Findings from an analysis of the characteristics and services of community rehabilitation providers (CRPs) in the early years of the 21st century are presented. Services provided by CRPs can be categorized along two dimensions: purpose (work, nonwork) and setting (facility-based, community). The number of individuals with disabilities present provides a third perspective for analysis. The majority of CRPs provided both work and nonwork services, and the majority of those that provide employment services offered both integrated and facility-based employment. Individuals with developmental disabilities were most likely to be supported in facility-based work (41%), followed by nonwork services (33%), and integrated employment (26%). Despite some changes in CRP characteristics, the goal of community membership has not yet been widely achieved.
Over the past 20 years, extensive research and practice has demonstrated that individuals with developmental disabilities can be successful members of the labor market, earning above minimum wage, receiving benefits, and having opportunities for career advancement (Beare, Severson, Lynch, & Schneider, 1992; Butterworth, Sullivan, & Smith, 2001; Wehman, Hill, Wood, & Parent, 1987; Wehman & Kregel, 1995; Wehman, Revell, & Kregel, 1998). Beginning with the introduction of supported employment in the Developmental Disabilities Act Amendments of 1984 and the Rehabilitation Act Amendments of 1986, best practices have evolved to incorporate person-centered career planning, systematic instruction, supported entrepreneurship, coworker supports, job creation and restructuring, workplace accommodations, and assistive technology. Concurrently, there has been increasing national emphasis on the participation of individuals with disabilities in the labor force and, more broadly, on community participation of people with disabilities. Recent public policy initiatives include the New Freedom Initiative; the Ticket to Work and Work Incentives Improvement Act of 1999; the Workforce Investment Act of 1998 (which includes amendments to the Rehabilitation Act); the Individuals With Disabilities Education Act; the Americans With Disabilities Act (2004) (including the U.S. Supreme Court Olmstead decision, 1999); the Developmental Disabilities Assistance and Bill of Rights Act of 2000; and the 2001 elimination of sheltered employment as a successful rehabilitation outcome by the Rehabilitation Services Administration.
Despite these initiatives, participation in employment continues to be limited for persons with developmental disabilities (Rusch & Braddock, 2004; Wehman, Revell, & Brooke, 2003; Yamaki & Fujiura, 2002). Longitudinal data collected as part of the Institute for Community Inclusion National Survey of Day and Employment Programs for People With Developmental Disabilities indicated that in FY 2001, only 24% of individuals funded or monitored by mental retardation/developmental disabilities (MR/DD) agencies were in integrated employment, and, despite modest growth in the number of individuals supported in integrated employment, the percentage in integrated employment had remained level since FY 1996 (Institute for Community Inclusion, 2005). Moreover, Mank (2003) found that individuals with developmental disabilities who do work often experience limited work hours and wages and that there has been little or no increase in wages over the past decade.
In this paper we provide a summary of day services and supports provided by community rehabilitation providers (CRPs) as identified in a national survey conducted in 2002 and 2003. Our purpose is to provide a national assessment of the current implementation of employment and day services and to identify policy and practice implications for state investment in employment and day services. The findings provide a national estimate of the distribution of services from data provided directly by CRPs based on a random sample of such providers drawn from a national list. The study also provides a follow-up to a 1991 national survey of CRPs (McGaughey, Kiernan, McNally, Gilmore, & Keith, 1994, 1996).
Community rehabilitation providers are the major source of employment and day services for persons with developmental disabilities. In the United States, over 8,100 CRPs (Menz, Botterbusch, Hagen-Foley, & Johnson, 2003) provide a broad range of services, programs, and supports to people with disabilities. Funded by a mix of sources, but primarily supported by federal and state monies, CRPs are an integral component of the current MR/DD service system.
The predominant population supported by CRPs is persons with developmental disabilities. McGaughey et al. (1994, 1996) found that CRPs reported serving an average of 168 individuals in day and employment services and that 78% of individuals served had a developmental disability. The survey covered four major types of day supports: integrated (supported) employment, facility-based (workshop) employment, facility-based nonwork, and community-based nonwork (see Table 1).
