Link to Figures Show Figures
	Article Text Figures References
	Send Link Send HTML Article

A Survey of Upper-Limb Prosthesis Users in Oxfordshire

Peter J. Kyberd, MSc, PhD, MISPO
Jane J. Davey, DipCOT, SROT
J. Dougall Morrison, FRCS

ABSTRACT

The design of a new device should embrace the needs of the target population. A survey of prosthetic limb users was performed at the Oxford Limb-Fitting Centre, Headington, Oxford, as part of the specification for an intelligent hand prosthesis. Users of prosthetic limbs were questioned about the positive and negative attributes of their limb and were asked about their wishes for the design of prostheses. A questionnaire was devised to address this and sent to 80 subjects; 69% responded. It was found that light weight, durability, ease of cleaning, longevity of operation (up to 12 hours), and suitability for driving were highlighted as priorities. This information was incorporated into the design of a new powered hand prosthesis. This article reviews the results of the survey and makes comparisons with similar studies. Finally, recommendations are made for the designs of any new prosthesis system.

Key Words: Hand, prosthetics, driving, attitudes, activities of daily living.

Introduction

The success of a new system often depends upon its ability to reflect the requirements of the target population. Consequently, knowing what users want is an essential part of the design process. To meet that need, it is important to canvass the opinions of present users. Their knowledge regarding deficiencies of current commercial devices and their suggestions for proposed improvements are invaluable.

When the European Community conceived an initiative to encourage research and development in the provision of assistive technological devices under the Technology Initiative for the Disabled and Elderly (TIDE) program, emphasis was placed on learning the opinions of the target populations. This article describes the results of a survey taken to obtain such information from present users of upper-limb prostheses at the local limb-fitting center. The overall project was the TIDE project Manipulation And Reaction Control under User Supervision (MARCUS).¹ The aim of the project was to produce practical prototypes of a more sophisticated hand prosthesis that would be operated under hierarchical control. It was the latest version of the family of Southampton Hand Prostheses.² Previously, these devices had limited applications in the field,³ but the intent of this phase was to design systems that would ultimately lead to wider applications in the field.

Survey

Design

We designed a four-page questionnaire that anyone, regardless of technical knowledge, could answer easily and quickly. The questions were designed to identify as wide a range of problems as possible and were divided into two sections: The first set of questions were close-ended and the second section contained longer, open-ended questions. The intention of the questions was both to provide a comparable data set and to allow users to express any feelings about the fit, function, or use of their various prostheses. A final section allowed further comments on any area that they chose. Detailed interviewing techniques were discounted because of time constraints: The requirement of the TIDE program was that the entire system specification should take no longer than three months. (Questions are listed in Appendix A at the bottom of this document.)

Procedure

The questionnaire was designed after interviews with a small sample of patients who routinely visited the Oxford prosthetic arm clinic and was later modified after discussions among the authors. During a brief pilot phase, the questions were asked orally to a number of clinic visitors and their responses used to hone the final set of questions. Ten questionnaires were then sent with a covering letter and a self-addressed, stamped envelope. This pilot had a 50% response rate and the results showed no substantive changes were needed. Another 70 subjects were chosen according to the criteria below, and the questionnaire was sent to them.

Sample Selection

A complete analysis was made of all the patients on the books of the Oxford Disablement Services Centre (DSC),⁴ which is based at the Nuffield Orthopaedic Centre, NHS Trust. The DSC is involved in the supply of artificial limbs and other aids and appliances. For the purposes of this survey, the date of last contact was recorded. The subjects were selected from this information.

The inclusion criterion was any person with an upper-limb loss who had made contact with the center in the past twenty-six months (80 individuals). The exclusion criteria were: (1) individuals with a partial hand loss (given the form of the intended prosthesis), (2) people older than 70 (to account for the increased chance of mortality in this group), (3) users younger than 17 (children need a different questionnaire), (4) patients deemed unsuitable for psychiatric reasons, and (5) people who reside outside the United Kingdom. Thus, the sample included active, long-established users and those recently referred.

