Quality of Life As a Potential Rehabilitation Service Outcome

Abstract

Vocational rehabilitation (VR) agencies, as well as other service providers, are under increased pressure to provide evidence of the effectiveness of services. The primary metric for evaluating services in the VR program is short-term employment outcomes. Although employment outcomes are crucial, they may serve as a poor proxy for the ultimate intended goal of services, namely, improved quality of life (QOL). In this study, a comprehensive framework (the International Classification of Functioning [ICF]) is used to assess QOL in two samples of adults with disabilities receiving educational and vocational services. The relationship between difficulty with work and daily living activities with QOL was compared with a more complex assessment based on the ICF framework, including other life areas such as social relationships and inclusion and environmental factors. Results indicated that the additional information provided by the ICF model substantially increases the prediction of QOL relative to the more traditional VR measures. Within-groups analysis provided more information specific to each sample.

Keywords

process and outcome rehabilitation counseling ICF model quality of life

Public vocational rehabilitation (VR) agencies are important providers of employment services to individuals with disabilities (Saunders, 2005). VR agencies aim to assist individuals to increase levels of independence, economic self-sufficiency, and community integration (Rubin, Chan, & Thomas, 2003). As state entities, VR agencies must document that they have provided effective services in an efficient manner (Rubin & Roessler, 2001) to ensure the continued funding of their programs (Rubin et al., 2003). The challenge to demonstrate that treatment provides clinically significant benefits to the client is common across sectors of health care, including mental health treatment and rehabilitation (Frisch, 2004). Demonstration of effectiveness is crucial in this era of accountability and evidence-based practice (Bishop, Chapin, & Miller, 2008; Jenney & Campbell, 1997). For a variety of reasons, including historical roots and legislative mandates, the outcome measurements of rehabilitation (particularly in the public sector) have largely been limited to employment and to a lesser extent independent living (Gilbride, Thomas, & Stensrud, 1998). However, in the literature, we find frequent suggestions that this narrow focus leaves us with a limited understanding of the results of rehabilitation services (Bishop et al., 2008; Chapin, Miller, Ferrin, Chan, & Rubin, 2004; Roessler, 1990) and a broader measure including components of quality of life (QOL) would be more useful (Bishop & Feist-Price, 2002; Chapin et al., 2004; Roessler, 1990).

QOL has long been held as the underlying goal of all rehabilitation interventions (Crewe, 1980). Researchers and policy makers have proposed that QOL is an important and useful way to measure the effect of services (Alexander & Willems, 1981; Cardus, Fuhrer, & Thrall, 1981; Chan, Rubin, Kubota, Chronister, & Lee, 2003; Frisch, 2004). The World Health Organization (WHO; 1998) defines QOL as “individuals’ perceptions of their position in life in the context of culture and value systems in which they live and in relation to their goals, expectations, standards, and concerns” (p. 2). This definition reflects subjective QOL, which is used interchangeably with subjective well-being or life satisfaction (Roessler, 1990). Over time, components of QOL have been investigated, confirmed, and organized into what are known as life domains. Common life domains in QOL assessment include physical health, psychological/emotional health, social support/social relationships, level of independence, employment/productive activity, environment, material/economic well-being, and spirituality/religion/personal beliefs (Bishop, 2005; WHO, 1998).

The challenges that individuals with disabilities face in securing employment, social and economic independence, and freedom to function at the highest possible level (Smart, 2001), suggests a relationship between these milestones and QOL. Chan, Wang, Muller, and Fitzgerald (2011) put opportunities for work and community participation into a broader context: “Without a doubt, lack of employment opportunities and work incentives excludes people with disabilities from full community participation, significantly affecting the quality of their lives” (p. 3). In this statement, authors imply that interrelationships exist between employment and other typical adult activities, disability, community integration or inclusion, and QOL. However, QOL is multidimensional, individually constructed construct (Cummins, 1996; Deiner, 1984), which may not be well captured by one aspect of individual experience (e.g., employment).

In responding to the challenges faced by individuals with disabilities in securing employment, social and economic independence, and freedom to function at the highest possible level (Smart, 2001), it is not surprising that VR agencies measure improvements to QOL by tangible impacts such as employment or increased independence. Labor participation rates for individuals with disabilities are consistently and substantially lower than the general population; recent figures include a participation rate of 64% for the general population compared with 18% for individuals with disabilities (Chan et al., 2011). Several authors have provided empirical support that connects employment and QOL, although additional factors also play a role. For example, Chapin and Holbert (2010) found a connection between subjective well-being and QOL ratings among former VR clients with spinal cord injuries who were employed compared with those who were not. Similarly, in a study of employment skills and subjective well-being, da Silva Cardoso, Blalock, Allen, Chan, and Rubin (2004) showed that functional skills were moderately correlated with subjective well-being; however, the life skills considered essential for work and independent living did not significantly correlate with psychological well-being. Authors stressed that other factors related to environment or interference with activity should be considered as part of rehabilitation interventions (da Silva Cardoso et al., 2004).

In addition to what has been found through research, both clients and providers of services (Chan et al., 2003) also suggest the inadequacy of short-term employment as the sole indicator of changes in life skills and QOL in assessment (Rubin et al., 2003). Although obtaining employment is crucial to QOL, it is likely that QOL is a much more complex, individually based perception than is indicated by this single outcome measurement (Cummins, 1996; Deiner, 1984).

The International Classification of Functioning (ICF)

The ICF model of health and disability provides an excellent conceptual framework for QOL because it acknowledges the interactive nature between function, activity, participation, and environment and the impact of disability (see Figure 1). Previous research has acknowledged the compatibility between the ICF and investigations related to QOL, particularly in the areas of health-related QOL (Cieza & Stucki, 2008; Kennedy, Lude, & Taylor, 2006; Whiteneck, Brooks, Harrison-Felix, & Gerhart, 2004). An important feature of the model is that it provides a comprehensive way to understand the impact of chronic illness and disability across cultures (Chan, Tarvydas, Blalock, Strauser, & Atkins, 2009), as well as its inclusion of environmental factors and emphasis on activities and community participation. Work and independent living are addressed in the model, but additional factors are also included. For these reasons, it was selected as a more comprehensive approach to conceptualizing QOL.

Figure 1.

The ICF model components and interactions.

The following research question was addressed in this study:

Research Question 1: What is the contribution of the ICF model (e.g., social relationships, environmental factors, personal factors, mobility, self-care) relative to typically derived outcomes of services (e.g., employment or independent living) in predicting QOL?

