This project used a participatory approach [44-47] that involved collaborators from municipalities and the community who were concerned with the daily lives of citizens with disabilities. The aim was to create a partnership that enabled the implementation of concrete actions and modifications that facilitated community participation among individuals with disabilities. A participatory approach was used to encourage the identification of solutions to overcome barriers to community participation, including improvements to MAT design; provision and training in MAT use; or the development of policies, regulations, actions, or services to improve the mobility of individuals with disabilities.

To qualify for the study, participants must have lived in Metro Vancouver, New Westminster, or North Vancouver, British Columbia, or in Quebec City, Quebec, and the surrounding area. Participants were required to use a MAT as their primary means of mobility, which could have been a cane, walker, crutches, manual wheelchair, power wheelchair, or scooter. Participant demographics (ie, age, type of disability, job, gender, or technological affinity) were not considered as eligibility criteria to facilitate the recruitment procedure. Individuals were excluded from the study if they were unable to communicate in French or English or if they could not provide informed consent. People who lived in nursing homes or residential care facilities were also excluded from the study.

To address the above-mentioned objectives, a mixed-methods approach was proposed. All participants took part in a semistructured interview, provided demographic information, and completed the standardized measures. Participants also had the option to participate in three additional methods: (1) GPS tracking of participants’ movement in the community combined with an activity diary, (2) PhotoVoice, and (3) physical environmental audits (see Table 1 for a detailed description of the relationship between the methods and the study objectives).

Participants’ mobility was recorded using a portable GPS (Travel Recorder XT, model BT-Q1000XT, Qstarz International Co). These data were supplemented with an Apple iPad mini (model ME280C/A, Apple Inc)-based activity diary app (ie, the customized Filemaker Go app) that allowed participants to describe the places they visited, their activities, the modes of transportation they used, and whether they were accompanied by others (see Multimedia Appendix 1 for details). The participants took part in a training session on how to use the app and the GPS. Participants were given a troubleshooting document for the devices and contact information for the research assistants in case of any difficulty with the devices and app or for emergencies. Research assistants were also available during the data collection process to answer any questions regarding the equipment and process. The participants used the GPS and app for a 1-week period. Upon completion of the data collection process, the participants returned the devices to the research assistants who reviewed the data (ie, tracks from GPS data and the diaries) with them. The research assistants then conducted a 20-minute, qualitative, semistructured interview with the participants regarding their main MAT and other MATs used.

To describe the sample, the research assistants also gathered quantitative data, including demographic information and the following outcome measures:

The data will also be combined with anxiety and depression data in predictive analyses and will be included in analyses to determine the influence of personal, MAT, and environmental factors on the activity spaces of people who use MAT using the GPS and trip diary data. Each method reported in this protocol requires separate analyses. To avoid bias, the interview guide (see Multimedia Appendix 2) was developed by team members prior to the data collection and all interviewers received training. The interview guide allowed participants to describe their own perception of barriers and facilitators to their mobility. The interview guide was developed to help participants provide their personal experience. According to participants’ addresses, a walking score was calculated using the website Walk Score [54], which is available for every address in the United States, Canada, and Australia. The tool ranks cities and neighborhoods according to the level of walkability, taking into account public transit, better commutes, and proximity to people and places. Our objective was to recruit 36 participants per site (n=72 total), representing 6 users from both sites per type of MAT (ie, cane, walker, crutches, manual wheelchair, power wheelchair, and scooter). Participants were invited to take part in other data collection methods.

PhotoVoice is a community-based participatory research method through which participants are asked to record visual images that capture their lived experiences [55,56]. It facilitates participant empowerment by considering investigators and participants as equal partners in the research process, such that participants are recognized as experts of their own experiences. Through a four-step PhotoVoice process, participants completed training, individual interviews, GPS data collection, and focus groups.