The catchment area for the center at the time corresponded approximately to the Oxford Regional Health Authority (ORHA), which comprises about 2.58 million individuals, half of which are male and half female.⁵ With its balance of urban and rural areas, the ORHA population is representative of the population of the United Kingdom as a whole. For this survey, the area is referred to as the "Oxford area." Patients are referred to the center where they are assessed by the staff. Routine maintenance is carried out by patients visiting the center personally, or by using the free postal service.

Results

Fifty-six people (69%) responded to the entire survey, a rate that was even greater than for the pilot.

Sample Characteristics

On February 1, 1992 the Oxford DSC had 341 living patients on their books (7 untraceable). Table 1 outlines the ratios and relates them to the survey and national values.⁴ In addition, it shows the comparison of the characteristics of the respondents with the entire center, including the nature of the loss (level and cause), the principal type of prosthesis used, and the gender. In the findings, some of the questions were interrelated, so the responses are grouped together.

Section A

Section A of the questionnaire contained the close-ended questions needing a simple yes or no response. In the following, "the population" refers to those who replied to the survey, and the percentages cited reflect their responses.

Use (Questions 1 and 7). Because of the selection procedure, the survey population probably were extensive users of their appliances, which is borne out by the results. Of the respondents, 89% said they used the limb "daily." Only 5.4% answered "occasionally" and 5.4%, "never." The last two groups reported extremely low use of their devices on an hourly estimate (Figure 1) .

Level of loss (Question 2). Forty-six percent of respondents had a below-elbow loss; the remaining had an above- or through-elbow loss.

Type of prosthesis (Question 3). Many subjects were users of more than one type of prosthesis, including purely cosmetic hands, split hooks, and devices that have a cosmetic appearance but a designed functional range (such as myoelectric hands). The latter category was included to illustrate the importance of appearance (Table 2) . Other research has shown^7-9 that the cosmetic arm forms an important part of the social standing of the user; but for work the user needs a functional hand. Our data reflect these findings.

Activities (Questions 5, 6, and 8). Respondents' reported occupations are listed in Table 2 . The different rates of use are shown in Table 3 . Twenty-nine people (53%) expressed a need for their arms during sporting activities, and 40 people (73%) found limited use in work environments, although some indicated that it was more for cosmetic than for functional reasons.

Problems (Question 4). While the survey asked about one area of greatest concern, respondents commonly flagged more than one area. Fit (14), cosmesis (13), and maintenance were all similarly reported by 20% of the sample. Only function was seen as being more of a problem (33%).

Desired qualities (Questions 9 and 10). At present, technical restrictions guarantee that prostheses described as "cosmetic" have limited functional range but are attractive. A lightweight, precise limb is often designed at the expense of strength or power, whereas a limb that is designed to be strong often turns out to be heavy and less precise. Most of the time these features are mutually exclusive. To understand how important each factor is and to learn which features users would want in a new design, respondents were asked to indicate their preference for a particular prosthesis type. The choices include (1) powerful, but less precise and heavy; (2) precise and light, but less powerful; (3) functional, but unsightly; and (4) cosmetic, but not particularly functional.

To easily view the results, the two factors are plotted on two orthogonal scales (cosmesis to function and heavy to light) in Figure 2 . The relative strengths of the responses can be seen easily from the relative lengths of the diagonal lines. Despite the request for a single response for both categories, a few individuals selected more than one category, or only answered one question. Consequently, these results lie on the horizontal or vertical scales.

Section B

Preferences. The responses fell into four main areas: cosmesis (18), weight (10, with 9 pleased with the lightness of their cosmetic hands), functional range, and other preferences.

Problems with fit. Whereas only 14 of the respondents identified the fit as a problem area in the first part of the questionnaire, 35 completed this section. Two of the replies described positive features even though the question had been designed with a negative bias to extract details of problems or difficulties. Overall, the three areas of trouble identified were the socket, straps, and discomfort with temperature (either too hot or too cold). Table 4 summarizes these results.