Method

Procedures

Two samples of adults with disabilities receiving employment, education, and/or residential services were recruited for this study. One sample included clients from a large community rehabilitation program (CRP) in the Midwest. The other sample included university students registered with the disability support center at a large university in the Midwest. Using the ICF model as a conceptual framework, information was collected in the following areas: (a) demographic information; (b) function, activities, and participation; and (c) environmental facilitators or barriers to participation. Information was collected on QOL as the outcome measure.

Demographic Information

Demographic information was collected in the following areas: area of residency (county), age, gender, highest level of education, marital status, occupational status (e.g., employment status), and type and duration of disability or chronic health condition. These demographic variables were found important in previous QOL research (cf. Cummins, 2003; Deiner, 1984) and included in this study as control variables.

Function, Activity, and Participation

The WHO Disability Assessment Schedule Version 2.0 (WHODAS 2.0) was selected as a way to operationalize function, activity, and participation because it is the only instrument of function and disability based on the ICF framework (Üstün, Kostanjsek, Chatterji, & Rehm, 2010). It is designed to measure levels of functioning, regardless of type of disability, as operationalized through questions on impairments, activity limitations, and participation restrictions (Üstün, Kostanjsek, et al., 2010). According to the WHO, the WHODAS 2.0 is “a practical, generic assessment instrument that can measure health and disability at the population level or in clinical practice” (p. 4). The WHODAS 2.0 captures information within six domains of functioning, including cognition, mobility, self-care, getting along with others, life activities, and participation, and provides a general measure of functioning and disability in a way that is applicable across nations and cultures (Üstün, Chatterji, et al., 2010).

Evaluation and Psychometric Properties of the WHODAS 2.0

The internal consistency (Cronbach’s alpha values) of the WHODAS 2.0 is reported as follows: Domain 1 (cognitive) = .94; Domain 2 (mobility) = .96; Domain 3 (self-care) = .95; Domain 4 (getting along) = .94; Domain 5a (daily life activities) = .94; Domain 5b (work/school activities) = .94; and Domain 6 (participation) = .05. These measures meet the standards of acceptable to very good internal consistency (Üstün, Kostanjsek, et al., 2010). A confirmatory factor analysis supported the independent structure of the domains and was similar across testing sites (coefficients ranged from .82 to .98 across domains). The overall intraclass coefficient for the model was calculated at .98, showing a high level of reliability. WHODAS 2.0 was found to be at least as sensitive to change across time as similar measures of social functioning, and results held across individuals from different socioeconomic and demographic groups indicating that the instrument is applicable to individuals from different cultures (Üstün, Kostanjsek, et al., 2010). The WHODAS 2.0 has been compared with other instruments measuring similar constructs. Results indicate that the WHODAS 2.0 measures similar constructs (.45–.65) as these instruments measures unique components too. Evidence of sensitivity to change after treatment and ability to differentiate samples of people with and without health problems support the construct validity of this instrument (Üstün, Kostanjsek, et al., 2010). High construct validity is important in the present study where we studied people with different types of disability from different life situations who were currently receiving rehabilitation services.

Environmental Factors

Environmental factors affecting community participation were assessed through a modified version of a section of the ICF checklist, which is a clinical interview form based on the ICF (WHO, 2003). The ICF checklist was found on the WHO website, and can be accessed at http://www.who.int/classifications/icf/training/icfchecklist.pdf. The environmental questions were adapted from the structured interview, and were selected based on applicability to the study. Participants were asked about the following environmental factors: products and technology (e.g., assistive technology, medication), natural environment (e.g., climate), relationships and social support (e.g., family, friends, personal care or health care staff), attitudes of others (e.g., family, friends, personal care, or health care staff), and services, systems, and policies (e.g., transportation, housing, employment, education). This section of the survey was designed to determine the person’s perception of the community and immediate social environment as well as whether these aspects support or hinder their community participation.

QOL

The outcome variable for this research question is self-reported QOL. The question that participants responded to was, “How would you rate your quality of life?” Responses were on a 5-point Likert-type scale ranging from “very poor” to “very good” with a neutral option. The single item global measure has been shown to match up well with other measures of QOL, be stable over time, and is not thought to be vulnerable to social desirability (Gill & Feinstein, 1994; Nolte, 2000).

Participant Recruitment and Data Collection

For the student sample, the survey was emailed by the director of the disability services center to 1,022 current university students. A total of 136 surveys (13.3%) were returned, of which 122 (11.9%) had sufficient data to be retained in the final sample. Two survey approaches were used for the CRP sample. During the in-person data collection, a total of 236 surveys were distributed. Of these, 200 were returned. Surveys were also distributed online (to 300 people), and 40 were returned. Of the 240 total surveys returned (44.8%), 224 (41.7%) had sufficient data to be retained in the sample. The final samples included 122 students and 224 CRP clients, for a total of 346 participants (22.2% overall response rate).

Missing Data

The researchers examined all survey responses for missing data, discarding the surveys of individuals who provided low-quality responses (e.g., same answer for entire survey). Approximately 28% of the sample had at least one missing data point. Participants who did not answer the question regarding their QOL were not included in the final sample. For the WHODAS 2.0 and Environmental sections of the survey, the researchers replaced missing values with a single imputation method (i.e., the median score for the sample) to maximize the number of participants who could be retained in the sample (McKnight, McKnight, Sidani, & Figueredo, 2007).

Participant Characteristics

A primary motivation for seeking out both samples was ensured that individuals reflected a range of individuals served by rehabilitation counselors. Having both samples also ensured participant variability in several key characteristics including age, education, work experience, life circumstance, and so on. Table 1 shows participant demographics.

Table 1.

Participant Characteristics Related to Personal and Disability Factors.