First, each participant took part in a training session on how to operate the camera feature of the app discussed in the previous method used. The training session allowed exchanges on ethical photo etiquette, specifically the importance of using a photo and video release form when taking pictures of other individuals. During the training session, participants worked with the researchers to identify potential images that they might purposefully set out to capture. Participants received a troubleshooting document for the provided device as well as the research assistants’ contact information in case of a problem with the devices and apps or an emergency. They were also provided with a folder containing a summary of the project and photo and video release forms for obtaining consent from individuals in their pictures.

Second, over a 2-week period, which could have been concurrent with the GPS data collection method, participants were asked to take pictures or videos of the mobility- and participation-related barriers and facilitators they encountered. They were encouraged to take the tablet with them at all times during this period. It was suggested that they use the app to record notes about each image, describing their reasons for taking it. At the end of the first week, the research assistants contacted the participants to check in and ask about any problems or difficulties the participants may have been experiencing.

Third, upon completion of the above-mentioned 2-week period, participants took part in an individual PhotoVoice interview to discuss their most significant photos (ie, a personal selection of a maximum of 10 images). The interviewer looked for common themes among the pictures, and discussed suggestions for improvements and how to facilitate them (see Multimedia Appendix 3). This interview lasted approximately 20-30 minutes, depending on the number of photos selected by the participant.

Finally, PhotoVoice focus groups were held until a total of 24 participants per site were recruited for this method of data collection. However, 36 participants per site completed the PhotoVoice procedure excluding the focus group, as they could have also participated in the GPS tracking, activity diary, and qualitative interview phase or they could have been different participants. After 5-7 participants who used various types of MATs finished taking pictures, they were asked to take part in a group discussion about the photos they had taken and a PhotoVoice focus group was planned (ie, approximately three focus groups per site). Participants were not required to participate in the focus groups. Participants who completed focus groups took turns sharing the most important images they had previously selected during their individual PhotoVoice interview. Participants were asked the following questions about their images and videos: “Please describe the photo/video you have chosen.” “Why did you select this photo/video for the interview?” and “Where was the photo/video taken?” Then, the entire group was asked, “Do members of the group have questions or comments about this picture?” After the photos were shared, the group was asked the following questions:

Participants that were unable to attend the focus groups were able to access the results if they were interested. The themes identified by the groups were shared among participants to gain a sense of how the findings resonated. The PhotoVoice focus groups were audio-recorded and transcribed verbatim. Images were numbered and referenced in transcripts. Each focus group lasted 2-2.5 hours.

The transcribed interviews will be analyzed using thematic analysis [57] through an inductive approach to coding. This process requires identification of relevant text in order to categorize data into emerging themes. The visual data collected from participants will be used to supplement the thematic analysis.

At the end of the study, a photo exhibition will be held to celebrate the participants’ work and raise public awareness of relevant issues. Family, friends, relevant stakeholders, and the general public will be able to view the pictures and listen to the stories behind them. The exhibition will be held in local libraries, community centers, or other public venues. Participants interested in taking part in the photo exhibition will be asked to select photos from the previous focus groups and write captions to accompany them. If the participants prefer, captions can be written by the researchers and approved by email to ensure the captions match their interpretations. The participants’ photos and captions will be presented in full consultation with them. Participants will also be integrated in the planning process of the exhibition regarding the type of presentation and the date(s) and place at which the exhibition will take place.

Two instruments were used as environmental audits. The first, a modified version of the Seniors’ Walkability Audit in Neighbourhood, renamed the Stakeholders’ Walkability/Wheelability Audit in Neighbourhood (SWAN) [58-60], was developed to collect objective data across five domains of the built environment: functionality, safety, destinations, aesthetics, and social aspects. The 98-item tool was designed to conduct microscale audits of street segments or blocks between two intersections. In addition to the checklist, participants were asked to take photographs to document barriers or facilitators to walkability and wheelability found in their neighborhood. The SWAN tool also included a secondary form to record additional contextual information that was not taken into account in the audit tool, such as general land use of the area. The original measure was pilot-tested with 24 older adults in three neighborhoods in Frankfurt, Germany, to assess its acceptability and utility and to collect pilot data on microenvironmental features (eg, sidewalk quality and street lighting) [61]. In this study, one research team member was paired with a MAT user to audit four segments in his or her neighborhood. Segments consisted of street sections between two intersections that were purposefully chosen by the participant to highlight barriers and facilitators to their daily mobility. In this way, the SWAN tool helped participants identify areas and elements of the built environment that could be improved for mobility and participation outcomes among MAT users. In our study, 24 participants per site were targeted for this method of data collection.

