41 Disability and Assistive Technology Systems

At the same time it is useful to be aware of the physical effects of impairments in a context of diversity and equality. This means recognition of physical and other differences, but that everyone is entitled to equal rights and equal opportunities. In addition it should be recognised that there is not a particular norm or way of being human that is better than others, but that the full range of diversity is equally valid and valuable.

The social model of disability can also be used to identify the following two areas of responsibilities for engineers and designers:

1.Design for all; that is designing and constructing devices and environments to be accessible and usable by as wide a range of the population as possible, including disabled people.

2.Design of assistive technology systems, for example, the design of devices to overcome existing environmental and social barriers, thereby extending the opportunities and options open to disabled people.

It should be noted that design for all is much wider than design to include disabled people and it aims to include all (or as many as possible) of the different groups in the population regardless of factors such as age, gender, size, ethnic origin and disability. This is compatible with the social model of disability which looks at removing barriers, since design for all seeks to remove barriers for a wide range of social groups in addition to disabled people.

1.2 Assistive Technology Outcomes: Quality of Life

In terms of the social model the aim of assistive technology is to overcome the gap between what a disabled person wants to do and what the existing social infrastructure allows them to do. It consists of equipment, devices and systems that can be used to overcome the social, infrastructure and other barriers experienced by disabled people and that prevent their full and equal participation in all aspects of society. Measurement of the impacts and outcomes of a particular technology can be used to determine whether the desired benefits have been achieved and to inform further developments. In the case of assistive technology, improved outcome monitoring could have an impact on improving the device-user match and the quality of service provision, which is particularly important in view of the evidence of high rates of device abandonment (Phillips and Zhau 1993). It could also increase accountability (DeRuyter 1997) and provide additional evidence to support demands for increased public funding for assistive technology.

However, relatively little attention has been paid to the assessment of assistive technology outcomes for a number of reasons, including the belief that its benefits are obvious, a focus on the performance of the technology, rather than the interaction of the user with the technology and relatively little demand from stakeholders (Fuhrer et al. 2003). As measurement of assistive technology outcomes becomes more common, it will become increasingly important that the focus is on the production of relevant results rather than completing routine documentation and that

the procedures involved are not time consuming for either assistive technology users or practitioners. This section will discuss the existing techniques for measuring the quality of life resulting from assistive technology use, how useful these quality of life measures are, and whether additional measures should be developed.

1.2.1 Some General Issues

Measurement of assistive technology outcomes requires consideration of the combined human plus assistive technology system, rather than just the performance of the technology on its own. Outcome measures should be designed for the contexts in which assistive technology is actually used. They should also take account of the facts that individuals frequently make use of several assistive technology devices and that the impacts of these devices depend on the application context and type of assistive technology as well as end user characteristics (Gelderblom and de Witte 2002). Outcomes measures should be adaptable to allow measurement of both the specific impacts of a particular assistive device on the quality of life of a given disabled person and the impacts of a particular technology or type of technology on a given group of disabled people.

Both objective and subjective measurements are relevant. Objective measurements include measures of the performance of human assistive technology systems and services, such as the time taken by the user to read a data set or find a particular webpage using a screen reader. Subjective measures include subjective assessments of the user’s satisfaction with the device and services and subjective measures of any resulting change in quality of life. In some cases it may be difficult to isolate changes due to the assistive device from those due to other causes (Smith 1996), such as a change in circumstances. Thus it may, for instance, be difficult to determine how much an improvement in quality of life is due to the provision and regular use of an assistive device and how much is the result of a move to a new house closer to family and friends, an improvement in health or getting a job.

Studies on subjective quality of life are generally based on surveys. The nature of the survey process means that it is highly unlikely that all the respondents will complete all the questions. However, the issue of how responses with missing answers to some questions should be treated have not really been resolved. Frequently such responses are deleted, but this can reduce the sample size significantly, for instance by 18.3% for a data set of 10 variables with 2% of the data missing at random (Kim and Curry 1977). Another important issue is respondent and selection bias. Selection bias can probably be avoided by following an appropriate methodology. However respondent bias could be a problem and further research may be required to determine whether people who are satisfied with their assistive devices are more likely to reply to quality of life questionnaires than those who are dissatisfied or who have abandoned them.

There is also a need to use several sources of information and triangulate methodologies to increase the reliability and validity of assessments. The development of quality of life indicators should also involve disabled people to ensure that indicators cover the areas they consider to be important, which may not be the same as those considered important by researchers (Hughes et al. 1995).

61 Disability and Assistive Technology Systems

1.2.2 Definition and Measurement of Quality of Life

A definitive definition of quality of life has not yet been obtained. For instance, a review of 87 studies from diverse literature found 44 different definitions (Hughes et al. 1995). Any measurement system for quality of life should ideally be based on an established body of theory and supported by empirical evidence. However, theory in this area is not very well developed, though it has been suggested that Maslow’s concept of a hierarchy of needs could provide a suitable framework (Maslow 1968; Sawicki 2002).

There is a considerable body of work on health related quality of life which is based on a medical model of disability and either tacitly or explicitly assumes that impairment reduces quality of life. However, studies show that satisfaction with life increases with the extent of social integration, employment and mobility and is not correlated with the degree of (physical) impairment (Fuhrer et al. 1992; Fuhrer 1996). This is in accordance with the social model of disability, which stresses the need to overcome social barriers and social discrimination. However, since the measurement of health related quality of life was developed before specific outcome measures for assistive technology, it will be discussed in Section 1.2.3.

