## Abstract

Self-directed home and community based services (HCBS) waiver services and supports for people with intellectual and developmental disabilities (IDD) have become a viable and widely used method of service provision in the United States. Grounded in theories of self-determination, previous literature on self-direction has suggested high satisfaction and positive outcomes for people who use self-directed programs as well as cost savings for state IDD service systems. This study explored the ways in which state IDD service administrators think about how self-direction may be used as a method of achieving cost savings while providing opportunities for people with IDD and their families to exercise choice and control. Informed by 54 high-level IDD service administrators in 34 states, and guided by a thematic analysis approach to data interpretation, the study found evidence that administrators typically see strong potential for self-direction to have cost-savings benefits, while also fostering choice. In the current political climate, the need for cautious fiscal stewardship may become a stronger driving force behind self-direction for people with IDD in the United States.

Self-directed supports and services (sometimes called participant-directed or consumer-directed) are a series of approaches to the delivery of services to people with intellectual and developmental disabilities (IDD) that place increased choice and control with the person who receives the long term services and supports (LTSS) and their family, compared with traditional agency-based LTSS service models. Although there is not a standard definition of self-direction, most authors have suggested that budgetary authority and employer authority are core elements (DeCarlo, Hall-Lande, Bogenschutz, & Hewitt, 2017; Gross, Blue-Banning, Turnbull, & Francis, 2015; Murphy, Selkow, Crisp, & Mahoney, 2012; Sciegaj et al., 2016). Budgetary authority refers to the ability of a user of self-directed supports to choose which supports and services they will use, usually within the confines of a menu of goods and services that have been approved by a particular state developmental disability or Medicaid agency. Employer authority speaks to the LTSS user's ability to recruit, hire, train, manage, and terminate their own support workers, also within guidelines that vary from state to state.

Since their emergence in the early 1990s, self-directed supports and services for have been brought to scale as one of the preferred methods of support for people with IDD and their families. Expansion of self-directed services and supports has proceeded at a rapid pace across states. As of 2015, 42 states offered people with IDD the opportunity to self-direct their long term services and supports (DeCarlo et al., 2017), an expansion of 310% over the 13 states that offered statewide self-direction for people with IDD in a study with similar methodology conducted just six years before (Walker, Hewitt, Bogenschutz, & Hall-Lande, 2009). Despite this expansion in the number of states that offer self-direction, the percentage of HCBS waiver users with IDD that states target for use of self-direction has remained relatively low. As of 2015, 63% of HCBS waivers that offer self-direction targeted between 1 and 20 percent of their total users for inclusion in self-directed programs (Friedman, 2018).

Self-direction is backed and promoted by the federal government in several important policies. In 2000, the New Freedom Initiative pushed states to develop more community-based LTSS that promoted choice and community integration (Heller, Arnold, van Heumen, McBride, & Factor, 2012). In 2001, the Centers for Medicare and Medicaid Services (CMS) added the option to include self-direction in states' 1915(c) waiver applications (Sciegaj et al., 2016). Additional waiver authorities were created to support self-direction in the mid-2000s, including the 1915(j), which enabled self-direction in state plan options (Sciegaj et al, 2016). In 2010, the Patient Protection and Affordable Care Act (2010) expanded the Medicaid funding authorities that states can use to support self-directed programs for people with IDD, including the Community First Choice 1915(k) waiver authority. Taken together, these policy changes have signaled a consistent federal commitment to the development and expansion of self-direction.

### Models of Self-Direction

Though self-direction is often treated as a single approach to the provision of services for people with IDD, it is important to note that there is significant variation within states' models of self-direction. Despite the variation in self-direction models across states, all models aim to support personal decision making and individual control in services and supports. In practice, self-direction is not one single approach, but a range of models with varying degrees of individual decision making and control. Some of the most notable variation exists within employer authority, where two models are currently dominant: the fiscal agent model and the agency with choice model (Murphy et al., 2012; Scherzer, Wong, & Newcomer, 2007).

In the fiscal agent model, a person with IDD holds responsibility for most employer functions for their paid direct support workers, including hiring, setting wages and benefits, training, scheduling, and terminating employment. They contract with a service provider that typically assists with management of the self-directed budget, including providing oversight to make sure budget caps are not exceeded, and complete payroll processing including tax withholding (Murphy et al., 2012; Scherzer et al., 2007). In the fiscal agent model, people with IDD maintain the ultimate authority and self-determination as an employer and they rely on the fiscal entity only for support in processing financial matters (Murphy et al., 2012; National Resource Center for Participant Directed Services, n.d.).

In contrast, people with IDD who use an agency with choice model of self-directed supports may have less authority as an employer, but also have to bear less of the responsibility and liability that are inherent in an employer arrangement. In agency with choice models, a person with IDD and a service provider are co-employers for direct support workers. The worker is officially an employee of the service provider, and is therefore subject to the wages, fringe benefit offerings, and policies of the provider. The person with IDD may recruit, schedule, and dismiss workers as they prefer, but only the provider has the authority to hire and terminate employment (Murphy, et al., 2012; National Resource Center for Participant Directed Services, n.d.; Scherzer et al., 2007). Agency with choice approaches to self-direction tend to be more expensive than fiscal agent models, but there are potential advantages in being able to provide fringe benefits (such as paid leave and medical insurance) more efficiently (National Resource Center for Participant Directed Services, n.d.).