Supported employment was further delineated as individual or group-based. The majority of CRPs (77%) offered both facility-based and integrated employment services, and CRPs were predominantly small organizations serving 100 or fewer individuals (56%). The largest service category identified by McGaughey et al. (1996) was facility-based work (41% of those receiving services), followed by facility-based nonwork (28%), group supported employment (9.8%), individual supported employment (10.9%), and competitive employment (9.8%). The proportion of participants with developmental disabilities varied across these services, ranging from 73% (facility-based work) to 84% (group supported employment) for most services, but only 47% for competitive employment services. At the time of this survey, some services were clearly emerging. The average length of operation of individual and group-supported employment services was only 4 years, whereas facility-based work services had been provided for an average of 16 years.
McGaughey et al. (1996) identified community-based nonwork or community integration services as an emerging service category that was reported by some CRPs despite not being a defined category in the survey. Community-based nonwork was added as an official category in later surveys, and in FY 2001, MR/DD agencies reported that 18% of individuals were supported in community-based nonwork services (Dreilinger, Gilmore, & Butterworth, 2001; Institute for Community Inclusion, 2005). There is currently limited data on the structure, activities, and outcomes of this service, and states have not established clear service expectations or quality assurance strategies (Sullivan, Boeltzig, Metzel, Butterworth, & Gilmore, 2004).
Services provided by CRPs can be categorized along two dimensions: purpose (work or nonwork) and setting (facility-based or community) (see Table 1). An additional dimension of services, the number of individuals with disabilities present, provides a third perspective for analysis. For example, as defined here, integrated employment includes both individual community jobs and small group models of supported employment. The literature suggests that group models are less successful than individual supported employment in achieving desired wages and levels of integration (Mank, Cioffi, & Yovanoff, 2000).
These three dimensions define alternative outcomes that can be placed in a larger context of the development of services and supports. Bradley (1994) defined three stages in the evolution of developmental disabilities services and supports. The first stage, the era of institutionalization and dependence, focused on separating people with disabilities from society by providing care in substantially separate settings. In the 1970s, a second era emerged: deinstitutionalization and community development, which focused on locating service settings in the community. Significant development and expansion of sheltered workshops and nonwork day programs occurred during this time—service models that were physically but not socially located in the community. This stage is represented by the facility-based work and facility-based nonwork cells in Table 1. The third stage, still in development, is the era of community membership, in which individualized supports are designed to enable people to be fully included in the larger community and to exert control over their own life choices. Within Bradley's framework, employment and day services should emphasize individual outcomes based on a supports paradigm. Full participation in the labor market, including both employment in individual integrated jobs and access to opportunities, such as self-employment, are central components of achieving community membership. The contribution of community-based nonwork services to community membership is less clear.
Looking at day services along these three dimensions, in the survey results presented here we address the following research questions: (a) What are the current characteristics and services of CRPs for people with developmental disabilities? (b) Do trends in services reflect that CRPs have embraced Bradley's (1994) era of community membership?
Our purpose in this study was to collect information about the characteristics and services of CRPs that provide employment and nonwork services to people with disabilities, with particular emphasis on people with developmental disabilities. The survey was adapted from the Institute for Community Inclusion's 1991 National Survey of Community Rehabilitation Providers (McGaughey et al., 1996), with additional items added to provide consistency with survey research conducted by the Rehabilitation Research and Training Center on Community Rehabilitation Programs at the University of Wisconsin-Stout.
The survey contained 22 items: Items 1 though 5 addressed organizational characteristics (e.g., type of organization, numbers served annually in employment and nonwork, respectively, and in both settings combined). Items 6 through 13 gathered data on employment services (e.g., funding source of services received, types of employment and nonwork services provided, and numbers of individuals supported by service and setting). In addition, Administrators of CRPs were asked about the activities and goals of their nonwork programs. Items 14 through 16 focused on CRP funding sources and annual operations budget for both employment and nonwork. Service trends over the past 3 years were the focus of Item 17. Items 18 through 22 provided information about CRP involvement with the Ticket to Work program, state and local Workforce investment Boards, and One-Stop Career Centers.
We pilot tested the survey with 10 organizations in the summer of 2001 and made revisions based on feedback before implementation. A shorter version containing 11 items (organizational characteristics, types of employment and nonwork services provided, number of individuals supported by service and setting, service trends over the past 3 years, and funding sources) was developed and offered in the third round of follow-up contact in an effort to increase responses.