Other approaches, which will not be considered here, have developed from increasing awareness of the importance of sustainable development, as well as recognition of the inadequacy of economic measures such as gross domestic product (GDP) (Hersh 2006). Indicators of this type have been used in public policy making, for instance in measuring areas of deprivation, sustainability and regional economic development in the U.K. (Sawicki 2000). There may therefore be a need for equivalent measures of the impact of assistive technology to support public policy making in this area.

Two of the approaches to defining quality of life for disabled people using assistive technology will be considered here. In the first, Hughes et al. (1995) obtained a consensus list of the following 15 dimensions of quality of life from a study of the disability literature:

1.Social relationships and interaction

2.Psychological wellbeing and personal satisfaction

3.Employment

4.Self-determination

5.Autonomy and personal choice

6.Recreation and leisure

7.Personal competence

8.Community adjustment and independent living skills

9.Community integration and normalisation

10.Support services received

11.Individual and social demographic indicators

12.Personal development and fulfilment

13.Social acceptance, social status and ecological fit

14.Physical and material wellbeing

15.Civic responsibility

The second approach due to Schalock (1996) has the following eight dimensions:

1.Emotional wellbeing

2.Interpersonal relations

3.Material wellbeing

4.Personal development

5.Physical wellbeing

6.Self-determination

7.Social inclusion

8.Rights

The second approach has the advantages of being more compact and solely concerned with the impacts on the person, rather than how these impacts are obtained, for instance through work, leisure, civic participation or relationships. It also seems to avoid an explicit or implicit assumption that impairment automatically reduces quality of life, whereas the categories of personal competence and support services received are less likely to be included in a quality of life assessment for non-disabled people. However, the second approach may require an additional category of social identity and status. In addition further research would be required to ensure that the way in which the categories have been framed is culturally neutral and therefore appropriate for use in a wide range of different countries and cultures.

1.2.3 Health Related Quality of Life Measurement

Quality of life is frequently referred to in healthcare literature and quality of life measurements are increasingly used as an endpoint in clinical trials and intervention studies (Bowling 1995; Bowling and Windsor 1999). Health related quality of life studies focus on the measurement of symptoms and function, including levels of impairment, with the aim of assessing health status as part of a clinical measurement of outcome. Unfortunately, this approach with its focus on fulfilling ‘normal’ roles leads to an automatic devaluing of groups such as disabled or unemployed people and the assumption that disabled people cannot have a good quality of life, which is not in fact the case (Doward and McKenna 2004). Since the introduction of quality of life research in the healthcare context in the mid-1970s, the volume of publications has grown exponentially and there are now over a thousand quality of life citations each year (Fuhrer 2000).

There is still considerable disagreement about the definition and measurement of quality of life (Beckie and Hayduk 1997). However, this has generally been based on traditional health-status measures, under the assumption that they are equivalent

81 Disability and Assistive Technology Systems

to quality of life, and a functional perspective in relation to the ability to perform daily living activities, with disability both considered and measured as something negative (Bowling and Windsor 1999). For instance, a 1994 study of a random sample of 75 articles (Gill and Feinstein 1997) found that most of the assessments used areas of functioning consistent with the biomedical model (Fuhrer 2000), imposing a value system on respondents and deciding independently of them which areas of life were worth being measured. This criticism is supported by evidence of a lack of agreement between individuals’ ratings of their own healthrelated quality of life and those of health care providers or relatives (Slevin et al. 1988; Spangers and Aaronson 1992).

An approach, which is more compatible with the social model of disability and empowering to disabled people would replace the idea of unaided functioning by consideration of independence and autonomy. Here independence is understood in the sense of ‘control of their life and choosing how that life is led…(and) the amount of control they have over their everyday routine’ (Brisenden 1986). Autonomy is defined as the ability to plan one’s own life, to enter into relationships with others and together actively participate in the construction of society. These definitions are applicable to both disabled and non-disabled people. A non-disabled person can be non-autonomous if he or she has trouble in planning his or her life, whereas a disabled person who uses assistive technology or a personal assistant will be autonomous if this technology and/or assistant enable him or her to plan his or her life, enter into relationships and participate in society. Similar comments hold for independence.

Studies also indicate little correlation between individuals’ ‘subjective’ judgements of wellbeing and ‘objective’ measures of income, educational attainment and health status (Diener 1984; Eid and Diener 2004). Health related quality of life measures have the further disadvantage of being formulated in the context of people who are considered ‘patients’, that is, have been receiving treatment for a health complaint. However, many disabled people are not receiving any medical treatment and, for those who are, this is by no means their whole identity.

Subjective quality of life (sometimes called wellbeing) has been defined as ‘the degree to which people have positive appraisals and feelings about their life, considered as a whole’ (Fuhrer 2000). Specific measures include the Patient Generated Index (PGI) and the Satisfaction with Life Scale. In the PGI, respondents are first asked to list the most important areas of their life affected by their impairments(s) and then to evaluate the effect in each area and weight the relative importance of the different areas (Fuhrer 2000). Therefore, although the person defines the areas to be considered, the approach is very much based on the medical model of disability.

The quality of life instruments database currently includes 1000 items, with full descriptions given for 454 of them (WWW1 2006). The use of the term instrument indicates a system for the measurement or assessment, in this case of quality of life, which could include a mixture of quantitative and qualitative measures. However, the majority of the instruments are medically based and seem to relate quality of life to wellness, lack of disability and/or functionality. Many of the instruments are specific to populations with particular illnesses or impairments. There seem to