Improving Quality of Care in Substance Abuse Treatment Using Five key Process Improvement Principles
Process and quality improvement techniques have been successfully applied in health care arenas, but efforts to institute these strategies in alcohol and drug treatment are underdeveloped. The Network for the Improvement of Addiction Treatment (NIATx) teaches participating substance abuse treatment agencies to use process improvement strategies to increase client access to, and retention in, treatment. NIATx recommends five principles to promote organizational change: (1) understand and involve the customer, (2) fix key problems, (3) pick a powerful change leader, (4) get ideas from outside the organization, and (5) use rapid cycle testing. Using case studies, supplemented with cross-agency analyses of interview data, this paper profiles participating NIATx treatment agencies that illustrate successful applications of each principle. Results suggest that organizations can successfully integrate and apply the five principles as they develop and test change strategies, improving access and retention in treatment, and agencies’ financial status. Upcoming changes requiring increased provision of behavioral health care will result in greater demand for services. Treatment organizations, already struggling to meet demand and client needs, will need strategies that improve the quality of care they provide without significantly increasing costs. The five NIATx principles have potential for helping agencies achieve these goals.
Clinicians are increasingly under pressure to implement evidence-based decision making and reduce inefficiency in medical care.1,2 The Institute of Medicine’s Crossing the Quality Chasm called for health care organizations to improve quality of care through system redesign.1 These recommendations were recently extended to services for alcohol, substance, and mental health problems3; many substance abuse treatment agencies have weak organizational structure.4 Incorporating process improvement approaches into addiction treatment services management may address this problem, allowing agencies to modify service delivery and enhance access and retention with only marginal expansions of infrastructure and resources.5, 6, 7 Demand for treatment services exceeds service capacity, and with current funding limitations, dramatic increases in treatment resources are unlikely. Thus, efficiency and effectiveness of existing service systems must improve if goals are to be met, particularly for admission processes and early engagement.6
Network for the Improvement of Addiction Treatment
The Robert Wood Johnson Foundation (RWJF) and the Substance Abuse and Mental Health Services Administration’s Center for Substance Abuse Treatment (CSAT) combined resources to support a demonstration project that used process improvement techniques within drug and alcohol treatment organizations. The project, entitled The Network for the Improvement of Addiction Treatment (NIATx), was developed and based at the University of Wisconsin – Madison and focused on five principles supported by a review of process improvement research8: understand and involve the customer, fix key problems, pick a powerful change leader, get ideas from the outside, and use rapid cycle testing.
The NIATx protocol begins with a “walk-through,” wherein a staffperson assumes the role of a customer to experience and record the organization’s treatment processes.9 The facilitators and barriers to treatment access and engagement are shared with a group of 3–6 staff who have agreed to play a role in process improvements. These groups, known as “change teams” are convened by a change leader to identify solutions and enact change projects designed to address problems identified in the walk-through. A cross-site evaluation of NIATx agencies found significant improvements including a 37% decline in days to admission, an 18% gain in retention in care between the first and second treatment session, and a 17% gain in retention between the first and third treatment session.6 A second NIATx cohort replicated these improvements.7
This paper elucidates the story behind these statistics, providing qualitative analyses and examples of processes adopted by agencies that successfully implemented NIATx principles. The following section reviews theoretical underpinnings behind the five principles.