Agency administrators were asked to report both annual and daily total numbers of people served in the different service settings. Data were requested on all individuals supported in services by the CRP and individuals with developmental disabilities supported by the CRP. For this survey, developmental disabilities was defined as including, but not limited to, mental retardation and sensory, neurological, and physical conditions acquired before age 22.
A nationally representative sample of CRPs that provided employment and/or nonwork services to individuals with disabilities, including those with developmental disabilities, was developed through a multistage sampling plan. Initially, a sample frame of 7,356 addresses was composed at the Rehabilitation Research and Training Center on CRPs, with input from project staff. The frame was cross-referenced with lists from other sources, including Goodwill, The Arc, United Cerebral Palsy, andthe Rehabilitation Accreditation Commission.
Sampling Method and Survey Response
Using simple random sampling, we drew a subsample of 25% (1,848 addresses) and randomly selected 400 addresses from the subsample for mailing in Fall 2001. Due to inaccuracies in the mailing list, many surveys were returned as undeliverable, and the survey yielded a very low response rate. We addressed these issues by randomly selecting 800 addresses from the subsample for address verification. When a current address for a CRP could not be tracked down, it was removed from the sample frame. This process yielded a final sample of 600 CRP addresses.
The 600 CRPs received a questionnaire and a postage-paid reply postcard that agencies could use to indicate whether they were not providing work and/or nonwork services. If a CRP was identified as ineligible (i.e., did not provide work and/or nonwork services), it was replaced with a CRP drawn from the remaining list. Replacement of ineligible CRPs was discontinued once the researchers had achieved the target number of responses and response rate (50%). Surveys returned as undeliverable by the United States Postal Service were checked to determine whether the provider still existed and/or to correct the address; surveys were sent again to the provider if correct information was obtained. Researchers conducted three rounds of follow-up telephone calls. Of the final sample of 507 eligible agencies, 254 returned the survey, yielding a 50% response rate. Of the 254 responding CRPs, 60 respondents completed the short version of the survey.
Strategies Used for Minimizing Nonresponse
Nonresponse error arises when some members of the sample population do not respond to the survey. Including a one dollar bill in the survey package was one of the strategies that we used to increase survey responses. In addition, copies of Institute for Community Inclusion publications were offered as a material incentive for survey completion. It is difficult to estimate the potential effect that those incentives had (or did not have) on the overall survey response rate. Timed follow-up reminders were another strategy employed. We conducted three rounds of follow-up telephone calls. Personal contact with survey respondents via telephone seemed to have a positive impact on the overall response rate. Offering a shorter version of the survey to CRPs in the third round of follow-up contact was another effort to increase responses.
Data Preparation and Analysis
We conducted the following postcollection processing steps: receipting, coding, data entry, editing, imputation, and data analysis. After designing a coding system, we went through the questionnaires to edit the answers and code those that were not already in numerical form (receipting and coding). Data entry personnel then entered the data from the questionnaires into an electronic file. We regularly checked the data entered to assure high quality of the survey data. Editing followed data entry: we inspected and, if needed, altered the data prior to statistical analysis through range edits, comparison to historical data (previous CRP surveys), checks of highest and lowest values in the data set, and consistency. Additionally, we checked the data for item nonresponse and adjusted for it (imputation). The survey data were then analyzed using a statistical analysis software package (SPSS 11.0). In this paper we report on the descriptive statistical results for the characteristics and services of CRPs.
Characteristics of Responding CRPs
Type of organization
The CRPs varied by type of organization. The vast majority were private organizations, with 86% nonprofit and 2% for-profit. Five percent reported that they were state-sponsored public agencies, and 6% indicated local sponsorship. The remaining 1% of CRPs identified as “other.”
Size of organization
The total number of individuals served in employment and nonwork services was used as an indicator of CRP size. The number of individuals supported in work and nonwork by the CRP on a selected date differed widely (see Table 2). Overall, survey respondents reported serving an average of 110 individuals in employment programs and an average of 61 in nonwork programs on a daily basis. Respondents were also asked to provide the number of individuals served on an annual basis in employment and nonwork services in order to reflect the total caseload for the year, including movement in and out of the program and individuals who do not receive daily services. On an annual basis respondents reported serving an average of 240 people with disabilities in employment programs and 180 individuals in nonwork programs.
Responding CRPs were broadly distributed geographically, with the largest number of CRPs (34%) located in the Midwest, followed by 27% in the South. Twenty-one percent of the responding agencies were in the Northeast, and 18% were located in the West.