	Students		CRP clients		Total
Variable	n	%	n	%	n	%
Gender
Female	95	78.5	126	57.0	221	64.6
Male	26	21.5	95	43.0	121	35.4
Age
25 or less	89	74.8	28	13.0	117	35.0
26–45	23	37.8	83	38.6	106	31.7
46 and older	7	5.9	104	48.4	111	33.2
Race/ethnicity
White	100	82.0	126	57.0	226	65.9
Black/African
American	10	8.2	53	24.0	63	18.4
Latino	3	2.5	14	6.3	17	5.0
Asian/Pacific
Islander	1	0.8	10	4.5	11	3.2
Multiracial	5	4.1	12	5.4	17	5.0
Other	3	2.5	6	2.7	9	2.6
Marital status
Never married	99	81.1	94	42.5	193	56.3
Cohabitating	8	6.6	5	2.3	13	3.8
Married	9	7.4	72	32.6	81	23.6
Separated	1	0.8	10	4.5	11	3.2
Divorced	5	4.1	31	14.0	36	10.5
Widowed	0	—	9	4.1	9	2.6
Disability type
Blind/visually
impaired	4	3.3	5	2.3	9	2.6
Deaf/hearing
impaired	4	3.3	5	2.3	9	2.6
Mobility	8	6.6	12	5.5	20	5.9
Brain injury	3	2.5	9	4.1	12	3.5
LD/ADHD	31	25.4	48	22.0	79	23.2
Mental Health	30	24.6	41	18.8	71	20.9
Chronic health	25	20.5	39	17.9	64	18.8
Autism	5	4.1	5	0.3	10	2.9
Other^a	12	9.8	54	24.8	66	19.4
Disability duration
<1 year	1	0.8	7	3.3	8	2.4
1–4 years	23	18.9	11	5.2	32	10.1
5–9 years	30	24.6	39	18.3	69	20.6
10 years and above	39	32.0	93	43.7	132	39.4
Since birth	29	23.8	63	29.6	92	27.5
Level of education
Less than HS	0	—	23	10.5	23	6.7
Some HS	0	—	25	16.1	25	7.3
HS/GED	15	12.3	81	37.0	96	28.2
Some postsecondary education (PSE)	56	45.9	57	26.0	113	33.1
Associate’s	9	7.4	15	6.8	24	7.0
Bachelor’s	28	23.0	13	5.9	41	12.0
Master’s and above	14	11.5	5	2.3	19	5.6
Living situation
Own apt/home	70	57.4	185	84.5	255	74.8
Other’s home	5	4.1	20	9.1	25	7.3
Group living	2	1.6	8	6.4	10	2.9
Shelter	0	—	3	1.4	3	0.9
College housing	41	33.6	0	—	41	12.0
Other	4	3.3	3	1.4	7	2.1
Occupational status^b
Paid employment	56	45.9	204	91.0
Self-employment	8	3.6	2	1.6
Nonpaid employment	17	11.5	4	1.8
Student	113	92.6	5	2.2
Retired	2	1.6	3	1.3
Unemployed—looking	11	4.9	4	3.3
Unemployed—not looking	4	3.3	1	0.0
Other	2	1.6	4	1.8

Note. Due to missing data, participant responses may not all add up to 346. CRP = community rehabilitation program; LD/ADHD = learning disabilities/attention-deficit hyperactivity disorder.

Of individuals indicating “Other,” 8 students and 26 CRP clients indicated multiple disability types, and 10 CRP participants did not indicate a disability type. ^bParticipants could indicate more than one occupational status.

The participants in the two sample groups differed by personal characteristics. Chi-square analyses or independent sample t tests were performed on each personal characteristic variable. Gender ratios were significantly different between samples (χ² = 15.807; p < .001; df = 1). The student sample included mostly females (78.5%), while the CRP sample had a higher proportion of females to males but was more evenly distributed by gender. Proportion of participants by race was also significantly different between samples (χ² = 24.141; p < .001; df = 5). The student sample was mostly White (82%). The CRP sample showed more diversity, with 43% of participants coming from racial/ethnic minority groups. A more detailed breakdown is available in Table 1. The students were younger than the CRP participants (students M = 25.23, SD = 8.91; CRP customers M = 42.89, SD = 13.45; t = 14.886; p < .001). Groups were also significantly different in terms of marital status (χ² = 61.513; p < .001; df = 5). As might be expected, 86% of the student sample was nonmarried or partnered. About two thirds of the CRP sample was not married or partnered, while one-third reported being married/partnered or living with a significant other. The student sample had levels of education ranging from high school graduate or equivalent to master’s degree or higher. The majority of students had completed some postsecondary education, with another 23% completing a bachelor’s degree. The CRP sample had a wider distribution of educational levels, with the largest group earning a high school diploma or the equivalent, and another 26% with some postsecondary education. The proportional difference in educational level was significant (χ² = 83.825; p < .001; df = 6).

There were no differences in proportion of reported disability type between the two samples (χ² = 13.114; p = .108; df = 8). The largest proportion of students reported learning disabilities or attention-deficit hyperactivity disorder (LD/ADHD) as their primary disability type, psychiatric or mental health, was second most common, and chronic health conditions as the third most common. Ten students reported “other” and the majority of these respondents (n = 8) indicated multiple disabilities. In the CRP sample, the majority of respondents indicated LD/ADHD as their primary disability type, with “other” (n = 54; 24.8%) as second most common. Of those reporting “other,” 26 individuals reported multiple disabilities and 10 chose not to indicate a disability type or indicated the kind of services they receive at the CRP. In each group, the largest proportion of participants reported having their disability or health condition for 10 years or longer. The second largest group in the student sample was “5 to 9 years.” In the CRP sample, the second largest group was “Since birth.” Differences in duration of disability between samples was significant (χ² = 21.427; p < .001; df = 4). The differences observed between groups support looking at the groups separately as well as overall.

With respect to occupational status, 91% of CRP respondents indicated being employed with only a small number of respondents indicating a different occupational status. Student responses were more varied; 92% of students indicated “student” as an occupation status with another 45% indicating some paid employment and 11% indicating nonpaid employment. Five percent of students indicated that they were unemployed but seeking work.

Although comprehensive comparative population data are not available, some information is available to provide a general context. For the student sample, information on disability type and grade level was available. The distribution of type of disability for the student sample was similar to that for the population of students with disabilities served by the resource center with one the exception—the sample contained a lower proportion of individuals with LD/ADHD (46% in the population, 25% in the sample) and a higher proportion of individuals reporting chronic illness (15% in the population, 20% in the sample). In addition, 82% of the students who received services from the resource center were undergraduates. The mean age and age range of participants in the student sample matches that of the overall university population (University Resource Center for Persons With Disabilities, 2010). The distribution of individuals for the broader university community by race/ethnicity is similar to the individuals in the sample. However, females were overrepresented in the student sample, relative to the broader university population which is nearly half male and half female (University Report, 2011).

