System and Individual Outcomes and Their Predictors in Services and ...

INTELLECTUAL AND DEVELOPMENTAL DISABILITIES

’AAIDD

2013, Vol. 51, No. 5, 298–315

DOI: 10.1352/1934-9556-51.5.298

System and Individual Outcomes and Their Predictors in Services and Support for People with IDD Rena´ta Ticha´, Amy Hewitt, Derek Nord, and Sherri Larson

Abstract The growth and advancement of community-based services for people with intellectual and developmental disabilities (IDD) have resulted in vast changes in the long-term services and support landscape as well as in expected outcomes of service systems for service recipients. Investments in IDD research have been made to provide a deeper understanding of these outcomes and to explain them. This article summarizes outcomes and their predictors through systems and individual lenses by examining the research and findings of the Administration on Intellectual and Developmental Disabilities’ Data Projects of National Significance that address residential services, employment services, costs of services, and individual outcomes. The article also discusses challenges and debates associated with outcome-related research and poses future research questions. Key Words: Projects of National Significance; outcomes; predictors

The 2012 Research and Training Center on Community Living State-of-the-Science Conference (SOSC) focused on outcomes people with intellectual and developmental disabilities (IDD) experience in their lives in areas such as living arrangements (from institutions to one’s own home), employment (from day habilitation to meaningful work contributions), social inclusion, self-determination, and health. People with IDD want to live in and participate fully in their communities; choose the supports they need and want; exercise control in how supports are provided; have stable, skilled support providers when needed; have satisfying lives and valued social roles; and be safe and have the assistance they need to manage life’s risks and be free from exploitation (The Arc, 2013). However, the extent to which individuals experience these desired outcomes varies depending on person-referenced characteristics, such as age or level of ID, as well as family/ community-related factors , including participating in community activities. Additionally, systemsreferenced characteristics, such as type of services, and global influences (e.g., the state or region in which the person lives), which affect the funding source for supports and policies guiding supports in the system and programs through which the person

298

is supported, also influence individual outcomes and quality of life. In this paper, the term and concept of level of ID is used, rather than support needs because level of ID is referred to most often as a variable available for analysis in large datasets discussed in this paper (i.e., National Core Indicators [NCI]). While the World Health Organization and American Association on Intellectual and Developmental Disability definitions of intellectual disability focus on support need there is no congruence with variables in the data reviewed and presented in this paper. In an examination of factors related to improved quality of life and providing services and supports for people with IDD, there is often not an obvious connection between research findings and practices implemented in the field. A framework that provides such a connection can aid in implementing practices and policies that are evidence based. In their framework on quality of life, Shogren, Bradley, Gomez, Yaeger, and Schalock (2009), outline the relationship between public policy and the desired outcomes for people with IDD. Shogren et al. (2009) argued that policy and practice in the disability field have been shaped over time by various social factors, including socialpolitical movements, court rulings, and research

Predictors of Outcomes for People With IDD

INTELLECTUAL AND DEVELOPMENTAL DISABILITIES 2013, Vol. 51, No. 5, 298–315

findings, all of which lead to more policy and effective service interventions. In this model, public policy is defined as rules, regulations, statutes, and funding, while practice is defined as diagnosis, classification, design, and delivery of supports. Equally important, however, is whether the policies and practices are based on sound research evidence. Consequently, a focus on predictors of outcomes that enhance quality of life for people with IDD along with a comparison of results across different service delivery systems is critical. According to this framework, policies, practices, outcomes and their predictors can be further organized as (a) person-referenced (e.g., self-determination and rights), (b) family/community-related (e.g., family interactions), (c) systems-referenced (e.g., types of services provided), and (d) global influences, such as living in a particular region or state (Shogren et al., 2009). A framework such as the one outlined by Shogren et al. (2009) provides a helpful organizer not only of studied variables related to outcomes for people with IDD but also of the level at which policy is being implemented and is impacting its recipients. This paper will summarize the content provided at the SOSC identifying and discussing what is known about common predictors of desired outcomes for people with IDD in the context of state IDD service systems. It will also highlight areas that need continued debate, and recommendations for future research in this area will be made. One of the challenges of this field is to be able to integrate the findings of the Projects of National Significance (PNS) and related studies to effectively inform policy makers, advocates and other stakeholders. This paper reflects, to an extent, the complexity of translating data findings to be able to make comparisons between the projects. One example of this challenge are variables, such as living arrangements, that are treated as outcomes in the Residential Information Systems Project (RISP), but as an important predictor of individual outcomes in the NCI project. This paper provides a unique opportunity to highlight the focus of administration on Intellectual and Developmental Disabilities (AIDD) PNS, including the outcomes and their predictors these projects have found to make a difference in the lives of people with IDD. The outcomes of interest are residential services, employment services, costs of services, and individual outcomes, such as choice

R. Ticha´ et al.

’AAIDD DOI: 10.1352/1934-9556-51.5.298

making, loneliness, and health. The PNS described here utilize different approaches to handling and analyzing their data. Residential services, employment services, and costs of services are outcomes reported in a descriptive way by RISP at the University of Minnesota (http://rtc2.umn.edu/risp/ main/); State of the States in Developmental Disabilities Project at the University of Colorado, Boulder (http://www.stateofthestates.org/); and Access to Integrated Employment: National Data Collection on Day and Employment Services for Citizens with Developmental Disabilities at the University of Massachusetts, Boston (http://www. statedata.info/). The NCI project, a collaborative effort between the National Association of State Directors of Developmental Disabilities Services (NASDDDS) and the Human Services Research Institute (HSRI), also funded by AIDD as a PNS, collects data on individual outcomes (http://www. nationalcoreindicators.org/). In addition, University of Minnesota’s RRTC partnered with NASDDDS and HSRI to conduct inferential analysis of the data to examine relationships between person-referenced, family/community-related, systems-referenced and global predictors, and individual outcomes for people with IDD through a project funded by the National Institute on Disability Rehabilitation Research. It is not the intent of this paper to provide comprehensive overviews of the PNS, but rather to highlight the outcomes related to residential services, employment services, costs of services, and individual outcomes, such as choice making, loneliness, and health. The NCI project results are described here in more detail because they provide outcome data that are related to the topics of the SOSC and because the authors had a key role in the inferential analyses of the NCI data.

