The Occupational Therapy Practice Framework: Domain and Process, 2nd Edition (Framework;AOTA, 2008) states that occupational therapists should support people’s participation in context. Effective assessment and training of mobility device skills is a vital component of increasing participation in people with mobility limitations. In accordance with Framework guidelines, researchers have argued that the most effective environment for training and assessment of mobility device skills needed for participation in the community is in a person’s natural environment (DeJong, 1979; Glass, 1998; Routhier, Vincent, Desrosiers, & Nadeau, 2003; Wade, 2003). Although training in a controlled clinical environment has not been proven to translate to a community context, current training and evaluation are being conducted almost exclusively in clinical settings (Wade, 2003). Little is known about the development, training, and evaluation of mobility skills in more natural settings, making it difficult for occupational therapists to discern whether mobility skills commonly assessed in clinical environments are adequate for people’s participation in community contexts.

The purpose of this pilot study was to develop and test a community mobility skills course (CMSC) to assess and relate people’s mobility device skills in the natural environment to their skills in a controlled clinical environment as measured by an inside mobility skills course (IMSC). This assessment was used to provide insight on mobility device skills needed in community settings that may be useful to occupational therapists when facilitating contextual participation for their clients.

More than 6.8 million Americans use assistive devices to help them with mobility (Kaye, Kang, & LaPlante, 2000). Of this group, 1.7 million people use a wheelchair or a scooter, and 6.1 million people use a device such as a cane, crutch, or walker. A significant portion of mobility device users experience limitations in participation: 84% of mobility device users report being limited in one or more activities, compared with 13% of people who do not use such devices (Kaye et al., 2000). Forty-three percent of mobility device users report being unable to perform a major life activity, compared with 3.5% of people who do not use mobility devices. Wheelchair and scooter users appear to be the most affected device groups: Approximately 66% and 60% report being unable to perform one or more major life activities (Kaye et al., 2000). Fifty-three percent of walker users, 45% of crutch users, and 38% of cane users report being unable to perform a major life activity (Kaye et al., 2000).

Mobility devices provide effective support to people with mobility limitations in their everyday participation (Kaye, Kang, & LaPlante, 2002), and increasing evidence has suggested that mobility devices may positively affect participation. Noreau, Fougeyrollas, and Boschen (2002), for example, surveyed people with spinal cord injury and found that assistive devices were perceived as a major facilitator for participation. Kilkens, Post, Dallmeijer, van Asbeck, and van der Woude (2005)  studied people with spinal cord injury and found that manual wheelchair skills were positively associated with participation.

Training mobility device skills so that they can be most efficiently used is an important component of increasing participation in people who use mobility devices. Evidence has suggested that skills training designed to help people overcome physical barriers in the environment and participate in life activities is beneficial to people who use a mobility device and may positively affect their independence (Kilkens, Dallmeijer, De Witte, Van Der Woude, & Post, 2004; Kilkens, Post, Dallmeijer, Seelen, & van der Woude, 2003). Simmons, Schnelle, MacRae, and Ouslander (1995)  found that in addition to increased feelings of independence, training can improve freedom of movement and quality of life for mobility device users.

Several tools for assessing mobility device skills exist. One such measure is the Wheelchair Skills Test Version 2.4 (WST; Kirby et al., 2004), used to assess wheelchair function as it applies to daily life. The WST consists of 50 wheelchair skill tests of varying levels of difficulty (Kirby et al., 2004). The Wheelchair Circuit, developed by Kilkens and colleagues (2004), assesses manual wheelchair mobility and tests eight wheelchair skills. The Obstacle Course Assessment (Routhier, Desrosiers, Vincent, & Nadeau, 2005) of wheelchair user performance consists of 10 obstacles and is designed to assess skills that a wheelchair user can and cannot perform. These assessments, however, test only power or manual wheelchair skills and do not include skill tests for other types of assistive devices, such as scooters, canes, crutches, and walkers. Moreover, each assessment is conducted in clinical settings. Although clinical settings are conducive to training interventions because the environment is convenient and controlled, such settings are less relevant to the mobility device user, and it remains unknown whether mobility device skills used in controlled settings generalize to the community (Wade, 2003).

Routhier et al. (2003)  argued that assistive device training should be available to people with mobility impairments using devices and should be carried out in natural environments, but little is known about mobility device skills training in community environments. The purpose of this study was to develop and pilot test a community-based mobility device skills assessment protocol tailored to each of the following six device groups: power wheelchair, manual wheelchair, scooter, cane, crutch, and walker. In addition, participants’ CMSC performances were compared with their IMSC performances to determine whether the IMSC provided adequate preparation for the mobility skills required in a natural environment.