According to the National Center for Educational Statistics (NCES), nationally approximately 11% of college students report having at least one disability (Snyder & Dillow, 2012). Among college students reporting a disability, approximately 57% are female, and 42% are male (Aud et al., 2012). This ratio differs from the gender ratios reported by the sample (78% female). NCES reported that 54% of college students with disabilities are between 15 and 23 years of age, 20% are between 24 and 29, and an additional 26% are 30 years of age and older (Snyder & Dillow, 2012). The present sample was a bit younger, with 62% aged 23 or younger, 16% between 24 and 29, and 20% aged 30 or older. NCES reports that employment figures for college students range from 40% to 51% of full-time students engaging in at least part-time employment (Aud et al., 2012).

The gender and race/ethnicity distributions for the respondents from the CRP client group are similar to those for the population from which they were recruited. Although exact comparison of disability type is problematic because the data from the CRP does not contain the same disability variables used in this study, study participants seem similar to the population with two exceptions: The population of CRP clients has a higher proportion of individuals with mental illness and chronic illness. The participants in the study had a higher level of education on average than the CRP client population (Program Report, 2011).

Exploratory Factor Analyses (EFA) were conducted on the WHODAS 2.0 and environmental questions with all participants to extract a common set of components, which allowed comparison of results across the two samples. A pooled within groups covariance matrix (university students and CRP clients were the two groups) using an oblique rotation was used for the EFA to permit calculation of a common set of components. The EFA approach was consistent with the assumption that the ICF model components were represented as latent variables as arranged by the instrument domains. According to the Kaiser-Meyer-Olin (KMO) measure, sample size met the criteria for this analysis (WHODAS EFA KMO = 0.910; Environmental section KMO = 0.903). Bartlett’s test was also significant for both sections, WHODAS 2.0: χ² = 7334.44 (443), p < .001; Environment: χ² = 3183.28 (153), p < 001, indicating that interitem correlations were substantial and adequate (Field, 2009). See Table 2 for information on the ICF model components and instrument domains.

Table 2.

ICF Model Components and Domains by Instrument Section.

ICF model components (Instrument section)	Domains
Personal factors	N/A—ordinal or nominal variables
Environment (ICF Checklist-Environment)	Products and technology
	Support and relationships
	Attitudes
	Services and systems
Function (WHODAS 2.0)	Cognitive
	Mobility
	Social
	Self-care
Activities (WHODAS 2.0)	Work-related activities
	Nonwork activities
Participation (WHODAS 2.0)	Participation

Note. ICF = international classification of functioning; WHODAS = World Health Organization Disability Assessment Schedule.

Function, Activity, and Participation

Using the items from the WHODAS 2.0, a six-factor solution explaining 68.56% of the variance was reached. Five items were dropped from the analysis because of low (<.4) item loadings (Ho, 2006). Standardized factor scores were generated using the regression score estimation method for use in the additional analyses. Compared with the original ICF model, which was the basis of the WHODAS 2.0 instrument, the factors resulting from the analysis were similar but not an exact match. Cronbach’s alpha measures of internal consistency were calculated for all factors and were in the very good to excellent range (Field, 2009). The first factor, called “social relationships and inclusion,” contained four items from the social dimension and two from participation dimension, specifically the items that indicated difficulty with feeling included in the community (α = .863). One item from the original social domain, “difficulty with sexual activities,” was dropped from the analysis because it cross-loaded with impact on self and family. An item from the original participation domain, “Dealing with barriers or hindrances in the world around you” dropped because it cross-loaded on the mobility/self-care domain and nonwork activities. The second factor, called “mobility and self-care,” contained all five items from the mobility domain, and two from the self-care domain (α = .885). The other two items from the original self-care domain, eating and staying alone, were removed. “Difficulty eating” cross-loaded on both self-care and social, and “difficulty staying alone” had low item loadings. The third factor contained the remaining four items from the original participation domain. It represents the impact of the disability or health condition on the person or their family and is called “impact on self/others” (α = .852). The fourth factor contained all four items from the original “nonwork activity” domain, it represents attending to household responsibilities (α = .942). The fifth factor contained five of the six items from the original cognitive domain (α = .823). The item “starting and maintaining a conversation,” was dropped from the analysis because it cross-loaded on both the social and cognitive components. The sixth and final factor, “work and school activities” (α = .885), included all four items from the original work-related activity domain. Table 3 below displays the variance explained and Table 4 shows the rotated pattern matrix. Table 5 displays the correlations between factors.

Table 3.

Total Variance Explained by the Six-Component Solution.

	Initial Eigenvalues			Extraction			Rotation
Factor	Total	% Var	Cum %	Total	% Var	Cum %	Total
Social/inclusion	11.429	36.868	36.868	11.429	36.868	36.868	6.454
Mobility/self-care	2.752	8.879	45.747	2.752	8.879	45.747	6.293
Impact	2.263	7.300	53.046	2.263	7.300	53.046	4.540
Nonwork	1.897	6.118	59.165	1.897	6.118	59.165	6.668
Cognitive	1.570	5.065	64.230	1.570	5.065	64.230	4.843
Work/school	1.342	4.329	68.558	1.342	4.329	68.558	6.158

Table 4.

Rotated Pattern Matrix of Factors.

	Factors^b
Item (ICF Domain^a)	1	2	3	4	5	6
Concentrating (C)					.743
Remembering (C)					.775
Analyzing problems (C)					.742
Learning new tasks (C)					.683
Understanding others (C)					.507
Standing for 30 min (M)		−.693
Standing up from sitting (M)		−.841
Moving inside home (M)		−.753
Getting out of home (M)		−.701
Walking long distances (M)		−.761
Washing self (S-C)		−.550
Getting dressed (S-C)		−.623
Dealing with strangers (S)	.786
Maintaining friendships (S)	.866
Getting along with those close to you (S)	.659
Making friends (S)	.843
Taking care of household tasks (NW)				−.835
Doing household chores well (NW)				−.825
Getting all housework done (NW)				−.827
Getting work done quickly (NW)				−.793
Work/school activities (W)						−.758
Doing tasks well (W)						−.736
Getting all tasks done (W)						−.784
Getting tasks done quickly (W)						−.767
Joining community activities (P)	.476
Living with dignity (P)	.527
Time spent on condition (P)			.689
Emotional consequences (P)			.686
Drain on family resources (P)			.717
Emotional consequences family (P)			.759
Independent leisure (P)			.565

Note. Removeditems: (1) starting conversations (C)—social and cognitive; (2) eating (S-C)—self-care and social; (3) staying alone (S-C)—low loadings; (4) sexual activities (S)—social and impact on self and family; and (5) dealing with barriers (P)—mobility/self-care and nonwork activities.

ICF = international classification of functioning.

