SSQ: study specific questionnaire; VAS: Visual Analogue Scale. *FIM does not measure motor function and movement function; therefore it is not certain what is measured. { Cognitive skills are reported, but FIM measures cognitive function. { Interrupted time series study: The study compares anticipated psychosocial impact (T1, T2) with the experienced psychosocial impact (T3, T4, T5) and is thus regarded as a descriptive study.

2 years

Tomita et al. [41]

Study

Study design and length of follow-up

Table I. (Continued).

Disabil Rehabil Assist Technol Downloaded from informahealthcare.com by Cea Fontenay Auz Roses on 02/11/11 For personal use only.

10 A˚. Brandt et al.

Disabil Rehabil Assist Technol Downloaded from informahealthcare.com by Cea Fontenay Auz Roses on 02/11/11 For personal use only.

Environmental control and smart home technology the studied tasks better than non-users. The pre-post study by Boman et al. [31] in which the participants were persons with brain injury showed that accomplishment of and satisfaction with performance of the most important activities increased after the ECS intervention. In the other two pre-post studies, no statistical analyses were performed and only results for each study participant were presented, but the results indicated that ECS intervention increased independence of the study participants [32], and that user satisfaction with performance of some activities was improved [42]. In the descriptive study by Harmer and Bakheit, participants compared their ability to perform a number of tasks with and without use of ECS [33] resulting in improvement in performance of less than half of the investigated tasks. In addition, a fourth of the participants found that ECS had made it possible for them to live alone and that they were able to socialise more. In the study by Ripat [37], experienced ECS users rated their performance of and satisfaction with their performance of important tasks using ECS at a high level, and 6–28% found that a number of important tasks were facilitated by ECS. In the descriptive study by Kanto-Ronkanen et al., qualitative data from users and carers showed improvement in socialising activities, independence in going in/out of the house and in opening windows [34]. Finally, in the discussion of the article by Ripat and Strock [38], it was stated in the discussion that the ability of the participants to perform activities was improved; however, no data were presented to support this. Quality of life as psychosocial impact. Only one of the effect studies concerned quality of life. Boman et al. [31] found improved self-perceived quality of life for participants with acquired brain injury and increase of psychosocial function after the ECS intervention. Four descriptive studies investigated psychosocial impact (competence, adaptability and self-esteem [7,36,37,39,44]) by means of the PIADS [43] (see Table I). The outcomes were positive and similar, and moreover, the study by Jutai [7] found that the study participants scored higher in competence than in adaptability and self-esteem. Two other descriptive studies: Harmer and Bakheit found that ECS increased feelings of self-worth, happiness and control over the environment for most participants [33], and Maguire et al. [35] found that nearly all study participants felt that the ECS had improved their quality of life, which was not specifically defined. User satisfaction with the technology. Two descriptive studies investigated the satisfaction of study participants with ECS, both by means of the QUEST 1.0. [17]. Vincent et al. [42] showed that the mean

11

satisfaction was high among five ECS users and Stickel et al. [39] showed that the global satisfaction (one item) of the study participants was high. In addition, Stickel et al. found that users were most satisfied with the effectiveness of ECS and least satisfied with cost [39]. See Table I.

Methodological quality of the included studies The internal quality of the studies varied substantially with no study fulfilling all nine criteria and most fulfilling 50–75% of the criteria (Table II). The study quality was best concerning adequate reporting of selection of population and psychometric properties of used assessment instruments, and congruence between aims and outcome measures and data presentation, respectively, in that nearly all studies fulfilled these criteria. Only a little more than half of the studies that followed study participants over time fulfilled the criteria concerning at least 4 month’s follow-up time, less than 20% drop-outs, and a sufficient description of the characteristics of the drop-outs. The study quality was very low concerning sufficiently large study size, which only the RCT had [40,41], and confounder analysis was not performed in any of the studies. The specific assessment of the methodological quality of the RCT [40,41] showed that it was low with only two of 11 criteria fulfilled, mainly because both articles failed in reporting basic information about methods of randomisation, treatment allocation and blinding (Table III). Only one effect study fulfilled all four external validity criteria [31] and two descriptive studies fulfilled three criteria possible to fulfil for this study design [7,33]. The remainder studies each failed to fulfil one criterion except from one study that did not fulfil any [34]. The studies mainly failed in fulfilling the criteria concerning description of participants and intervention/setting sufficiently well (Table IV). As to effect size, it was considered clinically important in three of the five effect studies.