Problems with function. The question triggered a range of responses, some general and some specific. Thirty-three respondents (60%) identified some aspect of the prostheses' function as a problem, 18% of whom were users of cosmetic arms.

These responses were dominated by the users of cosmetic arms. Six felt that it was other people who had trouble with the appearance. One found the hook "offensive" while another found that it stimulated conversation in social situations; this was seen as a positive effect, and was similar to comments recorded by Milstien.⁹ Similarly to the question on fit, there were 23 negative and six positive comments, although the form of the question itself was designed with a negative emphasis.

Problems with maintenance. This was interpreted by the respondents as both the service given by the Disablement Services Centre (DSC) and the durability of the devices. The respondents commented on the short life span and the discoloration of the cosmetic gloves.

Limitations of the Study

The data collected for the record of the entire survey population allow some analysis of the people who did not reply to the survey.⁴Table 1 shows that, proportionally, gender, cause of amputation, level of loss, and type of prosthesis have little bearing on the response rate. Far fewer subjects with bilateral amputations replied to the survey. This lower response is disappointing as their dependence on functional limbs makes their insights and desires particularly important.

An analysis of the age groups involved shows that there is a tendency for the younger users to be underrepresented and for the retired population to feature more strongly.

The group questioned in this survey are self-selecting, being those who regularly use the limb-fitting service. They are exactly the people who have much experience with using their prostheses and are most likely to reveal problems. It would be desirable to understand why some have rejected prostheses as an aid to living, but this type of survey is not able to gather such information.

Discussion

Analysis of Survey

The good response rate (69%) supports the choices made in the design of the questionnaire. Burroughs et al.¹⁰ report response rates of 10% to 50% as typical for this form of questionnaire. The survey provided a broad insight into the population of prosthesis users. The responses varied from brief to lengthy. While those who were generally happy with their devices appeared to have less to say, those who provided more comprehensive replies gave valuable insights into reasons for disuse. The strongly held beliefs of some respondents often were in contrast to others. The most significant features were cosmetic appearance and functional range. Individuals expressed clear desires for one factor without compromise, while the other was of no importance. This was true in both directions.

The form of selection can introduce bias into the results, although it did not affect the basic proportions of the population. A bias will exist toward those who use their devices and are motivated to reply, i.e., toward those with traumatic losses or above-elbow amputations (Table 1) ., As a group, however, they are more likely to use their prostheses extensively and so have useful insights into the life and performance of the current generation of prostheses, as well as an interest in any new developments. Generally the respondents used a device for at least some of the time. The survey did have one respondent who did not use a device because "it got in the way," implying it was of no functional use whatsoever.

Use

The largest number of users employ their devices for an entire day, or more than 12 hours. In response to a similar question, patients of the Cambridge DSC11 showed a similar trend, 58% used them for more than eight hours a day. It seems that people tend to use a device extensively once they start to wear it. This is supported by the finding that 65% of the Cambridge subjects used their arms for "social and work" situations.

Looking at other surveys, the majority of populations reports wearing a prosthesis for more than eight hours a day, (Table 5) . In addition, Weaver⁸ recorded an average use of 9.5 hours. However, it is important to note that most of the figures are for the time that the prosthesis is worn, not the time spent actually using the device. It is difficult to get a more accurate figure of use without closer monitoring of users.

As a consequence, attempts to relate the level of usage, as gauged by the reported hours worn, to type, level, or reported problems did not produce a correlation of any significance. If the occupations are scored using an occupation's workload-scale proposed by Kejella,¹² there is an association of heavy work with the hooks and lighter work with cosmesis. This association, however, is weak and does not strengthen when compared with reported daily use.

Leisure is an area in which specific terminal devices are most useful, and possibly most acceptable, because many sports involve the carrying and application of special tools. The limitations placed on swimming when using conventional prosthetic arms was noted.