Funding source by services provided
Respondents were asked to select from a list of funding sources all the ones that applied to their organization and to then choose the three largest funders. The list included federal, state, and local funding sources as well as other funders (e.g., business revenue, employers/industry). Overall, CRP services were funded by a mix of sources (see Table 3). The majority of CRPs that provided both work and nonwork services selected state departments of vocational rehabilitation (70%) and MR/DD (66%), business revenue (56%), and Medicaid Title XIX (55%) as funding sources. More than half of those providing work services only (66%) reported being funded by state departments of vocational rehabilitation. State departments of MR/DD were the most frequently reported funding source of providers of nonwork services only (67%).
Of those CRPs that provided both work and nonwork services, 29% reported Medicaid Title XIX as the largest funding source. For providers of work services, only 28% of respondents indicated state vocational rehabilitation agencies as their primary funder. This contrasted with providers of nonwork services only, of which 33% indicated state departments of MR/DD as their largest financial source. The second largest funder was state departments of vocational rehabilitation for work only (25%), state departments of MR/DD for work and nonwork (24%), and other funding sources (31%). The third largest funder was state departments of vocational rehabilitation for work only and work and nonwork (21% and 22%, respectively), and county/local/tribal government, self-pay plans for achieving self support, and impairment-related work expenses, and business revenue for nonwork only (20%).
In addition to funding sources, the survey also collected budgetary information from CRPs, whose operations budget for all employment and nonwork services for FY 2002–2003 ranged from a minimum of $15,000 to a maximum of $56 million, averaging around $2.7 million.
The majority of CRPs (69%) provided both employment and nonwork services. Of the remainder, 24% provided only employment services and supports compared to 7% that provided nonwork services only. These numbers indicate that the vast majority of CRPs (93%) provided employment services and supports, either alone or in conjunction with nonwork services.
In both employment and nonwork, most CRPs offered a mix of community-based and facility-based options. Of those CRPs that provided employment services, the majority (70%) offered both integrated and facility-based (sheltered) employment, with another 18% providing employment only in integrated settings, and 12% providing only sheltered employment. Likewise, the majority of CRPs (51%) that provided nonwork services served people in both facility- and community-based settings. Thirty-six percent used only facility-based nonwork settings, and 13% operated only community-based nonwork supports.
The three employment services most likely to be provided by organizations were competitive employment, individual supported employment, and sheltered employment, at 66%, 69%, and 71%, respectively (see Figure 1). Other service models were less prevalent, including transitional employment, enclaves, mobile work crews, and work-center-based employment. Entrepreneurial approaches were the least common, with only 14% of CRPs reporting provision of supports in this area. With respect to nonwork services, facility-based nonwork services were offered by 64% of CRPs, and community-based nonwork services were provided by 46% of CRPs.
In the overall category of integrated employment, individual employment models were more prevalent than group employment models. However, the findings suggest that enclaves and mobile work crews have continued to be actively-used models for employment of individuals with developmental disabilities, each provided by over one third of responding CRPs. In enclave and mobile work crew models of supported employment, a small group of workers with disabilities receives continuous support and supervision from CRP personnel. Responses to this survey showed that the average size of an enclave was 6 and the average size of a mobile work crew was 5. The total numbers served by enclaves and mobile work crews were 2,400 and 1,112, respectively. Respondents reported 430 enclaves and 231 mobile work crews.
Distribution of Individuals by Service
Respondents were asked to report the number of individuals served on a specific date of their choice that was recent and easy to report on. The CRPs reported that a total of 54,833 people were supported across employment and nonwork services by the 254 CRPs (see Table 4). Across all services and all populations served, the largest service categories were sheltered employment (28%) and facility-based nonwork (20%), followed by competitive employment and individual supported employment (12% each). Other types of integrated employment, such as enclaves, mobile work crews, transitional employment for people with mental illness, and entrepreneurship, were less populated (5%, 2%, 1%, and 0.1%, respectively). Only 66 individuals were reported as being involved in self-employment or other entrepreneurial options. Eight percent of the total served were in community-based nonwork, or less than half of those served in facility-based nonwork programs (20%).
