Outdoor Built Environment Barriers and Facilitators to Activity among Midlife and Older Adults with Mobility Disabilities

Abstract

Purpose: To gain better understanding of how the built environment impacts neighborhood-based physical activity among midlife and older adults with mobility disabilities. Design and methods: We conducted in-depth interviews with 35 adults over age 50, which used an assistive device and lived in King County, Washington, U.S. In addition, participants wore Global Positioning Systems (GPS) devices for 3 days prior to the interview. The GPS maps were used as prompts during the interviews. Open coding of the 35 interviews using latent content analysis resulted in key themes and subthemes that achieved consensus between coders. Two investigators independently coded the text of each interview. Results: Participants were on average of 67 years of age (range: 50–86) and predominantly used canes (57%), walkers (57%), or wheelchairs (46%). Key themes pertained to curb ramp availability and condition, sidewalk availability and condition, hills, aesthetics, lighting, ramp availability, weather, presence and features of crosswalks, availability of resting places and shelter on streets, paved or smooth walking paths, safety, and traffic on roads. Implications: A variety of built environment barriers and facilitators to neighborhood-based activity exist for midlife and older adults with mobility disabilities. Preparing our neighborhood environments for an aging population that uses assistive devices will be important to foster independence and health.

Nearly 40% of adults over age 45 have difficulty with physical movement (Altman & Bernstein, 2008). Twenty percent of the adult population residing in the United States has a disability compared with 51.8% of those over the age of 65 (Brault, 2008). People with disabilities are at risk for a variety of secondary health conditions that physical activity can prevent, including osteoarthritis, decreased fitness, pain and fatigue, weight gain and obesity, decreased quality of life, increased functional impairment, and depression (Boslaugh & Andresen, 2006; Rimmer, 1999). Current physical activity guidelines recommend all adults obtain at least 150min of moderate intensity activity or 75min of vigorous activity per week in addition to balance and strength training (U.S. Department of Health & Human Services, 2008). Yet, fewer adults with disabilities meet physical activity guidelines (37.7%) compared with those without disabilities (49.4%) (Centers for Disease Control & Prevention, 2007). In addition, as adults with disabilities age, their physical activity levels decline at a greater rate than those without disabilities (Altman & Bernstein, 2008; Boslaugh & Andresen, 2006). Means of curbing this decrease in physical activity are needed.

Most physical activity interventions among persons with disabilities have occurred through group, home, or rehabilitation-based programs (Rimmer, Chen, McCubbin, Drum, & Peterson, 2010). Yet, physical activity related to daily living can also be used as a means of counteracting disability as suggested by recent research (Welmer, Morck, & Dahlin-Ivanoff, 2011). Promoting lifestyle-based, unstructured activity and exercise that can be done in one’s own community is an underutilized approach among persons with mobility disabilities. This would include, for example, physical activities that are performed for transportation, such as walking to and from useful places, such as the public transportation stops or a grocery store.

An important barrier to performing neigh borhood-based physical activity for persons with disabilities is access to built environments that support active options (Rimmer, 1999). Ecological models emphasize the role played by social and physical environmental influences on behavior, including physical activity (Satariano & McAuley, 2003). Furthermore, the Health Impact Pyramid suggests that interventions that change the environment to make healthy choices the default have greater public health impact than those involving direct clinical care, counseling, and education (Frieden, 2010). The Disablement Process model also acknowledges the limitations imposed by the built environment and, in particular, how the built environment interacts with physical ability to promote disability (Imrie & Kumar, 1998; Verbrugge & Jette, 1994). In addition, the field of environmental gerontology recognizes the importance of place as people age and has posited models of how aging interacts with the environment (Kendig, 2003; Wahl, Iwarsson, & Oswald, 2012). Models of mobility, such as the comprehensive framework, acknowledge that the environment is one of several determinants of mobility across various life-space locations (e.g. outdoors, neighborhood) (Webber, Porter, & Menec, 2010). The built environment can be assessed with objective (e.g., using variables derived from Geographic Information Systems or audits) or perceived (e.g., Neighborhood Environment Walkability Scale) measures (Ding, Sallis, Kerr, Lee, & Rosenberg, 2011). Although objectively assessed built environment tends to show stronger relationships with outcomes (Ding et al., 2011), people’s perceptions of what is available in their environment is considered important and believed to impact behavior. Most researchers acknowledge that assessing both objective and self-reported built environment is ideal (Ding et al., 2011).