In addition to differences in employer authority, there is substantial state variation in budgetary authority from one state to another. While some states allow for expenditure of self-directed funds on a wide array of goods and services, other states permit spending on only a narrowly pre-defined menu of supports (DeCarlo et al., 2017). Since Medicaid is a primary funder of self-directed supports offered through LTSS in most states, the majority of approved expenditures must be reimbursable under Medicaid policies and guidelines (Sciegaj et al., 2016).

### Self-Determination Philosophical Orientation

Self-direction is rooted in theories of self-determination, which hold that people with IDD should be able to exercise choice and control in matters they deem to be important, with the support and assistance necessary. As a leading theorist on self-determination among people with IDD, Wehmeyer has written extensively about the importance of creating systems that enable people with IDD to exercise volition and to be an active agent in determining one's own life (Wehmeyer, 2005; Wehmeyer, Shogren, Little, & Lopez, 2017). While defining features of self-determination vary somewhat from one author to another, most contain an element of socio-ecological context that is particularly important to self-direction, as the ability to exert self-determination may be fostered or constrained by other people, policies, or program rules, or even the desire or ability of a person with IDD to exercise volition in a particular setting (Abery & Stancliffe, 2003; Shogren, 2016).

Self-direction's roots in self-determination are evident in the demonstration projects that began to bring self-direction to scale beginning in the mid-1990s, which were funded by the Robert Wood Johnson Foundation as self-determination demonstration projects (Bogenschutz, Hewitt, Hall-Lande, & LaLiberte, 2010; Sciegaj et al., 2012). Since that time, the potential for increased self-determination has been lauded as a benefit of self-directed supports by policymakers and people implementing policy (Bogenschutz et al., 2010, DeCarlo, 2016; Hall-Lande, Hewitt, Bogenschutz, & LaLiberte, 2012; Walker et al., 2009), as well as people with IDD and their families (Heller et al., 2012; Sciegaj et al., 2016).

### Fiscal Stewardship in Self-Direction

Perspectives on the potential cost-saving benefits of self-direction when compared with agency-directed services have also emerged in the literature over the past two decades. As advocates of self-directed supports cite philosophical themes of self-direction, policymakers and program administrators often look towards potential cost savings for such programs (Hall-Lande et al., 2012; Walker et al., 2009). In an early study of self-direction, Doty and colleagues (1996) were among the first to note that program administrators expressed a preference for self-direction because of the potential cost savings (Doty, Kasper, & Litvak, 1996). More recently, IDD service administrators at both the state (DeCarlo, Bogenschutz, Hall-Lande, & Hewitt, 2018; Walker et al., 2009) and local (Hall-Lande et al., 2012) levels have echoed the sentiment that self-direction is desirable, in part, because budget methodologies require that self-direction be less expensive than agency-based supports.

Research has indicated that, while budgetary models are partly to credit for the inherent cost savings of self-directed supports compared to agency-based supports, many administrators also credit people with IDD and their families for being cautious stewards of public funds. For instance, Hall-Lande and colleagues (2012) found that county-level IDD service administrators in Minnesota credited people who used the state's self-directed support system for spending funds wisely. Other research has highlighted additional potential contributing factors to cost savings such as lower overhead administrative costs and shifting responsibilities to unpaid caregivers, such as friends and family members (Doty et al, 1996).

### Outcomes in Self-Direction

To date, the most robust evaluation regarding self-directed LTSS came from the Cash and Counseling demonstration projects in New Jersey, Arkansas, and Florida, which piloted self-direction for people with physical disabilities and older adults. The Cash and Counseling evaluations found higher levels of satisfaction among individuals who used self-direction and health and safety outcomes that were at least equivalent to people who used professionally-managed supports (Foster, Brown, Phillips, Schore, & Carlson, 2003), findings that were also present in evaluations outside of the Cash and Counseling demonstrations (Doty et al., 1996; Benjamin & Mathias, 2001).

Specific to people with IDD, there have been relatively few studies to examine the experiences of individuals who use self-direction. Heller and Caldwell (2005) found that participation in self-directed services resulted in a reduction in institutional placements for people with IDD, compared to individuals who used agency-managed services. In addition, both employment and community participation were found to increase among users of self-direction, as compared to people with IDD who did not use a self-directed service delivery option (Caldwell & Heller, 2007).

Regardless of the population, the ability to directly hire and manage direct support workers is often reported as one of the most important aspects of self-direction from the perspective of both service users and administrators. People with IDD have reported high satisfaction with the ability to hire friends and family members as direct support workers (Bogenschutz et al., 2010; Heller et al., 2012), a perspective that was shared in early evaluations of self-directed supports for older adults (Matthias & Benjamin, 2008). IDD service administrators also noted the benefits of self-direction on the provision of direct support, specifically suggesting that having family and friends as direct support workers may help to improve the cultural match between a person with IDD and their worker and that worker stability and quality improved under self-direction (Hall-Lande et al., 2012).

### Research Questions

Although self-directed services and supports for people with IDD have continued to expand across states at a fairly rapid rate in recent years (DeCarlo et al., 2017), research on the implementation of such services has been limited. This study is the first in nearly a decade to seek perspectives directly from the state-level IDD program administrators who are charged with implementation of self-directed services and supports throughout the country. As such, the study sought to answer three core questions: (1) How do state-level IDD program administrators perceive the balance between fostering self-determination and stewardship of public funds in self-direction? (2) What are the primary drivers behind the expansion of self-direction for people with IDD? and (3) How do administrators perceive the roles and responsibilities of people with IDD and their families as participants in self-directed services and supports?