The SWAN data collection procedure was divided into three parts. The first part consisted of training the participants in how to use the SWAN tool. During the training sessions for the SWAN, participants received a Google Maps image of their own neighborhood. They were asked to identify four segments that they wanted to assess using the tool. The participants chose their own segments according to the following criteria: (1) segments that represented their neighborhood and had environmental features that had barriers or facilitators to maneuvering around with their MATs and/or (2) segments on streets that they frequently traveled.

The research assistants ensured that selected segments were auditable within a 2-hour period (ie, the chosen segments were not too far from one another). The four segments did not necessarily have to be in close proximity to the participant’s house nor were they required to have automobile traffic. The selected segments could be incomplete (ie, due to temporary maintenance work of one sidewalk or crosswalk).

The second part of data collection with the SWAN consisted of user-led data collection for each of the selected segments. Each participant was accompanied by a research assistant during data collection who helped take pictures of environmental factors, while conducting simultaneous audits of the segments. If participants wanted to change a previously selected segment on the day of data collection, they were permitted to do so as long as the new segment was not located at a distance too far from the other three.

The third and last part of data collection with the SWAN was a community forum. Community forums were held to share preliminary findings with the SWAN study participants and stakeholders. Although, it was not mandatory for each participant to attend the community forum, they were encouraged to do so. Invited stakeholders were selected from citizen committees, advocacy groups, city planners, or other community organizations working in the fields of disability, accessibility, and mobility. Members of the advisory committee who serve in the study sites were also invited. These forums fostered dialogue and discussion around methods for facilitating knowledge translation of SWAN findings and identifying potential intervention sites and strategies in each city.

The second audit tool was the Measure of Environmental Accessibility (MEA) [62], which required the research assistants to rate the indicators of the environments identified for improvement by the participants. The MEA was briefly presented to the participants during the PhotoVoice focus groups in Quebec City and during community forums in Vancouver. It included observable and measurable features of the built environment that were considered valuable indicators of accessibility to urban infrastructure for adults with disabilities. The MEA assessed exterior and interior urban built environments, including seven types of urban infrastructures: parking lots, pedestrian facilities, building access from the exterior, access to equipment, interior maneuvering areas, places for learning and leisure, and public restrooms [62]. The MEA labels were deconstructed to create three categories of information: (1) elements (ie, what was going to be evaluated), (2) components (ie, subcategories refining the description), and (3) criteria (ie, what needed to be measured) [62]. The rating scales included the following: (1) actual measures (ie, observable measures in the environment), (2) compliance (ie, regarding an observed measure with the criterion provided for each item—absent, compliant, or not compliant), and (3) observations and modifications (ie, explanations of the observations made and information on possible modifications to be made to improve accessibility) [62]. Most items had good-to-excellent interrater reliability indicators (626/882, 71.0%, using Gwet’s agreement coefficient) [62]. Participants who took part in the PhotoVoice focus groups in Quebec City and the community forums in Vancouver were asked whether certain pictures presented environments to be evaluated with this measure or if other environments that were not included in the pictures should be evaluated. The gathered data from the MEA evaluations will be analyzed by identifying the most recurrent obstacles and by comparing those found in Quebec City and in Vancouver.

The protocol for this study was approved by the Research Ethics Boards at the University of British Columbia (approval number H15-01340), the Centre intégré universitaire de santé et de services sociaux de la Capitale-Nationale (approval number 2015-424), and the regional health authorities of each site. All study participants provided informed consent.