When the built environment is supportive, individuals can use the outdoors for multiple activities including exercise and utilitarian purposes and can better access exercise facilities. Those aging with disabilities are likely more vulnerable to environmental constraints (Yen, Michael, & Perdue, 2009), but there are few studies that examine the impact of the built environment on the physical activity of older adults with disabilities or mobility impairments. A few studies were conducted in adult samples among people with various types of disabilities. One such study among adults with disabilities found that the presence of built environment buoys (i.e. facilitators of physical activities) was related to physical activity and active transportation (Spivock, Gauvin, Riva, & Brodeur, 2008). Curb ramps were noted to be important built environment barriers in another study among adults with disabilities (Rimmer, Riley, Wang, Rauworth, & Jurkowski, 2004). Problems with sidewalk pavement, puddles, and poor drainage were environmental barriers among people with visual and motor impairments; curb ramps and narrow sidewalks were also common barriers (Kirchner, Gerber, & Smith, 2008). Fewer studies have focused on older adults with mobility disabilities. In a recent study, adults over age 65 with higher levels of mobility impairment reported less physical activity and higher body mass index but were more able to do more physical activity for transportation when they lived in walkable neighborhoods (King et al., 2011). Among older adults with knee osteoarthritis-related functional limitations, Keysor found that high-community mobility barriers were related to activity limitations (Keysor et al., 2010). In a study of adults with disabilities, having sidewalks was associated with more physical activity and people with disabilities reported less access to community facilities for physical activity (Christensen, Holt, & Wilson, 2010). However, no known studies have adequately examined how older adults with mobility disabilities themselves view neighborhood barriers to physical activity.

Physical activity is considered one form of mobility (Prohaska, Anderson, Hooker, Hughes, & Belza, 2011). Few studies have examined how the broader concept of mobility relates to the built environment. A recent review concluded that higher street connectivity, safe street and traffic conditions, and proximity to destinations enhanced mobility, but the vast majority of studies used walking as the mobility outcome (Rosso, Auchincloss, & Michael, 2011).

The purpose of this study was to gain increased understanding of the barriers and facilitators to neighborhood-based physical activity from adults over 50 years old with mobility disabilities utilizing in-depth qualitative interviews. We defined mobility disability as use of an assistive device to ambulate. Our goal was to include anyone who used an assistive device to ambulate and was not condition or disease based. Our definition of assistive device was left broad so that users of assistive devices that are less commonly considered could be included. Also, we had no exclusion for people aging with specific conditions that lead to disability (e.g., multiple sclerosis) versus people aging with general conditions (e.g., decreased ability to walk due to arthritis or aging-related processes).

Our approach to better understanding built environment barriers to activity was qualitative for several reasons. Due to a dearth of existing research, an improved understanding of people’s perceptions of the built environment is needed, using bottom-up, rather than researcher-driven approaches. Qualitative research methods obtain detailed information that is difficult to measure in quantitative terms, such as understanding perceptions of an issue or decision-making processes. This information may be obtained by a variety of methods, including interviews, focus groups, and direct or indirect observation. Qualitative research findings can be used to tailor messages and materials, develop interventions, and develop or adapt measures (Nichter, Thompson, Shiffman, & Moscicki, 2002; Strolla, Gans, & Risica, 2006). Thus, such methods were deemed appropriate for the current investigation.

Design and Methods

Study Design

The Built Environment, Accessibility, and Mobility Study (BEAMS) was conducted from January–June 2011 in King County, Washington. King County has diverse typology including urban and suburban areas. We conducted home-based in-depth individual interviews supplemented with maps derived from Geographic Positioning System (GPS) devices. Interviews were selected to obtain detailed information about perceptions of the built environment in our study sample becuase prior studies are lacking.

Recruitment and Participants

We aimed to recruit 25–40 participants with the goal of reaching theme saturation (i.e., no further new concepts or themes are generated from additional participant interviews) (Sandelowski, 1995). To capture various view points, we attempted to recruit participants from diverse neighborhood types (high and low walkability; high and low income), that used a range of assistive devices including canes, walkers, manual wheelchairs, and powerchairs, and had a variety of disability types. Our main recruitment methods included announcements in relevant organizational e-newsletters (e.g., the MS [Multiple Sclerosis] Society, Arthritis Foundation, and senior center newsletters) and by distributing flyers at community events, senior centers, and senior housing facilities. Study inclusion criteria were 50 years of age and older, use of an assistive device to ambulate, reside in King County, Washington, speak and read English, leave home 3 days per week or more, and willing to have researchers come to their home.

People who learned about the study from our recruitment methods and who were interested in participating called the study phone line. The research coordinator spoke with each interested individual to explain the study, verify eligibility, answer questions, and obtain verbal consent. We screened 42 people for eligibility and interest in participating by phone. Seven did not enroll for various reasons including a death in the family (n = 1), not meeting eligibility criteria (n =1), health issues precluding them from participation (n = 3), and deciding not to participate (n =1).

