Evaluation of alternative computer input devices used by people with disabilities

Journal of Medical Engineering & Technology, Vol. 29, No. 3, May/June 2005, 119 129 Evaluation of alternative computer input devices used by people with disabilities G. TURPIN{, J. ARMSTRONG{, P. FROST{, B. FINE{, C. D. WARD and L. L. PINNINGTON {Derby Assistive Technology Evaluation Centre, Derby Hospitals NHS Foundation Trust, Uttoxeter Road, Derby, DE22 3NE, UK {AbilityNet, PO Box 94, Warwick, CV34 5WS, UK Division of Rehabilitation and Ageing, School of Community Health Sciences, University of Nottingham, UK The aim of this study was to evaluate a range of alternative computer input devices suitable for people with disabilities and to provide comparative data that will enable health care professionals and users to make informed choices when selecting products. Our focus was on the potential advantages and disadvantages of individual product features as they related to the abilities and needs of different users. A sample of 14 alternative keyboards and pointing devices commonly used by people with disabilities were appraised by 35 disabled adults. A multi-disciplinary panel of independent assessors also appraised the products. We identified key factors regarding the set-up, personal acceptability, ease of use, design features, compatibility and potential limitations of each device. We found that difficulties in accessing computers could sometimes be reduced or overcome by adjusting the existing workstation and customizing computer settings rather than through additional technology. However, successful computer access often requires a combined approach, as a single piece of equipment will rarely provide a complete solution. If alternative computer input devices are necessary, it is likely that the hardware settings will need customizing. 1. Introduction Computers are an increasingly important tool for social participation, education and recreation and have been integrated into everyday activities such as banking and shopping [1]. However, disabled people often have difficulty in accessing information technology [2, 3]. Among other factors, disabled people may encounter difficulties with the environment, data entry and information output, technical documentation or training and support [4]. When disabilities preclude conventional means of access such as a handoperated keyboard and mouse, alternatives must be sought, for example, special keyboards and pointers, switches, finger guards or wrist supports [1]. Successful use of such devices will depend on the degree to which the device or the product feature matches the needs of an individual user. Anson [5] has emphasized the importance firstly of compatibility and secondly of local support when selecting new hardware or software. Given the widening range of available products, it is likely that a good match will only be achieved if information about adaptive devices is categorized on the basis of the functional abilities for which they are appropriate [3]. Many rehabilitation professionals find it difficult to advise people who have problems with computer access. In the first place, professionals typically have limited knowledge or experience of the vast array of devices currently available or of new products that are appearing in the rapidly expanding information and communication technology market. In addition, many devices are supplied infrequently, are technically complex and have multiple setup options. There is a clear rationale, therefore, for increasing both professional and user awareness of the hardware and software options that are available to improve computer access [4]. *Corresponding author. E-mail: Gail.Turpin@derbyhospitals.nhs.uk Journal of Medical Engineering & Technology ISSN 0309-1902 print/issn 1464-522X online # 2005 Taylor & Francis Group Ltd http://www.tandf.co.uk/journals DOI: 10.1080/03091900500075317

120 G. Turpin et al. 2. Aims The ultimate aim of this study was to provide information in a form that would enable users to make informed choices when purchasing alternative input devices, and to assist professionals when making recommendations to those who find it difficult to access computers with standard equipment. The purpose of this study was therefore to evaluate a representative sample of alternative computer input devices that are most commonly used by disabled people and to highlight the potential advantages and disadvantages of individual product features and to relate these to the differing abilities and needs of users. 