What We Know About Outcomes System Costs and Outcomes—Data PNS Studies of system level outcomes provide important contextual information to support other types of research. The AIDD Data PNS collect longitudinal systems change data that identify the places in which people live (RISP at the University of Minnesota), Medicaid and other expenditures related to IDD services (State of the States in Developmental Disabilities Project at the University of Colorado, Boulder), and where people work or spend their day (Access to Integrated Employment: National Data

299


Collection on Day and Employment Services for Citizens with Developmental Disabilities at the University of Massachusetts, Boston). The following sections present findings on outcomes from these three projects as related to residential services, employment services, and costs of services. Where people with IDD live. There has been a dramatic change in where people with IDD live in the United States since deinstitutionalization began in the late 1970s. This shift has been guided by advocacy efforts and policy. Figure 1 shows the pattern of de-institutionalization in the United States. This rapid shift from institutional to community living was due to the availability of funding through the Medicaid Home and Community Based Services (HCBS) program in the early 1990s. Institutional living represented most often a state-operated facility with 16 or more residents. Community living, in contrast, includes non–stateoperated small group residential settings, living with family, or living independently with some supports. As a result, between 1982 and 2011, the number of Medicaid HCBS recipients with IDD increased from 1,381 to 616,491. A related change is the rapid growth in HCBS to support persons with IDD who live with family. In 1994, 29,806 HCBS recipients lived with family members; in 2011 the total had increased to 314,685. Nationally, an estimated 51% of all people with IDD receiving long-term services and support lived in the home of a family member, with 97,522 living in their own home (Larson et al., 2013).

’AAIDD DOI: 10.1352/1934-9556-51.5.298

Opportunities to expand community supports have led to significant increases in the number, variety, and flexibility of living options. Although there have been many criticisms of the Medicaid ‘‘institutional bias,’’ the overwhelming majority of Medicaid-funded services and supports are provided in the community. For persons with IDD, national trends are encouraging, but great disparities still exist between states in the quantity and quality of community supports. Although 11 states have closed all of their large institutions, and nationally only 12.2% of all Medicaid ICF/DD and HCBS recipients lived in a facility with 16 or more people, in 2010, 64% of recipients in Mississippi lived in places with 16 or more people (Larson et al., 2013). Figure 2 shows recent increases in people with IDD in state IDD systems who receive support while living in a home they own or lease or with family members. An examination of the state-level data reveals dramatic differences in access to community supports by state (see Table 1). State-by-state differences in the size of places people with IDD live often explains significant proportion of variability in outcomes experienced by people with IDD. Employment outcomes and services. In its interpretation of the Olmstead decision and Section 504 of the Rehabilitation Act, the U.S. Department of Justice (2013) stated that all people with IDD, including transition-age youth, should be provided the opportunity to work, and rather than be marginalized citizens, be fully embraced by their

Figure 1. Average daily population with IDD in large state-operated IDD and psychiatric facilities 1950 to 2011 (Source, RISP: Larson et al., 2013).

300



’AAIDD DOI: 10.1352/1934-9556-51.5.298

Figure 2. Changes in living arrangements for people with IDD since 1998 (Source, RISP; Larson et al., 2013).

community. This is supported by research that suggests participating in employment needs to begin early, by providing youth in transition with work experiences in school and immediately following; young people with IDD are more likely to be employed later in life (Carter, Austin, & Trainor, 2011). Not only do these experiences lead to jobs in the community, they tend to pay more (Cimera, 2011). Yet, while the majority of people with IDD seek economic independence and employment (Migliore et al., 2007), policies and investments in employment systems face serious challenges in successfully supporting them to obtain inclusive work. Recent system outcome statistics highlight these challenges, where only 26% of young adults with IDD are employed after high school (Carter et al., 2011). In addition, as the PNS data point out using different data sources, adult employment system outcomes also vary. Using Rehabilitation Service Administration case closure data, it was found that 51% of people

R. Ticha´ et al.

with ID who receive employment services and exit state VR programs are employed. The PNS also implements The National Survey of State Intellectual and Developmental Disabilities Agencies’ Day and Employment Services Survey across state developmental disability agencies, which served as a mechanism to capture trends in day and employment service from FY1988 to FY2011. This national survey found that of the 26 states that could report employment outcomes, 27.5% of people receiving day and employment services from state developmental disability agencies were employed in integrated settings (Butterworth et al., 2013). From 1988 to 2011, the number of people with IDD receiving day and employment services from state developmental disability agencies nearly doubled from 287,860 to 570,406. As seen on Figure 3, despite the growth in service use, the percentage of people receiving integrated employment services remained stagnant in recent years

301


’AAIDD

2013, Vol. 51, No. 5, 298–315

DOI: 10.1352/1934-9556-51.5.298

Table 1 Estimated Size of Residential Settings for Persons with IDD on June 30, 2011{ Estimated

Percentage in settings with

State

Settings

Residents

Residents per setting

1–15 residents

1–6 residents

1–3 residents

Alabama Alaska Arizona Arkansas California Colorado Connecticut Delaware District of Columbia Florida Georgia Hawaii Idaho Illinois Indiana Iowa Kansas Kentucky Louisiana Maine Maine Massachusetts Michigan Minnesota Mississippi Missouri Montana Nebraska Nevada New Hampshire New Jersey New Mexico New York North Carolina North Dakota Ohio Oklahoma Oregon Pennsylvania Rhode Island South Carolina South Dakota

1,130 937 2,071 1,414 31,867 4,450 2,937 372 599 7,341 3,517 637 2,333 4,712 3,678 2,076 1,176 2,232 2,925 1,525 3,695 6,433 9,738 4,596 378 1,926 1,012 2,490 1,029 1,414 2,386 2,503 15,450 3,676 1,370 16,449 1,622 2,081 11,336 1,024 927 1,147

3,678 1,201 4,282 3,782 56,522 5,395 6,901 1,043 1,341 15,938 6,192 1,018 4,459 21,946 8,777 9,083 4,609 4,944 7,212 2,802 7,581 12,059 20,614 14,599 3,224 7,049 1,875 4,053 1,802 1,888 12,051 3,462 46,775 8,624 2,154 25,699 4,366 6,395 24,733 2,212 4,841 2,454