Study Procedures

Human subjects’ approval was obtained from the Institutional Review Board, University of Washington. After we obtained verbal consent for participation, a consent form and a GPS device (Qstarz BT-Q1000XT, Qstarz International Co., Ltd., Taipei, Taiwan) with instructions for use were mailed out. Participants were instructed to wear the GPS device for 3 days, including 2 weekdays and 1 weekend day. The GPS and written consent forms were returned by mail to study researchers using a prepaid envelope. On receiving consent forms and the GPS device, the study coordinator contacted participants to schedule a home visit. Before the visit, study staff used www.walkscore.com (Carr, Dunsiger, & Marcus, 2011; Duncan, Aldstadt, Whalen, Melly, & Gortmaker, 2011) to obtain information about locations (e.g., grocery stores, parks, shopping) within 1–2 miles of the participant’s home to use as prompts during the in-depth interview. In addition, maps of the trips captured through GPS device use were printed in color to use as prompts during interviews.

Interview Procedures

Two study researchers visited participants in their homes to conduct a 2-hour in-depth interview. One study researcher served as the interviewer and the other assisted by taking detailed notes of the interview using a laptop. A structured interview protocol was developed by the study team. The protocol was based on urban planning and transportation literature, which suggests important environmental characteristics promoting physical activity, including residential density, access to useful destinations, street connectivity, facilities for walking (e.g., sidewalks), aesthetics, traffic, and crime safety (Saelens, Sallis, Black, & Chen, 2003). The protocol was reviewed by five experts for face validity and was pilot tested with one older adult. All interviewers were trained researchers with a minimum of master-level training in a health-related field or a registered nurse degree. All interviewers underwent a minimum of 5hr of training in using the interview protocol, including practicing with other study researchers and volunteers and observing the lead researcher (D.R.) conduct an interview. The note takers were instructed to take near verbatim notes and taught how to use an Excel template to fill in the answers to each item on the semistructured interview guide. The note takers underwent training and had to demonstrate their ability to accurately take notes during at least one practice interview session.

The semistructured interview guide utilized open-ended questions along with prompts to facilitate the interview process. The interview protocol consisted of four key portions centering on barriers and facilitators to (a) accessing and using the locations visited while using the GPS device (up to three locations were discussed using printouts of the participant’s GPS maps as a prompt); (b) use of indoor physical activity locations in the neighborhood; (c) use of outdoor physical activity locations in the neighborhood; (d) accessing utilitarian locations in the neighborhood (up to 10 locations were discussed) (interview protocol available on request from the lead author). In addition, if participants were deemed not to be at risk for falling using a brief screening questionnaire, we accompanied them on a brief tour of their outdoor immediate local area. The outdoor route was selected by the participant as a route they take at least once per week. During the outdoor interview, participants were prompted to provide their impressions of built environment features shown to relate to walking based on urban planning literature (Saelens et al., 2003) and literature examining built environments among adults with disabilities and older adults (e.g., curb ramps, sidewalk presence and conditions, safety, adequate crossings (Christensen et al., 2010; Kirchner et al., 2008; Rimmer et al., 2004; Yen et al., 2009). For this portion of the interview, detailed handwritten notes were taken and data entered into the Excel template immediately following the interview. Participants unable to go outdoor tour were asked the same questions while in the participant’s home. Following the interview, the team members who conducted the visit debriefed and double checked the content and accuracy of the interview notes.

Measurement of Demographic and Health Characteristics

To characterize the sample, median household income was obtained using Census 2000 data at the census tract level (the most current data available during this study). The website www.walkscore.com was used to estimate walkability of participants’ residential areas. Walkscore is validated to use as an estimate of proximity to walkable destinations (Carr et al., 2011; Duncan et al., 2011). Walkscore values of 0–49 are considered car dependent, 50–69 somewhat walkable, 70–89 very walkable, and 90–100 walker’s paradise (Walkscore, 2012). Participants self-reported their use of assistive devices (using a checklist), medical conditions, and race/ethnicity in a written survey. The survey also assessed the number of minutes and hours per day spent on walking or wheeling for exercise or pleasure.