## Method

This study utilized qualitative methods to gain an understanding of how state administrators of IDD services conceive of their role as fiscal stewards and promoters of self-determination in self-directed supports and services. All methods used to conduct this study were reviewed and approved by the Institutional Review Boards at each of the researcher's affiliated universities. All subjects participated voluntarily after approved informed consent procedures were followed.

### Data Collection

Data were collected over a six-month period in late 2015 and early 2016. The timing of data collection is significant because it took place after the introduction of the CMS Final Rule on Home and Community Based Services (CMS, 2014), which contained guidance that required many states to reshape the Medicaid waiver programs that most often support self-direction. While states continue to be in the process of implementing this guidance, the introduction of the Final Rule was a significant event that may have helped to shape participants' responses.

#### Interview protocol

A semi-structured interview protocol was developed based on investigation of the literature and a previous study on county-level IDD directors' perceptions of self-directed supports (authors redacted for peer review). The instrument contained a total of 16 questions, each of which had a set of potential follow-up questions that the interviewer would ask as needed to seek deeper information or clarification. Several of the questions sought general information about the state's self-directed programs, such as eligibility rules and program components. Other questions were designed to elicit deeper responses. These questions included “How do you balance the goal of fiscal stewardship with the goal of self-determination for individuals with IDD?” and “In your dream world, how would self-directed supports fit within the broader context of HCBS services for individuals with IDD?”

An initial draft of the interview protocol was developed by members of the research team. That draft was shared with three colleagues in other universities and by personnel at the National Association of State Directors of Developmental Disabilities Services (NASDDDS), who provided feedback for revisions. The revised interview tool was then piloted with a group of five state-level administrators. Based on that pilot, minor revisions were made to clarify question wording and the interview protocol was finalized.

#### Interview procedure

The research team conducted an initial document scan of all approved Medicaid waivers that were available on the CMS website to identify states that offered self-directed supports and services to people with IDD or autism through any Medicaid waiver program. In addition, the researchers searched state IDD services websites to corroborate findings and gather further information about each state's self-directed programs before the initiation of the interview stage of the study. The final document scan revealed 42 states that offered self-directed supports to people with IDD in a total of 102 waivers and state plans.

Following the document review, the researchers contacted IDD service administrators in each state that offered self-directed options for people with IDD. A description of the study, a copy of the interview protocol, and informed consent documents were sent to potential participants via email, along with a request for their participation. Potential participants contacted one member of the research team to express interested in participation, and a time to conduct an interview was agreed via email.

All interviews were conducted by the same member of the research team via phone. Telephone interviews are time-efficient and allow participants to feel at ease during the interview since they may be in a familiar space (McCoyd & Kerson, 2006; Novick, 2008), though there are disadvantages in not being able to gather contextual information from participants' gestures and settings (Garbett & McCormick, 2001; Novick, 2008). At the start of each call, the researcher reviewed informed consent documents with the participant before proceeding through the interview protocol. With each participant's verbal consent, interviews were digitally audio recorded and later transcribed verbatim. The transcripts became the texts for analysis. Interviews typically lasted for one hour, with a range between 25 minutes and 90 minutes.

### Participants

The sample for this study was taken from a publicly-available list of state IDD directors and waiver managers on the website of NASDDDS. Administrators in 34 of the 42 states that offered self-directed services to people with IDD participated in the study, representing all regions of the country (seven states in the Northeast, nine states in the Southeast, 10 states in the Midwest, and eight states in the West). In most instances, interviews were conducted one-on-one, although a few states included more than one individual on the call (e.g. administrators for multiple waivers, disability services administrator and Medicaid agency administrator, for instance). Interviews were conducted one-on-one in 18 states, with two administrators in eight states, and with three individuals from eight states, for 54 participants in total.

Participants represented all regions of the United States and a wide variety of experience with self-direction. Some states had long-established self-directed programs for people with IDD, while others had implemented programs more recently. Some states had large programs, while other states enrolled a modest amount of people in self-direction. Program rules varied from state to state. Taken together, the diversity of participants was representative of the varied experiences of states that provide self-directed options to people with IDD and their families.

### Data Analysis

The researchers used a thematic analysis approach to interpret the interview data, since the thematic analysis approach provides a guideline for how to inductively approach data, while allowing the analyst flexibility to pursue themes that emerge from texts (Braun & Clarke, 2006; Miles, Huberman, & Saldana, 2014). Specifically, Braun and Clarke's (2006) suggested approach to conducting thematic analysis was used to drive our analysis.

Analysis was conducted in Atlas.ti 7.5 (2014). Following Braun and Clarke's (2006) model, the thematic analysis began with one member of the research team deeply reading the transcripts before proceeding with the process of developing in-vivo codes that emerged from the data in a first round of analysis. The initial codes were reviewed with other members of the research team and together the researchers searched for themes in the coding scheme in a second round of coding and subsequent discussion. The themes were then reviewed against the original dataset which helped to refine and name the themes. The same member of the research team was the primary analyst in all three coding phases of the analysis with other members of the team providing peer review and consultation throughout.

## Results

State administrators of IDD services shared a multitude of perspectives about how they think about their roles as fiscal stewards and as promoters of choice and self-determination for people with IDD and their families. The results below are organized according to major themes from the thematic analyses, and include multiple perspectives on administrators' balance between self-determination and stewardship of public funds.

### Continuum of Stewardship and Choice

At the broadest level, administrators most typically reported seeking a balance between being cautious stewards of public funds and enabling choice for people with IDD and their families. Rather than portraying stewardship and individual self-determination as opposites, administrators most often expressed perspectives that suggested a continuum between the two, where they sought to strike a balance, though most administrators tended toward one side of the continuum or the other.