3.3 1.3 2.1 2.7 1.8 1.2 2.3 2.8 2.2 2.2 1.8 1.6 1.9 4.7 2.4 4.4 3.9 2.2 2.5 1.8 2.1 1.9 2.1 3.2 8.5 3.7 1.9 1.6 1.8 1.3 5.1 1.4 3.0 2.3 1.6 1.6 2.7 3.1 2.2 2.2 5.2 2.1

97 100 97 60 93 99 90 87 100 82 91 100 90 77 96 81 93 93 80 99 98 94 95 97 35 90 97 90 92 99 67 100 94 79 93 82 75 95 86 98 84 92

72 100 96 37 91 97 85 86 98 74 91 99 58 36 68 70 90 93 78 95 95 84 85 93 14 73 73 88 92 97 58 97 54 74 70 71 65 92 79 91 66 69

59 88 70 35 58 77 46 41 62 37 69 61 54 16 49 56 39 91 33 71 73 43 45 32 11 56 61 74 64 93 26 93 28 41 59 61 46 61 60 50 23 51

302



’AAIDD

2013, Vol. 51, No. 5, 298–315

DOI: 10.1352/1934-9556-51.5.298

Table 1 Continued Estimated State Tennessee Texas Utah Vermont Virginia Washington West Virginia Wisconsin Wyoming Estimated U.S. total

Percentage in settings with

Settings

Residents

Residents per setting

1–15 residents

1–6 residents

1–3 residents

2,276 8,862 1,847 1,402 2,680 3,041 1,083 7,790 427 200,019

6,056 29,009 3,379 1,638 7,973 7,044 1,854 12,723 1,286 460,597

2.7 3.3 1.8 1.2 3.0 2.3 1.7 1.6 3.0 2.3

86 84 78 100 67 85 97 94 94 88

74 82 73 100 60 83 70 78 86 75

64 52 63 95 44 53 61 56 51 49

{

This table includes estimates for states with incomplete reports. (Source, RISP; Larson et al., 2013).

(Institute for Community Inclusion, 2013). The use of integrated employment services varies greatly by state, ranging from 88% of people with IDD in Washington to 5% in Alabama, Nebraska, and Texas (Butterworth et al., 2013). Costs of services and supports. In FY2011, a total of $56.65 billion was spent in IDD services, 61% of which were paid for by way of federal revenue sources. An increasingly large funding source is Medicaid HCBS Waivers. The enrollment and expenditures related to HCBS Waivers increased rapidly, from $1.14 billion for 63,174 enrollees in FY 1992 to $18.3 billion for 627,270

enrollees in FY2011. Table 2 shows the HCBS funding across the nation averages $90 per capita, with wide variability from state-to-state, ranging from $284 per capita in New York to $14 per capita in Mississippi. Great disparity is also found in relation to investments in family support and supported living. A present, states average 20% of their IDD spending to these services. The table also highlights the progress made across the nation in community supports, where 80% of all IDD services are in the community settings with six or fewer people with IDD (Braddock et al., 2013a, 2013b, 2013c, 2013d).

Figure 3. Percentage of people in day and employment services receiving integrated employment supports across the nation (Institute for Community Inclusion, 2013).

R. Ticha´ et al.

303


’AAIDD

2013, Vol. 51, No. 5, 298–315

DOI: 10.1352/1934-9556-51.5.298

Table 2 IDD Spending and Services across the United States (Source: Braddock et al., 2013a)

State

HCBS per capita spending{

% of total IDD spending for family support and supported living{

% of total IDD spending on community services with six or fewer people§"

Alabama Alaska Arizona Arkansas California Colorado Connecticut Delaware District of Columbia Florida Georgia Hawaii Idaho Illinois Indiana Iowa Kansas Kentucky Louisiana Maine Maryland Massachusetts Michigan Minnesota Mississippi Missouri Montana Nebraska Nevada New Hampshire New Jersey New Mexico New York North Carolina North Dakota Ohio Oklahoma Oregon Pennsylvania Rhode Island South Carolina South Dakota Tennessee

$54 $156 $100 $53 $51 $65 $158 $99 $229 $45 $36 $74 $42 $44 $100 $110 $108 $74 $115 $209 $88 $100 $89 $224 $14 $70 $75 $107 $27 $141 $63 $133 $284 $53 $169 $102 $73 $150 $126 $207 $50 $110 $92

3 56 47 30 20 12 9 1 3 29 14 27 15 7 35 31 11 11 49 9 15 6 29 24 10 40 20 9 53 6 7 55 9 25 26 30 49 19 5 17 16 6 43

84 99 96 64 89 92 82 82 98 76 89 100 85 50 77 63 88 71 70 94 92 90 94 95 23 80 84 77 92 98 60 96 81 73 70 76 79 99 79 96 72 75 78

304



’AAIDD

2013, Vol. 51, No. 5, 298–315

DOI: 10.1352/1934-9556-51.5.298

Table 2 Continued

State

HCBS per capita spending{

% of total IDD spending for family support and supported living{

% of total IDD spending on community services with six or fewer people§"

$35 $48 $220 $59 $82 $135 $143 $170 $90

32 8 13 9 53 42 13 22 20

66 74 100 71 83 85 81 81 80

Texas Utah Vermont Virginia Washington West Virginia Wisconsin Wyoming United States {

Per capita denotes, per citizen of the general population. Spending for family support (children and adults in the family home) and supported living (adults in their own home) is divided by total IDD spending in each state. § Spending for community services in the states is provided by the HCBS Waiver in addition to other state and local government funding. Community services consists of a wide range of residential supports, supported living, supported employment, family supports, case management, service coordination, transportation, and other supports. Total IDD spending also includes public and private residential facilities serving 16 or more persons per setting, including ICF/ID spending (but excluding nursing facility spending) " Refers to residence, family support, supported employment, supported living, and related community day and support programs. {

Individual Outcomes and Their Predictors—NCI NCI. The NCI program is another PNS funded by AIDD (Bradley & Moseley, 2007). The NCI are standard measurement tools and processes used in 36 states (this number is expected to grow to 45 by 2014) to assess multiple outcomes of services provided to people with IDD. Unlike the other PNS, this project gathers data about and from individual service users and as necessary their legal representatives related to individual service user outcomes. The NCI-Adult Consumer Survey (NCI-ACS) was developed to measure state IDD agency performance across an array of individual, family, and systems outcomes with a goal of providing actionable information to state agency policymakers, program administrators, researchers, families, and other stakeholders to be used to benchmark progress, track performance over time, and compare outcomes across states. NCI does not address every possible element of system-wide performance. The indicators are intended to work in tandem with other performance tracking and monitoring systems that states operate. However, it