The NIATx five principles of change
Principle 1: Understand and involve the customer
Successful organizational innovations are customer-oriented.10, 11, 12, 13 The first principle calls for “continued commitment to understand the needs and expectations of people who can benefit from the products and/or services produced by the organization.”8 These organizations involve customers when developing organizational changes and assessing improvements.14 This principle can pose significant challenges resulting from clients’ disparate needs (e.g., co-occurring mental health disorders, criminal justice involvement, homelessness, domestic abuse, and child care) that require comprehensive, client-focused approaches.15, 16, 17, 18, 19
Principle 2: Fix key problems
Principle 2 directs change agents to focus on processes critical to the organization’s viability. Participants are encouraged to focus on the problems “that keep the Chief Operating Officer (CEO) up at night” because problems that are important to leaders are likely to have support necessary to lead to successful implementation. Specific targets can vary, but agencies are counseled to link changes to organizational objectives like reducing staff turnover or improving retention and agency financial status. One method is to reduce no-show rates; more clients showing up to treatment equate to more billable hours thus increasing revenue.14
Principle 3: Pick a powerful change leader
Principle 3 states that change leaders should have adequate authority and respect to assemble and motivate groups,8 have enough power and social capital to lead change efforts,20 are able to increase “buy-in” by winning the “hearts and minds” of staff,21 and provide teams with a clear, attractive vision.22 Change leaders must have access to and the support of the CEO, be able to allocate resources to the change effort, and be committed to change processes and goals.8 Staff resistance to change, a common barrier, can be mitigated by strong leaders who bridge “organizational segmentalism” (when departments do not engage or communicate well with each other) and barriers that prevent or slow change processes.23 Leaders who use consistent communication create a vision for change, promote engagement, and help overcome resistance.20,24
Principle 4: Get ideas from outside the organization
Process improvement’s emphasis on effectiveness and efficiency draws heavily on continuous improvement models developed to improve manufacturing processes.25,26 Applications in business management and health care processes have demonstrated the capacity of organizational change to enhance operational efficiency27 and strengthen patient outcomes.28,29 Change teams are encouraged to seek out and capitalize on other organizations’ successful improvements including production and business practices in other domains (e.g., transportation, hospitality, etc.).14
Principle 5: Use rapid cycle testing
NIATx promotes rapid cycle Plan–Do–Study–Act (PDSA) methods to identify problems and generate solutions (Plan), implement new processes (Do), measure and assess outcomes (Study), and institutionalize the change or make additional changes (Act).30,31 In a PDSA cycle, changes are piloted with a small sample of staff and patients to assess feasibility and effectiveness. If effective, the intervention is institutionalized. If partially or wholly ineffective, the initiative is modified and tested until successful, or until a decision is made to abandon the approach. The PDSA method involves paying attention to data, confirming the existence of problems, identifying opportunities for improvement, and evaluating the effects of any changes.8
Although a substantial body of theoretical knowledge about the application of these principles can be found for other settings (e.g., business and management literature),32 few clear examples exist in behavioral health. Consistent with Fernandopulle and colleagues33 (p. 182) suggestion that “case studies provide essential qualitative insight into attributes associated with quality performance”, this paper uses case studies of NIATx agencies, with cross-agency qualitative analyses, to illustrate successful applications of the principles and identify common pitfalls. Case study methods have become important tools in the study of health care delivery.34 For example, organizational improvement case studies within behavioral health include a benchmark study of the successful application of total quality management (TQM) in mental health settings,35 the use of program specification and accountability evaluation to monitor service delivery,36 and a multisite intervention to improve access to care for chronic drug users.37
Participant selection and characteristics
The sample includes agencies chosen to participate in NIATx. Eligibility was limited to nonprofit agencies with at least 100 admissions/year and serving at least 50% publicly funded clients. RWJF made ten awards in August 2003 and 15 more awards in January 2005 (awards were for 18 months). CSAT awarded funds to 13 grantees in September 2003, providing them each with 36 months of support. Further details about the initial round of applications and awards can be found elsewhere.9 Treatment agencies were located in rural and urban settings in 17 states, varied by size and services provided, and included multiple levels of care. Annual revenues ranged from less than $300,000 to over $11 million. (For additional information, see Ford and colleagues,9 and McCarty and colleagues6).
Site visits and interviews
Annual site visits and quarterly telephone interviews were completed with 38 participating NIATx agencies between October 2003 and May 2006. Individual interviews were conducted with change leaders, executive directors, and a variety of staffpersons; focus group interviews were conducted with front line staff, managers and therapeutic staff. Quarterly telephone interviews were completed with change leaders except when the annual site visit occurred. Executive directors and change leaders agreed to participate in interviews as part of the project. Staff were invited but not required to participate; participation or nonparticipation was kept confidential. All interviewees were informed about confidentiality, freedom to participate and the right to withdraw from the study at any point. The Oregon Health & Science University (OHSU) Institutional Review Board for the Protection of Human Subjects approved and monitored the study.
All interviews were conducted by the OHSU evaluation team and all interviewers held doctoral or masters level degrees. The qualitative interview guide was semistructured and covered the following dimensions: change initiatives and impacts, whether change initiatives were sustained, use of data to make decisions, change leader effectiveness, staff buy-in, staff empowerment, agency stability, effects on organizational culture, extent of customer focus, and learning from others. (Copies of interview guides are available upon request from the corresponding author.)
Each audio-recorded interview was about one hour. Interviewers summarized findings and documented reflections about interviews and site visits. After site visits, interviewers shared site-specific summaries of findings to programs, protecting confidentiality of interviewees, with the change leader as a member check to increase validity38 and requested clarifications and additional perspectives.