Data Analysis

The typed, detailed interview notes were uploaded into Atlas.ti version 6 (Berlin, Germany). A start list of outdoor built environment codes (Miles & Huberman, 1994) were created by study team members using an inductive approach after thoroughly reviewing the interview content. Codes were developed based on latent content analysis (i.e., analyzing themes and concepts) of participants’ perceptions of the outdoor built environment (across destinations) in the interview notes. We used primarily latent content analysis to determine themes introduced by study participants. The codes represented major theme groupings from participants’ perceptions of the built environment (e.g., sidewalks, street crossings, lighting). The two coders represented different training backgrounds (D.H., geriatric medicine; D.R., clinical psychology and public health) and independently coded all interview notes and compared results. One had not participated in interviews (D.H.), whereas the other participated in over one-third of the interviews (D.R.). Both coders reviewed all interview text to make initial codes. Similar to other qualitative studies that used in-depth interviews (Lingler et al., 2006), we used iterative procedures to conduct team meetings where coding was discussed until consensus was achieved. A detailed audit trail to track decisions and definitions for coding content was kept. The coders compared results and discussed each discrepancy to determine final codes. Agreement on codes ranged from 56% in initial coding rounds to 96% in later coding rounds; coding agreement was 100% after discrepancies were discussed. Due to the large number of broad themes coded, we grouped together subthemes of barriers and facilitators (e.g., debris in curb ramps, curb ramps too steep) within the larger coded themes (e.g., curb ramp conditions). This process was enhanced by a meeting with our project advisory board consisting of members from diverse backgrounds (nursing, rehabilitation medicine, urban planning, architecture, social work) that came from organizations that worked with adults aging with disabilities (e.g., MS Society, Arthritis Foundation). Discussions with our project advisory board helped us interpret the meanings in our findings. Descriptive data of assistive device use, medical conditions, self-reported physical activity, and race/ethnicity were examined using SPSS version 19.

Results

Thirty-five participants completed the in-depth interviews. Participants were on average 67 years old, predominantly women and white, and lived in areas considered somewhat walkable according to average Walkscore value (see Table 1). About one-third of participants reported no physical activity. The majority of participants used more than one type of assistive device (54%) and had multiple chronic conditions (mean = 3.88; standard deviation = 1.84; range 0–8). In addition to the most common conditions reported in Table 1, other conditions that participants reported included diabetes (n = 6), spinal cord injury (n = 6), rheumatoid arthritis (n = 5), cancer (n = 4), traumatic brain injury (n = 3), cerebral palsy (n = 2), chronic obstructive pulmonary disease (n = 2), multiple sclerosis (n = 1), postpolio syndrome (n = 1), fibromyalgia (n = 1), and amputation (n = 1). Canes, walkers, and wheelchairs were the most common assistive devices used by participants. Other types of assistive devices were rocker-soled shoes, a bicycle, a guide dog, and a specially adapted walker. The participant who reported using a bicycle as his assistive device stated that he would be unable to walk to important destinations due to arthritis in his knees, which made walking on hills difficult (he lived in a hilly area); his bicycle made it possible to go to local destinations.

Table 1.

Demographic Characteristics of Participants (N = 35)

Characteristic	Number (N) or mean (M)
Age, mean (range)	67 (50–86)
Female, N (%)	26 (74)
Walkscore, mean (range)	67 (18–98)
Engage in physical activity,a N (%)	21 (66%)
Minutes of physical activity/ week,b median (range)	270 (range 30–1260)
Median household income, mean (range)	46,199 (25,821 - 94,179)
Assistive device type, N (%)
Cane user	20 (57)
Walker or rollator user	20 (57)
Powerchair or scooter user	9 (26)
Manual wheelchair user	7 (20)
Other	4 (11)
Race/ethnicity, N (%)
White	30 (86)
African-American	2 (6)
Asian-American	1 (3)
Multiracial	1 (3)
Most common medical conditions, N (%)
High blood pressure	18 (51)
Visual problems	16 (46)
Osteoarthritis	13 (37)
Heart condition	10 (29)
Mental health condition	10 (29)
Fall in the past 6 months	10 (29)
Hearing problems	7 (20)
Stroke	7 (20)

^aOut of 32 participants that answered these items.

^bAmong those reporting any physical activity.

Built Environment Themes

Table 2 displays the identified themes from participant interviews in a list format. Key built environment features were curb ramps, parking, aesthetics, lighting, ramps, weather, crosswalks, sidewalks, amenities, walking paths/trails, safety, and geographical features (e.g., hills). In the table, only themes that received consensus (defined as four or more participants noting that theme) are described. For the most part, we summarized remarks across participants and across places they reporting both going and choosing not to go. Where we provide specific examples, participant numbers are identified in parentheses to denote the source of the remark.

Table 2.