Discussion Despite the fact that five effect studies supplemented with six descriptive studies were identified, it was not possible to determine whether ECS and SHT interventions improve activity and participation among persons with impairment. The main reason for this was that the studies differed regarding populations, interventions and instruments used, making it impossible to synthesise and compare studies. Besides, the level of evidence was low with insufficient fulfilment of quality criteria and a majority of descriptive studies. Main problems were

0 0 NA NA

NA NA NA NA 0 1

0 FU: 1 NF: 4 NA: 6

1

0 1

1

1

1 1 1 1

1

1 FU: 10 NF: 1

Study size 4 10 patient years

0 FU: 3 NF: 2 NA: 6

1

NA NA NA 1

NA

NA

0 NA

1

Follow up  4 months

NA: not applicable; FU: Fulfilled; NF: Not fulfilled. *If the study fulfilled the criterion it was assigned ‘1’, if not ‘0’.

Boman et al. [31] Croser et al. [32] Harmer & Bakheit [33] Jutai et al. [7] & Stickel et al. [39] Kanto-Ronkanen et al. [34] Maguire et al. [35] Rigby et al. [36] Ripat [37] Ripat & Strock [38] Sundar et al. [40] & Tomita et al. [41] Vincent et al. [42] In all

Reference

Selection of population adequate

1 FU: 3 NF: 3 NA: 5

0

NA NA NA 0

NA

0

1 NA

1

Dropouts 5 20%

1 FU: 4 NF: 2 NA: 5

0

NA NA NA 0

NA

1

1 NA

1

Description of dropouts

1 FU: 10 NF: 1

1

1 1 0 1

1

1

1 1

1

Outcome measures congruent with aims

1 FU: 10 NF: 1

1

1 1 0 1

1

1

1 1

1

Data presentation congruent with aims

0 FU: 10 NF: 1

0

0 0 0 0

0

0

0 0

0

Confounders adjusted in the analysis

Table II. Summary of internal validity assessment of included studies (modified from Borghouts et al. [27])*

Disabil Rehabil Assist Technol Downloaded from informahealthcare.com by Cea Fontenay Auz Roses on 02/11/11 For personal use only.

1 FU: 0 NF: 11

1

0 1 1 1

0

1

0 1

1

Psychometric properties of instruments reported

6/9

6/9

3/5 4/5 2/5 5/9

3/5

5/7

4/9 4/5

7/9

Fulfilled criteria

12 A˚. Brandt et al.

Environmental control and smart home technology

13

Table III. Study quality of the identified randomised controlled trial (Sundar et al. [40] and Tomita et al. [41]) (Van Tulder et al. [28]). Appraisal (Yes/No/Do not know/not applicable)

Criterion

Disabil Rehabil Assist Technol Downloaded from informahealthcare.com by Cea Fontenay Auz Roses on 02/11/11 For personal use only.

Was the method of randomisation adequate? Was the treatment allocation concealed? Were the groups similar at the baseline regarding the most important prognostic indicators? Was the patient blinded to the intervention? Was the care provider blinded to the intervention Was the outcome assessor blinded to the assessor? Were co-interventions avoided or similar? Was the compliance acceptable in all groups?

Do not know Do not know No (control group older, more illnesses)

Was the drop-out rate described and acceptable? Was the timing of the outcome assessment similar in all groups? Did the analysis include an intention-to-treat analysis?

Do not know Do not know Do not know Do not know No (7 of 53 selected for the treatment group did not want to use the high-tech intervention and 4 of 71 selected for the control group dropped out because they were not selected for the treatment group) (Not reported in Sundar et al. [40]) Yes: described, but not acceptable Yes No

Table IV. Summary of external validity and applicability assessment of included studies (adopted from Shekelle et al. [29]).