Coding and data analysis
The team began with initial read-throughs of a subset of interviews, making notes on possible themes for discussion at team meetings. Thematic codes were developed inductively and transcripts continually reviewed, allowing the data to dictate a majority of analytic categories.39 Through this iterative process of coding, review, discussion, and revision, researchers documented thematic codes and definitions in a codebook. This process was followed until no new categories emerged from the data and the codebook was finalized. Transcripts were then entered into ATLAS.ti 5.0 software,40 and the codes detailed in the code book were applied to the dataset. The contents of each coding category were reviewed to ensure agreement.
Evaluation staff took several steps to increase methodological rigor: multiple evaluators participated in data collection and analysis to ensure multiple viewpoints during code creation, evaluators considered rival explanations while analyzing data to validate the theoretical scheme, and agency representatives reviewed text to ensure clarity and provide additional detail (member checking). To ensure consistency across coders, we check-coded a 10% random selection of interviews using a subset of key codes, evaluating 526 coded passages. An additional coder reviewed inconsistencies between primary coders and check coders to resolve them; primary coders’ were judged accurate for 89.3% of the 526 passages.
For cross-agency analyses, thematic content was extracted from the database by reviewing text relevant to the five principles. A “best cases” case-study approach illustrates successful experiences from participating agencies; cross-agency findings from thematic analyses provide information about common experiences and pitfalls. To select cases, investigators reexamined each principle (as described in the “Introduction”), then collaboratively chose those with exemplary implementation of one or more principles. The type of case study approach used here allows researchers to explore individuals or organizations, simple or complex interventions, relationships, communities, or programs.41
Results are organized around the five NIATx principles. While agencies were encouraged to apply all five principles, each did so with varying degrees of success. As a result, it was impossible to select a single agency representing the “best case” for all of the principles. Pseudonyms are used in the following report of results to protect anonymity of agencies and individuals.
Principle 1: Understand and involve the customer
The customer focus aspect has been a big part of our change projects and change teams. All the change teams have had customers as part of their group — either current or former clients.
Results of this change cycle have been very positive: Client self-pay fees have increased from 22% per month to around 90%. Support staff are less burned out, and the reorganization has allowed [Esperanza House] to reduce [staff] hours [at their request]…[increasing] revenue. We were just looking for better service—we didn’t anticipate the increase in fee collections.
This agency also adjusted walk-in times to accommodate longer hours of operation, offered additional treatment modalities and techniques, including biofeedback, and restructured the orientation process to include sober peer support. They also trained staff to be more alert to client disengagement and implement motivational techniques to prevent early dropout.
Other NIATx agencies varied significantly on this dimension. Some made efforts to solicit client feedback and understand their needs, but most did little to involve customers directly in organizational changes or PDSA cycles. For example, clients were infrequently included as members of change teams or invited to provide input when changes were planned. The most common approach to involving the customer tended to be collecting client response after a PDSA cycle was initiated, although this was rarely institutionalized and often tended to be impromptu and informal.
Principle 2: Fix key problems
I am having to explain to people who want to come back that we are not the [New Life] we used to be. They are coming back with the expectation of routine and normalcy like [predisaster] and when they come they are confused and upset.
As a result, clients were leaving treatment prematurely. A PDSA cycle designed to facilitate a major cultural shift was proposed. The director felt staff spent too much time “babysitting” and conducting “detective” work to root out banned activities such as smoking. The change cycle called for staff to emphasize that clients should be “responsible for their surroundings and each other.” For instance, if a client was behaving disruptively, other clients were responsible for calling a community meeting so that all could discuss and resolve the problem. The goal of this change was to encourage engagement in treatment, increase retention, and reduce the amount of staff time devoted to enforcing rules.
[They] came up with a greetings system of their own and an orientation twice a week [for] new clients. Of the 50 people who were here last Friday, 40 of the 50 had been here for more than 30 days. That shows great success on their part…
When we started to involve the clients more in the change, there was a dramatic jump in continuation [retention in treatment]. We have also looked at our retention numbers for November [baseline] and December  and they were 60 and 69% respectively. About 75% have now [February 2006] been here more than 30 days.
New Life was exemplary, but some organizations did not progress beyond the stage of small changes. For example, a common pitfall was that initial change efforts targeted aesthetic issues such as cosmetic upgrades to waiting rooms, while more critical concerns, such as client retention, went unaddressed.