Summary of Built Environment Barriers and Facilitators

Barriers Concept/theme	Facilitators Concept/theme
Curb ramps
Lacking curb ramps in many places Subthemes: need to walk or wheel in street when there is a lack of curb ramps; curb ramps only on one side of the street	Presence of curb ramps promotes mobility Subthemes: certain areas have lots of curb ramps (downtown Seattle, senior friendly areas); curb ramp conditions are good
Dangerous curb ramp conditions Subthemes: curb ramps in poor condition, when using them must be careful, curb ramps are too steep, curb ramps leave you in the middle of the street outside crosswalks, bumpy textures can be dangerous, debris on curb ramps
Parking
Lack of handicapped parking spaces	Handicapped/disabled parking available Subthemes: enough spaces, having a disabled parking permit
Parking is too far from destination	Handicapped/disabled parking close to destination
Lack of driver awareness in parking lots Subthemes: difficulty with drivers entering/exiting parking; poor visibility of pedestrians, need to cross parking lot to access destination	Good driver awareness in parking lots Subthemes: good pedestrian visibility, adequate safety while walking in parking lot
Structural problems with parking lots Subthemes: limited space between vehicles in parking lot to enter/exit vehicle; no clear pedestrian route in parking lots, speed bumps, slopes	Positive structural features Subthemes: marked pedestrian pathways and handicapped parking spots, covered parking, smooth parking lot surfaces, clean surfaces
Aesthetics
Presence of unattractive features Subthemes: garbage, litter, graffiti, smoking	Presence of attractive features Subthemes: secluded areas, views, nature, trees, greenery, flowers, community gardens, water, well-maintained spaces, variety, animals (birds, ducks), artwork
Lighting
Inadequate light sources Subthemes: lamps out of order, shadows, inadequate number of lights, lack lighting in key places (e.g., on ramps, at crosswalks)	Adequate lighting Subthemes: presence of street lights, flood lights
	Use of personal lighting equipment Subthemes: carrying a flashlight or using headlamp when out at night
Ramps
Obstructions to usage Subthemes: blocked, slippery, too steep	Presence of ramps
Weather
Winter conditions Subthemes: snow, ice, cold, wind, rain, puddles, slippery pavement	Good weather
Summer conditions Subthemes: heat, sun
Activity dependent upon weather
Street crossings
Lack of crosswalks Subthemes: misplaced, long stretches between, not available at desirable crossing spots	Lots of places to cross
Difficult features of crosswalks Subthemes: crossing time too short, poor crosswalk markings, obstructed views of crosswalk, fast speed of car traffic at crossings	Clearly marked crosswalks Subthemes: presence of markings, push buttons, flashing lights, stop signs, and flags
History of accidents and associated fear Subthemes: drivers that are not attentive to people crossing the street, drivers that infringe on street crossers
Sidewalks
Lack of sidewalks	Presence of sidewalks
Inconsistent sidewalk surface quality and conditions Subthemes: uneven surfaces due to trees and roots, sidewalks undulate due to driveways, potholes, cracks, debris (e.g., foliage, litter)	Good sidewalk quality and conditions Subthemes: resurfaced/smooth sidewalks, flat/level sidewalks, clutter-free sidewalks
Sidewalk obstructions Subthemes: utility poles in the middle of sidewalks, sandwich boards blocking sidewalk, bicyclists on sidewalks, garbage cans, cars parked in driveways blocking sidewalk, overgrowth	Wide sidewalks
Sidewalks not wide enough	Presence of grass strip or other separation from traffic
Wet or icy sidewalks
Amenities (on streets, in parks)
Lack of shelter/covered areas	Covered/sheltered areas
Lack of places to rest	Places to sit and rest
	Accessible bathrooms
	Drinking fountains
Traffic
Poor visibility to drivers	Slow speed limit
Drivers go too fast and do not slow down for pedestrians	Low traffic roads
Busy streets	Take safety measures Subthemes: vigilance when crossing roads, wearing clothing to increase visibility, stay on safer streets, take familiar routes
Walking paths/trails
Conditions are poor Subthemes: walkway not kept clear, rough/uneven/unpaved paths, sloping paths	Availability of paved/smooth paths that are accessible
Paths lack amenities Subthemes: no places to rest, not accessible	Paths that are wide enough to pass others
Safety
Not feeling safe Subthemes: crime, dogs, rowdy people	Feeling safe Subthemes: neighbors look out, avoiding unsafe places, avoid going out after dark
Ground /geographical features
Hills (up and down), inclines, slopes	Level/flat ground
Slippery surfaces, grass, unlevel ground	Ground surfaces that are not a problem (grass, soft surfaces, earth)
Outdoor stairs Subthemes: presence of stairs, lack of handrails on stairs, no alternative entry to stairs, stairs with open slots

Curb Ramps.—The presence and condition of curb ramps were common themes. One participant noted that although some streets have curb ramps, she needed to lift her walker up or down when she came to streets without them and reported that it was getting more difficult as she got weaker (P23). Many participants noted that curb ramps were only on one side of the street or not continuous. Participants reported having to walk or wheel in the street when curb ramps were not available, which involved using driveways or the nearest curb ramp to access the street and travel in the street until the next curb ramp was reached. The conditions of curb ramps were important as well. Broken curb ramps or curb ramps that were too steeply cut were considered dangerous by many participants. Although people recognized the truncated domes (raised bumps) at curb ramps being helpful for some, others reported these became slippery in the rain. Curb ramps were an often cited facilitator for accessing destinations.