Participants for this pilot study were selected from a sample of 91 adults with mobility impairments who had previously completed the IMSC and Characteristics of Respondents (CORE) survey (Gray, Hollingsworth, Stark, & Morgan, 2008). Inclusion criteria required that participants lived in the community, had been using their mobility device for a minimum of 2 yr, and had been discharged from rehabilitation for at least 1 yr at the time of the study. Six people who met the criteria, each of whom used a different mobility device, were selected for participation. The participants were regular users of the community exercise facility where the IMSC was located and were selected in part because of their demonstrated reliability when attending exercise appointments. The devices used by the participants were a power wheelchair, a manual wheelchair, a scooter, a cane, a crutch, and a walker. Participants ranged in age from 28 to 63 and included 2 men and 4 women.

The IMSC course and protocol served as the basis for the CMSC course and protocol. The IMSC is designed to train participants to become proficient in skills required to use their mobility device efficiently. The IMSC was derived from the WST (Kirby, Swuste, Dupuis, MacLeod, & Monroe, 2002), a skills course used to identify a wheelchair user’s skills in a safe and controlled environment. IMSC tasks are specific to the mobility device used and include the following: opening and closing a door in the toward direction, opening and closing a door in the away direction, level driving or walking in the forward and backward directions, moving turns (commonly referred to as figure 8s), turning in place in the right (clockwise) and left (counterclockwise) directions, walking up and down two stairs with handrails on both sides, moving on a cross slope with both a right and left slope, going up and down a 1.5-in. curb, and driving or walking over uneven terrain (see Figure 1). Manual wheelchair users were asked to perform a “wheelie”1 in different situations; cane, crutch, and walker users were asked to ascend and descend a flight of stairs.

The IMSC tasks were scored on 4-point scales for completion and safety, and certain tasks were timed. A completion score of 4 indicated that the participant completed the task with no verbal or physical assistance from the evaluator; a completion score of 1 indicated that a participant was unable to complete a task. A safety score of 4 indicated that the participant completed the task safely, whereas a safety score of 1 indicated a maximum safety risk or that the evaluator had to intervene to prevent injury. Time was scored in seconds. Participants initiated a task on the evaluator’s instruction, and beginning and end markings for each task were used to ensure reliable and accurate time measurements.

Participants’ IMSC performances were assessed by a member of the research staff, which consisted of occupational therapists and recent occupational therapy graduates completing fieldwork. Each research staff member observed several assessments by other staff and was supervised administering several assessments before completing assessments on participants’ IMSC performances independently.

The CMSC was created in a St. Louis, Missouri, community at the Enabling Mobility Center, a collaborative effort between the Washington University Program in Occupational Therapy and Paraquad, a Center for Independent Living in St. Louis. All CMSC tasks were modified from the IMSC tasks that had been determined by the research staff to test skills that were helpful for participation in the community. CMSC tasks were designed using materials present in the community environment. Tasks consisted of opening and closing a door in the toward direction, opening and closing a door in the away direction, level driving or walking in the forward and backward directions, moving turns (figure 8s), turning in place in the right and left directions, moving on a cross slope in the right and left directions, going up and down a 1.5-in. curb, driving or walking over uneven terrain, and driving or walking up and down a ramp with a handrail (see Figure 2).

The same trained member of the research staff (either Walker or DeGroot) completed an evaluation form for each participant’s CMSC performance. A second member of the research staff (either Walker or DeGroot) was present during each evaluation to ensure participants’ safety while they performed the CMSC tasks. Like the IMSC evaluation, the CMSC evaluation included scores for completion, safety, and time. In addition, certain CMSC tasks were scored for task instruction. Task instruction allowed participants to instruct another person to complete the task for them without receiving a penalty for not completing the task independently. Task instruction was also scored on a 4-point scale: a score of 4 indicated that instructions were clearly provided by the participant and a score of 1 indicated that the participant was unable to give instructions about how to complete the task. Higher scores on task instruction indicated better performances.

The CORE Survey measures demographics, general health and wellness, assistive technology use, transportation use, and current participation in exercise. The CORE Survey was administered to participants at the time of their IMSC completion.

The Washington University human subjects committee approved all procedures of this study. Six qualified people, one from each device group and each of whom had previously completed the IMSC and CORE Survey, were contacted by Kimberly A. Walker and invited to participate in the current study. After informed consent was obtained, each participant was evaluated on the CMSC. Before starting each task, verbal instructions on how to complete the task were provided. If necessary, a member of the research staff provided a visual demonstration by walking through the task as an example. The participant was then scored on all tasks for time, completion, safety and, when appropriate, task instruction. During each task, the participant was carefully monitored by a member of the research staff to ensure safety. Once the CMSC had been completed, each participant’s scores on each CMSC task were compared with his or her scores on the equivalent IMSC tasks. The amount of time between a participant’s completion of the IMSC and his or her completion of the CMSC varied from the same day to 9 mo. Participants did not receive monetary compensation for completing the CMSC, but they were reimbursed $50 for their completion of the IMSC and the CORE Survey.