Reference Boman et al. [31] Croser et al. [32] Harmer & Bakheit [33] Jutai et al. [7] and Stickel et al. [39] Kanto-Ronkanen et al. [34] Maguire et al. [35] Rigby et al. [36] Ripat [37] Ripat and Strock [38] Sundar et al. [40] and Tomita et al. [41] Vincent et al. [42]

Study participants Intervention and Clinically relevant outcomes Size of the effect clinically described in detail setting described in detail measured and reported important (at least 10% gain) Yes No Yes Yes

Yes Yes Yes Yes

Yes Yes Yes Yes

Yes Yes NA NA

No Yes Yes No No Yes

No No No No No Yes

No Yes Yes Yes Yes Yes

NA NA Yes NA NA No

Yes

Yes

Yes

NP

NA: Non applicable; NP: Not possible to assess Criteria assessed: 1. Are the patients described in enough detail so that you can decide whether they are comparable to those that you see in your practice? 2. Are the interventions, control interventions, and treatment settings described sufficiently to enable you to provide the same for your patients? 3. Are the measured and reported outcomes clinically relevant? 4. Is the size of the effect clinically important (at least a 10% gain)?

small study sizes and lack of confounder analysis. However, despite these shortcomings the ECS and SHT interventions tended to facilitate independence of the study participants, instrumental activity of daily living, socialising and quality of life. User satisfaction was only investigated in two studies, but was found to be high. The identified studies varied substantially in terms of participant characteristics such as age, impairments and diagnosis. All interventions except for one [40,41] were categorised as ECS, but the functionality of ECS and SHT seemed to be overlapping; hence, differences between the two kinds of technologies have not been considered. Still, the applications

and their combinations varied among studies, and services differed in terms of adaptation, user training and follow-up. Even though it was not possible to synthesise the outcomes of the identified studies, the ECS and SHT seem to be promising technologies, since study participants tended to increase their abilities to perform the tasks that ECS and SHT interventions aim at enabling, such as identifying callers, locking and unlocking entrance doors, controlling lighting and controlling audiovisual equipment [33,36]. The technologies especially enabled IADL and social activities [31,36,37,40,41] and to a lesser degree basic ADL [31,36]. Furthermore, the interventions tended to

Disabil Rehabil Assist Technol Downloaded from informahealthcare.com by Cea Fontenay Auz Roses on 02/11/11 For personal use only.

14

A˚. Brandt et al.

enable persons with impairment to continue living in a private home as shown by the RCT [40,41], supported by two descriptive studies, which demonstrated that the users’ independence had increased [38] and the need for care had decreased [39]. The fact that one study showed that users with cognitive impairment experienced an increased quality of life [31] and that four studies showed that the interventions had a positive impact on the psychosocial function of users with physical impairment indicates that ECS and SHT interventions have the potential to increase the quality of life for persons with impairment [7,36–38]. In the four studies on the psychosocial impact of ECS and SHT interventions, the impact was higher than in findings in studies on other kinds of technological interventions, e.g., spectacles [44]. Psychosocial impact was measured by means of the PIADS, an instrument especially developed for assessing psychosocial impact of assistive device interventions [43], which probably explains its frequent and dominant use in the identified studies. Only two studies on satisfaction with the interventions was found [39,42], one of which was rather small [42]. So, even though the level of satisfaction was found to be high in two studies, it can hardly be concluded that user satisfaction with ECS and SHT interventions was high. It was interesting that one of the studies showed low satisfaction with cost, followup services and device compatibility, indicating that the technology does not always work well, requiring professionals to follow the users over time to ensure that the device continues to function [39]. Formerly, no systematic review on outcomes of ECS has been performed, but after we initiated our review Martin et al. [45] published a study on effects of SHT, which did not identify any study, while the present review identified five effect studies. Reasons for this discrepancy are easily identified. First, we included ECS [31,32,36,42]; second, our search covered all of year 2007 [41] and third, we also handsearched conference proceedings [40]. Still, the number of effect studies was disappointingly low, especially since ECSs have been available for many years, and therefore, more studies could be expected. One explanation may be that we did not include passive alarms, monitoring or telecare systems; technologies that presently attract much attention. But this is probably not the explanation for the lack of effect studies, since Demiris and Hensel [46] conducted such a review and identified numerous studies, but then again none of these were effect studies. A similar lack of effect studies was found in a systematic review of activity and participation outcomes of mobility device interventions, in which only eight studies were identified in spite of the fact that mobility device interventions are widespread [25].