R. Ticha´ et al.

is important to note that because of the nature of the random sampling and data collection used in the NCI program, it is not possible to follow over time the same subjects, and thus there is no capacity to longitudinally monitor individual outcomes. One key criterion for the selection of NCI measures was and continues to be the extent to which each indicator makes meaningful comparisons between states possible. Comparability is accomplished through the use of common data collection protocols and through common definitions of the particular phenomenon and data source addressed by the indicator. The indicators represented by survey items are divided into several domains including: health, employment, living situation, safety, social relationships, satisfaction with services, self-direction, inclusion, choices, rights and service access. The indicators have remained consistent over the last several years and thus can be used to analyze system-level trends over time. The NCI-ACS currently generates critical information on over 12,236 randomly selected people with IDD receiving publicly supported

305


’AAIDD

2013, Vol. 51, No. 5, 298–315

DOI: 10.1352/1934-9556-51.5.298

services. The NCI-ACS is used to interview individuals receiving services and other key informants and consists of three major sections. The Background Section collects demographic, diagnostic, functional, behavioral, medical, and earnings data from individual records and is usually completed by program administrators and case managers. Section I contains subjective questions and can be completed only by interviewing the service recipient (proxies are not allowed). Section II consists of questions that are objective and may be answered by the service recipient or someone who knows the person well (a proxy). Ten studies based on the NCI-ACS dataset have been published in peer-reviewed journals since 2008 on multiple outcomes. Table 3 provides key information about these studies and the outcomes that were studied in each. Based on the framework of Shogren et al. (2009), 8 of the 10 studies have included similar person-referenced predictors or covariates, seven have included a systems-referenced predictor (i.e. type of living arrangement), while four have included a global variable (i.e., state in which the person with IDD lives). Table 4 serves as an organizer of examined predictors of individual outcomes. Ordinary least squares or ordinal or logistic regressions were used to assess the impact of predictors, as a function of the type of an outcome

variable (study on overweight and obesity by Stancliffe et al., 2011, reported descriptive results only). Person-referenced characteristics (level of ID, age, mobility, and self-report ability to communicates verbally) accounted for significant differences in studies with outcomes of everyday and support-related choice making (Lakin et al., 2008a; Ticha´ et al., 2012), overweight and being diagnosed with Down syndrome (Stancliffe et al., 2012), preventative health care (Bershadsky et al., 2012), being diagnosed with autism and having relationships and being diagnosed with autism and other disabilities (Hewitt et al., 2012), choice of living arrangement (Stancliffe et al., 2011a), obesity and being overweight (Stancliffe et al., 2011b), and variations in expenditures for individuals with different characteristics and service needs (Lakin at al., 2008b). Additional person-referenced variables such as absence of problem behavior (Ticha´ et al., 2012) or absence of psychiatric diagnosis (Lakin et al., 2008a) were also significant predictors of everyday choice making. Type of living arrangement (systems-referenced predictor), even though not conceptualized consistently in all 10 studies, was a significant predictor in seven studies of the following outcomes: everyday and support-related choice making, preventative health care, choice of living arrangement, obesity and being overweight, and

Table 3 Studies Published in Peer-Reviewed Journals Based on the NCI Adult Consumer Survey Since 2008 Date

Journal{

Topic

Ticha´ et al. Stancliffe et al.

2012 2012

IDD JARID

Bershadsky et al. Hewitt, et al. Stancliffe, Lakin, Larson, Engler, Taub, et al. Stancliffe, Lakin, Larson, Engler, Bershadsky, et al. Stancliffe et al. Stancliffe et al. Lakin et al. Lakin et al.

2012 2011 2011 2011 2010 2009 2008 2008

PHR RSAD JIDR AJIDD AAC IDD AJMR IDD

Choice making Down syndrome and overweight Preventative health care Autism Living arrangements Obesity AAC Satisfaction Choice making Expenditures

Authors{

{

Full citations of the studies are included in the reference list. IDD, Intellectual and Developmental Disabilities; JARID; PHR, Public Health Reports; RSAD, Research in Autism Spectrum Disorders; JIDR, Journal on Intellectual and Developmental Disabilities; AJIDD, American Journal on Intellectual and Developmental Disabilities; AAC, Augmentative and Alternative Communication; AJMR, American Journal of Mental Retardation. {

306



expenditures. The study on people with Down syndrome and being overweight included type of living arrangement as a predictor but found no significant differences. Table 5 provides a summary of results of the effect of different types of living arrangement on each studied outcome using the NCI dataset. Most of the studies included a small agency residential setting with 1–15 people as the referent for comparison purposes. The NCI studies examined here have defined the type of living arrangement differently. In one study (Ticha´ et al. 2012) agency setting has been further categorized by number of individuals with IDD living there. Other studies (e.g., Stancliffe, Lakin, Larson, Engler, Bershadsky, et al., 2011) have differentiated only between agency settings with 1–15 people and institutions (typically 16+ people). Yet other studies grouped all congregate/agency-operated living arrangements into one category (e.g., Lakin et al., 2008). Individuals living in their own home (that they themselves owned or rented) were significantly more likely, for example, to make everyday choices than individuals living in any other type of residential setting. The only difference in making everyday choice that was not significant was between an individual living in a small agency residential setting with one to three people and living with own and host family (Ticha´ at al., 2012) or host family only (Lakin et al., 2008). A study by Bershadsky et al. (2012) on preventative health care for people with IDD used living with family as a referent and found that individuals living in agency-operated settings (especially those serving 16 or more people) were more likely to receive preventative health care than those living with their family. Four NCI studies examining differences in outcomes by state reported that state (global predictor) explained a significant proportion of variability over and above person-referenced and systems-referenced effects in what everyday and support-related choice they made, what type of preventative health care they received, and how much was spent on their services. Table 4 further illustrates which combinations of predictors and outcomes have been examined in previous NCI studies and of those that have been examined, which have demonstrated significant relationships. For example, Table 4 shows that level of ID has been included as a predictor in all but one of the previous studies and has been found significant in all but one (wellbeing and satisfaction

R. Ticha´ et al.

’AAIDD DOI: 10.1352/1934-9556-51.5.298

outcome) of the studies that included it. Table 4 also helps to organize NCI variables (predictors) into levels at which outcomes relate to policy as outlined by Shogren et al. (2009). Table 6 outlines predictors at the levels in the framework of Shogren et al. (2009) that have been significant in previous NCI studies. Person-referenced characteristics that have been consistent predictors of outcomes for people with IDD include: level of ID, age, mobility, mode of communication, challenging behavior, and psychiatric diagnosis. Systems-referenced predictor has been the type of living arrangement and the global domain predictor has been state. As illustrated in Tables 4 and 6, although important work has been accomplished to identify common predictors of selected individual outcomes for people with IDD, there are numerous areas that remain largely unexplored. Table 7 provides an organizer of these areas in the person-referenced characteristics, family/community related, systemsreferenced, and global domain as outlined by Shogren et al. (2009).

