How Does the Environment Influence Disability? Examining the Evidence
Julie J. Keysor*
How the environment influences disability is a crucial clinical and policy question. If facilitative environments decrease disability, then policies and clinical interventions could be implemented to support the environmental elements that are conducive to minimizing disability and optimizing participation in daily life activities. Recent conceptual frameworks of disability highlight the important role of the environment in the disablement process; that is, people are believed to interact with their environments to produce disabilities (IOM, 1997; Teel et al., 1997). Thus, the environment is identified as a contextual factor in the process of disablement (Letts et al., 1994; Fougeyrollas, 1995; WHO, 2000). Yet, there are many research questions that pertain to the proposed theoretical frameworks. If the environment does influence disability, how does it do so? Which environmental domains influence disability? Which facilitators and barriers influence disability?
This paper addresses these questions by examining the empirical evidence. First, recent environmental measurement approaches are reviewed
and critiqued. Second, the evidence on the environment-disability link among adults with mobility limitations is examined. Third, challenges to this area of research are discussed. A thorough review covering all areas of disability is beyond the scope of this paper. Instead, the focus is on the environment-disability link in adults with mobility limitations. Other important areas of the environment-disability link among children and individuals with hearing, visual, or learning impairments are not discussed here.
ENVIRONMENT MEASUREMENT APPROACHES
Researchers face several formidable challenges when they pursue environmental assessments. The first challenge is conceptual. To study the complex interplay between environmental factors and disability, researchers need to know how to identify and measure the environmental factors that are relevant to individuals. Fougeyrollas (1995) suggests that the organization and context of society contain social, cultural, and physical dimensions. Factors in these dimensions can become obstacles or supports to individual functioning. The taxonomy of environmental factors of Fougeyrollas and colleagues (1991) includes socioeconomic organization (e.g., family structure, political systems, and economic systems), social roles (e.g., law, values, and attitudes), nature (e.g., geography, climate, and time), and development (e.g., architecture, land development, and technology).
The International Classification of Functioning, Disability and Health (ICF), on the other hand, specifies five environmental domains: products and technology; natural environment and human-made changes; support and relationships; attitudes; and services, systems, and policies (WHO, 2000). The Craig Hospital Inventory of Environmental Factors (CHIEF) assesses five domains that are similar to those used in the ICF taxonomy: (1) attitudes and support, (2) services and assistance, (3) physical and architectural, (4) policies, and (5) work and school (Whiteneck et al., 2004c). Shumway-Cook and colleagues (2002, 2003), in contrast, focus on the physical domain of the environment and identify eight dimensions: (1) temporal, (2) physical load, (3) terrain, (4) postural transitions, (5) distance, (6) density, (7) attentional demands, and (8) ambient conditions. To date there is no consensus on which environmental domains or which elements of the domains should be measured to study the importance of the environment in the lives of people with disabilities.
The second challenge is one of measurement, with three general approaches currently being used. The first approach assesses an individual’s perceptions of the degree to which environmental factors influence his or her participation in daily life. Four instruments assess the environment in this manner: (1) CHIEF (Whiteneck et al., 2004c); (2) the Measure of the Quality of the Environment (Fougeyrollas et al., 1997); (3) the Facilitators
and Barriers to Participation for People with Mobility Impairments and Limitations (Gray et al., 2005); and (4) Fange and Iwarsson’s (1999) self-assessment of the physical housing environment.
CHIEF is a 24-item self-report instrument that assesses the frequency with which people encounter environmental barriers related to attitudes and support, services and assistance, physical and architectural domains, policies, and work and school and the impact of each environmental factor on daily life, as perceived by the individual (Whiteneck et al., 2004c). The respondents are asked to indicate how often various barriers in the environment have been a problem over the past 12 months. If environmental factors have been a problem, the respondents are asked to indicate whether the factor has been a big or a little problem. The reliability and validity of the instrument are acceptable for measurement of the effects of all domains.