ICF Domains: C = cognitive; S = social; M = mobility; S-C = self-care; W = work activities; NW= nonwork activities; P = participation. ^b1 = social relationships and inclusion; 2 = mobility/self-care; 3 = impact on self/family; 4 = nonwork activities; 5 = cognitive; 6 = work/school activities.

Table 5.

Correlations Between WHODAS 2.0 Factors.

	1	2	3	4	5	6
1	1.000
2	−.287	1.000
3	.202	−.213	1.000
4	−.325	.349	−.281	1.000
5	.377	−.186	.171	−.235	1.000
6	−.381	.306	−.185	.415	−.288	1.000

Note. WHODAS = World Health Organization Disability Assessment Schedule.

Key:

= social relationships & inclusion

= mobility/self-care

= impact on self/family

= nonwork activities

= cognitive

= work/school activities.

Environment

The ICF checklist included four distinct aspects of environment: products and technology, support and relationships, attitudes, and services. Three factors were extracted that explained 60.16% of the variance, including “services,” “relational support and attitudes,” and “assistive technology and accessibility.” One item was removed from the analysis (products for personal use) because the communality statistic was below acceptable minimums (.308), and another (services related to communication) was removed because of low (<.4) factor loadings (Ho, 2006). Cronbach’s alpha scores were calculated for each factor and were all within range of good to very good internal consistency: Services α = .868 (three items); Support and attitudes α = .872 (seven items); and Assistive technology and access α = .788 (three items). Table 6 below displays the variance explained and Table 7 shows the rotated pattern matrix. Table 8 displays the correlations between factors.

Research Question 2: What is the contribution of the ICF model (e.g., social relationships, environmental factors, personal factors, mobility, self-care) relative to typically derived outcomes of services (e.g., employment or independent living) in predicting QOL?

Table 6.

Total Variance Explained by the Three-Component Solution.

	Initial Eigenvalues		Extraction		Rotation
	Total	% Var	Cum %	Total	% Var	Cum %	Total
Services	7.267	42.747	42.747	7.267	42.747	42.747	5.428
Supp./attitudes	1.680	9.884	52.631	1.680	9.884	52.631	5.704
AT and access	1.280	7.529	60.161	1.280	7.529	60.161	4.079

Table 7.

Rotated Pattern Matrix of Factors.

	Factors^b
Item (ICF checklist domain^a)	1	2	3
Products for mobility/transportation (P&T)			.811
Products for communication (P&T)			.808
Public and private buildings (P&T)			.751
Relationships with immediate family (S&R)		−.761
Relationships with friends/acquaintances (S&R)		−.759
Relationships with personal care staff/healthcare workers (S&R)		−.529
Attitudes of family (A)		−.865
Attitudes of friends/acquaintances (A)		−.823
Attitudes of personal care staff/healthcare workers (A)		−.651
Societal attitudes in the community (A)		−.617
Accessible housing (S)	.441
Accessible transportation (S)	.642
Legal services (S)	.588
Social security (S)	.569
Health services (S)	.849
Education and training (S)	.824
Employment (S)	.773

Note. ICF = international classification of functioning. Removed items = (1) Products for personal use (low communality), and (2) services for communication (low loadings).

P&T = Products and Technology; S&R = Support and Relationships; A = Attitudes; S = Services. ^b1 = services; 2 = relational support & attitudes; 3 = assistive technology & accessibility.

Table 8.

Correlations Between the Environment Factors.

	Services	Support and attitudes	AT and accessibility
Services	1.000
Support and attitudes	−0.485	1.00
AT and accessibility	0.402	−0.436	1.00

Analysis

To answer this question, a block entry regression was used to regress QOL on the independent variables entered as follows: Block 1—personal factors (demographic characteristics), Block 2—work/school and nonwork activities (work and household tasks), and Block 3—the remaining ICF components. This analysis compared the variance explained in QOL by the information typically available to counselors when providing rehabilitation services (personal factors, disability information, and difficulties with work or life activities) with the full ICF model (Block 3). The analysis was carried out first with the entire sample using the mean-centered predictors based on pooled within groups variance. Data were mean centered to account for any differences in variance in the two samples. Because differences between samples were observed, an additional within-group analysis was also conducted using the pooled variance component scores along with the original demographic data. The results are presented below: first the entire sample, then student sample, and then CRP client sample.

Combined Sample

When QOL was regressed on the demographic variables (Block 1), the variance explained was 3.8%. Adding the work/school and nonwork activities (Block 2) resulted in 7.8% of the variance explained (an increase of 4.0% in R² over Block 1). The full model (Block 3) explained 25.0% of the variance in QOL (an increase of 17.2% in R² over Blocks 1 and 2), a substantial improvement in variance accounted for in other life areas beyond simply personal factors, work/school, and nonwork activities. Table 9 shows the model summaries.

Table 9.

Model Summaries for the Block Regression-Combined Samples.

Model	R ²	Adjusted R²	ΔR²	SE
Demographics only (age, gender, race, marital education, duration, disability)	.084	.038	—	.905445
Demographics + Activities (work/school and nonwork)	.127	.078	4.0%	.886451
Demographics + Activities+ All components from ICF (social relationships/inclusion, mobility/self-care, cognitive, impact on self/family, services and community attitudes relational support and attitudes, AT and accessibility)	.307	.250	17.2%	.799619

Note. ICF = international classification of functioning.

Student Sample

When QOL was regressed on the demographic variables (Block 1), the variance explained was 1.4%. Adding Block increased the variance explained to 17.8% (an increase of 16.4% in R²). The full model (Block 3) explained 25.8% of the variance in QOL (an increase of 8.0% in R²). For students adding the work/school and nonwork activities along with the rest of the factors from the ICF model adds substantially to the predictive power of the model; however, the second block shows the greatest increase in variance explained. Table 10 shows the model summaries.

Table 10.

Model Summaries for the Block Regression-Student Sample.

Model	R ²	Adjusted R²	R ²	SE
Demographics only (age, gender, race, marital education, duration, disability)	.132	.014	—	.925061
Demographics + Activities (work/school and nonwork)	.291	.178	16.4%	.844625
Demographics + Activities+ All components from ICF (social relationships/inclusion, mobility/self-care, cognitive, impact on self/family, services and community attitudes relational support and attitudes, AT and accessibility)	.404	.258	8.0%	.802358

Note. ICF = international classification of functioning.