One reason for lack of effect studies may be that assistive devices apparently work well for the users and therefore do not seem to require effect studies. Another reason may be that assistive devices, to a higher degree than, e.g., medical interventions, are based in the social sector where the demand for evidence of effects of interventions exists to a lesser degree than in the medical sector, and thus there is less tradition for effect studies within the field of AT.

Methodological study quality The RCT design is regarded as the optimal design in outcomes research [47], but only one such study was identified, and unfortunately the study quality was rather low, reducing its level of evidence. In addition, the number of study participants was limited, encompassing less than 50 study participants, and in eight of the 11 studies the number was less than 20, which hampers the generalisability of the findings and may cause insufficient study power for determining effects. The problem of small study sizes was further deteriorated by too short follow-up time: in the two pre-post studies, it was 2 wk [32] or 3 mo [42], with study sizes of eight and five users, respectively, resulting in too small study sizes. Besides more than half of the studies were descriptive, representing a low level of evidence. The level of evidence of the identified trend towards positive outcomes is therefore regarded as low. One reason for the small study sizes may be that the target group is mainly persons with impairments, perhaps making it difficult to recruit a sufficient number of study participants. In addition, a large drop-out rate is often a problem in disability cohort studies, because individuals in the target group often get illnesses; this is especially problematic in studies that follow the participants over a longer period of time [48]. This kind of difficulty is illustrated by the identified RCT [40,41], in which 113 persons within the inclusion criteria initially joined the study, but after 2 year only 78 persons were left. It was problematic that confounder control had not been carried out in any of the studies, nor had there been an attempt to avoid bias in other ways, except that control groups were included in two studies. In small studies, it is not possible to perform confounder control, but in the RCT [40,41], it could have been a method for controlling for age and illness differences in the intervention and control groups, which occurred due to different drop-out rates in the two groups. Even though the special appraisal of the RCT study quality showed that it was low, mainly because basic information about study methodology was lacking in both articles and information about the study was not totally identical in the two papers,

Environmental control and smart home technology

Disabil Rehabil Assist Technol Downloaded from informahealthcare.com by Cea Fontenay Auz Roses on 02/11/11 For personal use only.

e.g., the number of study participants differed, the study quality of the RCT after all was better than the other identified studies by having a randomised control group, a sufficiently large sample size and a long follow-up time. In only two [31,33] of the 11 studies ethical considerations and approval were reported, which is surprising since it is a usual requirement when publishing empirical studies, and it is troubling that is has been possible to publish articles without reporting on ethical issues.

Strengths and weaknesses of the review The review was carried out in a systematic and transparent way with no language limitations, and by applying a comprehensive search in databases as well as hand search and snowballing; it resulted in a large number of studies that were screened by two researchers independently. We believe that we have found nearly all studies within our inclusion criteria. As opposed to most systematic reviews, this review did not have any restrictions concerning study design. One reason was that we expected to identify only a few studies, and therefore, we needed to include all studies that could elucidate this under-researched area following the principle that even though evidence may be weak, there is always some information which should be used in decision making [49]. In fact, the descriptive studies did offer some information about important outcomes representing user perspectives of and experiences with AT such as psychosocial impact of and satisfaction with AT interventions, which effect studies do not necessarily provide. Our study focused on outcomes regarding activity and participation, quality of life and user satisfaction. However, this was not the main focus of all the identified studies and was only a minor aspect in some studies. For example, the aim of one study was to evaluate the provision of ECS, while assessing quality of life outcomes of AT only was a secondary aim [35], and in another the focus was to investigate the ability of persons with cognitive impairments to learn and use ECS [31]. Subsequently, the studies were not primarily designed as outcomes studies and it may not be totally fair to judge them as such. However, the studies were within our inclusion criteria and did provide useful information on ECS and SHT outcomes.

