|
Abstract
While early independent mobility may have positive effects on a child's
development, it is still difficult for a clinician to determine when a
child is developmentally ready to operate a powered wheelchair. The Rehabilitation
Engineering Research Center at Rancho Los Amigos Medical Center in Downey,
California has undertaken a project to develop a cognitive assessment battery
to predict a young child's functional performance in a wheelchair. The
first phase of this project - the development of a powered mobility program
(PMP) - is presented here. A 34-item assessment battery was developed through
a task analysis and input from professionals who train children in mobility
skills. The battery includes basic, structured and unstructured wheelchair
skills. A flexible approach for administering the PMP is presented, as
well as findings from twenty-four children between the ages of 18 and 36
months who were evaluated using the PMP.
Keywords: Powered mobility program; Young children; Cognitive
readiness; Wheelchair training
It has been documented in research with non-disabled children that a child's cognitive and psychosocial development is influenced by his or her ability to move about in the environment independently (Berenthal, Campos and Barrett, 1984; Campos and Berenthal, 1987). This self produced locomotion allows spontaneous exploration of the environment and provides an opportunity to experience successful mastery of events that indirectly shape self esteem and motivation. Other research with children with disabilities found differences between able-bodied children and children with disabilities in cognitive and perceptual skills; differences that increase with age (Verberg, 1987). Finally, Tatlow (1980) documented passive attitudes and vague body images in children with disabilities who could not move independently. There are many factors that influence a child's ability to learn to maneuver a powered wheelchair functionally. Some of these factors include physical access, cognitive developmental 'readiness', temperament factors (e.g. attentiveness, persistence, motivation), and dynamic sensorimotor integration (i.e. processing sensory input, motor planning, reacting in a timely manner). As a result of improved power wheelchair technology, physical access to powered mobility is no longer a primary impediment. Proportional joysticks can be changed from standard to short throw for children with limited range of motion; joystick tension can be decreased for those with weak strength; and smaller, remote joysticks can be easily mounted inside the armrest, at midline or in any location where the child has consistent and accurate control. For children who have motor control problems (e.g. those with cerebral palsy or brain injury), powered wheelchair controls can be configured as unidirectional switches instead of proportional joysticks. Pediatric-sized wheelchair frames and custom seating systems provide the support and stability needed to ensure consistent access to the controls. In addition, new powered wheelchair electronics provide adjustable parameters to alter wheelchair performance to fit each child’s ability. Despite improvements in wheelchair technology, it is still difficult for clinicians to determine whether a child has developed the cognitive skills and/or has the temperament to operate a powered wheelchair safely unless the child is actually placed in the wheelchair itself. Even with actual powered wheelchair trials, a dynamic sensorimotor problem is difficult to distinguish from a cognitive developmental delay or from temperament issues. Therefore, clinicians may attempt to 'train' a child for months or even years without determining the source of the primary deficit and may delay prescribing a powered wheelchair without an appropriate focus of intervention in the interim. On the other hand, there are those young children who are cognitively capable and would potentially benefit from independent powered mobility but may not be given the opportunity at an early age. To begin to address these issues, the Rehabilitation Engineering Research Center at Rancho Los Amigos Medical Center (RLAMC) in Downey, California has undertaken a project to develop a cognitive assessment battery to predict a young child's functional performance in a powered wheelchair. The project consisted of three phases: (1) development of a powered mobility program; (2) development of a cognitive assessment battery; and (3) validation of the cognitive assessment battery in predicting powered mobility performance. The current paper describes the development of the powered mobility program. 2. The Powered Mobility Program The Powered Mobility Program (PMP) was developed to introduce and evaluate powered mobility skills in young children ages 18-36 months through exploratory play in the wheelchair. The skills were identified by clinicians as those needed to maneuver a powered wheelchair safely. The battery of items developed to assess these skills is presented below. 2.1. Item development Based on a review of the literature and on input from professionals working in the field of powered mobility, several domains of wheelchair performance were established. These domains reflect a hierarchical nature of complexity of skill level and include basic wheelchair skills (e.g. starting/stopping, directional control), complex transitional wheelchair skills (e.g. sequential combinations of basic skills to perform tasks) and functional maneuvers in structured and unstructured environments. A task analysis was performed of the skills commonly trained and evaluated during initial exposure with a wheelchair. This resulted in 28 wheelchair 'tasks' (e.g. 'Turns to the right from a resting position,' 'Stops at a door with footrests within 12 inches without hitting the door'). These tasks were then reviewed by four professionals in the field of powered mobility training at RLAMC and several tasks were added and/or modified, resulting in a 37-item battery. This battery was then reviewed by eight professionals outside of RLAMC and, based on their feedback, was reduced to 34 items. This battery was then pilot tested with three young children and minor revisions were made to several items. The final version is comprised of 34 items across three domains (i.e. basic skills, integration of skills in structured environments, performance of skills in unstructured environments). Seventeen tasks assess basic skills, including cause and effect association (e.g. pushes joystick to engage wheelchair in motion), directional control (e.g. navigates wheelchair in a forward direction for 10 feet) and speed control (e.g. changes speed to respond to commands such as 'slow down' or 'let's go faster'). These basic skills are incorporated into eleven tasks assessing functional skills in a structured environment, such as driving the wheelchair across an unoccupied walkway and into six tasks assessing performance in an unstructured environment, such as maneuvering in the community, clinic building or shopping mall. A copy of the 34 tasks in the powered mobility assessment is presented in Appendix A. The skills assessed in the PMP are presented as discrete tasks for the purpose of facilitating a more objective scoring process. However, it is recognized that the actual wheelchair experience is less structured and the tasks often blend together as the child is allowed to play and explore in the wheelchair. 