3. Method 3.1. Product selection To establish which types of equipment were used most commonly and which most often presented difficulties, we used a semi-structured questionnaire of a convenience sample of seven professionals with backgrounds in occupational therapy, speech and language therapy, education and ergonomics. This preliminary information was used to inform further consultation meetings with larger groups of professionals, manufacturers and suppliers of information technology (IT) equipment. Following consultation with professionals in local health, social services and education, we narrowed the focus of the evaluation to keyboards and pointing devices most commonly used by people with physical disabilities. The intention was to describe products that can provide access to computers for the largest proportion of the disabled population. In collaboration with AbilityNet, an independent UK charity that specializes in the field of IT for disabled people; we identified keyboards and pointing devices most commonly used by people with physical disabilities. Available products were grouped according to principal features. Where two or more products had similar design features we evaluated the product with the higher specification, thereby incorporating products with all lesser functions in the evaluation. A further criterion for inclusion was that the information supplied by manufacturers was adequate to ensure compatibility with existing equipment. Where a product was available from more than one supplier the purchase was made from the least costly among a cross section of suppliers. Using these methods we selected three compact keyboards, three expanded keyboards, a keyguard and seven alternative pointing devices for full evaluation. 3.1.1. Compact keyboards. These are reduced in size relative to a standard keyboard, but include a full range of keys. Some keys are slightly smaller, as is the space between them. Three keyboards with a standard QWERTY key layout were selected. Two of these had dual function keys in place of the numeric keypad, one incorporated a trackball pointing device and in another the keys were mounted on a ramped, stepped surface (see figure 1). Figure 1. Three examples of compact keyboards designed to assist users with limited trunk control/movement, dexterity or who have limited desk space. (a) Cherry G84-4100 a slimline and ultra low compact keyboard; (b) Datalux Spacesaver a ramped compact keyboard; and (c) Trackboard a compact keyboard with integral trackball. 3.1.2. Expanded keyboards. These are usually larger and have keys or spaces between them that are larger than those found on standard keyboards. Three expanded keyboards were selected and these featured a range of key styles: raised keys, recessed keys and flat (membrane) keys. The key layouts were: standard QWERTY, ABC and Frequency-of-Use (where the most frequently used keys are placed together). Two of the keyboards had interchangeable key layouts and two incorporated a mouse function (see figure 2). 3.1.3. Keyguards. These can provide support for hands and arms and are designed to improve accuracy in key selection. One detachable keyguard was included in the evaluation (see figure 3).

Evaluation of alternative computer input devices 121 Figure 3. An example of a keyguard designed to assist users with impaired control of movement (the Maxess Products Keyguard). functions, an adjustable sized mouse suitable for both left and right hand users, and a wireless infrared head mounted pointing device operated by a reflective dot worn by the user (see figure 5). 3.2. Participants Prospective participants aged 18 years or more, identified from the Derby Assistive Technology Evaluation Centre volunteer register, were invited to take part in trials of the products. Criteria for inclusion in the study were that participants should (1) be current or previous users of computers for education, work or leisure purposes; (2) possess some communication skills; and (3) have adequate cognitive and visual function to use and appraise the devices. The Local Research Ethics Committee approved the study and all participants gave informed consent. 3.3. Product appraisals Figure 2. Three examples of expanded keyboards designed to assist users with impaired vision or control of movement. (a) IntelliKeys a flat membrane expanded keyboard with QWERTY, ABC or customized layout; (b) BigKeys Plus LX an expanded keyboard with raised keys arranged in QWERTY or ABC layout; and (c) USB King Keyboard an expanded keyboard with recessed round keys arranged in a frequency of use layout. 