Challenges, Debates, and Unresolved Issues The PNS cited in this paper, including RISP; State of the States in Developmental Disabilities Project; Access to Integrated Employment project; and the NCI, provide a wealth of information about where people live, work and about their outcomes related to community living, self-determination, health, etc. and the costs of these services and supports. These projects are administered and managed by separate institutes and cover their own area of focus utilizing their particular expertise. Large datasets that are the products of these projects (published yearly or every other year) have an important descriptive value and provide interesting contextual information to states about services and support for people with IDD through a systems lens. The NCI datasets with its variety of outcomes measured about individuals with IDD also provide an opportunity for inferential analysis that evaluate significant differences between subgroups of participants on a variety of outcomes and establish significant relationships between predictors and outcomes. The purpose of this paper is to provide an explicit organizational framework for moving forward in research and policy making as efficiently as possible not only by highlighting the already-made contributions but also by outlining the areas of need

307


’AAIDD

2013, Vol. 51, No. 5, 298–315

DOI: 10.1352/1934-9556-51.5.298

Table 4 The Effects of Different types of Predictors on National Core Indicator Outcomes for People with IDD

Predictors/outcomes Person-referenced Level of ID (milder) Age (older) Gender (female) Ethnicity Mobility (walks) Problem behavior (none) Mental health diagnosis (none) Autism Down syndrome Vision or hearing impairment Physical disability (none) Cerebral palsy (none) Alzheimer’s Self-report Communicates (verbally) Physically active Better health

Everyday Everyday choice choice

Support- Support- Down synrelated related drome and choice choice overweight

Preventative health care

‘ ‘ n.s X ‘ ‘ ‘ n.s X n.s X X X ‘ ‘ X X

‘ n.s n.s X ‘ n.s ‘ n.s X n.s X X X ‘ X X n.s

‘ n.s n.s X n.s n.s n.s n.s X n.s X X X ‘ ‘ X X

‘ n.s n.s X ‘ ‘ ‘ n.s X n.s X X X ‘ X X n.s

~ ~ ‘ X ~ X X X ‘ X X ‘ n.s X X ~ X

‘ ‘ n.s X ‘ n.s n.s X X X X X X X n.s X ‘

sig

sig

sig

sig

n.s

sig

sig 19 8,892 44%

sig 6 2,398 50%

sig 19 8,892 31%

sig 6 2,398 46%

X 25 12,381 11%

sig 20 11,569 18%

Systems-referenced Type of living arrangement Global domain State Number of states Number of participants Total variability explained

Note. ‘ 5 significant predictor of an increase in outcome; ~ 5 significant predictor of a decrease in outcome; sig 5 significant; n.s. 5 not significant; X 5 not examined; refer to Table 5 for more detailed information about the effect of type of living arrangement; type of living arrangement and state were dummy coded. for further work focused on person-referenced characteristics, family/community-related variables, systems-referenced factors, and global factors (Shogren et al., 2009) that affect lives of people with IDD. The following section provides a summary of some of the challenges the PNS projects face to make the information they generate of the highest quality and accessible to its readers and users. Dissemination. Given the wealth of information generated by the PNS, there is a challenge to effectively disseminate the project findings to stakeholders who use them to create, update, and implement policies for people with IDD. This

308

challenge includes (a) presentation of findings in ways that they can be easily interpreted and understood by varied target audiences including advocates, practitioners, and policy makers, (b) presence in easily accessible publications and other outlets, and (c) a possibility to easily manipulate the data as needed by the user. The primary means of dissemination of the PNS information are through technical reports and websites. Some of the PNS use social media as a dissemination tool, and one project provides a summary report specifically for self-advocates. When results of PNS are published in peerreviewed journals that require subscription or in



’AAIDD

2013, Vol. 51, No. 5, 298–315

DOI: 10.1352/1934-9556-51.5.298

Table 4 Extended Autism and Autism other and disability relationships

Choice of where to live

Choice of whom to Overweight live with or obesity

Loneliness

Wellbeing and satisfaction Expenditures

‘ ‘ ‘ X X X X ~ X X X X X X X X X

‘ ‘ ‘ X X X X ~ X X X X X X X X X

‘ n.s n.s X X X ‘ n.s X n.s ‘ n.s X ‘ n.s X X

‘ ‘ ‘ X X X ‘ n.s X n.s n.s n.s X ‘ ‘ X X

‘ X ‘ n.s X X X X ‘ X X X X X X X X

X n.s X X X X X X X X X X X X n.s X X

n.s n.s n.s X X n.s ‘ n.s X X X X X X X X ‘

~ ‘ n.s X ~ n.s ‘ ‘ X n.s X X X X X X ~

X

X

sig

sig

sig

n.s

sig

sig

X 25 12,382 6%

X 25 12,382 16%

X 26 6,778 31%

X 26 6,778 26%

X 20 8,911 NA

X 25 12,766 12%

X 6 1,885 7%

sig 4 1,421 56%

publications of limited circulation or through dense technical reports, accessibility by both the professional and consumer community to these findings is restricted. The less accessible these publications are to the community, the less understanding, awareness, and dialogue are generated between policy and the most current findings. The PNS system is expanding its reach by adapting the dissemination to respond to the wider audience needs, through briefs, social media, and web-based report building features to facilitate more effective advocacy. Accessibility and ease of results interpretation can also be enhanced by the use of data display technologies (e.g., Tableau software, Spotfire, and Domo). Such methods enable the user to display figures and trends customized to their needs and

R. Ticha´ et al.

interests, while allowing for users to drill down to answer research questions of their own. Psychometric characteristics of measures. To collect data of high quality and relevant to the time, place, and situation to people with IDD and their support providers, it is necessary to work with data collection instruments that have good psychometric properties, including interobserver reliability, internal consistency, differential item functioning, social, content and construct validity, and sensitivity to change over time. It can be challenging to keep revising data collection instruments because such changes affect the ability to make trend comparisons. Instrumentation development and refinement are, however, an important aspect of an efficient and effective study of the impact of

309

310

2012 2012 2011 2011 2011 2010 2009 2008 2008

Stancliffe et al. Bershadsky et al. Hewitt et al.