Principle 3: Pick a powerful change leader
Beth is director of program development at a large, urban, multisite treatment organization, “Sober Living.” Beth took the lead on change initiatives and, over time, encouraged change team members to select their own targets. Her approach included: (a) attention to collecting and analyzing data, (b) ability to communicate clearly about change cycles, (c) support from the agency’s CEO, and (d) unfailing enthusiasm.
Beth routinely incorporated the collection and analysis of data into the change team’s process improvement strategy. Data-driven process improvements were based not only on quantitative data collected from change cycles but also face-to-face interviews with clients. Beth encountered resistance to process improvements, especially within the Sober Living’s detoxification unit. Staff described how oppositional the unit had initially been to Beth, given her outsider status, with comments like: “Who is she coming in here and talking about possible changes? She doesn’t even work in this unit.” Beth gained acceptance and buy-in through effective communication: she engaged detoxification staff, explaining PDSAs and how the change team worked. She elicited and addressed concerns, encouraging two-way communication. As a result, Beth was able to help the unit improve the quality of information provided to clients, streamline intake paperwork so that it was less burdensome, and cross-train staff to address all components of the intake process in order to avoid delays.
Other NIATx agencies indicated a wide range of leadership methods and styles. These were met with varying levels of success. Primary pitfalls included unclear expectations for staff, lack of clear communication with change team members, turnover within the change leader position creating instability, and lack of adequate authority.
Principle 4: Get ideas from outside the organization
Two agencies illustrate how organizations can leverage knowledge gained from others. These agencies were from the same state but participated in different NIATx cohorts. “Step-Ahead,” an agency funded in the first RWJF cohort, had been engaged in process improvements for a longer period of time than “Choices” (an agency funded in the second RWJF cohort). Initially, the flow of information was designed to be unidirectional—from the more experienced agency to the newer agency, but staff soon discovered the value of bidirectional exchanges.
We decided at that point that we did not want to [manage walk-in clients] the same way that they did. But it was very helpful information and helped us make that decision. Shortly thereafter, we piloted our walk-in [procedures].
We both exchanged ideas — they got more from us because we were in the process for over a year, but we got ideas from them…as far as efficiency. We loved having them. Later, I heard [Choices’ executive officer] speak of how much they were able to do so quickly because they were able to observe us and get peer-to-peer support; it helped them grasp the [NIATx] concept quicker.
Step-Ahead and Choices continued to collaborate and share information with one another over the course of their participation. Our cross-agency analyses suggested that “getting ideas from the outside” was the least incorporated of NIATx’s five principles. This appeared to be due to lack of time, staff shortages, and the concentration of effort on change projects within agencies.
Principle 5: Use PDSA rapid cycle change projects
“Transformations” treats several hundred patients a year in a northeastern city. It provides inpatient and outpatient treatment, methadone maintenance, and detoxification services to a mostly publicly funded population. Transformations used a series of PDSA cycles to improve retention. Baseline data showed early treatment dropout was a serious problem in their outpatient services: 50% of the clients who completed intake appointments did not return for the first treatment appointment. The team decided to train intake staff in Motivational Interviewing (MI)42 and implement MI techniques during the intake appointment (Plan). The intervention was implemented (Do) and data were analyzed (Study). Over the following month, retention increased from 50% to 90%. At that time, the change team agreed on an appropriate action (Act): They would adopt MI training for all intake staff and make MI processes a standard procedure.
Staff initiated a second MI-based PDSA cycle in their residential facilities. In June 2005, the baseline residential retention rate (the proportion of admitted clients who completed one week of treatment) was 80%. By October, continuation rates had risen to 100%. The change was adopted as a standard procedure in the residential unit as well.
To increase staff interest and buy-in for the changes, the change team posted week-by-week tallies of change-related data for all staff to see. The progress through these initial successes increased staff attention to customer needs and inspired new change cycles designed to further increase retention in care including providing a tour of the facilities, creating a “health management group” for clients on psychiatric medications, changing medication administration times to allow clients an extra hour of sleep, and instituting an “open door policy” to allow clients to express concerns to staff
[The PDSA cycles] have made a big difference in how we approach change. [They] provided a structure for us to make changes and improvements [that are] measurable and…easy for people to use. The changes feel less overwhelming because of the process.
Cross-agency analyses showed a wide variety of responses to implementing PDSA cycles. Some agencies had difficulty implementing cycles, while others adopted the process quickly. Challenges included identifying appropriate outcome measures [Planning], accurately tracking pre- and post-change [Doing], interpreting differences in pre–post measures [Studying], and determining appropriate next steps [Acting].
