Skip Navigation
Policy Innovation Profile

Major Health Systems Collaborate on Organizational Structures and Policies, Enhancing Access to Care and Reducing Uncompensated Care Costs for the Uninsured

Tab for The Profile



The Voices of Detroit Initiative brought together Detroit’s four major health care systems (Henry Ford Health System, Detroit Medical Center, St. John Health, and Mercy Hospital) with local Federally Qualified Health Centers and the Detroit Public Health Department to strengthen the health care safety net for uninsured residents. These systems created an organizational structure to facilitate collaborative decisionmaking and develop policies and goals to enhance cooperation and ensure that uninsured individuals had access to services across the care continuum. Policies were developed for screening and enrolling patients in medical homes, providing a full range of primary care and specialty services, raising funds to support program services, and sharing data to facilitate care coordination and program evaluation. The initiative has increased enrollment in public insurance, enhanced access to primary care–based medical homes and specialty care, and reduced the costs of uncompensated care for local health systems.

Evidence Rating (What is this?)

Suggestive: The evidence consists of post-implementation data and anecdotal information related to the following: the number of patients enrolled in the program and/or assisted in applying for Medicaid between 1999 and 2010, how participating systems have increased access to primary and specialty care as a result of the program, and estimates of the reduction in uncompensated care costs for local health systems during the first 5 years of the initiative.
begin do

Developing Organizations

Advantage Healthcare Centers; Community Health and Social Services, Detroit; Detroit Community Health Connection; Detroit Health Department; Detroit Medical Center; Health Centers Detroit; Henry Ford Health System; St. John Health, Detroit; Wayne State University
end do

Date First Implemented

The initial grant period ran from 1999 to 2005. Since 2005, the program has continued in a modified format with different funding.begin pp

Patient Population

Vulnerable Populations > Impoverished; Medically uninsured; Insurance Status > Uninsured; Vulnerable Populations > Urban populationsend pp

Problem Addressed

More than 1 in 6 Americans (50 million individuals) lack health insurance, with those living in urban areas most likely to be uninsured. Being uninsured has negative consequences not only for those without coverage (who suffer poorer health outcomes), but also for other stakeholders that bear the cost of providing their care, including providers, local businesses, and insured patients.
  • Many uninsured, especially in urban areas: Nationwide, nearly 17 percent of Americans lack health insurance. In some urban areas, however, the percentage is much higher. In Detroit, hard hit by the recent recession, as many as 1 in 3 residents lives at or below the poverty level, and recent census figures estimate that more than 30 percent of adults aged 18 to 64 have no insurance.1
  • Negative health outcomes: Adults without health insurance have worse health than those who have insurance. Often they do not receive preventive services and will skip doctor's visits or effective therapy even when symptoms arise. They suffer worse health outcomes from conditions such as cancer, heart disease, diabetes, and asthma, which leads to lower quality of life and premature death.2
  • Large burden on other stakeholders: Other major stakeholders in the system, including providers, local businesses, and even insured patients, bear much of the cost of caring for the uninsured, as outlined below:
    • Provider burden through uncompensated care: The care provided to uninsured Americans totals roughly $86 billion a year, with patients paying about $30 billion of these costs. Hospitals and other safety net providers provide the remaining $56 billion as uncompensated care. While various government subsidies cover some of this cost,2 the American Hospital Association reports that uncompensated care costs the average hospital $7.9 million, or 5.8 percent of hospital expenditures.3 In some areas, uncompensated care costs can be much greater. For example, Detroit has no public hospital and no tax structure to support uncompensated care, making local health systems the de facto safety net providers. By law, emergency departments (EDs) cannot turn away a patient unable to pay for care. This structure raises the cost of uncompensated care for Detroit providers. For example, the Henry Ford Health System has found that between 12 and 20 percent of hospital expenditures go to uncompensated care.4
    • Employer and community burden: Financial pressures caused in part by the rising costs of uncompensated care can affect the availability and costs of health care services for insured patients, businesses, and the community at large. Financial pressures can cause hospitals to discontinue certain services or even close. (In fact, Mercy Hospital, one of the four original partners in this program, closed its doors in 2000.) Hospitals often have little choice but to pass along the costs of uncompensated care to insured patients, resulting in higher health care premiums for businesses and individuals. These higher premiums affect corporations’ ability to provide insurance to their employees and/or the cost of doing business.5