Individuals with developmental disabilities were more likely to be in facility-based or nonwork services than individuals not identified as having a developmental disability. Of the total served by the 254 CRPs, 70% (38,298) identified clients as having developmental disabilities. The largest number of individuals with developmental disabilities was in facility-based employment, including sheltered employment and work-center-based employment (41%), followed by facility-based nonwork services (21%), indicating that facility-based programs were still the predominant service model for people with developmental disabilities. Overall, individuals with developmental disabilities participated at a lower rate in integrated work compared to other populations receiving supports from CRPs (26% vs. 45%), at a higher rate in facility-based employment (41% vs. 18%), and at a relatively equal rate in nonwork services (33% vs. 37%).
If CRP services and supports have entered Bradley's (1994) era of community membership, their emphasis should be on employment (purpose), taking place in the community (setting), and occurring on an individualized basis (number). In terms of purpose, the vast majority of CRPs (93%) provided employment services and supports, either alone or in conjunction with nonwork services, and over two thirds of the individuals they served were in work services, confirming that employment programs had a central role in the CRP system. A closer look, however, reveals that the majority of individuals participated in facility-based employment, group-supported employment, or both. Only 18% of individuals with developmental disabilities and 24% of individuals from other populations were being supported in individual integrated jobs. Individuals with developmental disabilities were more likely to be in facility-based and nonwork services, with 74% participating in facility-based work, center-based employment, or nonwork services. These findings are consistent with data collected at the agency level (Institute for Community Inclusion, 2005; Rusch & Braddock, 2004).
Although individual and group supported employment services have transitioned from a new innovation to a standard service offering, participation in integrated employment has not changed substantially since McGaughey et al.'s (1991) survey. McGaughey et al. (1996) reported that approximately 31% of all individuals supported by CRPs were in integrated employment, and in the present study 32% were reported in integrated employment. Emphasis has shifted slightly from group supported employment to individual jobs. McGaughey et al. (1996) reported that 9.8% of individuals were in group supported employment compared to 7% in the current study. There has also been a decline in participation in facility-based and work-center based employment from 41% to 34%, suggesting that the role of nonwork services has grown since 1991. In fact, McGaughey et al. reported that only 28% of individuals were in facility-based nonwork services, and .4% were in community-based nonwork services in 1991. In the current study, 34% of individuals were reported to be in nonwork services, including 8% in community-based nonwork. This trend suggests that the value for employment as a central outcome of state supported services has declined since 1991.
The use of facility-based settings was prevalent in both work and nonwork programs. Both the majority of agencies providing employment services and the majority of those providing nonwork day activities used facility-based settings. Likewise, the majority of individuals supported were in facility-based settings. The continued reliance on facility-based settings for day services is a critical sign that CRP services have yet to reach the community membership era.
Finally, even in community-based services, there was evidence of considerable use of group services. In community employment, both enclaves and mobile work crews had a strong presence, with over one third of providers providing them and close to one third of the individuals in community employment participating in those models.
Overall, the findings of the 2002–2003 survey demonstrate that the nature of CRP services has not substantively changed since McGaughey et al. (1996) conducted a similar survey over 10 years earlier. The majority of services offered and individuals served continue to be in facility-based or group-based models. Individuals with developmental disabilities, although comprising a majority of all individuals served, continue to be underrepresented in competitive employment and overrepresented in group- and facility-based settings. This lack of fundamental change indicates that despite policy and philosophical directions toward community membership, CRP services remain in the era of deinstitutionalization and community development.
Over a decade has passed since the “emerging era of community membership” was identified, and that advance apparently continues to elude CRP services. Our findings show that CRPs are in a predicament similar to deinstitutionalization, as observed by Holburn (2000): “The comparison [of data for 1967 and 1998] is unsettling because it suggests that our field is promoting a social-inclusion model of supports and services and, at the same time, maintaining a thriving segregated sector” (p. 531). The lack of change in the percentage of individuals participating in integrated employment since 1991 is particularly disturbing.
If we are striving to gain community membership for people with MR/DD, our concern lies in the effectiveness of public policy at directing resources toward employment opportunities that support community membership, strengthening the capacity of the national network of CRPs to support integrated employment outcomes and strengthening the opportunity for individuals to consider a full range of life outcomes and self-direct supports to achieve personal goals. Although the survey results presented here suggest that as a nation, resources and priorities have not been sufficiently realigned to expand employment opportunities, individual states and CRPs have been successful in expanding community employment. In FY 2001, 3 states (Connecticut, New Hampshire, and Washington) reported supporting 50% of their total day and employment service caseload in integrated employment, and 6 additional states reported more than 40%. Similarly, some CRPs have successfully shifted emphasis from facility-based services, including closing one or more facility-based programs (Butterworth, Fesko, & Ma, 2000; Murphy, Rogan, Handley, Kincaid, & Royce-Davis, 2002).