15

satisfaction outcomes of ECS and SHT interventions was weak and no recommendations as such can be given, still there was a tendency towards positive outcomes, indicating that ECS and SHT interventions most likely are beneficial for some individuals with impairment and the interventions should be considered. However, it is important to follow ECS and SHT interventions over time in the users’ homes since continuous adaptation and technological support often seem to be required [39]. It is evident that there is a need for high quality studies on outcomes of ECS and SHT interventions. The RCT study design for effect studies should be aimed at [47], but the design is often challenging [48]. One such challenge is constituted by user preferences and is illustrated by the included RCT [40,41]. In this study, random allocation was applied to avoid systematic bias and differences between intervention and control group [47], but was corrupted because the preferences of the study participants differed between the experimental and control intervention, resulting in different drop-out rates in the two groups. Another problem with the RCT design concerns ethics, since many kinds of AT have been available for a long time, for example ECS, which persons with impairments in many countries are eligible to receive for free if they are considered to benefit from the intervention (e.g., in the Nordic countries). Subsequently, for ethical reasons in these countries it is only possible to perform studies that compare ECS or other kinds of AT with services or devices that are expected to result in better outcomes than usual services [50,51]. In addition, more knowledge about the magnitude of effects is required regarding for whom the AT in question works and for whom not, negative consequences, postponement of need for help, etc. Often obtaining such knowledge requires longitudinal studies [52], which unfortunately are bound to have high drop-out rates among participants with impairment [48]. Future research needs to enrol a large number of participants given the challenges described above. One solution is multi-centre studies, which can recruit larger groups of study participants and attract high quality researchers. However, specific funding is necessary for these purposes, and hopefully outcomes research of information and communication technologies will be a part of research programmes internationally, especially because much focus is on the future demographic development and subsequent strive for independent living [53].

Recommendations for praxis and future research

Conclusion

Even though the identified evidence concerning activity and participation, quality of life and user

The review identified five effect studies and six descriptive studies. One study was about effects of

Disabil Rehabil Assist Technol Downloaded from informahealthcare.com by Cea Fontenay Auz Roses on 02/11/11 For personal use only.

16

A˚. Brandt et al.

SHT and the others were about ECS, but since the functionalities of the technologies seemed to be overlapping, the findings probably apply both types of technologies. Applications and characteristics of the study participants differed among studies, even though most were younger than usual in AT outcome studies and most had physical impairments. ECS and SHT interventions tended to increase activity and participation, especially IADL and socialising, and study quality of life of the participants. In addition, user satisfaction was found to be high in two studies. Despite the positive outcomes, no recommendations can be given as such due to the diversity of the studies not making it possible to synthesises study results and because of low level of evidence: study quality was mostly insufficient, mainly in terms of small study sizes and lack of confounder control, and in addition, less than half of the identified studies were effect studies. There is a need for larger, high quality and preferably multicentre effect studies applying RCT designs or longitudinal cohort studies with baseline data and longer follow-up time.

Acknowledgements The authors thank the librarians at Stakes (now National Institute for Health and Welfare), Helsinki, Finland, for substantial help with the electronic searches. The authors also thank Jamie B. Vickers for reviewing the language of the manuscript. Declaration of interest: We are grateful to the Nordic Development Centre for Rehabilitation Technology (now Nordic Welfare Center) for financial support. We declare that we do not have any conflict of interest in the performed research. References 1. Shakespeare T. The social relations of care. In: Lewis G, Gewirtz S, Clarke J, editors. Rethinking social policy. London: Open University and Sage; 2000. pp 52–65. 2. Barker RN, Kendall MD, Amsters DI, Pershouse KJ, Haines TP, Kuipers P. The relationship between quality of life and disability across the lifespan for people with spinal cord injury. Spinal Cord 2009;47:149–155. 3. Cooper RA, Dicianno BE, Brewer B, LoPresti E, Ding D, Simpson R, Grindle G, Wang H. A perspective on intelligent devices and environments in medical rehabilitation. Med Eng Phys 2008;30:1387–1398. 4. Stenberg L, Matthiassen N-E, Jordansen IK, Salminen AL, Kotiranta P, Palsdottir B, Mørk T, Flø R, Leczinsky C, Estreen M, editors. Provision of assistive technology in the Nordic countries. 2nd edition. Va¨llingby, Sweden: NUH – Nordic Centre for Rehabilitation Technology, 2007. 5. International Organization for Standardization. Assistive products for persons with disabilities – classification and terminology. DS/EN ISO 9999: 2007(E). Geneva: ISO Copyright Office; 2007.