CRP Client Sample

When QOL was regressed on the demographic variables (Block 1), the variance explained was 7.9%. Adding Block 2 to the model attained 9.2% of the variance was explained (an increase of 1.3% in R²). The full model (Block 3) explained 26.5% of the variance in QOL (an increase of 17.3% in R²). For the CRP clients, the ICF model added significantly to the prediction of QOL. Table 11 shows the model summaries.

Table 11.

Model Summaries for the Block Regression-CRP Sample.

Model	R ²	Adjusted R²	R ²	SE
Demographics only (age, gender, race, marital education, duration, disability)	.143	.079	—	.880221
Demographics + Activities (work/school and nonwork)	.165	.092	1.3%	.873832
Demographics + Activities+ All components from ICF (social relationships/inclusion, mobility/self-care, cognitive, impact on self/family, services and community attitudes relational)	.349	.265	17.3%	.786300

Note. CRP = community rehabilitation program; ICF = international classification of functioning.

Summary

The purpose of this study was to compare how well QOL could be predicted by using information on difficulty with employment and independent living tasks only, compared with additional information on other life areas as is found in the ICF model. For the two samples combined using the full ICF model added substantially to the variance explained in QOL (an increase of 17.2% in R²). For both students and the CRP client sample, the full model explained a significantly larger proportion of the variance than the reduced models (Blocks 1 and 2). For the student sample, the work/school and nonwork activities (Block 2) also contributed significantly to the variance accounted for in QOL. Given the larger number of CRP participants compared with students, the results from the combined samples may be more reflective of the CRP sample.

Overall, results indicated that the typical outcomes measures used by rehabilitation services (e.g., employment and independent living tasks) were relatively weak indicators of QOL, although for students, these factors showed a stronger relationship. Instead, other life areas and environment (the full ICF model) provided a better way to account for reported QOL than just the demographic information and information related to functioning with work and school tasks and household activities.

The findings from the present study were consistent with previous efforts to connect levels of function in life areas to measures of subjective well-being. These studies, like the present study, suggest that additional factors beyond employment and household management skills are important to well-being and subjective QOL. da Silva Cardoso et al. (2004) connected life skills (e.g., employability skills, self-care, communication, etc.) with measures of subjective well-being (e.g., physical, psychological, social, and environmental). Findings indicated a moderate correlation between life skills and well-being. Authors concluded that rehabilitation services aimed at improving life skills would lead to an improvement in subjective QOL. However, the level of required for work did not seem to correlate significantly with psychological well-being. Authors suggested that other factors may be at play, including stress, family support, and interference in activities.

Heinemann and Whiteneck (1995) used an earlier version of the ICF (the ICIDH) to compare aspects of function, disability, and handicap with subjective well-being and found that social and productive activities were the strongest predictors. Participants included 758 adults with traumatic brain injury, recruited through the state head injury association, rehabilitation facilities, state VR agencies, and additional community sources. In the current study, the addition of social factors and inclusion, environment, and additional life areas seemed important in predicting QOL as well. In the Heinemann and Whiteneck’s study, the model only explained 13% of the variance in QOL, and environmental factors were not included. In an earlier study, Fuhrer, Rintala, Hart, Clearman, and Young (1992) also found an indirect negative relationship between level of perceived “handicap” and life satisfaction; however, in the predictive model of life satisfaction, the significant predictors were self-assed health, perceived control, and social support indicating that these other factors were more prominent.

Previous studies where results showed relationships between life satisfaction, various types of life skills, and perception of handicap provide us with a nice comparison to the present study. Instead of “life skills” or perceived “handicap,” the focus here was on broader dimensions of functioning as well as including environmental supports. The common theme is that it appears that function plays a role in QOL, but the greater picture is a combination of other elements related to the interaction of the person in their environment.

Study Limitations

Although this study has several strengths, some limitations must be noted. First, the two samples recruited for this study were not adequate to generalize to all people with disabilities, or even all CRP clients and university students with disabilities. The combination of an incentive offered and the low response rate in the student sample makes it likely that some response bias is observed in this group of participants. However, the samples were quite comparable to the populations from which they were drawn. The student sample was compared with national data on college students. Although many of the demographics (e.g., gender, age) were different from the current sample, the proportion reporting also having a job was similar. A larger overall sample size would have been desirable to increase generalizability, as would confirmation that the individuals who chose to participate did not differ from individuals who did not. These limitations suggest taking care when considering the implications of the findings for college students with disabilities in particular. Replication of these results in a larger student sample would provide more confidence in the applicability of the findings. The full model (all three blocks entered) contained 23 predictors, for which the student sample is only 5 participants per variable, which is below the generally accepted guideline of 10 cases per predictor (Field, 2009). The combined groups analysis and the CRP sample analysis are both closer to the generally accepted criteria (15 cases per variable and 9.7 cases per variable, respectively). Another limitation concerns the self-report methodology, particularly for information on function or difficulty with particular tasks. Although other researchers have found “subjective health assessments” useful in ascertaining health and function (e.g., Albrect, 1996), there is no way to confirm validity of self-responses.

Implications for Rehabilitation Administrators and Counselors

Wright’s seminal rehabilitation counseling text states, “rehabilitation is a facilitative process enabling a person with a [sic] handicap to attain usefulness and satisfaction with life” (Wright, 1980, p. 3). Wright goes on to explain that handicap results from a combination of disability itself, as well as cultural, financial, or educational disadvantage. The goal of rehabilitation is not limited to employment; it includes activity considered personally useful and satisfying. For a variety of reasons, including historical roots and legislative mandates, the outcome measurements of rehabilitation (particularly in the public sector) have largely been limited to employment and to a lesser extent independent living. However, in the literature, we find frequent suggestions that this narrow focus leaves us with a misleading impression of the results of rehabilitation services (Bishop et al., 2008; Chapin et al., 2004). Results from this study supports claims made by Bishop and Feist-Price (2002) and Chapin et al. (2004) that employment and independent living are not sufficient to capture the broader picture of QOL.

The different results from the two samples (e.g., the contribution of Block 2 for the student sample) might be attributed to some demographic differences observed between the students and the CRP clients. It is possible that factors such as the younger average age, the higher level of education, or the prominent role as student might have contributed to the variation in results between the samples. Further study into particular factors that are relevant to the QOL of students as well as CRP clients is warranted to highlight factors that might be of importance when providing services.