Among administrators who tended to lean towards enabling greater self-determination, it was often philosophical agreement with the key tenets of self-determination, such as individual choice, that drove their responses. As one administrator from the Northeast explained,

Philosophically, we are very much into having individuals make individual choices about their lives, but we definitely are cognizant of the fact that we are stewards of public money. It's really what we're trying to do now to balance that out. The bottom line is the most important thing are choices and individual self-directing.

Toward the other side of the continuum, many administrators discussed how they put safeguards in place to maintain control of spending in their states' self-directed programs. Controls such as spending caps, lists of approved services, and guidelines for hiring direct support workers were all commonly cited as ways in which administrators exercised their roles as stewards. Administrators who had been working with self-directed programs since their inception sometimes noted that the trend was moving more toward fiscal control as self-directed expanded, as noted by a participant from a Northeastern state:

I think the other challenge for us because we were involved from the beginning days was it really initially started as a self-determination movement not as a self-direction movement and when that term self-direction came about and really began to focus more on the fiscal authority we really struggled with that… it took away from the person centeredness and… really got us into just focusing on how was the money spent and were people doing it in ways that could be justified through the scrutiny of that public lens.

The sections that follow detail the ways in which state IDD service administrators discussed the factors that may be moving self-direction toward greater fiscal responsibility, as well as the ways in which self-directed programs support the self-determination of people with IDD and their families.

### Political Environment

A number of administrators talked about how the political environment in their specific states dictated a fiscally conservative approach to managing self-directed services and supports. In the most general sense, some administrators suggested that the political trends were that, in the words of an administrator from the Southeast, “States are trying to cut Medicaid funding all over the place, and it's hard, you know.” More specifically, some administrators noted that state lawmakers did not have an appetite for spending on the Medicaid waivers that fund most self-directed services, including one study participant from the West who stated, “It's all tied to legislative funding as to how many people we can allocate to a waiver at any given time. And so, we'd all love to serve more people but we know we only have so many funds.”

Other administrators noted that the administration of self-directed supports had become more cumbersome over the years, as policymakers acted to put more controls on public spending. Such burdens often fell on local administrators and support coordinators, who were responsible for processing service plans and completing authorization paperwork. This frustration led an administrator from the Northeast to comment:

We have to provide a number of health and safety assurances which require a lot of extra work on the part of our support brokers… there's a lot more that goes into working with an individual who directs his or her own services to comply with all of those assurances.

Despite state financial constraints and the perceived administrative burden of self-directed supports, study participants also noted that self-direction was likely to expand in coming years, due to updated federal rules, most notably the 2014 CMS Final Rule on Home and Community Based Services. Simply put by a participant from the Southeast, “Self-directed. No problem. Especially with the Final Rule, it's something we're going to start seeing more and more of all over.”

### Cost Savings

Administrators often spoke about the importance of self-directed services and supports having cost savings compared to traditional agency-based services for people with IDD. In the most general sense, self-direction was observed to have cost savings over traditional services since a provider agency was not involved as an intermediary, resulting in lower overhead and administrative costs. An IDD administrator from a Southeastern state explained this logic: “You get to knock out some of that middle money that goes to the provider that helps cover their overhead and you take that and you give that to the client.” There were a number of specific means by which such cost savings were obtained, including the development of spending caps, the integration of budget algorithms into the assessment process, the spending approval process for goods and services, and the introduction of regulations for hiring direct support workers.

Spending caps were among the most common cost savings mechanisms noted by participants, though the form of those caps varied from state to state. In some states, there were annual caps for all services used in self-direction, often based on the service recipient's age. In other states, there were sub-caps to limit spending in certain categories of goods and services. Regardless of the form of spending caps, many administrators noted that the caps were unpopular among advocates and service users who found the limits to be arbitrarily set and ultimately worked against the individual's right to self-determination in designing the services that were right for them. The objections of advocates were often overruled by administrators and policymakers. One administrator from the Midwest shared a common experience, saying, “Instead of having the set cap, some advocates have asked that we be able to just set the monthly allocation individually based on each individual's service plan. I think our General Assembly and administration are hesitant to do that.”

Budget algorithms, as described by administrators, had similar budgetary impact as spending caps but were more complicated procedurally. Based on an individual's level of need determined in the assessment, the budget algorithm incorporated the average rate within a state for that service into the dollar amount assigned to the participant's budget. The budget algorithm was a weighted regression formula that allowed administrators to better control costs. An administrator from the Midwest described their “cost neutrality formula” thusly:

We know that if you authorize 100 hours of [a support] in any given month, not all 100 hours is used…If we have 100 units of [a support] is the average of what we'll be using. If the average is 10 dollars per hour but we know that only 80% of the services that are authorized in a plan actually get used, we would reduce the average statewide cost by 20%. We would give the member 80% of the statewide average cost. In addition to direct cost savings, budget algorithms allowed the state to project what the entire body of participants would spend in a given year, providing predictability in a complex decentralized service delivery system. One administrator from the Southeast viewed this as bringing the money “closer to truly where the needs are… [however] we are in constant lawsuits…over the algorithm” because it was perceived to decrease spending flexibility for people with IDD.

Another common method of achieving cost savings described by IDD service administrators was the need for people with IDD and their families to go through an approval process to spend allocated funds on selected goods or services. By requiring approval for some expenditures, states were able to prevent spending on some expensive goods and services. An administrator from the West described their process:

With our goods and services they need to be based on a participant's assessed need, so we do go through an authorization process which includes a clinical review… and the participant would need to demonstrate an assessed need. There's some other criteria around being cost-effective and whatnot.