The expansion of employment options for persons with disabilities will require that federal, state, and local policies and practices are aligned such that all are viewing employment as a not only viable but also desirable outcome for persons with disabilities. Federal and state policy needs to clearly establish and measure employment as a key systems outcome, direct resources toward employment and away from other service models, and invest in the development of new leaders and a skilled workforce.
Federal policy and funding
Although employment is an established policy goal in federal initiatives, Medicaid, principally under the Home and Community Based Services (HCBS) Waivers, is the most significant source of federal support for individuals with developmental disabilities. Currently, data suggest that only 15% of individuals receiving day habilitation services under a Medicaid Home and Community Based Waiver are supported in employment (West, Hill, Revell, Kregel, & Campbell, 2002). Rizzolo, Hemp, Braddock, and Pomeranz-Essley (2004) reported that in 2002, Medicaid funding for day activity and other segregated nonemployment programs totaled $488 million, more than four times the $108 million allocated that year for HCBS Waiver funding for supported employment. Establishing a clear mandate that states include employment goals in existing and new Waivers and achieve gradually rising benchmarks for employment outcomes would establish a federal expectation for increases in employment outcomes as would establishing a uniform employment data-collection requirement across Medicaid supported programs.
State policy and practice
State MR/DD agencies that have been effective at encouraging the movement of individuals into integrated employment have clearly defined employment goals and a visible employment outcome data reporting system, strong agency leadership, interagency collaboration, ongoing training and outreach for CRPs, local control, flexibility, and a respect for innovation (Cohen, Butterworth, Gilmore, & Metzel, 2003; Novak, Rogan, Mank, & DiLeo, 2003). In a study of local MR/ DD service areas, Connolly (1999) reported that a clear statement that integrated employment is the preferred outcome for individuals, combined with outcome standards for integrated employment in contracts, predicted higher rates of employment.
As with the federal government, states can communicate priorities by requiring employment outcomes as part of the contracting process and establishing incentives in rate structures that support integrated employment over other alternatives. State priorities are also communicated by collecting and publishing employment outcome data and investing in training and technical assistance to improve employment outcomes. Several state MR/DD agencies have recently adopted employment first policies. In Washington a recent policy statement went a step further by establishing a policy on working age adults that “establishes employment supports as the primary use of employment/day program funds for working age adults” (Washington Division of Developmental Disabilities, 2004). Tennessee's employment-first policy requires that employment be the first day program option considered and that individuals who are not working in the community receive a community-based situational assessment at least once every 3 years (personal communication, director of day services for the Tennessee Division of Mental Retardation).
The expansion of nonwork services and the emergence of community-based nonwork as a service option raise a particular challenge for state agencies and suggest that states have not clearly articulated systems goals. In particular the expansion of the community-based nonwork alternative may be competing with the expansion of employment opportunities, and states need to clarify the intent and goals of those services and their relationship to the goals of community membership articulated by Bradley (1994).
This document was supported in part by Cooperative Agreement 90ND0204 from the Administration on Developmental Disabilities, Administration for Children and Families, and the U.S. Department of Health and Human Services. Points of view or opinions do not necessarily represent official Administration on Developmental Disabilities policy. The authors thank Margot Birnbaum, Jennifer Bose, Ann Downing, John Halliday, William Kiernan, and Joe Marrone for their invaluable assistance with this work. We also thank Fred Menz and the staff of the Research and Training Center on Community Rehabilitation Programs at the University of Wisconsin-Stout, who provided assistance in developing the sample used in this project. Requests for reprints should be sent to the third author.
Deborah S. Metzel, PhD, Department of Environmental, Earth, and Ocean Sciences; Heike Boeltzig, MS, John Butterworth, PhD (email@example.com); Jennifer Sullivan Sulewski, PhD; and Dana Scott Gilmore, MS, Institute for Community Inclusion, University of Massachusetts Boston, 100 Morrissey Blvd., Boston, MA 02125