This study strongly suggests that using the ICF model as a framework for client assessment, which connects functional skills, environmental supports, and opportunities for activity and participation to QOL is useful for clients at the beginning and the end of service provision. The ICF along with a brief assessment of QOL (e.g., “How would you rate your quality of life?”) could provide practitioners with self-reported and subjective information about how the person is doing when first introduced to the agency, and thus provide some structure and direction for service plan development that includes areas of life connected to QOL as well as the goal of services. The guidelines for services within agencies are broad, and examples range from counseling and guidance, vocational training/education, durable medical equipment to rehabilitation technology, supported employment, or transition services as stipulated in the act (Rehabilitation Act of 1973, as amended, Title I, Section 103). Counselors may need to focus their counseling and guidance efforts in some of these other life areas or environmental matters that are related to QOL to meet the individual needs of clients. For example, counselors might address social relationships or sources of support, aspects of the individuals home or work environment, or community access through counseling and guidance or referral to more intensive services by more appropriate providers if necessary. This more holistic model is consistent with the philosophical underpinnings of rehabilitation (Wright, 1980), and ensures that attention is paid to environmental supports as well as person-level factors. Given the curricular requirements set by the Council on Rehabilitation Education (CORE), master’s level rehabilitation counselors graduating from accredited programs should be prepared to address these additional life areas in their counseling (CORE, 2012).

In addition, the information at both intake and exit could provide the agency with data that reflects changes in life areas such as employment or independent living as well as functional ability, environmental support, and perceived QOL. At a time where accountability is essential to program funding (Leahy, Thielsen, Millington, Austin, & Fleming, 2009), more comprehensive information about the nature of the impact of services can only help agencies prove their merit or worth (Frisch, 2004). Ensuring that the benefits of rehabilitation services are well understood by funders as well as the general public is critical to program viability (McFarlane, Schroeder, Enriquez, & Dew, 2011).

Implications for Rehabilitation Counseling Research

As one of the few examples of cross-disability QOL research with practical connections to rehabilitation services, the results of this study would be strengthened with replication in various settings with samples representative of the general population of clients of rehabilitation services. Continued work is needed to understand the connection between the areas of the ICF model of health and disability and effective service planning and provision. QOL is one outcome measure to consider, but others relevant to individual service providers might also reveal important findings related to implementing the ICF as a comprehensive model of assessment. For example, are employment and customer satisfaction improved when the ICF model is used as a framework for assessment and service planning? Another area of future research is to assess changes in personal perception in multiple life areas (e.g., included in the ICF) as a method of determining the impact of services. Focusing only on employment or educational outcomes may underestimate the broader impact of services on QOL.

Conclusion

Findings of this study indicate that QOL can be explained as a combination of function, difficulty with work and daily living activities, community participation, and environmental support. The ICF model is much better at estimating QOL than the more typical measures of rehabilitation services outcomes, employment and independent living. To better help clients attain QOL, a shift in how we consider individuals, plan interventions, and measure outcomes may be beneficial at both the individual and agency level in targeting resources to areas personally meaningful as well as providing more complete information on the value of service interventions. The present study provides us with initial results to that support the utility of the ICF for conceptualizing disability and its impact in a way that is inclusive of both personal and environmental factors, and providing a more comprehensive picture of QOL.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported in part by a dissertation grant from Michigan State University.

References

Albrect

(1996). Using subjective health assessments in practice and policy-making. Health Care Analysis, 4, 284–292.

Alexander

J. L.

Willems

E. P.

(1981). Quality of life: Some measurement requirements. Archives of Physical Medicine and Rehabilitation, 62, 261–265.

Aud

Hussar

Johnson

Kena

Roth

Manning

Zhang

(2012). The condition of education. Arlington, VA: National Center for Educational Statistics. Retrieved from http://nces.ed.gov/programs/coe/

Bishop

(2005). Quality of life and psychosocial adaptation to chronic illness and acquired disability: A conceptual and theoretical synthesis. Journal of Rehabilitation, 71, 5–13.

Bishop

Chapin

M. H.

Miller

(2008). Quality of life assessment in the measurement of rehabilitation outcome. Journal of Rehabilitation, 74, 45–54.

Bishop

Feist-Price

(2002). Quality of life assessment in the rehabilitation counseling relationship: Strategies and measures. Journal of Applied Rehabilitation Counseling, 33, 35–42.

Cardus

M. D.

Fuhrer

M. J.

Thrall

R. M.

(1981). Quality of life in benefit-cost analysis of rehabilitation research. Archives of Physical Medicine and Medical Rehabilitation, 62, 209–211.

Chan

Rubin

S. E.

Kubota

Chronister

Lee

(2003). Counselors’ and consumers’ derived criteria for assessing the effectiveness of rehabilitation services. International Journal of Rehabilitation Research, 26, 335–337. doi:10.1097/00004356-200312000-00014

Chan

Tarvydas

Blalock

Strauser

Atkins

B. J.

(2009). Unifying and elevating rehabilitation counseling through model-driven, diversity-sensitive evidence based practice. Rehabilitation Counseling Bulletin, 52, 114–199. doi:10.1177/0034355208323947

10.

Chan

Wang

C. C.

Muller

Fitzgerald

(2011). Vocational rehabilitation outcomes: A multi-level analysis of economic indicators, VR agency characteristics, and RSA-911 data. Phase one report. Madison, WI: Rehabilitation Research and Training Center on Effective VR Service Delivery Practices.

11.

Chapin

M. H.

Holbert

(2010). Employment at closure is associated with enhanced quality of life and subjective well-being for persons with spinal cord injuries. Rehabilitation Counseling Bulletin, 54, 6–14.

12.

Chapin

M. H.

Miller

S. M.

Ferrin

J. M.

Chan

Rubin

S. E.

(2004). Psychometric validation of a subjective well-being measure for people with spinal cord injuries. Disability and Rehabilitation, 26, 1135–1142. doi:10.1080/09638280410001714772

13.

Cieza

Stucki

(2008). The international classification of functioning disability and health: It’s development process and content validity. European Journal of Physical and Rehabilitation Medicine, 44, 303–313.

14.

Council on Rehabilitation Education. (2012). Accreditation manual for masters level rehabilitation counselor education programs. Schaumburg, IL: Author.

15.

Crewe

N. M.

(1980). Quality of life: The ultimate goal of rehabilitation. Minnesota Medicine, 63, 586–589.

16.

Cummins

R. A.

(1996). The domains of life satisfaction: An attempt to order chaos. Social Indicators Research, 38, 303–328. doi:10.1007/BF00292050

17.