The Measure of the Quality of the Environment is a 72-item self-report instrument that assesses the extent to which various barriers and facilitators of the environment influence peoples’ daily lives in six domains: (1) support and the attitudes of family and friends; (2) income, job, and income security; (3) governmental and public services; (4) the physical environment and accessibility; (5) technology; and (6) equal opportunity and political directions (Fougeyrollas et al., 1997).
The Facilitators and Barriers to Participation for People with Mobility Impairments and Limitations is also a self-report instrument that assesses the perceived accessibility of elements of the home and community environments (Gray et al., 2005). Fange and Iwarsson (1999) take a similar approach in their recently developed self-report assessment of the physical housing environment. The 31-item instrument assesses eight dimensions of the physical environment, including (1) overall satisfaction with housing conditions, (2) suitability, (3) security and safety, (4) importance, (5) privacy, (6) social contacts, (7) flexibility, and (8) accessibility. The reliability and content validity of the instrument are acceptable (Fange and Iwarsson, 1999).
Instruments like those reviewed above are important for identifying relevant environmental domains as well as important barriers and facilitators. However, a perception of an environmental impact is not direct evidence of its actual influence on a person’s level of participation. A person’s perception of the impact that the environment has on his or her participation may be strongly correlated with his or her participation, thereby resulting in inflated measures of effect. Thus, measurement of a person’s perception of environmental barriers does not allow the researcher to examine empirically whether the presence or the absence of such a factor in a person’s environment is directly associated with variation in a person’s level of participation. A more direct measure of the actual environment is needed. To examine how environmental factors affect peoples’ involvement in daily
activities, the extent to which a person’s environment contains elements that could facilitate or restrict participation needs to be characterized and correlated with the level of disability.
Two approaches described in the current literature attempt to characterize the elements of people’s environments. One approach, developed by Shumway-Cook and colleagues (Shumway-Cook et al., 2003), assesses the extent to which people avoid and encounter barriers and facilitators in the physical environment by using the self-report Environmental Components of Mobility Questionnaire. The questionnaire assesses eight dimensions of the physical environment: (1) temporal, (2) physical load, (3) terrain, (4) postural transitions, (5) distance, (6) density, (7) attentional demands, and (8) ambient conditions. Shumway-Cook and colleagues’ self-report approach has been shown to be reliable (Shumway-Cook et al., 2003); however, the limitation of this approach is that asking people to ascertain the extent to which they avoid or encounter barriers may not necessarily correspond to the factors that are in their environments. For example, if someone with mobility limitations is unable to negotiate stairs, he or she may state that he or she avoids stairs, but the person’s environment may not have stairs. In addition, whether people avoid aspects of their environment is likely to be confounded by function; people will avoid stairs if they are unable to negotiate stairs.
Shumway-Cook and colleagues developed an observational approach that corresponds to the Environmental Components of Mobility Questionnaire in which specific factors of the environment comprising the physical domain are assessed independently of a person’s level of participation (Shumway-Cook et al., 2002). The administration of the measure involves a structured observational encounter between a researcher and a participant; that is, a researcher observes and videotapes the participant as he or she performs community mobility activities, such as going to a grocery store or seeing a health care provider. Although this approach provides the investigator with an independent assessment of environmental factors, the limitations are the protocol’s cost and the time that it takes to administer it. Although it has attractive features, the protocol of Shumway-Cook et al. is not feasible in large-scale field studies, in which feasibility, study costs, and participant burden are paramount concerns.