3.1.4. Alternative pointing devices. Seven alternative pointing devices were selected. Trackballs contain a rotating ball on the upper surface of the device. The pointer is controlled by rotating the ball, using a finger, thumb or foot. Trackballs have the advantage that they are used in a static position and may be operated away from the desk, e.g. in the hand or on a wheelchair tray. Models with a variety of configurations were selected; two had programmable buttons for common tasks such as double-click and were suitable for use in either hand and the third was a cordless thumb-operated infrared trackball (see figure 4). Other pointing devices selected included a joystickshaped mouse, a touchpad capable of performing all mouse An expert panel comprising 7 independent assessors, with backgrounds in occupational therapy, physiotherapy, ergonomics and information technology, carried out appraisals of (1) the product as supplied; (2) set-up procedures including methods of connection; and (3) clarity, content and format of instructions and other product information. The weight, dimensions and compatibility of each device with the computer operating system were recorded together with methods used to interface the devices with computer hardware. We also identified the needs met by each product, along with limitations. Finally, four participants with no specialist knowledge of IT repeated these appraisals to determine whether the set-up and operating procedures were clear to novice users. 3.4. User trials User trials were undertaken at our centre unless a participant requested a home visit. Trials were carried out with one to five participants at a time. We assessed the computer access needs of each participant; to ensure that the trials were carried out under optimal conditions and using appropriate equipment. We adjusted computers and workstations to accommodate each individual and device settings

122 G. Turpin et al. Users appraised the performance of products in the following standard sequence of tasks: (1) opening and closing files; (2) typing a specified sentence which incorporated all the letters in the alphabet as well as the numeric keys; (3) highlighting a section of text; (4) cutting, pasting and moving text; and (5) scrolling up and down a page of text. We asked participants to compare each device with the others they evaluated and also with the device they were currently using. A semi-structured questionnaire was administered to gather their views about the equipment and ease of use, advantages and disadvantages, comfort and acceptability. This combination of methods allowed both quantitative and qualitative information to be gathered about each device and its features. 4. Results and discussion 4.1. Participants Figure 4. Trackball pointing devices are available in a range of shapes and configurations. (a) The Kensington Expert Mouse operated by hand using a central trackball; (b) the Kensington Turboball operated by hand using a central/ anterior trackball; and (c) the Logitech Cordless Trackman Wheel a cordless thumb-operated device. were customized to meet personal preferences. Participants evaluated only those products (or product features) that matched their needs and that they were physically capable of operating. Each person evaluated up to five products, presented in random order and balanced across the group. Each product was evaluated by five people. We recruited 35 participants (18 women and 17 men) for the user trials. The mean age of the participants was 53 years (range 32 77 years). Of the participants 34 were current users of computers and one was a past user; in addition 10 had neurological conditions, 11 had rheumatological conditions, eight had musculoskeletal problems affecting the upper limb and the remainder had other problems affecting computer access. Some impairments arose from more than one condition. The characteristics of the participants are summarized in table 1. Most participants 28 (80%) were aware of the accessibility options available as standard in most personal computers and 22 (62%) knew that adjustments can be made to a standard keyboard and pointing device or mouse to improve access. However, they had little or no knowledge of how to access these options on the computer. Difficulties reported by participants in using adaptive devices during the trials could sometimes be alleviated by such adjustments and in some cases access options obviated the need for any adaptive equipment. The following sections present findings for each product group, derived from both the user trials and professional appraisals. The observed and reported benefits of individual product features and alternative computer input devices are highlighted and mapped to the needs of users with different impairments and abilities in tables 2 and 3. The features, methods of connection and compatibility options for each input device are summarized in tables 4 and 5, respectively. 4.2. Compact keyboards Users with reduced lateral trunk movement, musculoskeletal pain or impaired control of the upper limb, found compact keyboards useful. As the keyboards are small, this reduced the need to abduct the shoulder whilst reaching for the mouse. Users who needed to operate a keyboard from an articulated arm, a wheelchair tray or in other environments where space is limited also appreciated these keyboards.