Stancliffe, Lakin, Larson et al.

Stancliffe, Lakin, Larson et al. Stancliffe et al. Stancliffe et al. Lakin et al.

Lakin et al.

Everyday choice Support-related choice Down syndrome and health Preventative health care Other disability Relationships Choice of where to live Choice of who to live with Overweight or obesity Loneliness Wellbeing and satisfaction Everyday choice Support-related choice Expenditures

Outcome

‘ ‘ ‘

‘ ‘ ~

‘ ‘

~

‘ X

X X

Ref

‘

‘ ~

‘ ‘

X

Ref Ref Ref

Ref Ref Ref

Ref Ref Ref ‘

~

~

~ ~ ‘

‘

~ ~

Own Host Agency 1–3 Agency 4–6 Agency 7–15 Agency 16+ home Family family residents residents residents residents

Note. Ref 5 referent; ‘ 5 outcome is better in that setting; ~ 5 outcome is worse in that setting; ~ 5 significant for nursing facilities; X 5 not examined.

2012

Date

Ticha´ et al.

Authors

Type of living arrangement

Table 5 The Effect of Different Types of Living Arrangement on National Core Indicator Outcomes for People with IDD


2013, Vol. 51, No. 5, 298–315 DOI: 10.1352/1934-9556-51.5.298

’AAIDD



’AAIDD

2013, Vol. 51, No. 5, 298–315

DOI: 10.1352/1934-9556-51.5.298

Table 6 Predictors Found Significant in Previous NCI Studies Domain Person-referenced

Systems-referenced Global domain

Predictors Level of ID Age Mobility Mode of communication (self-report) Challenging behavior Psychiatric diagnosis Type of living arrangement State

predictors on outcomes. Here is an example of a need for an instrument refinement with respect to internal consistency of survey items, which directly relates to its content validity. In the NCI survey there is a section on rights of people with IDD that includes four questions; for example, Do people read your mail or email without asking you first? or Have you ever participated in a self-advocacy group meeting, conference, or event? In the survey, however, there are other questions that are related to rights not included in this rights section of the survey; for example, Do you choose what you buy with your spending money? in the section on choices. Consequently, when analyzing and interpreting the survey data, researchers need to understand and identify the underlying constructs within the survey

Table 7 Topic Areas in Need of Further Study by Domain Outlined by Shogren et al. (2009) Domain Person-referenced

Needing further study

NCI examples

Health/preventative health care Rights-human and legal

Impact of health and preventative health care Right to privacy Access to services Having paid employment Having sick time/paid vacation Number of hours worked Type of job Wages Work satisfaction Satisfaction of services Satisfaction of living arrangements Feeling safe Having friends Having an intimate relationship Having control over finances Having control over where you work Having control over where you live Able to see family Living with family Having employment or participating in other activities Having transportation Participating in self-advocacy group Payer of employment or day supports Types of services receiving Satisfaction with services Living/working in a location under specific jurisdiction

Material well-being

Life satisfaction

Emotional well-being

Self-determination

Family/community-related

Family Employment/daily activities Community

Systems-referenced

Support agency of program Services

Global domain

State or region

R. Ticha´ et al.

311


using the actual data via statistical techniques, such as factor analysis or structural equation modeling. Conceptualization of variables. An additional topic of debate related not only to instrumentation but also to data interpretation and its application to policy is the conceptualization of predictor and outcome variables, so that results can be compared across studies. Within the NCI publications, for example, different researchers (who work together) have been using a different conceptualization of residential settings. Stancliffe et al. in their research studies (e.g., Stancliffe, Lakin, Larson, Engler, Bershadsky, et al., 2011) have categorized living arrangements for people with IDD into institution, group home, agency apartment, own home, family home, and host/foster home. Other studies categorized living arrangements for people with IDD by type and their size but did not distinguish between agency-operated group home and agency-operated apartments of the same size (e.g., Ticha´ et al., 2012). Such differences in the conceptualization of variables are a hindrance in comparing results across studies for studying the effect of predictors on outcomes and in applying such information to creating policies for people with IDD. Data analysis sophistication. Only one published study using the NCI data set has linked state policy to individual outcomes findings (Hewitt et al., 2011). The PNS provide a longitudinal look at important systems level indicators that could be successfully used as predictors of outcomes using the NCI data. However, for answering research questions and questions about policy for people with IDD related to outcomes and their predictors, there are different levels of analyses that are required; for example, individual, residential setting, and state levels. A challenge in the field of IDD research and policy work is to keep educating its workforce on the most appropriate and sophisticated methodologies and data analytic techniques to be able to interpret and explain the meaning of the collected data pertaining to each question (e.g., hierarchical linear modeling). Another challenge is the linking of data sets. For example, PNS data sets contain information about individual quality of life outcomes and statelevel service investments and outputs. By better aligning data sets, researchers can more effectively investigate the relationships between costs and services to individual outcomes. These challenges apply to teams working on PNS as well as researchers affiliated with universities and other institutions

312

’AAIDD DOI: 10.1352/1934-9556-51.5.298

involved with studying outcomes and their predictors for people with IDD. Similarly, the research teams need to have the capacity to evaluate the psychometric properties of their instruments. Need for a common framework. A common framework that organizes variables and their constructs at person-referenced, family/community-related, systems-referenced, and global domain levels is needed across PNS and other projects investigating predictors of outcomes for people with IDD to allow for a more systematic and coordinated effort (Shogren et al., 2009). Such a framework would provide an organizer for variables and their constructs that have and have not yet been addressed (or not in sufficient detail) as related to the level of each of the variables; that is, person-referenced, family/community-related, systems-referenced, and global domain (Shogren et al., 2009). The vision for this paper is to bring to attention the unique as well as overarching contributions of the PNS and their findings to readers in varied areas of work related to improving the outcomes for people with IDD. By having a summary of outcomes and their predictors that need further exploration organized by level of analysis, readers who do future research in this area are more likely to systematically select those topics for their work that have been outlined. Similarly, policy makers can use this framework as a guide for making decisions based on the richness or the lack of evidence in each of the domains affecting lives of people with IDD.