Additionally, many administrators discussed their state's approach to paying direct support workers as a source of cost savings, since hourly rates for direct support workers were often capped or set at a standard rate. In some states, direct support workers had to be registered with the state's Medicaid agency, which may have been a deterrent to hiring family and friends to perform direct support work.

### Scaling Back Choice

In their role as stewards of public funds, administrators were often faced with decisions about what types of spending were appropriate for people with IDD and their families who used self-directed supports. Administrators often reported changes in how they have approached the idea of appropriate spending of self-directed funds over time. In many states, based on participants' narratives, self-directed services began as a widely flexible program, that allowed for spending on any number of goods and services, provided that such spending was supported by the individual service plan of the person with IDD. Over time, however, many programs have become more restrictive in their spending flexibility, often with lists of authorized expense categories, as explained by a participant from a Midwestern state:

We had to realize over time we had to provide more technical assistance and clearer parameters to our agencies and families about the appropriate use of Medicaid and that just because you have an individualized budget doesn't mean it's your money and you can spend it in any way that you want. It still has to meet the requirements of the waiver.

As indicated by this administrator, the availability of Medicaid as a source of payment for self-directed services has become increasingly more important for many states. With state budgets often stretched thin, Medicaid has become the payer of choice for many states, as indicated by an administrator from the Southeast who said, “I will admit, we do look at Medicaid as this infinite bank to some degree… if we can get someone out of the state program and into Medicaid, let's go for it.” Since Medicaid will only pay for certain goods and services, administrators who shared this perspective were often forced to put spending restrictions on people who used self-directed supports.

In addition to overall restrictions in the types of goods and services people with IDD and their families could use with funds from self-direction, administrators also discussed regulations about who could provide services, particularly direct support. States varied widely in their rules about whether a close family member could be paid for the provision of direct support, how rates were set for direct support workers, and whether the direct support worker had to be registered as a provider with the state's Medicaid agency. In states with more restrictions on the provision of direct support, the thinking was that it may be unwise to spend public funds when a family member could potentially provide support at no cost to taxpayers. One Southeastern administrator discussed their state's debate about whether family members could be paid for providing direct support using self-directed funds:

There had been a strange little balance there between natural support versus paid employees. There's a philosophical argument there by administrators to say, ‘Wait a minute. This seems like it's a waste because they were already doing it for free. Why pay them to do it now.' Of course, the argument on the other side would be, ‘Look. That's a valuable thing. I feel like I need to pay the person.'”

Other administrators saw this tension more pragmatically in terms of whether paying families to provide direct support could help to support the development of people with IDD. An administrator from the Midwest expressed concerns, which ultimately informed decisions about the state's flexibility in hiring practices in their self-directed program for people with IDD:

We do have concern that there are families who are keeping the individuals at home just for the income that it generates through both the SSI payment and then being able to become a provider and having us pay them to maintain the individual. Just the fact that it's self-directed doesn't necessarily mean they're getting that community integration.

### Participants as Stewards

Administrators typically saw the state as having the primary role for stewardship of public funds in self-directed services and supports for people with IDD and their families. However, many also saw a role for people with IDD and their families in carefully managing public funds. In total, the perspectives of study participants varied when the role of services users as stewards was considered.

On one end of the spectrum, some participants were impressed that people with disabilities and their families often managed their budgets more carefully than many human service agencies, finding ways to stretch their funds, and often leaving some of their allotted funds unspent. This perspective was summarized by an administrator from a Southeastern state who said:

We get scenarios where [self-directed program participants] will mention how they were accessing it from an agency that was charging like say, upwards of $10,000 a year for … nutritional supplements, for example. They found that they can go out and buy it at Walmart for let's say …$2,000. They were raving about that. They were saying, ‘Hey let's save money.' We sit back sometimes saying, ‘Wow. They're thinking of the state dollar too. That's great!'

Most often, state IDD service administrators spoke of trying to educate people with IDD and their families about the importance of responsible spending. Some administrators did so through support coordinators or fiscal agents, while others reported that personnel from the state took a more direct role in educating users of self-directed services about their responsibilities when spending public funds. With a degree of frustration, one administrator from a state in the Southeast shared:

If they want to sit back and allow it to happen to them instead of you know work for it then [self-direction] may not be the best place. If you are going to be the employer then you have to take the employer responsibility. If you want to manage your money then that means you need to be able to reconcile your account.

Though most administrators saw some role for people with IDD and their families as stewards of public funds, others shared less of that perspective, sometimes holding the opinion that users of self-directed programs were concerned only with the ability to have choice in how they spend their allotted budget. This position was succinctly summarized by an administrator from the Midwest who stated, “Consumers, individuals, advocates, they don't necessarily care about fiscal stewardship unless it impacts them personally.”

While the narratives of the majority of state IDD services administrators concentrated on their roles, and the roles of program participants, as fiscal stewards, many also noted the important role of self-direction in promoting self-determination. Many administrators pointed to the central role that the philosophy of self-determination played in shaping self-directed services and supports, yet, many administrators in this study saw the relationship between self-determination and self-direction as being more circular. For example, they indicated that self-direction promoted self-determination in program participants, and that participants' self-determination then placed more demands on the state to build better models of self-direction. The notion that self-direction codifies and enhances self-determination was common in the participants' narratives, including that of an administrator from a state in the Southeast who said, “They're the ones making decisions and, you know, this is going to give them a better quality of life and, you know empower them to take charge of their service needs.”