Cummins

R. A.

(2003). Normative life satisfaction: Measurement issues and a homeostatic model. Social Indicators Research, 64, 225–256. doi:10.1023/A:1024712527648

18.

da Silva Cardoso

Blalock

Allen

C. A.

Chan

Rubin

S. E.

(2004). Life skills and subjective well-being of people with disabilities: A canonical correlation analysis. International Journal of Rehabilitation, 27, 331–334.

19.

Deiner

(1984). Subjective well-being. Psychological Bulletin, 95, 542–575.

20.

Field

(2009). Discovering statistics through SPSS (3rd ed.). London, England: SAGE.

21.

Frisch

M. B.

(2004). Quality of life assessments and interventions: The quality of life inventory. In Maruish

M. E.

(Ed.), The use of psychological testing for treatment and outcome assessment (3rd ed., Vol. 2, pp. 1277–1332). Mahwah, NJ: Lawrence Erlbaum.

22.

Fuhrer

M. J.

Rintala

D. H.

Hart

K. A.

Clearman

Young

M. E.

(1992). Relationship of life satisfaction to impairment, disability, and handicap among persons with spinal cord injury living in the community. Archives of Physical Medicine and Rehabilitation, 73, 552–557.

23.

Gilbride

D. D.

Thomas

J. R.

Stensrud

(1998). Beyond status code 26: Development of an instrument to measure the quality of placements in the state-federal program. Journal of Applied Rehabilitation Counseling, 29, 3–8.

24.

Gill

T. M.

Feinstein

A. R.

(1994). A critical appraisal of the quality of quality of life assessments. Journal of the American Medical Association, 272, 619–627. doi:10.1001/jama.272.8.619

25.

Heinemann

A. W.

Whiteneck

G. G.

(1995). Relationships among impairment, disability, handicap, and life satisfaction in persons with traumatic brain injury. Journal of Head Trauma Rehabilitation, 10, 54–63. doi:10.1097/00001199-199508000-00006

26.

(2006). Handbook of univariate and multivariate data analysis and interpretation with SPSS. Boca Raton, FL: Chapman & Hall. doi:10.1201/9781420011111.ch7

27.

Jenney

M. E. M.

Campbell

(1997). Measuring quality of life. Archives of Disease in Childhood, 77, 347–354. doi:10.1136/adc.77.4.347

28.

Kennedy

Lude

Taylor

(2006). Quality of life, social participation, appraisals and coping post spinal cord injury: A review of four community samples. Spinal Cord, 44, 95–105.

29.

Leahy

M. J.

Thielsen

V. T.

Millington

M. J.

Austin

Fleming

(2009). Quality assurance and program evaluation: Terms, models, and application in rehabilitation administration. Journal of Rehabilitation Administration, 33, 69–82.

30.

McFarlane

F. R.

Schroeder

F. K.

Enriquez

Dew

(2011). How do we lead when change is constant? Journal of Rehabilitation, 77, 4–12.

31.

McKnight

P. E.

McKnight

K. M.

Sidani

Figueredo

A. J.

(2007). Missing data: A gentle introduction. New York, NY: Guilford.

32.

Nolte

(2000). Disablement, context, and quality of life after brain injury: Measuring what matters (Unpublished doctoral dissertation). Retrieved from Dissertation Abstracts Online. (UMI No. 800-52f-0600).

33.

Program Report. (2011). Human services 2010-2011 program outcome summary report. Lansing, MI: Author.

34.

Rehabilitation Act of 1973, Public Law No. 93-112, 87 Stat. 355 (codified as amended in scattered sections of 15 U.S.C., 20 U.S.C., 29 U.S.C., 36 U.S.C., 41 U.S.C., and 42 U.S.C.).

35.

Roessler

R. T.

(1990). A quality of life perspective on rehabilitation counseling. Rehabilitation Counseling Bulletin, 34, 82–91.

36.

Rubin

S. E.

Chan

Thomas

D. L.

(2003). Assessing changes in life skills and quality of life resulting from rehabilitation services. Journal of Rehabilitation, 69, 4–9.

37.

Rubin

S. E.

Roessler

R. T.

(2001). Foundations of the rehabilitation process (5th ed.). Austin, TX: Pro-Ed.

38.

Saunders

J. L.

(2005). Case management in public rehabilitation. In Chan

Leahy

M. J.

Saunders

J. L.

(Eds.), Case management for rehabilitation health professionals (pp. 267–282). Osage Beach, MO: Aspen Professional Services.

39.

Smart

(2001). Disability, society, and the individual. Austin, TX: Pro-Ed.

40.

Snyder

T. D.

Dillow

S. A.

(2012). The digest of educational statistics 2011. Arlington, VA: The National Center for Educational Statistics. Retrieved from http://nces.ed.gov/programs/digest/d11/

41.

University Report. (2011). Diversity and inclusion: Annual progress report (Executive Summary). East Lansing, MI: Office of Inclusion and Intercultural Issues. Retrieved from http://www.inclusion.msu.edu/Outreach/Diversity_Report.pdf

42.

University Resource Center for Persons With Disabilities. (2010). Student information and trends. East Lansing, MI: Author.

43.

Üstün

T. B.

Chatterji

Kostanjsek

Rehm

Kennnedy

Epping-Jordan

Pull

(2010). Developing the World Health Organization Disability Assessment Schedule 2.0. Bulletin of the World Health Organization, 88, 815–823. doi:10.2471/BLT.09.067231

44.

Üstün

T. B.

Kostanjsek

Chatterji

Rehm

(2010). Measuring health and disability: Manual for disability assessment schedule (WHODAS 2.0). Geneva, Switzerland: World Health Organization.

45.

Whiteneck

Brooks

Harrison-Felix

Gerhart

(2004, June). Exploring the relationship between participation and environmental factors: A look at three studies using the Craig Handicap Assessment and Reporting Technique (CHART) and the Craig Hospital Inventory of Environmental Factors (CHIEF). Presentation at North American Collaborating Centre Tenth Annual Conference. Retrieved from http://secure.cihi.ca/cihiweb/en/downloads/Gale_Whiteneck.pdf

46.

World Health Organization. (1998). WHOQOL User manual. Geneva, Switzerland: Author.

47.

World Health Organization. (2002). Towards a common language of functioning, disability, and health: ICF. Geneva, Switzerland: Author.

48.

World Health Organization. (2003). ICF Checklist: Version 2.1a Clinical form. Geneva, Switzerland: Author. Retrieved from http://www.who.int/classifications/icf/training/icfchecklist.pdf

49.

Wright

G. N.

(1980). Total rehabilitation. Boston, MA: Little, Brown, and Company.