Another means of characterization of the environment is description of the actual elements of the environment. This measurement approach is taken by Keysor and colleagues (2005) in the Home and Community Environment (HACE) instrument. HACE is a 36-item instrument that assesses barriers and facilitators in six conceptual domains: (1) home mobility, (2) community mobility, (3) basic mobility devices, (4) communication devices, (5) transportation factors, and (6) attitudes (Keysor et al., 2005). The home mobility domain assesses the extent to which people have barriers in the
home, such as the number of steps at the main entrance or the number of stairs inside the home. Facilitators are also assessed, including ramps, elevators, and chairlifts. The community mobility domain assesses the extent to which a person’s community has barriers such as uneven sidewalks and curbs without curb cuts. Basic mobility devices, communication devices, and transportation facilitators are assessed to indicate whether the respondent has a mobility or communication device and the type of transportation, irrespective of its use. Reliability was acceptable for all subscales. This approach allows the person to characterize the environment irrespective of his or her level of participation or functional ability. The investigators found, however, that participants were unable to characterize environmental domains related to policies and services. Thus, it is questionable whether participants can validly characterize the availability of services and policies for individuals with limitations in daily activities, which is a potential limitation to this assessment approach.
Taking a similar approach of characterizing the environment but focusing on the home environment, Freedman and Agree (2005) report on a new instrument that can be used to assess the “existence,” “acquisition,” and “use” of environmental modifications and assistive technologies. The self-report survey covers five general areas: (1) the home environment, (2) mobility and other devices, (3) the effectiveness of features and devices, (4) the presence of computers and telephones, and (5) residual activity of daily living and instrumental activity of daily living. The reliability and validity of the instrument are acceptable (Freedman and Agree, 2005). The Enabler is another instrument that characterizes the environment irrespective of the level of function (Iwarsson, 1997). This instrument is completed by a trained professional. The level of functional impairment and the presence of barriers in the home and community environment are assessed, and a combined score indicates the degree of accessibility problems.
In summary, a new generation of instruments that can be used to assess the environment has been developed over the past several years. Researchers appear to take three general approaches, with each approach having strengths and weaknesses (Table D-1).
THE EVIDENCE: THE DISABILITY-ENVIRONMENT LINK
Medline and the Cumulative Index of Nursing and Allied Health Literature were reviewed to identify research articles examining the environment-disability link among adults with mobility limitations. “Environment,” “environmental,” “technology,” and “device” were referenced to title searches and cross-referenced with the terms “disability,” “disablement,” “handicap,” and “participation.” Research articles pertaining to the role of the environment on disability among adults with mobility limita-
TABLE D-1 Comparison of Approaches for Measurement of the Environment
Determination of perceived impact
Self-report assessment of the degree to which the environment influences participation; includes accessibility
Seems to be a reliable approach for people to self-report barriers and facilitators
May create a statistical bias: artificially correlated with disability (participation)
Perceived barriers or facilitators may not be actual elements of the individual’s environment
Determination of barriers encountered or avoided (perceived or actual)
Self-report and observational approach that assesses the extent to which barriers are avoided
Self-report and observational methods seem to be reliable
A time-consuming observational method that is not feasible for large epidemiological studies; assessment is confounded by functional ability
Characterization of the environment
Self-report or observational assessment of actual elements of the environment
Representation of the environment irrespective of level of function or disability
Some domains or aspects of the environment may be difficult to reliably self-report; observational methods are time-consuming and not feasible for use in large epidemiological studies
tions, including stroke, arthritis, spinal cord injuries, and general aging, were reviewed.
Several studies show that individuals with mobility limitations report the presence of barriers and facilitators in their environment that restrict participation. Barriers are reported in the built environment domain, as well as in the domains of social attitudes, social institutions, cultural norms, and technology (Meyers et al., 2002; Noreau et al., 2002; Gray et al., 2003; Whiteneck et al., 2004a). On the other hand, individuals with mobility limitations report that important facilitators in the environment are the positive attitudes of individuals in the community; social support; and the availability of technology, devices, and products and accessible transportation (Meyers et al., 2002; Noreau et al., 2002; Gray et al., 2003).