Next Research Questions Even though a significant amount of research has been conducted on changes in inclusive practices for individuals IDD in the areas of living arrangements, employment, community inclusion, and supports, through the work of PNS using large surveybased datasets, many questions remain unanswered. Ongoing research is needed to further explore which predictors have a significant impact on outcomes at person-referenced, family/community-related, systems-referenced, and global levels. As guided by NCI, a Project of National Significance that is an example of having utilized inferential analyses to establish relations between different types of predictors of outcomes for people with IDD, Table 7 provides suggested topic areas that need further investigation either as predictors, outcomes, or both. For example, at the person-referenced level, additional work is needed in the area of employment of



people with IDD. More specifically, what personreferenced, family/community-related, systems-referenced, and global domain characteristics of people with IDD predict having paid employment, having sick time/paid vacation, the number of hours worked, type of job, and wages. Conversely, further exploration is needed to determine how having paid employment, having sick time/paid vacation, the number of hours worked, type of job, and wages predict other quality of life outcomes of people with IDD. Additional areas in need of further exploration include but are not limited to person’s rights and emotional wellbeing, family involvement, impact of different services as well as the specific characteristics of states or regions on individual outcomes. As further work is planned and conducted with the goal to explore the predictors of outcomes for people with IDD, one needs to consider the nature of the predictor categories discussed in this paper and outlined by Shogren et al. (2009). Personreferenced predictors, such as level of ID, age, or even having a physical disability, that have been studied with more frequency than predictors at other levels are the least likely to be changed as a result of an intervention. Although there is a need for additional investigation of predictors at the person-referenced level that can be affected with an intervention (e.g., feeling safe or having control over finances) studies are particularly needed that focus on predictors at the family/communityrelated, systems-referenced, and global levels. For example, knowing how availability, frequency, and patterns of local, regional, or state transportation tend to affect material well-being and life satisfaction of people with IDD would be of value.

Conclusion At this point in the scientific endeavor of improving the quality of lives of people with IDD by uncovering the patterns of outcomes over time and exploring the effectiveness of predictors on outcomes, the SOSC highlighted the necessity for better synergy between PNS, policy making, and implementation. It is apparent that even though the PNS have played a critical role in informing the field about the most recent trends in residential services, employment services, costs of services, and individual outcomes for people with IDD, there is room for improvement. The utilization of available resources and findings can be maximized in order for research, policy making, and advocacy communities

R. Ticha´ et al.

’AAIDD DOI: 10.1352/1934-9556-51.5.298

to become more efficient and effective in serving people with IDD nationwide by implementing a framework for organizing the results of PNS, such as the one outlined by Shogren et al. (2009), training researchers in cutting-edge descriptive and analytic methods, improving dissemination strategies, as well as increasing communication between researchers, trainers, research and training centers, and policy makers.

References Anderson, L., Larson, S. A., Lakin, K. C., & Kwak, N. (2002). Children with disabilities and social roles: An analysis of the 94/95 NHIS-D. DD Data Brief, 4 (1). Minneapolis: University of Minnesota, Research and Training Center on Community Living. Bershadsky, J., Taub, S., Bradley, V., Engler, J., Moseley, C., Lakin, K. C., Stancliffe, R., Larson, S., Ticha´, R., & Bailey, C. (2012). Place of residence and preventive health care for developmental disabilities service recipients in 20 States. Public Health Reports, 127(5), 475–485. Braddock, D., Hemp, R., Rizzolo, M. C., Tanis, E. S., Haffer, L., Lulinski, A., & Wu, J. (2013a). The State of the States in Developmental Disabilities 2013: The Great Recession and Its Aftermath (Preliminary Edition). Boulder: Department of Psychiatry and Coleman Institute, University of Colorado and Department of Disability and Human Development, University of Illinois at Chicago. Braddock, D., Hemp, R., Rizzolo, M. C., Tanis, E. S., Haffer, L., Lulinski, A., & Wu, J. (2013b). Home and Community Base Services (HCBS) waiver federal-state spending per capita. Retrieved from http://www.stateofthestates.org/index.php/ intellectualdevelopmental-disabilities/interactivecharts#chartControls Braddock, D., Hemp, R., Rizzolo, M. C., Tanis, E. S., Haffer, L., Lulinski, A., & Wu, J. (2013c). Sixor-fewer person community spending as a percent of total I/DD spending. Retrieved from http://www. stateofthestates.org/index.php/intellectualdevelop mental-disabilities/interactive-charts#chartControls Braddock, D., Hemp, R., Rizzolo, M. C., Tanis, E. S., Haffer, L., Lulinski, A., & Wu, J. (2013d). Public spending for family support and supported

313


living as a percentage of total I/DD. Retrieved from http://www.stateofthestates.org/index.php/ intellectualdevelopmental-disabilities/interactivecharts#chartControls Bradley, V. J., & Moseley, C. (2007). National Core Indicators: Ten years of collaborative performance measurement. Intellectual and Developmental Disabilities, 45(5), 354–358. Butterworth, J., Hall, A., Smith, F., Migliore, A., Winsor, J., Domin, D., & Sulewski, J. (2013). State data: The national report on employment services and outcomes 2012. Boston: University of Massachusetts Boston, Institute for Community Inclusion. Carter, E., Austin, D., & Trainor, A. (2011). Predictors of postschool employment outcomes for young adults with severe disabilities. Journal of Disability Policy Studies, 20, 1–14. Cimera, R. E. (2011). Does being in sheltered workshops improve the employment outcomes of supported employees with intellectual disabilities? Journal of Vocational Rehabilitation, 35, 21–27. Dane, A. V., & Schneider, B. H. (1998). Program integrity in primary and early secondary prevention: Are implementation effects out of control? Clinical Psychology Review, 18, 23– 45. Hewitt, A. S., Stancliffe, R. J., Johnson Sirek, A., Hall-Lande, J., Taub, S., Engler, J., Bershadsky, J., Fortune, J., & Moseley, C. (2011). Characteristics of adults with autism spectrum disorder who use adult developmental disability services: Results from 25 US states. Research in Autism Spectrum Disorders, 6(2), 741–751. Human Services Research Institute & The National Association of State Directors of Developmental Disabilities Services. (2012). Consumer outcomes: Final report: 2010–2011 NCI adult consumer data. Cambridge, MA: Author. Institute for Community Inclusion. (2013). StateData.info. Retrieved from http://www.statedata. info. Lakin, K. C., Doljanac, R., Byun, S., Stancliffe, R. J., Taub, S., & Chiri, G. (2008). Choicemaking among Medicaid HCBS and ICF/MR recipients in six states. American Journal of Mental Retardation, 113(5), 325–342. Lakin, K. C., Doljanac, R., Byun, S., Stancliffe, R. J., Taub, S., & Chiri, G. (2008). Factors associated with expenditures for Medicaid home and community based services (HCBS)