Parking.—Although parking was not an anticipated barrier to physical activity, many participants noted that parking-related issues hindered their mobility. Although the majority of participants in this study did not drive (n = 26/35), participants reported many issues with parking when they drove, received a ride from someone else, or had to navigate parking lots. A main facilitator was having disabled parking available, particularly when it was close to the destination. One participant noted some destinations (e.g., stores) were located above parking areas, but only accessible via stairway, which prevented the use of such destinations (P10). Navigating parking lots due to parking being far from destinations was noted to cause problems because it was expressed that drivers operating vehicles seemed to lack awareness toward pedestrians. Participants noted that nonaccessible parking is not wide enough to accommodate assistive device use (e.g., getting a walker out of the car).

Aesthetics.—Nature was an important aspect for promoting activity. Noted features included less concrete and more greenery, flowers and foliage, community gardens, animals (ducks, birds, dogs), and children playing. Such features promoted getting outside and enjoying the walk rather than just going to a destination. Features such as gardens were noted to promote social interaction.

Lighting.—Adequate sources of lighting were important for participants, as nearly half of the sample reported visual impairment. Several participants noted that when street lighting was inadequate and poor visibility was a concern, they coped by carrying a flashlight or using a headlamp when they went out at night. Other participants reported depth perception problems with poor lighting and not going out after dark in part because of this problem.

Weather.—Participants reported rain as a barrier and tended to stay inside on such days. Bad weather caused difficulty with slippery pavement, places to rest becoming wet, problems with traction on hills, and metal gratings embedded in sidewalks becoming slick. Heat and sun exposure were also barriers to going outside and walking particularly when there was not enough shade or places to rest. A few participants reported having good rain-protective gear so that they felt comfortable going out in the rain, but others reported that it is tricky to carry an umbrella while using a walker or cane, stopping them from going outside in rainy weather. One participant reported carrying blankets in her walker during cold weather in case she had to wait for a ride outside (P12).

Street Crossings.—Difficulties with street crossings were a prominent theme. The lack of crossings at convenient spots was a barrier to walking or wheeling in the neighborhood. Difficulty viewing the crosswalk was a barrier when vehicles parked close to the curb, making it hard to see oncoming traffic. Crossing signal times were often reported as too short for those using assistive devices. Participants reported being pressured to go as quickly as possible without tripping, which made participants afraid of falling. One participant suggested that it would be nice to have two signal buttons—one button for those who need more time to cross and a button for those who can cross easily (P23). Fear of drivers failing to see them crossing the street was a concern. One woman noted that a pedestrian was killed at the crosswalk outside her apartment complex by a driver using a cell phone who did not see the pedestrian (P11). Participants coped with these barriers by avoiding certain crossings, being vigilant in watching for vehicles when crossing, and by staying home. On the other hand, facilitators included having adequate places to cross which made the trip efficient and less exhausting for those who are mobility impaired.

Sidewalks.—Participants reported having to walk or wheel in the street where there were no sidewalks and felt unsafe doing so. Some reported not being able to walk or wheel due to the lack of sidewalks. Sidewalk surfaces and quality posed several barriers to walking, including difficulty negotiating assistive devices over cobblestone and brick sidewalk surfaces. Metal grates embedded in sidewalks were also a problem. One participant reported that her assistive shoes do not grip metal well, so she needed to try to walk around these grates (P26). Uneven sidewalks were tricky to navigate while using walkers. Participants also reported many types of obstructions were barriers to getting around with an assistive device, such as sandwich boards that narrow available sidewalk space, making it difficult to pass. Bicyclists using sidewalks were also noted to pose problems when sharing sidewalks that were not wide enough. Facilitators for use included newly resurfaced smooth sidewalks, wide sidewalks with enough passing room, and the presence of a grass strip to separate participants from traffic.

Amenities.—Having shelter from the weather (primarily rain in King County) available while waiting for buses or transportation while resting at parks or on local streets was noted to be important. Places to rest while at parks, on trails, or in line at the food bank were very important as participants noted becoming fatigued when standing or walking for long periods. Participants reported they would walk more on local streets if there were more resting spots. Resting spots were deemed necessary by many, who reported needing to walk very slowly or limitations on the amount of time they could walk between resting points. For example, one woman who used a walker reported that it takes her 30–40min to go three blocks to the hair dresser because she needs to stop and rest along the way (P14). It was possible for her to do this on occasion because there were enough places to sit and rest. For those using walkers, some noted that using the walker’s seat was helpful, whereas others noted that it was not a desirable sitting place due to it being uncomfortable, for example, because the seats are too high. Drinking fountains were also noted to be helpful, as well as accessible restrooms in the local neighborhood.