Agencies’ experiences elucidate how simple strategies produce valuable improvements in delivery of substance abuse treatment. While agencies were encouraged to apply all five principles, each did so with varying degrees of success. As a result, it was impossible to select one single agency as representing the “best case” for all of the principles. Each agency had its own strengths and weaknesses related to the NIATx principles and for some, these strengths and weaknesses changed over time. For instance, interviews suggest that agencies with changes in leadership or unstable financial environments went through periods where they were unable to engage fully in NIATx change efforts. Similarly, organizations encountering staffing, accreditation, or financial stress found it difficult to allocate staff time and energy to organizing changes around all five principles. We also found that smaller agencies had more difficulty implementing changes because they had fewer organizational resources of all kinds. Each agency, however, focused on the principles that were most important to its leadership and staff, taking into account their capacity for change.
It is also noteworthy that successful process improvements are not without costs. Smaller agencies struggled with changes such as higher clinical caseloads when treatment access was improved. Similarly, increased workload or changes in hours of operation can put a strain on already overburdened staff. Moreover, stress can result from simply working in an environment where things are continually changing, even if those changes are improvements.
The primary limitation of this study is that the findings are confined to NIATx agencies—agencies that recognized the value of process improvement prior to applying for funding from the RWJF or CSAT, successfully completed walk-through exercises, and submitted successful grant applications. Thus, they likely represent agencies more amenable to process improvement strategies from the outset. At the same time, the experiences reported across NIATx-participating organizations and as part of NIATx funding applications9 suggest that lessons learned from these case studies and cross-agency analyses may have broader applicability. Finally, we deliberately selected best example cases, which are not representative of all participating agencies. Nonetheless, they provide valuable opportunities to learn from successful change efforts. By improving administrative and clinical functions, agencies increase the likelihood that they will improve client access to and retention in treatment, thus improving outcomes.
Implications for Behavioral Health
Improving the quality of behavioral health care is a challenge. Quality improvement techniques have been applied in various health care settings, but initiatives in drug and alcohol treatment lag behind. Recent approaches emphasize continuously improved treatment processes as means of improving care quality. These efforts will be increasingly needed as the environment for behavioral health care changes. The Patient Protection and Affordable Care Act includes provisions for mandated mental health and substance abuse treatments that will increase the level of services offered and extend those services to a wider population.43 Systems of care are expected to document the effectiveness of care delivery and use data to proactively manage care so that processes are more efficient and more effective. Behavioral healthcare providers will face important decisions in effectively designing benefits, service delivery, and outreach and enrollment programs to meet the needs of newly eligible adults. As demand for treatment services grows beyond current service capacity and resources, process improvement becomes even more important for patients and staff in behavioral health care settings. NIATx principles offer innovative ways for programs to meet demands while improving care and maintaining a client focus. Incremental improvements may seem modest but, when aggregated over time and across sites, can lead to substantive improvements. Modest gains can be a viable strategy for making large social problems more manageable.44 NIATx is an opportunity for treatment programs to learn process improvement. These case studies demonstrate that process improvement strategies can contribute to enhanced quality of care for alcohol and drug disorders.
We appreciate the support and participation of the treatment programs that have participated in NIATx. The Network for the Improvement of Addiction Treatment (NIATx) was funded by grants from the Robert Wood Johnson Foundation and cooperative agreements from the Substance Abuse and Mental Health Services Administration, Center for Substance Abuse Treatment. The National Evaluation Team at Oregon Health and Science University was supported through awards from the Robert Wood Johnson Foundation (46876 and 50165), the Center for Substance Abuse Treatment (through subcontracts from Northrop Grumman—PIC-STAR-SC-03-044, SAMHSA SC-05-110), and the National Institute on Drug Abuse (R01 DA018282). National Program Office activities at the University of Wisconsin were supported through awards from the Robert Wood Johnson Foundation (48364), and the Center for Substance Abuse Treatment (through a subcontract from Northrop Grumman—PIC-STAR-SC-04-035). We are especially grateful for the cooperation and collaboration from the 38 members of NIATx.
Conflicts of interest
The authors have declared that no conflict of interest exists.