Individuals with mobility limitations may also avoid physical barriers in the community. Shumway-Cook and colleagues (2003) showed that people with mobility limitations were more likely to report that they avoided long-distance ambulation, crossing streets with traffic lights, crossing busy streets, using stairs and escalators, walking on curbs or uneven surfaces, and going out in icy conditions. There were no differences between individuals with mobility limitations and individuals without mobility limitations in the rates of going outside when it was dark, snowing, very hot or cold, wet or noisy. However, when actual behavior was observed, there were fewer differences between individuals with mobility limitations and those without (Shumway-Cook et al., 2002). Individuals with mobility limitations were more likely than age-matched individuals without mobility limitations to use the elevator and were less likely to ascend two flights of stairs or walk on uneven surfaces. However, there were no differences between the two groups in the average distance walked; the rate of crossing streets with traffic lights; or the temperature, light conditions, and level of precipitation during a walk.
Although this descriptive approach shows that people with mobility limitations do perceive that aspects of their environment limit or enhance their participation, it does not provide strong evidence showing how the environment influences disability. To address this question, the environment needs to be correlated with measures of disability. Surprisingly, only a few published studies have assessed the effects of the environment on disability outcomes. The few studies that have been published in the literature are reviewed below.
Three studies of adults with mobility limitations showed a small but significant association of the environment on disability. Whiteneck and colleagues (2004a), in a large population-based study of 2,726 individuals with spinal cord injuries, correlated the results obtained with CHIEF environment scales with measures of participation and life satisfaction. The CHIEF environment scales (i.e., physical-structural, services-assistance, and
attitudinal-support) showed a small but significant association with the total Craig Handicap Assessment and Reporting Technique (CHART) score, although the CHIEF environment scales accounted for only 4 percent of the variance in participation. Demographics, injury-related factors, and activity limitations each accounted for approximately 20 percent of the variance in participation. Similar patterns emerged when the findings obtained with the individual CHIEF subscales were correlated with those obtained with the CHART subscales. Physical and structural barriers were weakly correlated with physical independence, mobility, and occupation; services and assistance were weekly associated with physical independence, cognitive independence, mobility, occupation, and social integration; and the attitudinal domain and support were associated with physical independence and cognitive independence. The CHIEF subscales were more strongly associated with life satisfaction, accounting for approximately 10 percent of the variance in life satisfaction.
Similar findings have been reported in other studies (Badley et al., 1998; Rochette et al., 2001). Badley and colleagues (1998), in a cross-sectional population-based study of 16,017 individuals with self-reported, arthritis-associated disability, reported that modifications to the kitchen were associated with less dependence in external household activities and domestic activities of daily living, whereas a modified bathroom, as well as a cane and other moving aids, was associated with more disability in domestic and personal care activities of daily living. The environmental factors explained 8 percent of the variance in dependence in daily activities (i.e., disability), with functional limitation explaining more than 40 percent of the variance in disability. However, the combination of functional limitation and environmental barriers explained a larger amount of the variance in disability than the combination of functional impairment and personal or sensory factors. In a cross-sectional study of 51 stroke patients selected from a rehabilitation unit, Rochette and colleagues (2001) showed that environmental barriers explained 6 percent of the variance in disability, as measured by the Assessment of Life Habits.
One study showed that individuals with spinal cord injuries believed that the biggest factor limiting their daily life activities was related to technology. In a cross-sectional study of 70 individuals with spinal cord injuries who used a wheelchair, Chaves and colleagues (2004) found that limitation due to one’s wheelchair was most strongly correlated with perceived participation limitations, followed by the environment and physical impairments.