6. Lange ML. The future of electronic aids to daily living. Am J Occup Ther 2002;56:107–109. 7. Jutai J. Psychosocial impact of electronic aids to daily living. Assist Technol 2000;12:123–131. 8. Smart house definition. Internet. 2010. http://architecture. about.com/od/buildyourhous1/g/smarthouse.htm. Last accessed June 2010. 9. Petterson I. Significance of assistive devices in the daily life of persons with stroke and their spouses [Doctoral dissertation]. ¨ rebro, Sweden: O ¨ rebro University; 2006. O 10. WHO. International classification of functioning, disability and health: ICF. Geneva: WHO; 2001. 11. Jutai J, Fuhrer MJ, Demers L, Scherer M, DeRuyter F. Toward a taxonomy of assistive technology device outcomes. Am J Phys Med Rehabil 2005;84:294–302. 12. Roland M, Torgerson DJ. Understanding controlled trials: what are pragmatic trials? Br Med J (BMJ) 1998;316:285–286. 13. Gartlehner G, Hansen RA, Nissman D, Lohr KN, Carey TS. Criteria for distinguishing effectiveness from efficacy trials in systematic reviews. Technical review 12 (prepared by the RTIInternational-University of North Carolina evidence-based practice center under contract no. 290-02-0016). AHRQ publication no.06-0046. Rockville, MD: Agency for Healthcare Research and Quality; 2006. ˚ , Kreiner S, Iwarsson S. Mobility-related participa14. Brandt A tion and user satisfaction: Construct validity in the context of powered wheelchair use. Disabil Rehabil Assist Technol 2010;5:305–313. 15. Harper A, editor. WHOQOL-BREF. Introduction, administration, scoring and generic version of the assessment. Field trial version 1996. Geneva: Programme on Mental Health, World Health Organization. pp 7. 16. Rust KL, Smith RO. Satisfaction with assistive technology: what are we measuring? In: Resna, editor. Conference proceedings from the 27th International Conference on Technology & Disability. Arlington, US: Resna; 2004. 17. Demers L, Weiss-Lambrou R, Ska B. Quebec user evaluation of satisfaction with assistive technology QUEST version 2.0. An outcome measure for assistive technology devices. Webster, NY: The Institute for Matching Person & Technology; 2000. 18. Chan M, Este`ve D, Escriba C, Campo E. A review of smart homes—present state and future challenges. Comput Methods Programs Biomed 2008;91:55–81. 19. Cheek P, Nikpour L, Heather N. Aging well with smart technology. Nurs Adm Q 2005;29:329–338. 20. Craig A, Tran Y, McIsaac P, Boord P. The efficacy and benefits of environmental control systems for the severely disabled. Med Sci Monit 2004;11:RA32–RA39. 21. Ding D, Cooper RA, Kaminski BA, Kanaly JR, Allegretti A, Chaves E, Hubbard S. Integrated control and related technology of assistive devices. Assist Technol 2003;15:89–97. 22. Topo P. Technology and the needs of people with dementia: a literature review. In: Eizmendi G, Azkoitia JM, Craddock G, editors. Challenges for assistive technology. Amsterdam: IOS Press; 2007. pp 254–258. 23. Schu¨nemann HJ, Jaeschke R, Cook DJ, Bria WF, El-Solh AA, Ernst A, Fahy BF, Gould MK, Horan KL, Krishnan JA, Manthous CA, Maurer JR, McNicholas WT, Oxman AD, Rubenfeld G, Turino GM, Guyatt G; ATS Documents Development and Implementation Committee. An official ATS statement: grading the quality of evidence and strength of recommendations in ATS guidelines and recommendations. Am J Respir Crit Care Med 2006;174:605–614. 24. Merlin T, Weston A, Tooher R. Extending an evidence hierarchy to include topics other than treatment: revising the Australian ‘level of evidence’. BMC Med Res Methodol 2009;9:34.

Disabil Rehabil Assist Technol Downloaded from informahealthcare.com by Cea Fontenay Auz Roses on 02/11/11 For personal use only.