2.2. PMP administration Each child participates in 6 one hour powered mobility training sessions, during which the wheelchair is introduced as a means of play and exploratory movement. Basic skills such as turning, moving forward and stopping are learned through the child's own trial and error play with the wheelchair. For example, some children play with the joystick and engage it continuously to the right or left, to experience the sensation of spinning in a circle. By allowing the child to experiment with the joystick, he is able to understand the effect on his movement in the wheelchair. A child who appears intimidated may initially sit in the therapist's lap in the wheelchair and observe activation of the wheelchair to retrieve a toy from a parent. During the actual training sessions, verbal instructions are kept to a minimum. Short phrases such as 'Let's go', 'Where's mom?', 'Follow me' and 'Come here' are used instead of specific directional instructions such as 'Go forward', 'Turn right, now left.' As basic skills are learned, more functional skills are introduced in controlled environments, such as moving along a quiet hallway and passing through a doorway. Some children have even learned directional control by following birds walking on the ground and trying to catch them. Once functional skills are familiar, they are introduced in environments which are distracting and unpredictable, such as a busy clinic. At this point, basic 'rules of the road' such as staying on one's own side of the hall, are also emphasized. Each session is conducted on hospital grounds under the supervision of a single physical therapist to ensure consistency with the powered mobility training approach, the tasks presented and the environment explored. Sessions are designed to last approximately one hour, beginning once the child had been properly positioned in the wheelchair. In general, the first session was somewhat shorter (approximately 20 min) to minimize a sense of frustration or fatigue. Subsequent sessions typically lasted the full hour, with some medically fragile children requiring a short five to ten minute break in the middle for food, drink or rest. Children were always asked whether they wanted to continue in the wheelchair, to allow them a sense of control and to ensure that the experience was a positive one to look forward to in the next session. 2.3. PMP scoring The final wheelchair driving session is videotaped for later review and scoring. Each task is scored according to the amount of hands-on assistance and/or verbal cueing that occurred during the final session. The tasks are scored according to the following scale: 0 - Task not attempted. 1 - Maximal hands-on assistance of joystick with verbal cueing. 2 - Minimal hands-on assistance of joystick with verbal cueing. 3 - Direct stand-by guarding with verbal cueing. 4 - Verbal cueing only. 5 - Age appropriate supervision. Each value on the scale has specific, well-defined criteria describing the amount and type of assistance and verbal cueing given to facilitate consistency in scoring. Item scores are summed and divided by the total number of items. The child's overall score on the PMP reflects the average amount of assistance and/or verbal cueing needed to safely maneuver the wheelchair through all 34 tasks and thus is used to reflect and index of the overall level of independent mobility across tasks.
Table I Etiologies of participants
2.4. PMP testing To date, a total of 24 children (19 male, 5 female) between the ages of 20 and 36 months (x = 28.9 months) have participated in the powered mobility program. All have severe physical orthopedic disabilities which affect independent mobility, but do not have cognitive impairments or learning disabilities (Table 1 presents etiologies. This was intended to minimize the influence of sensorimortor problems that may result from injuries involving central nervous systems disorders such as cerebral palsy. All children were unable to ambulate and had severely limited ability to propel a manual wheelchair. All children used an Everest and Jennings pediatric 'Barbie' wheelchair (12" wide by 12" deep) with a Special Health Systems seating system; both the wheelchair and the seating system were donated for use in this study. The wheelchair frame was modified with a foot bumper guard around the footplates to protect the children's feet and legs from injury if the child bumped into anything and a temporary ventilator tray mounted to the back for use by children who were ventilator dependent. The speed, acceleration and deceleration parameters were consistent for all children; however, the speed during the initial sessions may have been reduced to allow the child to become accustomed to movement. Prior to beginning the PMP, motor access was established using a proportional joystick as the wheelchair input device; all children were able to use proportional joysticks as long as their individual needs were evaluated. For example, the joystick mounting may have been located at midline or on the arm rest and the tension may have been decreased. All children preferred to use their hand to manipulate the joystick, even if they had no active control at this site. They were able to do so because their body size was small enough to allow the use of trunk movement to augment limited or absent upper extremity control. The joystick was individually modified for each child's needs to ensure consistent and accurate access to all directions of movement. The tension of the joystick was decreased for those children with muscle weakness (e.g. those with spinal muscular atrophy, polio). The actual terminal device on the joystick was also important. A round knob on the top was most commonly used. However, for children with impaired hand function (e.g. those with arthrogryposis), a sideways or upright mounted U-shaped bracket allowed easier control. If a child was stronger pulling backward then pushing forward on the joystick, then the, joystick was reversed so that pulling backward moved the wheelchair forward. While it remains uncertain as to whether a proportional joystick is more difficult than unidirectional switches, for the purpose of this project a standard wheelchair input device was used across all children to control for this variable.