Traffic.—Not feeling visible to drivers, fast moving traffic, and lack of driver attention toward pedestrians were the main barriers related to car traffic. Participants reported difficulties such as overgrowth or parked cars blocking drivers from seeing pedestrians. Cars turning too quickly around corners or driving over the speed limited were problems. Participants reported especially being on alert for drivers making free right-hand turns, who may not be looking for pedestrians. Vigilance when crossing streets and wearing bright colors to increase visibility were tactics many participants used to promote their use of local streets. Streets with slow moving traffic were preferred and used more often when possible.

Other important outdoor built environment facilitators were presence of ramps and feeling safe from crime. Noted additional outdoor built environment barriers included hills, outdoor stairs, and walking paths and trails in poor condition. Participants noted that destinations with stairs are difficult to access and must be avoided. Stair handrails were important to many. Availability of ramps was an important facilitator of mobility, but problems using ramps included them being blocked, poorly situated (e.g., making you hold a door open while trying not to slide down the ramp), slippery, and inconvenient to access, forcing one to go out of the way to use it.

Discussion

The focus of our study was to examine perceptions of how the built environment impacts neighborhood-based physical activity among late middle-aged and older adults with mobility disabilities. Several prominent built environment themes emerged, including the role of street crossings, curb ramps, sidewalks, weather, lighting, aesthetics, parking, traffic, and amenities such as places to rest and shelter. Many of our findings are consistent with previous research on perceptions of the built environment among populations with disability or older adults. For example, barriers to activity included curb ramps among people with disabilities (Rimmer et al., 2004) and outdoor temperatures and lack of street lights among women with arthritis (Brittain, Gyurcsik, McElroy, & Hillard, 2011). Among the general older adult population, Chaudhury, Mahmood, Michael, Campo, and Hay (2012) found that safety and security, accessibility to public transportation and neighborhood facilities, comfort of movement (e.g., seating, railings, handrails, ramps, stairs, water fountains), and peer support were important to promote physical activity. However, it seems that more themes related to the detailed features of the built environment were explicitly mentioned as barriers by our participants. For example, rather than just the presence of sidewalks being important, many specific features of the quality of sidewalks were important such as bumps, gratings, debris, obstructions (such as sandwich boards and overgrowth), and width of sidewalks. The same was true for curb ramps. And while Chaudhury and colleagues (2012) sample of older adults found ramps to be supportive, our sample reported many barriers to the use of ramps (e.g., they are often blocked or too steep; see Table 2). Such differences in findings are likely a result of our inclusion of a sample that already has mobility disability. Our findings suggest that current accessibility standards need further refining to ensure usability by people with disabilities.

Interestingly, although we asked about safety concerns, most participants mentioned feeling safe and no one mentioned feeling more vulnerable due to their disability. In addition, some themes were mentioned by some participants as barriers and others participants as facilitators. For example, the truncated domes (raised bumps) covering curb ramps (which are yellow in King County) were noted to become slick in rain and dangerous, which impacted participant mobility. However, others found the bumps to be helpful in notifying them that they were entering an intersection. Such findings have implications for universal design in which certain built environment adaptations may be helpful for some and not others.

Although barriers to being active was the focus of the study, participant barriers and facilitators tended to center around how the built environment impacted their general mobility and ability to get to important destinations such as grocery stores, volunteer activities, senior centers, drug stores, and the post office. For example, parking was noted to be a key built environment theme even though it is more related to nonactive transportation. This illustrates the importance to our study sample of including all types of mobility, rather than merely active forms of mobility such as walking and wheeling in studies examining the built environment (Webber et al., 2010). Because physical activity is known to benefit the health of older adults and people with disabilities, promoting active transportation among people aging with mobility disabilities may be important. Yet, the broader needs of people aging with mobility disabilities also need to be considered, such as how are they accessing destinations for food, medical care, and other tasks that are potentially more important than being physically active (e.g., if one has difficulty obtaining food, which is vital to survival, this is more important than encouraging them to be physically active). Thus, there are complex relationships between mobility, built environment, and health, which need to be better understood in this population and cannot be understood by examining only one form of mobility (e.g., driving or physically active transportation). As our population is aging and rates of mobility disability are expected to increase, studies that focus more broadly at relating built environment features to various types of mobility among older adults with mobility limitations are needed. One recent study showed that community mobility barriers were related to daily activity limitations and frequency, so this is worthy of more investigation (Keysor et al., 2010).