When a comparison is made between the device chosen and the level of loss, there are broadly equal numbers in each group. While there is a trend toward using more functional appliances for the above-elbow group with amputations and toward using cosmetic appliances for those with below-elbow losses, the differences are small and have little significance (Table 6) .

Driving. The absence of a left hand in the UK is likely to be more inconvenient when the user has to drive unconverted cars. Few people seemed to be prevented from driving because of limb loss, though for a unilateral loss, automatic shifting is a clear advantage. In a survey conducted in Manchester,¹³ the proportion of drivers who reportedly did not use their left prosthesis to drive was higher than in this survey. Nationally, 33% of the adult population do not hold a current driving license.¹⁴ Only 22% of the survey population stated they did not drive. This might be a slightly younger population and so it is conceivable that there would be a higher proportion of drivers in a cohort group, but it appears that upper limb absence does not restrict the ability to drive.

Fit

Fit is well known to be an important part of the prosthesis supply and the survey results support this. It is, therefore, especially important to continue the existing efforts to improve methods of suspension. However, comparing the basic types of device with the main areas of concern (Table 7) no category is the province of any single group.

Function

All groups commented on functional limitations, which are not the exclusive concern of those who use functionally specific devices such as split hooks. Function is also a concern of those who use purely cosmetic devices as well. Therefore, the inevitable conclusion is that if a person wears a prosthesis, he or she finds a use for it. Any additional function that can be added to any prosthesis, even to cosmetic ones, will be used if the device can be operated easily and conveniently.

Maintenance

The failure to ask an unambiguous question about maintenance resulted in a small number of persons who interpreted the question to be about the service they received from the limb-fitting center. An equal number of respondents praised and criticized their maintenance. A survey conducted at the Cambridge limb-fitting center identified communication as an area needing improvement,¹¹ such as keeping users up to date with changes and improvements to devices. At Oxford, "communications" with the center received four comments, three of which were unfavorable. None was concerned with the changes in performance for new devices, but with the direct communications with center personnel. The general comments were more widely spread eight favorable, 10 unfavorable. These suggest a range of individual concerns rather than any particular failure in the service or any major areas that need improvement.

Cosmesis and Function

Four respondents stated that function was the overriding concern and indicated that they believed this to be true for most of their fellow users, whereas another found cosmesis to be "paramount." A comparison of their responses shows that the former group drove and worked using their prostheses, employing functional devices exclusively. The other responses of the latter individuals were consistent with their indicated desire to use the prosthesis as a cosmetic device.

Since appearance is the main design feature of a cosmetic arm, it is unfortunate that over half of such users are unhappy in some way with the appearance. Only six users of cosmetic devices did not comment at all and six made favorable remarks. They also identified a range of subtle factors that influence cosmesis, such as the way the arm hangs or moves or the color of the gloves.

The need for precision during manipulation was identified by users of both cosmetic and noncosmetic arms. Therefore, functional improvements should be the target of cosmetic and nonanthropomorphic terminal devices. Questions 9 and 10 force a choice along the line of the paradigm that says cosmesis is the opposite of function and that the two are mutually exclusive. The comments made by users confirm that this is not so clear cut. Users of purely passive arms also described functional problems with their devices (Question 11). In addition, some users found that the fingers broke on their cosmetic hands. These failures are highly suggestive that the hands are used in an active role during manipulation.

Another aspect of cosmesis is the appearance of straps. It was noted that only white straps have been supplied to the adults; pink are provided for children. The former stand out too much for some users. Similarly, the straps may not lie flat against the skin so that they are visible beneath clothing.

Conclusions

The survey findings can be summarized as follows:

• Both appearance and function are important, separately and in the same device
• Weight is important
• Straps used for actuation and suspension create discomfort and adversely affect appearance
• If a prosthesis is worn, it is worn for most of the day
• Cosmetic gloves discolor too quickly, and they should be available in a wider range of skin colors
• Sports and leisure activities are areas for the use of specialist terminal devices
• Driving is an important activity.