Evaluation of alternative computer input devices 123 Figure 5. Other pointing devices. (a) Anir Mouse Pro a joystick style hand operated device; (b) Whale Mouse designed to be operated by a flat hand; (c) Cirque SmartCat designed to be touch operated; and (d) SmartNav an infrared head pointing device. Table 1. Characteristics of the participants (mean age = 53; age range = 32 77). Number Percentage Gender Female 18 51 Hand dominance Right 30 86 Not able to use hands 1 3 Computer use Current user 34 97 Previous user 1 3 Duration of computer use (for work, education 10 years or more 19 54 and/or leisure purposes) 5 9 years 10 29 1 4 years 2 6 Less than 1 year 4 11 Diagnosis Progressive neurological condition Multiple sclerosis 5 14 Transverse myelitis 1 3 Charcot-Marie-Tooth disease 1 3 Friedreich s ataxia 1 3 Non-progressive neurological condition Encephalitis 1 3 Cerebral palsy 1 3 Rheumatological condition Rheumatoid arthritis 6 17 Ankylosing spondylitis/spondylosis 5 14 Other musculo-skeletal condition or symptom with an impact on upper limb function Carpal tunnel syndrome, Dupuytren s contracture, tennis elbow, restricted movement and pain 8 22 Other condition with no impact on upper limb function 9 25

124 G. Turpin et al. Table 2. Matrix of product features and user impairments/abilities. Features weakness no movement reduced range of movement reduced dexterity impaired coordination Dominance: left handed Impaired trunk control/ movement Visual impairment Limited literacy skills e.g. dyslexia KEYBOARDS: Expanded 3 3 3 3 3 Compact 3 3 3 3 On screen 3 3 3 3 3 3 Dual function 3 3 3 with pointing device Keyguard 3 3 3 3 3 Key layout: QWERTY 3 ABC 3 3 Frequency of 3 3 3 3 use Custom overlay 3 3 3 3 3 3 Key type: Raised Recessed 3 3 3 Flat membrane 3 3 3 Coloured 3 3 Auditory 3 feedback Tactile feedback 3 POINTING DEVICES: Size/design: Alternative sizes 3 3 3 3 Adjustable size 3 3 3 3 Trackball 3 3 3 3 3 Mouse 3 3 3 3 Flat 3 3 3 3 Buttons/functions: Programmable 3 3 3 3 3 buttons Scroll function 3 3 3 3 3 Dwell/switch 3 3 3 3 3 3 3 clicking Usual method of operation: Head or other body part 3 3 The reader should be aware that the benefits of individual features may differ between products and when selecting an input device consideration should also be given to personal settings. 1. Key size and the reduced space between keys can affect key selection and could be uncomfortable. 2. Height, angle and depth of the keyboard were relevant factors those with a fixed height and angle could be difficult to use comfortably. 3. The keyboard with an incorporated pointing device was not suitable for both left- and right-handed use. 4. Separate numeric keypads allow independent positioning of the device; some users preferred this option to dual function keys. 4.3. Expanded keyboards We found that expanded keyboards could be helpful for users with visual impairments, limited literacy skills or reduced dexterity. Larger or differently coloured keys were often arranged in a simplified layout (e.g. alphabetically). Frequency-of-use keyboard layouts reduced the effort required for typing by decreasing the distance required to reach commonly used keys, which were centrally located. Larger and/or recessed keys, and the option of a keyguard, were beneficial to users with reduced dexterity or coordination. Membrane keyboards provided users with an easily accessible choice of keyboard and pointing device layouts.