- 1.Institute of Medicine. Crossing the quality chasm: a new health system for the 21st century. 1 ed. Washington, DC: National Academy Press; 2001.Google Scholar
- 2.Institute of Medicine. To err is human: Building a safer health system. Washington, DC: National Academy Press; 2000.Google Scholar
- 3.Institute of Medicine. Improving the quality of health care for mental and substance-use disorders: Quality chasm series. Washington, DC: National Academy Press; 2006.Google Scholar
- 5.Capoccia VA, Cotter F, Gustafson DH et al. Making “stone soup”: Improvements in clinic access and retention in addiction treatment. Joint Commission Journal on Quality & Patient Safety 2007;33(2):95–103.Google Scholar
- 10.Cooper R. Project NewProd: What Makes a New Product a Winner? An Empirical Study of Succesful Versus Unsuccesful Industrial Innovation. Montreal, Quebec: Quebec Industrial Innovation Centre; 1980.Google Scholar
- 12.Gerstenfled A. A study of successful projects, unsuccessful projects and projects in progress in West Germany. IEEE Transactions on Engineering Management EM–23 1976;116–123.Google Scholar
- 13.Maidique M, Zirger BJ. A study of success and failure in products innovation: The case of the US electronics industry. IEEE Transactions on Engineering Management EM-31 1971;192–203.Google Scholar
- 16.Westermeyer J. Cross-cultural studies on alcoholism. Alcoholism: Biomedical and genetic aspects. Elmsford, NY: Pergamon; 1989.Google Scholar
- 18.Substance Abuse and Mental Health Services Administration CSAT. Screening and assessment for alcohol and other drug abuse among adults in the criminal justice system. Treatment Improvement Protocol (TIP) Series 7. Washington DC: US Government Printing Office; 1994.Google Scholar
- 19.Substance Abuse and Mental Health Services Administration CSAT. Comprehensive Case Management for Substance Abuse Treatment: Treatment Improvement Protocol (TIP) Series 27. Rockville, MD; 1998.Google Scholar
- 20.Kotter JP. Leading change: Why transformational efforts fail. Harvard Business Review 1995;59–67.Google Scholar
- 21.Hill S. How do you manage a flexible firm: The total quality model. Work, Employment and Society 1991;397–415.Google Scholar
- 22.Kouzes J, Posner B. The leadership challenge. San Francisco: Jossey-Bass; 1988.Google Scholar
- 23.Burns T, Stalker G. The management of innovation. New York: Oxford University Press; 1961.Google Scholar
- 24.Kotter JP, Schlesinger L. Best of HBR: Choosing strategies for change. Harvard Business Review 2008;3–13.Google Scholar
- 25.Deming WE. Out of the crisis. Cambridge, MA: MIT-CAES; 1986.Google Scholar
- 26.Juran JM. Juran’s quality control handbook. New York: McGraw-Hill; 1988.Google Scholar
- 28.O’Connor GT, Plume SK, Olmstead EM et al. A regional intervention to improve the hospital mortality associated with coronary artery bypass graft surgery. The Northern New England Cardiovascular Disease Study Group. JAMA 1996;275(11):841–846.Google Scholar
- 30.Gitlow H. Tools and methods for the improvement of quality. Homewood, IL: Irwin; 1989.Google Scholar
- 31.Shewart W. Statistical method from the viewpoint of quality control. New York: Courier Dover Publications; 1986.Google Scholar
- 32.Gustafson DH, Cats-Baril WL, Alemi F. Systems to support health policy analysis : Theory, models and uses. Ann Arbor, MI: Health Administration Press; 1992.Google Scholar
- 36.Savas SA, Fleming WM, Bolig EE. Program specification: A precursor to program monitoring and quality improvement. A case study from Boysville of Michigan. Journal of Behavioral Health Services & Research 1998;25(2):208–216.Google Scholar
- 38.Lincoln YS, Guba EG. Naturalistic inquiry. Beverly Hills, CA: Sage; 1985.Google Scholar
- 39.Glaser BG, Strauss AL. The discovery of grounded theory: Strategies for qualitative research. Chicago: Aldine; 1967.Google Scholar
- 40.User’s manual for ATLAS.ti 5.0 [computer program]. Berlin: ATLAS.ti Scientific Software Development; 2004.Google Scholar
- 41.Yin RK. Case study research: Design and methods. 3rd ed. 2003. Thousand Oaks, CA, Sage Publications. Applied Social Research Methods SeriesGoogle Scholar
- 42.Miller WR, Rollnick S. Motivational interviewing: Preparing people to change addictive behavior. 2nd ed. New York, NY: The Guilford Press; 2002.Google Scholar
- 43.Mental Health America. Fact sheet: Medicaid expansion. http://www.mentalhealthamerica.net 2011; Available at http://mentalhealthamerica.net/go/action/policy-issues-a-z/healthcare-reform/fact-sheet-medicaid-expansion/fact-sheet-medicaid-expansion. Accessed August 8, 2011.