314

’AAIDD DOI: 10.1352/1934-9556-51.5.298

and intermediate care facilities for persons with mental retardation (ICF/MR) services for persons with intellectual and developmental disabilities. Intellectual and Developmental Disabilities, 46(3), 200–214. Lakin, K. C., Gardner, J., Larson, S. A., & Wheeler, B. Y. (2005). Access to support for community lives, homes and social roles. In K. C. Lakin & A. P. Turnbull (Eds.), National goals and research for persons with intellectual and developmental disabilities (pp. 179–215). Washington, DC: American Association on Mental Retardation. Lakin, K. C., Larson, S. A., & Kim, S. (2011). Behavioral outcomes of deinstitutionalization for people with intellectual disabilities: Third decennial review of US studies, 1977–2010. Policy Research Brief, 21 (2). Minneapolis: University of Minnesota, Institute on Community Integration. Larson, S. A., Lakin, K. C., & Hill, S. L. (2012). Behavioral outcomes of moving from institutional to community living for people with intellectual and developmental disabilities: U.S. studies from 1977 to 2010. Research and Practice for Persons with Severe Disabilities, 37(4), 1–12. Larson, S. A., Salmi, P., Smith, D., Anderson, L., & Hewitt, A. S. (2013). Residential Services for persons with intellectual or developmental disabilities: Status and trends through 2011. Minneapolis: University of Minnesota, Research and Training Center on Community Living, Institute on Community Integration. Larson, S. A., Salmi, P., Smith, D., Wuorio, A., & Webster, A. (2012). Residential services for persons with developmental disabilities: Status and trends through 2010. Minneapolis: Research and Training Center on Community Living, University of Minnesota. Migliore, A., Mank, D., Grossi, T., & Rogan, P. (2007). Integrated employment or sheltered workshops: Preferences of adults with intellectual disabilities, their families, and staff. Journal of Vocational Rehabilitation, 26, 5–19. Qian, X., Larson, S., Ticha´, R. Stancliffe, R., & Wuorio, A. (2013). Predicting levels of engagement of individuals with intellectual and developmental disabilities residents who live in community group homes. Manuscript in preparation.



Shogren, K. A., Bradley, V. J., Gomez, S. C., Yeager, M. H., & Schalock, R. L. (2009). Public policy and the enhancement of desired outcomes for persons with intellectual disability. Intellectual and Developmental Disabilities, 47(4), 307–319. Stancliffe, R. J., Lakin, K. C., Larson, S. A., Engler, J., Bershadsky, J., Taub, S., Fortune, J., & Ticha´, R. (2011). Overweight and obesity among adults with intellectual disabilities who use intellectual disability/developmental disability services in 20 U.S. states. American Journal on Intellectual and Developmental Disabilities, 116(6), 401–418. DOI: 10.1352/19447558-116.6.401 Stancliffe, R. J., Lakin, K. C., Taub, S., Chiri, G., & Byun, S. (2009). Satisfaction and sense of wellbeing among Medicaid ICF/MR and HCBS recipients in six states. Intellectual and Developmental Disabilities (formerly Mental Retardation), 47(2), 63–83. Stancliffe, R. J., Larson, S. A., Auerbach, K., Engler, J., Taub, S., & Lakin, K. C. (2010). Individuals with intellectual disabilities and augmentative and alternative communication: Analysis of survey data on uptake of aided AAC, and loneliness experiences. Augmentative and Alternative Communication, 26(2), 87–96. Stancliffe, R.J., Lakin, K.C., Larson, S.A., Engler, J., Taub, S., & Fortune, J. (2011). Choice of living arrangements. Journal of Intellectual Disability Research, 55, 746–762. Stancliffe, R. J., Lakin, K. C., Larson, S. A., Engler, J., Taub, S., Fortune, J., & Bershadsky, J. (2012). Demographic characteristics, health conditions and residential service use in adults with Down syndrome compared to

R. Ticha´ et al.

’AAIDD DOI: 10.1352/1934-9556-51.5.298

adults with other intellectual or developmental disabilities in twenty-six US states. Intellectual and Developmental Disabilities, 50(2), 92–108. The Arc of the United States. (2013). FINDS Survey – Family and individual needs for disability support. Retrieved from http://www.thearc.org/ researchreport Ticha´, R., Lakin, K. C., Larson, S. A., Stancliffe, R., Taub, S., Engler, J., Bershadsky, J., & Moseley, C. (2012). Correlates of everyday choice and support related choice for 8,892 randomly selected adults with intellectual and developmental disabilities in 19 states. Intellectual and Developmental Disabilities, 50(6), 486–504. U.S. Department of Justice. (2013). Interim settlement agreement—Providence, Rhode Island. Retrieved from http://www.ada.gov/olmstead/ documents/ri-providence-interim-settlement.pdf This article was supported in part by Grant H133B080005 from the National Institute on Disability and Rehabilitation Research, U.S. Department of Education to the Research and Training Center on Community Living at the University of Minnesota, and through cooperative agreements from the Administration on Community Living, U.S. Department of Health and Human Services #90DN0291-01 and #90DN0297.

Authors: Rena´ta Ticha´ (e-mail: [email protected]), Research and Training Center on Community Living, University of Minnesota, 150 Pillsbury Dr. SE, Minneapolis, MN 55455, USA; Amy Hewitt, Derek Nord, and Sherri Larson, University of Minnesota.

315