Evaluation of alternative computer input devices 125 Table 3. Matrix of alternative computer input devices and user impairments/abilities. weakness no movement reduced range of movement reduced dexterity impaired coordination Dominance: left handed Impaired trunk control/ movement Visual impairment Limited literacy skills e.g. dyslexia Compact keyboards: Cherry 3 3 3 Datalux 3 3 3 Trackboard 3 3 Expanded keyboards: IntelliKeys 3 3 3 3 3 3 BigKeys 3 3 3 3 3 USB King 3 3 3 3 3 3 Keyguard: Maxess 3 3 3 3 3 Trackball pointing devices: Expert 3 3 3 3 Turboball 3 3 3 3 Logitech 3 3 3 Other pointing devices: Anir Mouse 3 3 Whale 3 3 3 3 Mouse Cirque 3 3 3 3 SmartNav 3 3 3 3 3 3 3 3 The reader should be aware that when selecting an alternative input device, consideration should also be given to personal computer settings. Layouts were manufactured in the form of overlays and could therefore be customized. 1. To meet the needs of individual users the entire configuration had to be considered, including key size, shape, spacing, layout and profile (e.g. flat, recessed or raised). 2. The pressure required to activate keys varied between raised, recessed and membrane keyboards. In the membrane keyboard we evaluated, the pressure required could be adjusted. The option to adjust pressure sensitivity was beneficial for users who found it difficult to exert sufficient pressure and for those who activated keys inadvertently. 3. Two of the expanded keyboards we evaluated had integral mouse controls, which could not be operated simultaneously with the keyboard. This proved time consuming for some users. However, a separate pointing device could be connected to the keyboard. 4. Users found the size of the keyboards limited their workspace, so that positioning a separate pointing device was more difficult. 5. Functionality was affected by keyboard dimensions (e.g. size, height and angle), which varied between products. Some keyboards could be adjusted in height or angle, making comfortable positioning easier to achieve. 4.4. Keyguard The keyguard we evaluated was valued by users with weak upper limbs as they were able to stabilize their hands directly above the keys and rest them on the guard, when necessary. Participants felt a keyguard would improve typing accuracy by allowing one key to be pressed at a time. Users with tremor or reduced manual coordination found that a keyguard reduced unwanted keystrokes. 1. Some users found that access to the keys was restricted. 2. Users found it difficult to strike the spacebar with the thumb as the keyguard had a circular rather than an elongated opening. However, this design feature was advantageous for users who rested their hands in the area of the spacebar. Users reported that circular openings would be helpful to people who use a pointing stick to strike keys. 3. To ensure compatibility, the keyboard and the keyguard should be purchased together: even standard keyboards vary in shape and configuration.

Table 4. Matrix of alternative computer input devices and product features. Compact keyboards Expanded keyboards Keyguard Trackball pointing devices Other pointing devices Features Cherry Datalux Trackboard IntelliKeys Big Keys USB King Maxess Expert Turboball Logitech Anir Mouse Whale Mouse Cirque KEYBOARDS: Expanded 3 3 3 Compact 3 3 3 3 On screen 3 Dual function 3 3 3 with pointing device Keyguard 3 3 Key layout: QWERTY 3 3 3 3 3 3 3 ABC 3 3 Frequency of use 3 Custom overlay 3 Key type: Raised 3 3 3 3 3 Recessed 3 Flat membrane 3 Coloured 3 3 3 Auditory 3 3 feedback Tactile feedback 3 POINTING DEVICES: Size/design: Alternative 3 sizes Adjustable size 3 Trackball 3 3 3 Mouse 3 3 Flat 3 Buttons/functions Programmable 3 3 3 3 3 3 buttons Scroll function 3 3 3 3 3 Dwell/switch 3 clicking Usual method of operation Hand 3 3 3 3 3 3 3 3 3 3 3 3 3 Head or other 3 body part SmartNav 126 G. Turpin et al. These products should be considered in conjunction with personal computer settings.