Most specifically, some administrators pointed to the authority to hire direct support staff of the person with IDD's choosing as the practice that most embodied self-direction's potential to enhance self-determination. The ability, in the terms of an administrator from the West, to enable a person with IDD or their legal representative to, “have the ability to determine who comes into their home and provides support to them” was a vitally important exercise in self-determination that self-direction enabled for service users, and which was often deemed to be absent from traditional services.

The circular nature of the relationship between self-direction and self-determination was reflected by administrators, including one from the Southeast, who noted that the increased choice and control enabled through self-direction leads to less need for services, which then allows the state to save money and offer support to more individuals:

The more you put responsibility and ownership on individuals, the less behaviors, and the less issues you seem to have. I think the more that we transition that way the better off people we serve will be and the better off the money savings, and all-around people will be better off.

## Discussion

Self-direction for people with IDD may be reaching a crossroads between the philosophical roots of self-determination and the benefits of cost containment, based on the narratives of the study participants. While administrators continue to acknowledge the role of self-determination in self-direction, consistent with earlier studies (Doty et al, 1996; Walker et al., 2009), most state administrators of self-directed programs with whom we spoke saw as their primary role being stewards of public funds in an environment in which Medicaid and state funding are often perceived as being under review and up against a threat of potential contraction. Administrators also frequently saw self-direction as a tool to serve more people with IDD with less funds, which was necessitated by long wait lists for HCBS waiver services in their states. At the same time, they reported sometimes stringent measures they took to assure that spending by people with IDD would fall beneath pre-set caps and within a predetermined list of goods and services sometimes limiting the ability of an individual with IDD to exercise their self-determination in choosing the services that would optimize their community living and integration. Based on the narratives of participants, at this crossroads of self-determination and fiscal stewardship as the driving force behind self-direction, it appears that fiscal interests may be the dominating factor at this juncture in time.

### Enrollment Expansion and Budget Contraction

Consistent with previous findings specific to IDD (DeCarlo et al., 2017; Friedman, 2018) and related to other populations (Sciegaj et al., 2016), IDD administrators who informed this study discussed the rapid expansion of the availability of self-directed services and supports in many of their states, which did not have self-direction a few years earlier (Walker et al., 2009). Further, they suggested that enrollments in self-directed services would likely continue to increase in coming years, suggesting that self-direction is increasingly becoming a preferred method of providing services for people with IDD. The potential reasons for this expected expansion in enrollment included the reported cost savings as well as policy changes such as the CMS Final Rule on Home and Community Based Services (2014) and the ability of states to provide self-direction under Medicaid expansion enabled by the Affordable Care Act, which some administrators noted would provide further support for the move toward more self-direction. Although the number of programs is likely to rise and administrators seek to increase program size, a recent study, based on data from around the same time as this research, found that the expansion of available self-directed programs did not necessarily translate into enrollment of more people with IDD in those programs (Friedman, 2018).

While the number of self-direction programs was on the rise, based on the reports of study participants, there were indications that categories for expenditures of self-directed funds may be contracting from the more flexible spending rules in earlier years (Hall-Lande et al., 2012). The contraction of approved spending categories was previously noted by Hall-Lande and colleagues (2012) as one of the challenges facing local-level IDD administrators, especially since self-direction had been considered instrumental in shifting the philosophy of personnel working in the IDD service system toward greater self-determination for people with disabilities. Likewise, the narratives of participants in this study suggest a national trend toward retrenchment of choice in self-direction with many administrators, particularly those who have worked with self-directed programs for several years, noting that their states' programs had taken a turn away from self-direction's philosophical roots in self-determination and moved toward more rigid, standardized rules and program policies which were often aimed at compliance with Medicaid guidelines and cost containment.

### Importance of Political Climate

The relationship between the political climate and self-direction was evident in the majority of our interviews with IDD administrators. Participants were clear in sharing their perspective that the current political environment is driving the evolution of self-direction, and that the fiscal and regulatory climate is currently demanding more cautious stewardship of more limited public funds than had been observed in the early years of self-direction. While cost savings over traditional models of LTSS has always been a hallmark of self-direction (Doty et al., 1996; Hall-Lande et al., 2012; Sciegaj et al., 2016), our study's participants frequently noted that state legislatures are becoming more conservative in making Medicaid funds available, and that state IDD agencies are placing more regulations on how Medicaid funds may be expended. Participants suggested that self-direction may be a way to serve more people with IDD using fewer Medicaid and state funds, thereby making public dollars go farther in meeting the needs of people with IDD and their families, even if the extent and flexibility of services may not be as robust as it was in self-direction's earlier iterations.

It is possible that political climate may be driving not only concerns about cost, but broader structural issues related to self-direction as well. Although we did not explicitly ask administrators about the emergence of Medicaid Managed Care models or the expansion of agency with choice approaches to self-direction, for example, such changes to the service structure would seem to be congruent with administrators' narratives about how self-direction has become increasingly oriented toward fiscal stewardship. Managed care approaches to the provision of HCBS waiver services, including those that are self-directed, are designed to reduce expenditure of public funds, and though often more expensive than fiscal agency models of self-direction (Murphy et al., 2012), agency with choice provides an additional layer of accountability and oversight in how funds are spent by people with IDD, giving states greater control over the use of public funds.