3. Results 3.1. PMP scores The children's PMP scores ranged from 0.03 to 4.85 (on the PMP scale from 0-5), with an average score of 2.37 (S.D. = 1.9). Scores of 4 (verbal cueing only) and above were considered to represent functional performance with the wheelchair, while scores of less than 3 (reflecting some degree of hands-on assistance) were considered to indicate that a child was not able to drive the wheelchair functionally. Scores in the 3-4 range (direct stand-by guarding with some verbal cueing) were considered to be marginally functional. In these cases, other mediating factors, such as the child's level of attention, motivation and parental supervision, need to be strongly considered. Table 2 shows the frequencies of scores presented by amount of assistance provided. Table 2 Frequencies by amount of assistance
In examining the pattern of performance across the three different task domains (basic, structured, unstructured), it was found that every child who was able to master the basic skills with no hands-on assistance proceeded to pass the PMP with a final score of 4 or greater, while 6 of the 14 who required minimal to maximal hands-on assistance with the basic tasks were unable to even attempt the structured tasks. Overall PMP scores were not associated with prior independent mobility such as walking, crawling or rolling (X2 = 2.17, NS), however, there was a significant positive correlation with chronological age (r = 0.41, P < 0.05). The youngest child to demonstrate functional performance on the PMP was 25 months.
3.2. Inter-rater agreement Videotaped PMP tasks from a subset of nine children were reviewed by two independent raters. Inter-rater agreement was determined by the Kappa statistic. In order to establish consensus, the PMP scoring scale described earlier (ranging from 0 - task not attempted, to 5 - age appropriate supervision) was subdivided into half-point intervals (e.g. 0-0.5, 0.51-1.0, etc.) along the entire continuum. The overall average PMP scores for the two raters were compared for each child. Overall scores that fell in the same half-point interval were considered in agreement. The overall Kappa scores for inter-rater agreement was 0.87, P < 0.001. Landis and Koch (1977) indicate that scores above 0.75 represent excellent agreement above chance. The scores for the two independent raters were highly correlated (r 0.99, p < 0.01).
3.3. Rater drift In order to establish that raters were consistent over the course of the study, consistency in intra-rater scoring over time was also evaluated. A single rater re-evaluated eight videotaped mobility performances 8-12 months after initial review. Consistency in scoring was determined in the manner described above for inter-rater agreement and the Kappa statistic was used. The Kappa for intra-rater agreement was 0.52, P < 0.01. Landis and Koch (1977) indicate that values between 0.40 and 0.75 represent fair to good agreement above chance. Pearson correlation coefficient between the two scoring times was r = 0.99, P < 0.01, however, scores tended to be slightly lower during the second scoring period.
