Consumer Evaluation of Powered Feeding Devices

Richard M. Mahoney, Anna Phalangas Applied Science and Engineering Laboratories (ASEL) University of Delaware/Alfred I. duPont Institute Wilmington, Delaware USA

Abstract

The project described here was formulated to evaluate commercially available powered feeding devices. This project was carried out by the Rehabilitation Engineering Research Center on Rehabilitation Robotics of the University of Delaware and the A.I. duPont Institute in Wilmington, Delaware, in collaboration with the Division of Rehabilitation of the A.I. duPont Institute. The study began in October of 1994 and the subject training and evaluation was recently completed in October, 1995. The discussion and results portrayed in this paper represent preliminary results of this project.

Introduction

Many individuals with severe manipulation disabilities require assistance with the task of eating. Because of its primary importance in a person's life, much effort has been exerted to design a robotic or automated device to aid individuals to eat independently. Although some devices have been fully developed, and several have become commercially available, there has been little acknowledged success in providing individuals with a powered feeding device {1}. This project has identified three commercially available powered feeding devices (as shown in Figure 1): the Beeson Feeder, manufactured by Maddak, Inc. of Pequannock, New Jersey, the Handy 1, manufactured by Rehab Robotics, Ltd. of Staffordshire, England, and the Winsford Feeder, of Winsford Products, Inc. of Pennington, New Jersey. Each device operates on a similar principle. A mechanical arm is controlled by the user to scoop food from a desired section of a plate. The food is then presented to the user at a suitable height.

Motivation

Although powered feeding aids have been available for approximately ten years, a small percentage of the user population utilize them, and few Rehabilitation Specialists are trained to prescribe them. Criticisms of the devices vary widely, but no formal evaluation of powered feeding devices has been carried out. Because of the lack of published information about the efficacy of powered feeding devices, the project staff carried out an informal phone survey of 12 Occupational Therapy Departments from major rehabilitation hospitals throughout the United States. Table 1 below summarizes the questions asked in the survey and the responses obtained. The response totals provide only a gross indication of the number of patients being served by the institutions. It is striking, however, that almost every institution reported that none of their patients make use of powered feeding devices. Each institution was also asked to provide an open-ended comment on their use of powered feeding devices. The comments generally indicated that the available feeders are too expensive, too difficult to use, unreliable, inconvenient, difficult to transport, unattractive, and time-consuming to set up. Several institutions make no use of powered feeding device.

Experimental Design

The experimental design was guided primarily by the desire to provide subjects with the opportunity to use the feeding devices in the environment in which they normally eat their meals. Information was obtained from the subjects through two mechanisms, a log sheet, which recorded the use of the devices in the home or institutional setting, and questionnaires, which were administered to both the subjects and the subjects' assistants. The questionnaires were administered after the use of each device and recorded the subjects' and assistants' ratings of various criteria of the machines. Presentation of the devices to the subjects was balanced to reduce bias.

Subjects

Subjects for this study were identified through the Institute's Department of Occupational Therapy and through a mailed solicitation to local schools and rehabilitation facilities. Disability groups represented by the 12 subjects included cerebral palsy and arthrogryposis. Several candidates were not entered into the study due to a lack of sufficient head and neck stability. The subjects ranged in age from 9 to 40 and were split roughly 60% male and 40% female.

RESULTS

Compilation of the log sheets reveals that the results obtained are based on approximately 110 uses of the devices, totalling almost 90 hours in feeding time. Average meal times were roughly 33 minutes for all three devices, with set up and clean up time averaging under 5 minutes for all devices. On average, each meal required 1.8 interventions by an assistant.

Only preliminary results are provided here based mainly on the feedback acquired from the questionnaires administered to the subjects and their assistants during the evaluation sessions. Figure 1 shows bar graphs which compare the subject and assistant results across each device (the questionnaires administered to each group were the same). The dark bars represent the average response of the subjects and the white bars represent the average response of the assistants. In some cases, where the assistant information is missing, further data analysis is required.

Further information will be forthcoming after the data analysis is completed. In addition to the results shown here, information was also obtained regarding the subjects' and assistants' relative rankings of the devices.

Subjects were evaluated by the Occupational Therapy staff of the project. Some preliminary observations have indicated that some of the subjects exhibited an improved sense of self-esteem, improved lip closure, and improved posture after the two day home use, which corroborates other earlier findings [2].

The complete results of this study will be compiled into a series of technical reports. In addition to a general report, each manufacturer will be sent a report which specifically highlights the strengths and weaknesses of their product as revealed during the period of the study.

Discussion

It was generally found that the subjects both enjoyed using the feeding devices and found them relatively easy to learn how to use. Although Figure 1 provides some relative measurement of the subject preferences across devices, in almost all cases, the subject and assistant impressions were above average.

Although no record of how long an intervention lasted, it was generally found that these lasted less than a minute. It was found then, that, on average, subjects were able to eat independently for up to one-half hour.

Based on the subjects who were not admitted into the study, it was found that the current feeders are less suitable for individuals with high-level spinal cord injuries or advanced muscle wasting disorders. In both cases, the barrier for the subject was a lack of stable head and trunk control.

Conclusions

Based on the usage statistics in this study, it is believed that powered feeding devices could play a larger role in supporting people with severe manipulation disabilities to eat more independently. Each device was found to have its own strengths and weaknesses. It is hoped that communication of this information to the manufacturers will assist in improving future versions of the devices. It is also hoped that the data in this study will lead therapists and physicians to prescribe feeding devices.

References

[1] K. Einset, J. Deitz, F. Billingsley, S. Harris (1989). The electric feeder: an efficacy study. The Occupational Therapy Journal of Research 9(1): 38-52.

[2] L.L. Pinnington, J.R. Hegarty (1992). Achieving behavioural change through robot-assisted eating. In Proceedings of IEE Colloquium on Robotics and Its Role in Helping Disabled People, Digest No: 1992/108, pp. 5/1 - 5/8. London, England: Institution of Electrical Engineers, Science Education and Technology Division.

Acknowledgments

The authors gratefully acknowledge Adam Krass of Maddak, Inc., Mike Topping of Rehab Robotics, Ltd., and Bill Morwood of Winsford Products, Inc. for their support of this study. Contributions to this work were also made by Michael Alexander, Kim Pierson, Kathy McConnell, Heather Kasumovic, Michaelle DeBaecke, Wendy Harron, Carrie Strine, and Marnie King. Gratitude is extended to the subjects who participated in this study.

Funding for this project was provided by the Nemours Foundation and National Institute on Disability and Rehabilitation Research Grant #H133E30013, the Rehabilitation Engineering Research Center on Rehabilitation Robotics.

For further information or to obtain a copy of the final report, please contact Richard Mahoney, ASEL, PO Box 269, Wilmington, Delaware, 19899, 302-651-6830,mahoney@asel.udel.edu. Consumer Evaluation of Powered Feeding Devices