While participants noted that state legislatures were becoming more concerned with conservation of public funds, they also noted that federal rules, most notably the CMS Final Rule on Home and Community Based Services (2014), were pushing states to develop more self-directed service options for people with IDD, such as those that may be enabled through Medicaid expansion and under state plan options. In coming years, it will be important for researchers, people with IDD and their advocates, and administrators of IDD services to understand how the complex interplay of forces in the political climate, federally and at the state level, continue to shift how self-direction is implemented, and what the effects are on the health, service satisfaction, self-determination, and community integration of people with IDD.

### Research Limitations

This research was conducted after the introduction of the CMS Final Rule on Home and Community Based Services (2014), which prompted several states to make plans to change their Medicaid waiver programs that were pertinent to self-direction. Because data collection occurred only shortly after the rule was placed in effect, it is important to note that some state IDD services administrators may not have had adequate information about how self-direction may have changed in their states as a result of the HCBS Final Rule. Due to this timing, and since self-direction is a rapidly emerging and changing approach to service delivery, the findings of this study must be considered within their context as a cross-sectional snapshot in time.

Although the study does account for the perspectives of administrators from a majority of states that offered self-direction to people with IDD through their state IDD agency at the time of data collection, we were not successful in securing interviews with administrators in all states where self-direction was present. Since approaches to self-direction vary considerably from one state to another, it should be acknowledged that, while substantially representative, these results may not represent an exhaustive perspective of administrators across the United States. Additionally, we did not endeavor to interview administrators in states that did not offer self-direction to people with IDD, and their perspectives may present additional insight into the emerging landscape of self-direction.

With the dearth of research on self-direction focused directly for people with IDD, there is ample opportunity for future study. Very little has been written about outcomes for people with IDD who use self-direction, in terms of health, social inclusion, community living, or employment. Virtually none of what has been written to date accounts for differing models of self-direction. Such examinations that would account for the emerging systematic difference in the administration and implementation of self-direction would be helpful in assisting administrators to develop service options that can optimize service outcomes, in the context of cost containment. Likewise, we know little about how outcomes for people with IDD who self-direct compare with outcomes among users of other HCBS programs, offering another potentially fruitful line of future inquiry.

This study also incorporated the perspectives of only one stakeholder group with an interest in self-direction: state-level IDD system administrators. The position of administrators, as actors within the government, may differ significantly from other stakeholder groups, including people with disabilities, their families and advocates, support staff, and service providers. While understanding the perspectives of administrators is vital, since they exercise the greatest control over design of IDD service systems, it would also be valuable for future research to strive for better understanding of the perspectives of other stakeholder groups and how they perceive self-direction's benefits and limitations.

## Conclusion

Self-direction has emerged rapidly as one of the preferred methods of providing publicly-funded supports for people with IDD. Yet, despite the growth of self-direction programs and supports across states, there are still many important questions around both effective and responsible practices for promoting individual choice and control within the current policy climate. This study has illustrated some of the challenges that have arisen from this rapid expansion in the availability of self-direction, as well as some of the opportunities that have come with self-direction's emergence. IDD administrators face a difficult task in balancing choice for people with IDD and the need to be careful stewards of public funds, and this continuum between choice and stewardship has largely shaped self-direction.