Stronger evidence on the impact of the environment on disability will come from prospective studies. No published prospective studies met the search criteria established for this paper. However, the author and colleagues recently conducted a 12-month longitudinal study of adults who were discharged from the acute-care hospital setting with a diagnosis of
neurological disorder, lower-extremity orthopedic trauma, or medically complex conditions (Jette et al., in press). Data for 345 participants who completed 1-month and 6-month interviews were analyzed to examine the impact that the environment had on disability. HACE was used to assess home and environmental barriers and facilitators (Keysor et al., 2005), the Medical Outcomes Study-Social Support was used to assess social support (Sherbourne and Stewart, 1991), and the Participation Measure for Post-Acute Care was used to measure disability (Gandek et al., in press). In multivariate analyses that adjusted for age, sex, educational attainment, race, disease severity, physical and mobility activities, and applied cognition, it was found that at 1 month greater numbers of community mobility barriers were statistically significantly associated with more community disability, with transportation facilitators approaching statistical significance. On the other hand, more home mobility barriers were associated with less social and home participation, whereas more community barriers, more mobility technology facilitators, and more social support were associated with more social and home participation. With the exception of social support, however, environmental barriers and facilitators were not predictive of community participation and social and home participation at 6 months (Keysor et al., submitted for publication).
The strongest evidence of the environment-disability link comes from an 18-month prospective randomized controlled trial conducted by Mann and colleagues (1999). One hundred four community-dwelling elderly people were randomized either to a treatment group that received resources for assistive technology (e.g., a walker or a wheelchair) or environmental modifications (e.g., the addition of ramps or the removal of scatter rugs) that were evaluated as being necessary by trained clinicians or to a control group of individuals who received the usual care for their conditions. There were no differences in disability between the treatment and the control groups, as measured by CHART. However, the treatment group had significantly less decline in function over the 18-month period and lower health care costs related to institutionalized care.
On the other hand, Whiteneck and colleagues (2004b), in a study of 73 individuals with traumatic brain injuries, found that the environment was an important determinant of participation. Transportation, the surroundings, government policies, attitudes, and the natural environment were perceived to have the most impact on people’s lives. The total CHIEF score and the subscales of attitudes and services, which represent the impacts of environmental barriers, were associated with more disability, as measured by the overall CHART score. More physical and policy barriers, as measured by two subscales of CHIEF, were associated with the mobility and occupa-
tion subscales of CHART. More barriers in attitudes and services were associated with less cognitive independence.
SUMMARY AND CRITIQUE OF CURRENT EMPIRICAL EVIDENCE
Research on how the environment influences disability has been limited. The majority of studies have small sample sizes; and many studies draw samples from a convenience clinical population, thereby limiting the generalizability of the findings of the studies. The study designs are cross-sectional, thereby limiting the causal inferences of the effects of the environment on restricting or enhancing participation that can be made. Lastly, the majority of study samples comprise individuals whose primary mode of locomotion is a wheelchair, so little or nothing is known about the impact of the environment on other populations.
The studies that have been published have provided evidence that individuals with mobility limitations perceive that particular barriers and facilitators of the environment—particularly in the physical and social domains—restrict or enhance their participation in life activities. In addition, individuals with mobility limitations seem to avoid physical barriers in the community, although their actual behavior patterns may differ from self-reported perceptions.
Despite the findings that people with mobility limitations report the presence of barriers in their environment, it is not clear how the environment influences disability. A few studies have reported a small association of the environment with disability, but this explains less than 10 percent of the variance in disability (Badley et al., 1998; Rochette et al., 2001; Whiteneck et al., 2004a). On the other hand, one study of individuals with traumatic brain injuries showed a stronger association of the environment with disability (Whiteneck et al., 2004b), a finding that could suggest that the environment-disability link could be dependent on the specific type of disability.
Researchers in the field of environment-disability face several challenges in the upcoming years. The first challenge pertains to measurement. Further work is needed on conceptualizing the domains of the environment that are relevant to disability as well as what elements of each domain need to be assessed—particularly in relation to the social, political, and attitudinal domains of the environment. Additional work is also needed to illuminate the best measurement approach for assessment of the environment in a manner that can be used to examine whether and how the environment influences disability. Does assessment of the perceived impact of the envi-
ronment bias associations with participation; that is, is the perceived impact of the environment artificially correlated with participation? Is ascertainment of characteristics of the environment irrespective of disability a better approach to examination of the impact of the environment on disability? On the other hand, as Fange and Iwarsson (1999) advocate, should both approaches be used? Can participants reliably self-report characteristics of their environment? Are the measures valid? Lastly, can the diverse domains of the environment be assessed in a parsimonious manner to minimize respondent burden?