Table 5. Methods of connection and compatibility options for each computer input device. Compact keyboards Expanded keyboards Keyguard Trackball pointing devices Other pointing devices Cherry Datalux Trackboard IntelliKeys Big Keys USB King Maxess Expert Turboball Logitech Anir Mouse Whale Mouse Cirque SmartNav Methods of connection: PS / 2 3 3 3 3 3 3 3 3 3 3 3 USB 3 3 3 3 3 3 3 3 3 Serial 3 3 3 3 3 3 AT 3 3 3 3 3 Compatibility options: IBM PC 3 3 3 3 3 3 3 PC 3 3 3 3 MS DOS 3 3 3 3 3 3 Windows 3.1 3 3 3 Windows 95 3 3 3 3 3 3 3 3 3 3 3 Windows 98 3 3 3 3 3 3 3 3 3 3 3 3 Windows 3 3 3 3 3 3 Millennium Windows 2000 3 3 3 3 3 3 3 3 Professional Windows 2000 3 3 Server Windows NT 3 3 3 3 3 3 3 3 3 3 Windows XP 3 3 3 3 3 3 Laptop 3 Apple MAC 3 3 3 Apple MAC OS 8.1 3 Cherry 3 Keyboard G84 4100 Evaluation of alternative computer input devices 127

128 G. Turpin et al. 4.5. Trackballs Users with limited control or range of upper limb movement and who had difficulties in using a mouse found trackballs helpful. Users unable to hold a mouse and click a button simultaneously, or to use their index finger to do this, could overcome these obstacles by customizing the programmable buttons. Trackballs were sensitive and could be difficult to control when making small movements. However, users found that some unintentional movements of the cursor were overcome by adjusting the sensitivity of the device. 1. Trackballs are operated from a static position and therefore allow use on or away from the workspace. 2. Trackballs varied in size and shape, and the configuration of buttons differed. It was necessary for users to check, therefore, that they could reach the buttons comfortably. 3. Functions of the programmable buttons, and sensitivity, could be adjusted to suit individual needs. 4.6. Other pointing devices 4.6.1. Joystick style pointing device. This device operated like a conventional mouse, the direction of the cursor responding to the individual s arm movement. A button on top of the device could be controlled by the thumb with a rocking motion to right and left click, and there was another programmable button on the shaft. This device was helpful to users requiring an alternative working position to that provided by a standard mouse. 1. The orientation of the cursor needed to be adjusted using the software provided. This is important as the cursor direction was found to be relative to the movement of the user s arm. 2. Shoulder movement was required to operate this device and the manufacturer s instructions recommended an ideal arm position that was sometimes compromised by a lack of available workspace on the desk. 4.6.2. Adjustable sized mouse. The product included in the evaluation is designed for use with a flat hand and straight wrist. It can accommodate varying hand sizes and can be operated by either hand. The operation is similar to that of a standard mouse and it includes a third scroll button. This device was valued by users who have limited grip and prefer the hand to remain flat with the wrist in a neutral position, who want the heel of the hand supported whilst still being able to access the buttons, and who prefer the concept of a standard mouse but do not get enough comfort or support from a standard device. 1. This device is suitable for multiple users, i.e. in the home, education or workplace, as it accommodated left- and right-handed users and various hand sizes. 2. Successful operation required a flat hand. 4.6.3. Flat touchpad. This device is similar to that found on a laptop. Movement of the finger over the touchpad area operates the cursor. All functions, e.g. drag, draw and highlight are activated by tapping the pad or use of the programmable buttons at the side and below the touchpad surface. Other features are one-touch scroll and zoom and speed and sensitivity adjustment. This device is operated from a static position, it can be used away from the workspace, e.g. on a wheelchair tray or held in the hand. Users who are willing and able to hold the hand relatively flat thought touchpads were useful. Similarly, participants who find it easier to make light touches rather than firm button presses to perform click functions and who prefer to control the mouse pointer with a static device recognized the benefits of a flat input device. 1. Effective use of this device was dependent on controlled, fine hand movement and the ability to maintain a relatively flat hand. 2. Left- or right-handed users could operate it, on or away from the workspace. 4.6.4. Infrared head mounted pointing device. The product included in the evaluation operates by a wireless infrared sensor which tracks a tiny disposable dot placed on the user, usually on the face. Movement is transmitted to the cursor on the screen. Using an on-screen keyboard, selections are made by positioning the cursor over the required key for a set period of time (this is known as dwell clicking). Selection can also be made via a switch or with a standard mouse. Minimal head movement was required to operate this device successfully, as was an element of head control. The device has pre-set or custom sensitivity settings, therefore it was possible to accommodate the head movement achievable by each user. The device could also accommodate mild head tremor. This device was helpful to users who had limited or no arm movement and who had adequate control of slight head movements. The participants felt it was less obtrusive than a headset. 1. This device could store the personal settings of more than one user and is therefore suitable for multiple users. 2. It was necessary to experiment with placement of the dot and the receiver. 3. On-screen keyboards reduced visibility of open documents on the screen. 4. Varying amounts of practice were required to operate the device successfully.