Another clear finding was that regardless of the features present, our sample was highly adaptive in figuring out ways to get to places they needed to go. Some adaptations enabled physical activity to occur (e.g., wearing bright colors, being willing to travel in the road when sidewalks or curb ramps were not available), whereas others involved sacrificing an opportunity for physical activity (e.g., getting a door-to-door ride from a friend). When neighborhoods are designed in ways that provide a default opportunity for people of all ages and abilities to be active while engaging in daily living, physical activity can become an inherent part of daily life. However, it is unlikely that the built environment barriers our participants mentioned can always be remedied. Municipalities may not have enough money to continually maintain sidewalks and crosswalks, build new sidewalks, and so forth. Therefore, ensuring means of adequate transportation to areas that are walkable and wheelable will be crucial. For example, one participant was frustrated that all our interview questions focused on her home neighborhood. She reported that her home neighborhood was completely inaccessible for her powerchair and all of her neighborhood streets lacked sidewalks. However, she spent the majority of her days volunteering at a senior center several miles away from home which she reached using the local paratransit service. Then she could do all of her errands in that more walkable and wheelable part of town (P8).

In addition to our qualitative findings, this study provides useful information on the ability to use GPS devices among aging adults with mobility disabilities, a novel aspect of this study. There were few problems with using our mail-based procedures, only one device malfunctioned, and participants overwhelmingly reported having no problems using the device for 3 days. The retrieved maps provided the study team with useful information to tailor interviews before the home visit.

Our main study limitation was the lack of recording and transcribing interviews. This is typically the procedure for in-depth interviews, but we opted not to do so because early in the recruitment process several participants indicated concern with being recorded. A number of potential study participants received government services for housing, income, medical coverage, and transportation due to their disability. Participants voiced concerns about having these services taken away if they were found to be complaining. Based on this issue, the decision was made to use a detailed note taking approach. We do not feel this decision affected our findings as we reviewed them with a diverse research team with backgrounds in medicine, nursing, clinical psychology, and public health, as well as with our project advisory board. We also employed training procedures among the note takers. Because of this procedure, we were not able to provide direct quotes from participants and rather opted to include specific examples given by participants to ensure that we do not inaccurately misquote.

Additional study limitations include that we interviewed participants during the winter and spring, so participant responses may be more relevant for these times of year. The interviews were conducted in one urban and suburban county in Washington. Thus, findings may not apply to other regions both nationally and internationally. Additional qualitative research in various regions with a variety of built environments would be helpful or future work. One of the major barriers in our study was hills, whereas other geographic locations do not have such terrain. Also, because the study was focused on built environment barriers, we may have missed other important influences such as the kindness of strangers. In addition, we focused on the neighborhood built environment, but the home environment is very important to consider for its role in daily activity (Iwarsson et al., 2007). Studies that examine the role of both indoor and outdoor environments on physical activity among older adults with mobility disabilities would be useful. Also, although our data analysis team came from various fields, all researchers came from a public health perspective, whereas other perspectives (e.g., environmental gerontology) could provide useful view points on the themes presented. Finally, we focused on defining mobility disability as use of an assistive device; however, there are many ways to define mobility disability, including people’s perception of whether they have difficulty with mobility which may be more relevant (Verbrugge & Jette, 1994).

The main strength of our study is the use of qualitative methods, which allowed the viewpoints of participants to emerge. For example, if we had only conducted a survey examining built environment relationships to physical activity, we would have missed the importance our participants placed on achieving broader mobility. Although research shows that physical activity has very important mental, physical, and cognitive benefits for older adults and people with disabilities (Physical Activity Guidelines Committee, 2008), it would be shortsighted to ignore potentially more important concerns of this population.

Future qualitative research should focus on built environment barriers to mobility, including a focus not only on physical activity but also on broader conceptualizations of mobility (e.g., transportation, life space), access to food, social interaction, and access to health care. Although we used GPS devices to bolster our qualitative interview approach, such data could be used in future studies to better understand where people with mobility disabilities go, types of transportation used to get to destinations, and how much time is spent outside the home. We need to better examine ways people with mobility disabilities preplan their outings, as well as how they develop creative adaptations in their built environment. It would be valuable to conduct a study with a larger sample to compare differences in perceived barriers based on demographic and functional characteristics. Additional research on this topic could lead to policy interventions that allow older adults with mobility limitations to live healthier and stay mobile in their communities (Rosso et al., 2011).

Funding

This work was supported by the CDC Prevention Research Centers Program, through a grant to the University of Washington Health Promotion Research Center [cooperative agreement #U48-DP001911], The Built Environment, Accessibility and Mobility Study, Basia Belza, and Dori Rosenberg, co-principal investigators. Dr. Huang receives salary support from the Department of Veterans Affairs Advanced Fellowship in Geriatrics. The contents of this paper are the responsibility of the authors and do not necessarily represent the official view of the funders.