Environmental control and smart home technology ˚ , Samuelsson K, To¨yta¨ri O, Malmivaara 25. Salminen A, Brandt A A. Mobility devices to promote activity and participation: a systematic review. J Rehabil Med 2009;41:697–706. 26. Cicerone KD. Evidence-based practice and the limits of rational rehabilitation. Arch Phys Med Rehab 2005;86:1073– 1074. 27. Borghouts AJ, Koes BW, Bouter LM. The clinical course and prognostic factors of non-specific neck pain: a systematic review. Pain 1998;77:1–13. 28. van Tulder M, Furlan A, Bombardier C, Bouter L; The Editorial Board of the Cochrane Collaboration Back Review Group. Updated method guidelines for systematic reviews in the Cochrane collaboration back review group. Spine (Phila Pa 1976) 2003;28:1290–1299. 29. Shekelle PG, Andersson G, Bombardier C, Cherkin D, Deyo R, Keller R, Lee C, Liang M, Lipscomb B, Spratt K, Weinstein J. A Brief Introduction to the Critical Reading of the Clinical Literature. Spine 1994;19(18)(Suppl):2028S– 2031S. 30. van Tulder M, Malmivaara A, Hayden J, Koes B. Statistical significance versus clinical importance: trials on exercise therapy for chronic low back pain as example. Spine 2007;32:1785–1790. 31. Boman IL, Tham K, Granqvist A, Bartfai A, Hemmingson H. Using electronic aids to daily living after acquired brain injury: a study of the learning process and the usability. Disabil Rehabil Assist Technol 2007;2:23–33. 32. Croser R, Garrett R, Seeger B, Davies P. Effectiveness of electronic aids to daily living: increased independence and decreased frustration. Aust Occup Ther J 2001;48:35–44. 33. Harmer J, Bakheit AM. The benefits of environmental control systems as perceived by disabled users and their carers. Brit J Occup Ther 1999;62:394–398. 34. Kanto-Ronkanen A, Leino E, Salminen A. The benefits of environmental control units in everyday life, view of users and helpers. In: Craddock G, McCormack L, Reilly RB, Knops H, editors. Assistive technology—shaping the future. Amsterdam: IOS Press; 2010. pp 596–601. 35. Maguire SM, McCann JP, Swallow M. An audit of the provision of environmental control systems in Northern Ireland, 1992–1997. Clin Rehabil 2001;15:320–323. 36. Rigby P, Ryan S, Joos S, Cooper B, Jutai J, Steggles E. Impact of electronic aids to daily living on the lives of persons with cervical spinal cord injuries. Assist Technol 2005;17:89–97. 37. Ripat J. Function and impact of electronic aids to daily living for experienced users. Technol Disabil 2006;18:79–87. 38. Ripat J, Strock A. Users’ perceptions of the impact of electronic aids to daily living throughout the acquisition process. Assist Technol 2004;16:63–72. 39. Stickel MS, Ryan S, Rigby PJ, Jutai J. Toward a comprehensive evaluation of the impact of electronic aids to daily living: evaluation of consumer satisfaction. Disabil Rehabil 2002;24:115–125. 40. Sundar V, Tomita M, Stanton K. Effects of currently available smart home technology on frail elders. In: Mann WC, Helal A, editors. Promoting independence for older persons with disabilities. Amsterdam: IOS Press; 2006. 41. Tomita MR, Mann WC, Stanton K, Tomita AD, Sundar V. Use of currently available smart home technology by frail elders: process and outcomes. Top Geriatr Rehabil 2007;23:24–34. 42. Vincent C, Drouin G, Routhier F. Examination of new environmental control applications. Assist Technol 2002; 14:98–111. 43. Day H, Jutai J. Measuring the psychosocial impact of assistive devices: the PIADS. Can J Occup Ther 1996;9:159–168. 44. Day H, Jutai J, Strong G. Visual aids retention and abandonment. Vision ‘99 Conference Proceedings, 1999.