Scores for completion, safety, and task instruction did not differ for any participant between the IMSC and the CMSC. With the exception of Participant 5, who received a score of 3 for safety on the task of turning in place in the right direction, all participants received scores of 4 for completion, safety and, when appropriate, task instruction unless they chose not to complete the task on the IMSC and CMSC. The prevalence of scores of 4 may have been because of the high level of experience participants had with their devices.

Time scores differed between the IMSC and CMSC performances and are therefore reported in detail. The opening and closing a door in the toward direction and the opening and closing a door in the away direction tasks were not timed on the IMSC and therefore could not be compared with CMSC times for these tasks.

Participant 1 was a 40-yr-old woman who had used a power wheelchair for 15 yr after incurring a C5-level spinal cord injury as a result of a motor vehicle accident. Participant 1 completed the IMSC 9 mo before she completed the CMSC. She completed each of the eight tasks more quickly on the CMSC than she did on the IMSC (see Table 1). Although she did not ask for any assistance for the IMSC tasks, Participant 1 asked for assistance in completing the tasks of opening and closing a door toward and opening and closing a door away on the CMSC. Participant 1 completed the task of going up a 1.5-in. curb on the IMSC but refused to attempt the task on the CMSC because she was not comfortable going up a curb in the community environment. She stated that if she encountered this obstacle in her community travel she would find another route to use.

Participant 2 was a 57-yr-old man who had used a manual wheelchair for 30 yr since incurring a T7-level spinal cord injury as a result of a gunshot wound. Participant 2 completed the IMSC 6 mo before completing the CMSC, and he completed 10 tasks on both the IMSC and the CMSC. Of the 10 tasks, Participant 2 completed 6 tasks more quickly on the CMSC than on the ISMC: level propelling forward, moving turns, going across a cross slope (both right and left directions), going up a 1.5-in. curb and going down a 1.5-in. curb (see Table 1). He completed turns (in both right and left directions) and inclines (both ascending and descending) slower on the CMSC. While moving down the incline, Participant 2 ran into the handrail on the right side, increasing the time it took to complete this task. On the CMSC, Participant 2 attempted to open the door away independently, but was unable to complete this task. When prompted by the evaluator to ask for help, he was able to give the evaluator instructions to open the door. He refused to attempt the uneven grass terrain on the CMSC because he was not comfortable completing the task in his device. Because of heterotrophic ossification, Participant 2 sits in a semireclined position in his wheelchair and is unable to perform wheelies. When asked by the evaluator how he would handle this obstacle if encountered in the community, he stated that he would find another route to use.

Participant 3 was a 42-yr-old woman with athetoid cerebral palsy since birth who had used a scooter for mobility for 12 yr. Participant 3 completed the IMSC the same day she completed the CMSC and performed all 12 tasks on both the IMSC and the CMSC. She performed 9 tasks faster on the CMSC than the IMSC, with the exception of (1) cross slope in the right direction and going down a 1.5-in. curb, which she completed more quickly on the IMSC, and (2) propelling on uneven terrain, on which she showed no time difference (see Table 1).

Participant 4 was a 28-yr-old woman who had used a quad cane for 2 yr since she was diagnosed with multiple sclerosis. Participant 4 completed the IMSC 2 mo before she completed the CMSC, and completed eight tasks on both the IMSC and the CMSC. She completed moving turns, going up and down a 1.5-in. curb, and going up an incline faster on the CMSC than she had on the IMSC (see Table 1). However, she completed the left and right turns in place, uneven terrain, and going down an incline slower on the CMSC than on the IMSC. Participant 4 declined to complete the IMSC tasks of maneuvering a flat surface in both directions and cross slopes because of fatigue.

Participant 5 was a 35-yr-old man who was born with an unknown type of cerebral palsy and who had used a single forearm crutch for 28 yr. Participant 5 completed the IMSC the same day he completed the CMSC. He completed 9 tasks on the IMSC and 11 tasks on the CMSC. He refused to complete the task of walking in reverse on both the IMSC and the CMSC because of balance issues when walking backward. He also refused to complete the tasks of moving across a cross slope in the left direction and going down a curb on the IMSC. He completed eight of the tasks more quickly on the IMSC; he moved across the uneven terrain on the CMSC more slowly than he did on the IMSC (see Table 1). When turning in the right direction, Participant 5’s crutch went outside the allowed area, causing him to receive a safety score of 3 for this task.