Recommendations

A number of recommendations can be made for the next stages of prosthetic research: 1. Any new form of prosthesis must be lightweight; power is a lesser concern. 2. An externally powered hand must be capable of operating for more than 12 hours, either on a single charge or by easy charging. Electric power sources are the only practical solution, so their batteries must be easily changed. 3. With any new device the user must be able to drive a car easily, preferably without any modification to the vehicle. 4. Gloves should be easily cleaned, long lasting, and attractive.5. Any new design of prosthesis should be more functional than existing devices, but available in both anthropomorphic and nonanthropomorphic forms. 6. Cosmetic hands should be stronger and more functional. 7. Suspension and actuation are areas that require further improvement. 8. A closer study of activities is needed to make choices about the design of a hand and the gripping surfaces and geometries. The recommendations have been carried through into the design of the next two generations of Southampton Hands.^1,15

Acknowledgements

This project was supported by the European Community, TIDE initiative, TP150.

The authors thank the staff of the Prosthetic Service, Mary Marlborough Centre at the Nuffield Orthopaedic Centre, Oxford, for their help, advice and encouragement.

The authors particularly wish to thank Barbara Marks of the Oxford Orthopaedic Engineering Centre and Karen Sharpe of the Prosthetics Service for their help, Paul Bagwell for his assistance in conducting the survey, and Ann Dunn for help with manuscript preparation.

References:

1. Kyberd PJ, Holland O, Chappell PH, Smith S, Tregidgo R, Bagwell P, Snaith M. MARCUS: A two degree of freedom prosthesis with hierarchical grip control. IEEE Transactions on Rehabilitation Engineering. 1994;3:1:65-69.
2. Chappell PH, Kyberd PJ. The prehensile control of a multifunction hand prosthesis by a microcontroller. Journal of Biomedical Engineering 1991;13:9:363-369.
3. Kyberd PJ, Chappell PH. The Southampton Hand: an intelligent myoelectric prosthesis. Journal of Rehabilitation Research and Development 1994;31:4:326-334.
4. Kyberd PJ, Beard DJ, Morrison JD. The population of users of upper limb prostheses in Oxfordshire. Prosthetics and Orthotics International 1997:21:2:85-91.
5. Oxford (UK) Regional Health Authority. The health of the people of the Oxford region 1991.
6. McColl I. Review of artificial limb and appliance centre services: the report of an independent working party under the chairmanship of Professor lan McColl. London: DHSS 1986.
7. Atkins DJ, Meier RH (eds). The comprehensive management of the upper-limb amputee. New York: Springer-Verlag New York, Inc. 1988.
8. Weaver SA, Lange LR, Vogts VM. Comparison of myoelectric and conventional prostheses for adolescent amputees. American Journal of Occupational Therapy 1988;42:2:87-91.
9. Milstien SG, Heger H, Hunter GA. Prosthetic use in adult upper limb amputees. Pros Orth Intl 1986;10:27-34.
10. Burroughs SF, Brook JA. Patterns of acceptance and rejection of upper limb prostheses. Orthotics and Prosthetics 1985;39:2:40-47.
11. Fraser C. A survey of users of upper limb prostheses. British Journal of Occupational Therapy 1993;56:5:166-168.
12. Kejella GH. Consumer concerns and the functional value of prostheses to upper limb amputees. Pros Orth Intl 1993;17:157-163.
13. Verrall T, Kulkarn JR. Driving appliances for upper limb amputees. Pros Orth Intl 1995;19:124-127.
14. UK Department of Transport. Transport Statistics for Great Britain: 1995. HMSO 1996.
15. Kyberd PJ, Evans M. Intelligent Control of a Prosthetic Hand. Paper presented at the International Conference on Rehabilitation Robotics, Bath, UK, April 1997.

Website built by oandp.com