The second challenge pertains to disentangling the causal relationship of the environment on disability. The evidence to date shows that some environments that are more facilitative are associated with less disability, whereas other environments that are facilitative are associated with more disability. This finding may reflect a human adaptation process, with people modifying their environment because of their health status. It is likely that people adapt to their environment—i.e., they change how they accomplish activities. In addition, people may adapt or change their environment to meet their needs. This element of human adaptation poses particular challenges for understanding how the environment influences disability. Prospective longitudinal studies are needed to develop evidence of whether the changes in the baseline environment are related to changes in disability. Experimental clinical trials will also provide evidence of whether the manipulation of various aspects of the environment influences disability.
The third and perhaps the most important challenge will be to implement complex analytical approaches in study designs. Disability theoretical frameworks posit that disability is the result of disease-specific person-level factors and the environment in which a person lives, that is, the interaction of the person within his or her environment. Studies to date have focused on assessment of the direct impact of the environment on disability and have not examined statistical interaction effects. Perhaps there is a certain threshold of functional limitation in combination with a certain type of restrictive environment that is conducive to greater disability. For example, people who are able to walk one or two blocks may be more strongly affected by restrictive communities than people who are able to walk several blocks. On the other hand, perhaps people with a combination of cognitive and physical impairments will be more adversely affected by restrictive environments than people with physical impairments alone. Large sample sizes and complex analytical strategies will be needed to assess interactive effects, and this could be imperative to understanding how the environment influences disability.
Lastly, the role of the environment on disability is likely to be quite complex, and important factors may need to be taken into consideration that are not clearly articulated in the current environment-disability litera-
ture. First, the role of behavioral factors in the person-environment interaction may be imperative. Do people adapt to barriers and remain active in their lives? Does one’s ability to resolve the barriers and advocate for his or her needs mediate the relationship between restrictive environments and disability? Second, is disability the optimal outcome in the examination of person-environment interactions? Is the environment a determinant of disability, or does the environment influence other outcomes, such as satisfaction and quality of life, as some evidence suggests (Richards et al., 1999; Whiteneck et al., 2004a)?
Badley EM, Rothman LM, et al. (1998). Modeling physical dependence in arthritis: the relative contribution of specific disabilities and environmental factors. Arthritis Care and Research 11(5):335–345.
Chaves ES, Boninger ML, et al. (2004). Assessing the influence of wheelchair technology on perception of participation in spinal cord injury. Archives of Physical Medicine & Rehabilitation 85(11):1854–1858.
Fange A and Iwarsson S (1999). Physical housing environment: development of a self-assessment instrument. Canadian Journal of Occupational Therapy. Revue Canadienne d Ergotherapie 66:250–260.
Fougeyrollas P (1995). Documenting environmental factors for preventing the handicap creation process: Quebec contributions relating to ICIDH and social participation of people with functional differences. Disability and Rehabilitation 17(3–4):145–153.
Fougeyrollas P, Bolduc M, et al. (1991). Use of the Health and Activity Limitations Survey in Relation with Conceptual Framework of the ICIDH. Ottawa, Ontario, Canada: Canadian Society of the ICIDH and Quebec Committee of the ICIDH.
Fougeyrollas P, Noreau L, et al. (1997). Measure of the quality of the environment. ICIDH and Environmental Factors International Network 9(1):32–39.
Freedman VA and Agree E (2005). Linking measures of assistive technology and disability. Workshop on Improving Survey Measures of Late-Life Disability. Sponsored by the Office of the Assistant Secretary for Planning and Evaluation, U.S. Department of Health and Human Services, Washington, DC.