Evaluation of alternative computer input devices 129 5. Professional appraisers commented that it might not be suitable for those with excessive uncontrolled movements of the head or trunk. 4.7. Instructions, set up and use Of the 14 products evaluated, 13 were supplied with instructions; however, the quality, content and format of these varied considerably. Errors in installation and setting up of devices were common to both professional and novice users, and not all products had details of how to obtain technical support. Eight products were set up and used during the trials without any difficulties, but we encountered problems with the remaining six devices. Six instruction sheets included troubleshooting information about problems that are known to occur and provided advice to overcome them. Eight instruction sheets included a telephone help/support number; five of these were for contacts outside the UK. 5. Conclusion This study has produced detailed information on a series of product groups designed to facilitate basic computer operations for people with the most common types of impairment. By focusing primarily on product features rather than on comparisons between specific products, our trials have highlighted ergonomic issues that are relevant to a wider range of products and have identified specific points to consider when matching products to individual needs. Our study confirms that to enable successful computer access a combination of approaches may be required. A single piece of equipment will rarely be a complete solution in itself. Customizing computer settings to the needs of the individual can resolve some of the most common difficulties experienced by users when operating a keyboard or mouse. Where alternative hardware is required, computer and device settings will often need to be customized to enable an individual to operate the device successfully and with optimal efficiency. One of the best ways to increase the ability of disabled people to access information and communication technologies is to provide information both to users and to professionals. Our work has made a contribution to this end. On the basis of our findings we recommend that manufacturers and suppliers should improve the quality of instructions for setting up their products. We suggest that products would be more effectively marketed and would be more likely to meet the needs of individual users if product information highlighted specific features such as those considered in our study. Acknowledgements The authors would especially like to thank volunteer users and DATEC staff who participated, professional appraisers including those from AbilityNet, Sarah Woodbridge (Ergonomist and Hand Therapist) and the Medical Illustration Department, Derby Hospitals NHS Foundation Trust. Finally, we are appreciative of the Medicine and Healthcare products Regulatory Agency, an executive agency of the Department of Health (UK) for funding this work. References [1] Fraser, B.A., Bryen, D. and Morano, C.K., 1995, Development of a physical characteristics assessment (PCA): A checklist for determining appropriate computer access for individuals with cerebral palsy. Assistive Technology, 7, 26 33. [2] IBM Accessibility Center, 2001, Identifying reasons for producing accessible content and products. Available online at: http://www- 306.ibm.com/able/access_ibm/reasons.html (accessed on 13 January 2004). [3] Burkhead, E.J., Sampson, J.P. and McMahon, B.T., 1986, The liberation of disabled persons in a technological society: Access to computer technology. Rehabilitation Literature, 47, 162 168. [4] Merrow, S.L. and Corbett, C.D., 1994, Adaptive computing for people with disabilities. Computers in Nursing, 12, 201 209. [5] Anson, D., 1994, Finding your way in the maze of computer access technology. American Journal of Occupational Therapy, 48, 121 129. Copies of the report entitled MHRA 2004. Specialist computer controls: An Evaluation of Alternatives to the Standard Keyboard and Mouse (04110) can be obtained from The Medicines and Healthcare products Regulatory Agency, Hannibal House, Elephant & Castle, London SE1 6TQ. Tel: + 44 (0)20 7972 8226, Web: www.mhra.gov.uk