17

45. Martin S, Kelly G, Kernohan WG, McCreight B, Nugent C. Smart home technologies for health and social care support. Cochrane Database Syst Rev 2008;4:CD006412. doi:10.1002/14651858.CD006412.pub2. 46. Demiris G, Hensel BK. Technologies for an aging society: a systematic review of ‘‘smart home’’ applications. IMIA Yearb Med Inform 2008:33–40. 47. Altman DG. Practical statistics for medical research. London: Chapman & Hall/CRC; 1991. 48. Kunz R, Oxman AD. The unpredictability paradox: review of empirical comparisons of randomised and non-randomised clinical trials. BMJ 1998;317:1185–1190. 49. Malmivaara A. Effectiveness and efficiency of assistive technology. STAKES, Helsinki, Finland: Nordic Assistive Technology Seminar. Outcomes of Assistive Technology – Supporting Practice and Development, 2008. 50. Lilford RJ. Ethics of clinical trials from a Bayesian and decision analytic perspective: whose equipoise is it anyway? BMJ 2003;326:980–981. 51. Djulbegovic B, Hozo I. At what degree of belief in research hypothesis is a trial in humans justified? J Eval Clin Pract 2002;8:269–276. 52. Schmoor C, Caputo A, Schumacher M. Evidence from nonrandomised studies: a case study on the estimation of causal effects. Am J Epidemiol 2008;167:1120–1129. 53. Bierhoff I, van Berlo A, Abascal J, Allen B, Civit A, Fellbaum K, Kemppainen E, Bitterman N, Freitas D, Kristiansson K. Smart home environment. In: Roe PR, editor. Towards an inclusive future. Impact and wider potential of information and communication technologies, COST 219ter. Amsterdam: IOS Press; 2007. pp 110–156.

Appendix 1. The Ovid MEDLINE search strategy 1.

2.

3.

4.

(‘assistive technolog*’ or ‘assistive device*’ or ‘assistive product*’ or ‘self help device*’ or ‘technical aid*’ or ‘empowering technolog*’ or robot* or ‘safety control system*’ or ‘environmental control system*’ or domotics).mp. [mp ¼ ti, ab, tx, kw, ct, ot, sh, hw, it, nm] (‘electronic aid*’ or ‘home environmental intervention*’ or robot* or ‘smart home technolog*’ or ‘information technolog* for assisted living at home’ or ‘home-based assistive technolog*’ or ‘smart house*’ or ‘smart home*’ or ‘electronic assistive technology’).mp. [mp ¼ ti, ab, tx, kw, ct, ot, sh, hw, it, nm] (outcome* or effect* or cost* or cost-effectiveness* or impact).mp. [mp ¼ ti, ab, tx, kw, ct, ot, sh, hw, it, nm] (activit* or occupation* or participation* or user satisfaction or quality of life* or independence or autonomy or security or safety or assistance or empowerment or ADL or PADL or IADL or self efficacy or control).mp. [mp ¼ ti, ab, tx, kw, ct, ot, sh, hw, it, nm]

A˚. Brandt et al.

18 5. 6.

Disabil Rehabil Assist Technol Downloaded from informahealthcare.com by Cea Fontenay Auz Roses on 02/11/11 For personal use only.

7. 8. 9.

10.

1 or 2 (home automation or remote control device*).mp. [mp ¼ ti, ab, tx, kw, ct, ot, sh, hw, it, nm] limit 6 to humans limit 7 to yr ¼ ‘‘1993 – June 2009’’ (assistive technolog* or assistive device* or assistive product* or self help device* or technical aid* or empowering technolog* or robot* or safety control system* or environmental control system* or domotics).mp. [mp ¼ ti, ab, tx, kw, ct, ot, sh, hw, it, nm] (electronic aid* or home environmental intervention* or robot* or smart home technolog* or information technolog* for assisted living at home or home-based assistive technolog* or smart house* or smart home* or electronic assistive technology).mp. [mp ¼ ti, ab, tx, kw, ct, ot, sh, hw, it, nm]

11.

12.

13. 14.

15. 16. 17. 18. 19. 20.

(outcome* or effect* or cost* or cost-effectiveness* or impact).mp. [mp ¼ ti, ab, tx, kw, ct, ot, sh, hw, it, nm] (activit* or occupation* or participation* or user satisfaction or quality of life* or independence or autonomy or security or safety or assistance or empowerment or ADL or PADL or IADL or self efficacy or control).mp. [mp ¼ ti, ab, tx, kw, ct, ot, sh, hw, it, nm] 9 or 10 (home automation or remote control device*).mp. [mp ¼ ti, ab, tx, kw, ct, ot, sh, hw, it, nm] limit 14 to humans limit 15 to yr ¼ ‘‘1993–2008’’ 3 or 4 5 and 17 limit 18 to humans limit 19 to yr ¼ ‘‘1993–2008’’ ¼ 4 6572