Gandek B, Sinclair J, Jette A, Ware J. (In press). Development and initial testing of the participation measure for post-acute care (PM-PAC). American Journal of Physical Medicine & Rehabilitation.
Gray D, Gould M, et al. (2003). Environmental barriers and disability. Journal of Architectural and Planning Research 20(1):29–37.
Gray D, Hollingsworth H, et al. (2005). Participation and Environment Measurement System: PATS and FABS. Paper prepared for the American Public Health Association 133rd Annual Meeting, December 10-14, Philadelphia, PA.
IOM (Institute of Medicine) (1997). Enabling America: Assessing the Role of Rehabilitation Science and Engineering. Washington, DC: National Academy Press.
Iwarsson S (1997). The Enabler. A Method of Analyzing Accessibility Problems in Housing. Dalby/Lund, Sweden: Lund University.
Jette A, Keysor J, et al. (in press). Beyond function: predicting participation outcomes in a rehabilitation cohort. Archives of Physical Medicine & Rehabilitation.
Keysor J, Jette A, et al. (2005). Development of the home and community environment (HACE) instrument. Journal of Rehabilitation Medicine 37(1):37–44.
Keysor, J, Jette A, et al. (Submitted for publication). Does a prospective examination of the environment show a link with participation?
Letts L, Law M, et al. (1994). Person-environment assessments in occupational therapy. American Journal of Occupational Therapy 48(7):608–618.
Mann W, Ottenbacher K, et al. (1999). Effectiveness of assistive technology and environmental interventions in maintaining independence and reducing home care costs for the frail elderly: a randomized controlled trial. Archives of Family Medicine 8(3):210–217.
Meyers AR, Anderson JJ, et al. (2002). Barriers, facilitators, and access for wheelchair users: substantive and methodological lessons from a pilot study of environmental effects. Social Science & Medicine 55(8):1435–1446.
Noreau L, Fougeyrollas P, et al. (2002). The perceived influence of the environment on social participation among individuals with spinal cord injury. Topics in Spinal Cord Injury 7(3):56–72.
Richards JS, Bombardier CH, et al. (1999). Access to the environment and life satisfaction after spinal cord injury. Archives of Physical Medicine & Rehabilitation. 80(11):1501–1506.
Rochette A, Desrosiers J, et al. (2001). Association between personal and environmental factors and the occurrence of handicap situations following a stroke. Disability & Rehabilitation 23(13):559–569.
Sherbourne C & Stewart, A. (1991). The MOS Social Support Survey. Social Science and Medicine 32:705–714.
Shumway-Cook A, Patla AE, et al. (2002). Environmental demands associated with community mobility in older adults with and without mobility disabilities. Physical Therapy 82(7):670–681.
Shumway-Cook A, Patla AE, et al. (2003). Environmental components of mobility disability in community-living older persons. Journal of the American Geriatrics Society 51(3):393–398.
Teel C, Dunn W, et al. (1997). The role of the environment in fostering independence: conceptual and methodological issues in developing an instrument. Topics in Stroke Rehabilitation 4:28–40.
Whiteneck G, Meade MA, et al. (2004a). Environmental factors and their role in participation and life satisfaction after spinal cord injury. Archives of Physical Medicine & Rehabilitation 85(11):1793–1803.
Whiteneck G, Gerhart KA, et al. (2004b). Identifying environmental factors that influence the outcomes of people with traumatic brain injury . Journal of Head Trauma Rehabilitation 19(3):191–204.
Whiteneck G, Harrison-Felix CL, et al. (2004c). Quantifying environmental factors: a measure of physical, attitudinal, service, productivity, and policy barriers. Archives of Physical Medicine and Rehabilitation 85:1324–1335.
WHO (World Health Organization). (2000). International Classification of Functioning, Disability, and Health. Geneva, Switzerland: World Health Organization.