Participant 6 was a 63-yr-old woman diagnosed with arthritis who had used a walker for 8 yr. Participant 6 completed the IMSC 7 mo before she completed the CMSC. Participant 6 completed all 12 tasks on the IMSC and on the CMSC. She completed most tasks more quickly on the CMSC than the IMSC, with the exceptions of (1) turning in the left direction and moving across uneven terrain, which were completed more slowly on the CMSC than the IMSC, and (2) moving turns, which showed no time difference (see Table 1).

In general, participants completed CMSC tasks more quickly than they completed the corresponding IMSC tasks. Not surprisingly, performance on tasks varied from participant to participant. The power wheelchair user showed improvement on every task she performed on both the IMSC and the CMSC, whereas the cane user showed improvement on the fewest tasks, improving on 50% of the tasks she performed on both courses. The crutch user showed improvement on almost 90% of tasks, the scooter and cane users showed improvement on 75% of tasks, and the manual wheelchair user showed improvement on 60% of tasks. These results suggest that the amount of mobility device skill training needed for community mobility may depend on the mobility device used.

On some tasks, there were universal improvements in time from the IMSC to the CMSC. Driving on a level surface in forward and reverse directions and moving turns showed a decrease or no increase in time for participants who completed these tasks on both courses. These findings suggest that these skills may be generalized from a clinical setting to a community setting for mobility device users independent of the kind of mobility device used.

Other, more difficult tasks, including opening and closing a door, turning in the left and right directions, moving across uneven terrain, and going up and down inclines and curbs, however, did not show uniform improvement, supporting claims of researchers who have argued that the skills needed for device use in clinical settings differ from those required for device use in community settings (DeJong, 1979; Glass, 1998; Routhier et al., 2003; Wade, 2003).

The power wheelchair user refused to ascend a 1.5-in. curb on the CMSC, although she had done so without problem on the IMSC. The manual wheelchair user refused to attempt uneven terrain on the CMSC, although he had completed this task on the IMSC. The wheelchair users were the only participants who refused to complete tasks on the CMSC, supporting the findings of Kaye et al. (2000)  that wheelchair users are more likely to be limited in activities than other device users. Both participants cited safety concerns as their refusal reason, indicating that skills sufficient to meet safety concerns in controlled contexts may be insufficient in community settings for some mobility devices. Indeed, for a manual wheelchair user to ascend or descend a curb or to propel over uneven terrain, he or she may need to perform a wheelie—a higher level skill that not all manual wheelchair users have mastered. A power wheelchair user, who often has a more significant impairment than a manual wheelchair user, must be aware of the possibility that the wheelchair could tip while the user ascends or descends a curb or propels over uneven terrain. The predictability of a controlled environment may help wheelchair users feel safe when attempting these tasks, but they may be dissuaded from attempting more challenging tasks in a community setting. Training and practicing certain skills in community settings, then, may help to decrease safety risks that prohibit participation in community contexts.

Several factors limit the inferences that can be made from this study’s findings. First, because this project was a pilot study, the sample size was small. In addition, the time between ISMC and CMSC completion was not standardized for participants, and it is not clear what impact, if any, the variation in time had on the results. The CMSC contained more open space, and participants were therefore less likely to run into walls or furniture than they were on the IMSC, which may have partially contributed to faster scores on CMSC tasks. Finally, participants completed the CMSC during the winter, and the cold weather may have been an especially motivating factor for completing the CMSC tasks as quickly as possible.

Some mobility device skills assessed in clinical environments appear to transfer to community settings, and some higher-level skills appear to be unique to the community setting. Moreover, challenging community tasks appear to differ from device user to device user, suggesting community mobility skills training should be specialized for the mobility device used. Further research is needed to determine whether the differences found in this study are sustained for larger groups of mobility device users and to understand the factors linked to these differences. Using a larger sample, researchers could examine useful information for the generalization of mobility skills taught in clinical settings to improve performance in the community.

The CMSC developed during this study is feasible and easy to use. The assessment requires no costly equipment, takes a short time to complete, and should be simple to duplicate in other community environments. The CMSC provides a tool for identifying problematic barriers and skills that should be trained to improve safe and efficient use of mobility devices in community settings, and ultimately have a positive influence on increasing community participation of people with mobility limitations.

This project was supported by the following people: Sue Tucker, Carla Walker, Jessica Dashner, Melissa Lyles, Polly Gray, and Denise Curl. This study was funded by the National Institute on Disability and Rehabilitation Research (Grant No. H133A010701).

1A wheelie results from the front wheels of a wheelchair coming off the ground when sufficient torque is applied to the rear wheels. Wheelies allow mobility device users to increase obstacle-negotiating abilities (MacPhee et al., 2004).