4. Discussion The PMP represents one component of a project to determine when very young children have the cognitive skills necessary to operate a powered wheelchair. The PMP is an approach designed to introduce young children to a wide range of wheelchair skills necessary for safe, functional use of a powered wheelchair. In addition, the program allows flexibility in introduction and training of new skills, as it begins with spontaneous exploration of movement through basic skills and progressively introduces tasks to promote integration of these skills to develop functional mobility. During the course of the PMP, a child progresses from a level of spontaneously initiated exploratory play in the wheelchair to moving about with functional intent. The approach also allows for the use of creative approaches and incentives to motivate children to try new skills. For example, one child learned to 'back up' by rolling backward over a balloon filled with water in order to burst it. In comparing the patterns of performance for children who were successful with the PMP with those who were not, it was found that children who performed well on the basic skills were able to integrate these skills and master the structured and unstructured tasks, although, as would be expected, they usually required more assistance for the higher level tasks. In contrast, those children who were not successful on the PMP were found to average 1.15 in their basic skills, reflecting a need for maximal hands-on assistance for even these fundamental tasks; most were not able to attempt the higher level skills. When these children were placed in a powered wheelchair, they were able to initiate some types of spontaneous movement (e.g. spinning in a circle) but were not consistent, goal-directed or safe in exploring their surroundings; assistance was consistently needed to stop them before hitting an object or wall. Interestingly, while most of these children were not able to stop on verbal command consistently, several were able to veer to avoid objects or to stop when reaching for a toy. Thus, it appeared that tasks that had a more concrete cause-effect aspect were learned more easily than more abstract verbal commands such as 'stop.' This was especially true when this verbal command was given for reasons not apparent to the child, such as at the bottom of a ramp leading into the parking lot. The PMP represents a continuum of mobility skills, from an exploratory (not necessarily goal oriented) type of movement, transitioning to a more functional level of mobility needed for independence in the home, school and community environments. Mastering the basic skills is believed to be a necessary but not sufficient step in the ultimate development of functional mobility in the community. Some children who attained the basic skills at a level between 3 and 4 still had difficulty integrating these and accomplishing the more difficult, functional tasks. It appeared that children who required assistance with basic skills such as stopping, veering and directional control were not developmentally ready to combine these fundamental skills in tasks in structured and unstructured environments, but were at the stage of exploring their environment through more random movement. In essence, they were interested in exploratory movement, but safety and judgement had not yet been incorporated into their mobility patterns. For these children, other options are available to introduce mobility in a safer, less expensive device. For example, powered mobility toys, adapted for children with disabilities, have allowed children as young as 12 months the opportunity to spontaneously explore their environment (Cooper, 1992; Place and Lipka, 1992). Recently, a transitional powered mobility aide (TPMA) has also been developed which is lower to the ground and uses simple switches to operate (Wright and Barker, 1992). The TPMA was designed, not to replace the powered wheelchair, but to allow children to transition from being non-mobile to acquiring skills needed to control a powered mobility device and ultimately, a powered wheelchair. This mirrors a course of development similar to that of a non-disabled child learning to transition from a non-mobile state to exploratory crawling and finally, to goal-directed independent ambulation. Continued documentation of very young children's abilities to operate power wheelchairs will aid in the clinician's decision-making process and subsequent justification when recommending a power wheelchair to funding agencies. Independent mobility allows young children with physical disabilities to be integrated into appropriate educational programs and can lead to enhanced psychosocial and cognitive development. It is anticipated that the current project will identify the cognitive developmental skills that influence successful power wheelchair mobility. These cognitive skills can be targeted and developed by therapists and parents through appropriate developmental activities. The power mobility program presented here builds on the developmental cognitive skills present in the child so that through exploration and play, she can learn functional mobility skills at her own pace.
Acknowledgements Funding for research was provided by the National Institute on Disability and Rehabilitation Research, U.S. Department of Education, Grant No. H133E00015. Opinions expressed in this paper are those of the authors and should not be construed to represent opinions or policies of NIDRR.
Appendix A Powered Mobility Program tasks Basic mobility skills
References Berenthal, B., Campos, J. and Barrett, K. (1984) Self-produced locomotion: An organizer of emotional, cognitive and social development. In: Emde R. and Harmon R. (Eds.). Continuities and Discontinuities in Development. New York: Plenum, pp. 175-209. Butler, C., Okamoto, G.A. and McKay, T.M. (1983) Powered mobility for very young disabled children. Dev. Med. Child Neurol. Campos, J. and Berenthal, B. (1987) Locomotion and psychological development in infancy. In: Jaffee (Ed.), Proceedings of the RESNA First Northwest Regional Conference. Childhood Powered Mobility: Developmental, Technical and Clinical Perspectives. Washington, DC: RESNA, pp. 11-42. Cooper, R. (1992) An inexpensive approach to early power mobility. Proceedings from the 8th International Seating Symposium, Vancouver, Canada 101-102. Landis, J.R. and Koch, G.G. (1977) The measurement of observer agreement for categorical data. Biometrics 33, 159-174. Place, P. and Lipka. D. (1992) Powered Assisted Mobility for Pre-School Children with Severe Motor Impairments. Proceedings of RESNA International '92 Conference. Toronto, 1992. Tatlow, A. (1980) Towards a comprehensive motor education in the treatment of cerebral palsy.Physiotherapy 66, 322-336. Verberg, G. (1987) Predictors of Successful Powered Mobility Control. In: Jaffee (Ed.), Proceedings of the RESNA First Northwest Regional Conference, Childhood Powered Mobility: Developmental, Technical and Clinical Perspectives. Washington, DC: RESNA, pp. 11-42. Wright, C. and Barker, P. (1992) Personal communication and mobility day camp. Rehabilitation Engineering Center, Lucile Salter Packard. Children's Hospital at Stanford. Stanford. CA. |