## References

References
Abery,
B.,
&
Stancliffe,
R.
(
2003
).
An ecological theory of self-determination: Theoretical foundations
.
In
M. L.
Wehmeyer,
B. H.
Abery,
D. E.
Mithaug,
&
R. J.
Stancliffe's
(
Eds.
),
Theory in self-determination: Foundations for educational practice
(pp
.
25
42
).
Springfield, Illinois
:
Charles C. Thomas Publisher
.
Atlas.ti (Version 7.5) [Computer software]
(
2014
).
Scientific Software Development GmbH
.
Benjamin,
A. E.,
&
Matthias,
R. E.
(
2001
).
Age, consumer direction, and outcomes of supportive services at home
.
The Gerontologist
,
41
,
632
642
. doi:
Bogenschutz,
M.,
Hewitt,
A.,
Hall-Lande,
J.,
&
LaLiberte,
T.
(
2010
).
Status and trends in the direct support workforce in consumer directed supports
.
Intellectual and Developmental Disabilities
,
48
,
345
360
. doi:
Braun,
V.
and
Clarke,
V.
(
2006
).
Using thematic analysis in psychology
.
Qualitative Research in Psychology
,
3
,
77
101
. doi:
Caldwell,
J.,
&
Heller,
T.
(
2007
).
Longitudinal outcomes of a consumer-directed program supporting adults with developmental disabilities and their families
.
Intellectual and Developmental Disabilities
,
45
,
161
173
. doi:
Centers for Medicare and Medicaid Services
(
2014
).
Fact Sheet: Summary of key provisions of the 1915(c) Home and Community Based Services (HCBS) waiver final rule (CMS 2249-F/2296-F)
.
Baltimore, MD
:
Author
.
DeCarlo,
M.
(
2016
).
Implementation of self-directed supports for individuals with intellectual and developmental disabilities: A political economy analysis (Doctoral dissertation)
.
DeCarlo,
M.,
Bogenschutz,
M.,
Hall-Lande,
J.,
&
Hewitt,
A.
(
2018
).
Implementation of self-directed supports for people with intellectual and developmental disabilities in the United States
.
Journal of Disability Policy Studies
.
DeCarlo,
M.,
Hall-Lande,
J.,
Bogenschutz,
M.,
&
Hewitt,
A.
(
2017
).
State of the states in self-direction for individuals with intellectual and developmental disabilities. Policy Research Brief, 26(1)
.
Minneapolis, MN
:
Research and Training Center on Community Living at the University of Minnesota
.
Doty,
P.,
Kasper,
J.,
&
Litvak,
S.
(
1996
).
Consumer-directed models of personal care: Lessons from Medicaid
.
Millbank Quarterly
,
74
,
377
409
. doi:
Foster,
L.,
Brown,
R.,
Phillips,
B.,
Schore,
J.,
&
Carlson,
B. L.
(
2003
).
Improving the quality of Medicaid personal assistance through consumer direction
.
Health Affairs
,
22
,
162
175
. doi:
Friedman,
C.
(
2018
).
Participant direction for people with intellectual and developmental disabilities in Medicaid Home and Community Based Services Waivers
.
Intellectual and Developmental Disabilities
,
56
,
30
39
. doi:
Garbett,
R.,
&
McCormack,
B.
(
2001
).
The experience of practice development: An exploratory telephone interview study
.
Journal of Clinical Nursing
,
10
,
94
102
. doi:
Gross,
J. M.,
Blue-Banning,
M.,
Turnbull,
H. R.,
&
Francis,
G. L.
(
2015
).
Identifying and defining the activities of participant direction programs: A document analysis
.
Journal of Disability Policy Studies
,
25
,
220
232
. doi:
Hall-Lande,
J.,
Hewitt,
A.,
Bogenschutz,
M.,
&
LaLiberte,
T.
(
2012
).
County administrator perspectives on the implementation of self-directed supports
.
Journal of Disability Policy Studies
,
22
,
247
256
. doi:
Heller,
T.,
Arnold,
C. K.,
van Heumen,
L.,
McBride,
E. L.,
&
Factor,
A.
(
2012
).
Self-directed support: Impact of hiring practices on adults with intellectual and developmental disabilities and families
.
American Journal on Intellectual and Developmental Disabilities
,
117
,
464
477
. doi:
Heller,
T.,
&
Caldwell,
J.
(
2005
).
Impact of a consumer-directed family support program on reduced out-of-home institutional placement
.
Journal of Policy and Practice in Intellectual Disabilities
,
2
,
63
65
. doi:
Matthias,
R.,
&
Benjamin,
A.
(
2008
).
Paying friends, family members, or strangers to be home-based personal assistants: How satisfied are consumers?
Journal of Disability Policy Studies
,
18
,
205
218
. doi:
McCoyd,
J.,
&
Kerson,
T.
(
2006
).
Conducting intensive interviews using email: A serendipitous comparative opportunity
.
Qualitative Social Work
,
5
,
389
406
. doi:
Miles,
M. B.,
Huberman,
A. M.,
&
Saldana,
J.
(
2014
).
Qualitative Data Analysis: A methods sourcebook. (3rd Ed.)
.
Thousand Oaks, CA
:
Sage
.
Murphy,
M. G.,
Selkow,
I.,
Crisp,
S.,
&
Mahoney,
K. J.
(
2012
).
Agency with choice: Key components for practical implementation while maintaining participant choice and control
.
Boston
:
National Resource Center for Participant-Directed Services
.
National Resource Center for Participant Directed Services
. (
n.d.
).
Costs in the two models
.
Novick,
G.
(
2008
).
Is there a bias against telephone interviews in qualitative research?
Research in Nursing & Health
,
31
,
391
398
. doi:
Patient Protection and Affordable Care Act, 42 U.S.C. § 18001 et seq
. (
2010
).
Scherzer,
T.,
Wong,
A.,
&
Newcomer,
R.
(
2007
).
Financial management services in consumer-directed programs
.
Home Health Care Services Quarterly
,
26
,
29
42
. doi:
Sciegaj,
M.,
Mahoney,
K. J.,
Schwartz,
A. J.,
Simon-Rusinowitz,
L.,
Selkow,
I.,
&
Loughlin,
D. M.
(
2016
).
An inventory of publicly funded participant-directed long-term services and supports programs in the United States
.
Journal of Disability Policy Studies
,
26
,
245
251
. doi:
Shogren,
K.
(
2016
).
Self-determination and self-advocacy. In Critical issues in intellectual and developmental disabilities: Contemporary research, practice, and policy
.
Washington, D. C
.:
American Association on Intellectual and Developmental Disabilities
.
Walker,
P.,
Hewitt,
A.,
Bogenschutz,
M.,
&
Hall-Lande,
J.
(
2009
).
Implementation of consumer-directed services for persons with intellectual and developmental disabilities: A national study. Policy Research Brief, 20(1)
.
Minneapolis, MN
:
University of Minnesota, Institute on Community Integration, Research and Training Center on Community Living
.
Wehmeyer,
M. L.
(
2005
).
Self-determination and individuals with severe disabilities: Re-examining meanings and misinterpretations
.
Research and Practice for People with Severe Disabilities
,
30
,
113
120
. doi:
Wehmeyer,
M. L.,
Shogren,
K. A.,
Little,
T. D.,
&
Lopez,
N.
(
2017
).
Introduction to the self-determination construct
.
In
M.
Wehmeyer,
K.
Shogren,
T.
Little,
&
S.
Lopez
(
Eds.
).
Development of Self-determination Through the Life-Course
.
New York
:
Springer
.

## Author notes

Development of this manuscript was supported by Grant #90RT5019-01-01 to the Research and Training Center for Community Living from the National Institute on Disability Independent Living and Rehabilitation Research (NIDILRR), U.S. Department of Health and Human Services. Grantees undertaking projects under government sponsorship are encouraged to express freely their findings and conclusions. Points of view or opinions do not therefore necessarily represent official NIDILRR policy.