What They Did

Back to Top

Description of the Innovative Activity

The Voices of Detroit Initiative (VODI) brought together Detroit’s major health care systems to strengthen the safety net for uninsured residents. These systems created an organizational structure to facilitate collaborative decisionmaking and developed standard policies and goals to enhance cooperation and ensure that uninsured individuals had access to a standard set of services across the care continuum. Policies related to screening and enrolling patients in medical homes, providing services, raising funds to support program services, and sharing data. Key elements of this initiative are outlined below:
  • Organizational structure to facilitate collaborative decisions: To build the partnership and convert competitors into collaborators, chief executive officers (CEOs) and other senior administrators from the participating systems created an organizational structure to facilitate collaborative decisionmaking while still respecting the individual cultures of the partners. The structure served to break down the ambitious goals of the project into manageable pieces while still directly involving key individuals from each organization. To that end, the partners created several committees to facilitate decisionmaking through discussion, debate, and consensus rather than voting. The central VODI office provided support to these committees, which are described below:
    • Oversight committee: Made up of the CEOs from each health system, a representative of Wayne State University, and the Mayor of Detroit, this committee functions as the board of directors for the project. It met quarterly during the first 4 years to review and approve decisions made by other committees. (After the grant period ended, this committee became the Board of Directors and meets only twice a year.)
    • Executive committee: This group includes high-ranking representatives from the participating health systems, Federally Qualified Health Centers (FQHCs), and the community at large. The group meets monthly (and sometimes more often) to develop policies related to project leadership, financial issues, program strategies, and evaluation. The committee brings its recommendations (some of which are contentious) to the oversight committee for approval. This committee also hears and compiles reports from the participating health systems and the workgroups (see below), with these reports often feeding into the committee's policy recommendations. The committee ensures that the health system partners carry out agreed-to policies, and oversees the program’s executive director and evaluation process.
    • Workgroups: Five workgroups met regularly during the grant period to identify best practices and develop proposed policies and procedures related to specific aspects of the program. A representative of one of the participating health systems or the Detroit Health Department acted as chair of each group. Workgroups brought their recommendations to the executive committee for consideration. (After the grant period, the five workgroups were replaced with ad hoc committees that meet as needed to address specific issues as they arise. For example, an ad hoc committee was recently formed to address the problem of ED "frequent flyers.") The five workgroups that met during the grant period included:
      • Community steering: This committee included representatives of community-based organizations, such as mental health organizations, senior centers, the local school system, cultural groups, neighborhood associations, institutions of higher education, and religious organizations. It engaged the community for input and guidance related to the program. For example, this group outlined services needed by the community and recommended making anyone earning 250 percent of the Federal poverty level or less eligible for program services.
      • Network development: Made up of representatives of the partner organizations, this group looked at care models, disease management issues, and strategies for better integrating service delivery in the Detroit area.
      • Risk pool and evaluation: This group defined the target population, with the goal of ensuring that program services reach this group. To that end, the group set parameters with respect to age, race, gender, and health status, and identified specific subpopulations to be targeted, including the working poor, those transitioning from welfare to work, and the chronically uninsured.
      • Dental/oral health coalition: This group addressed issues related to dental health and care for uninsured residents during the grant period.
      • Learning laboratory and evaluation: This group designed and supervised research and evaluation of the project.
  • Common policies to enhance cooperation, ensure patient access to common services: Each participating system agreed to common policies designed to enhance cooperation and ensure that uninsured individuals had access to a common set of vertically integrated services across the care continuum, regardless of where they entered the system. Policies related to enrollment targets, screening and enrollment procedures, provision of a common set of services in a medical home, and data sharing, as outlined below.
    • Mutually agreed-upon enrollment goals: To make a significant impact without overwhelming the health systems, the partners agreed on the goal of enrolling 27,500 Detroit residents in VODI during the 5-year grant period (1999 to 2004). Each participating system agreed to take a portion of those patients, with grant monies allocated based on the number of patients enrolled in the program. Each system reported monthly on its progress toward its enrollment goal.
    • Formal contracts: During the grant period, the Executive Committee developed an annual contract with each participating system for each year of the grant period. The contract outlined how much of the grant money would go to that health system and specified goals for enrollment, data collection and sharing, and service delivery. The CEO of each system signed these agreements, ensuring his or her involvement and knowledge of the process. In addition to this overall contract, VODI used memorandums of understanding to outline data sharing and other procedures.
    • Common screening for eligibility: Participating systems adopted a common process for screening patients for eligibility, using a series of seven questions to determine if the patient may be eligible for Medicaid, other government-funded health insurance, or VODI. Screening takes place when a patient without insurance presents for care in an ED at any of the participating health systems. During the grant period, enrollment specialists (employed by the individual health system but funded by VODI) asked the questions; currently, ED staff take on this responsibility. If the patient qualifies for Medicaid, the enrollment specialist or staff member assists in filling out the application. Those waiting for Medicaid coverage to begin (which sometimes can take more than 180 days) also qualify for VODI services.
    • Common enrollment procedure and assignment to medical home: If the patient does not qualify for Medicaid and meets other VODI eligibility requirements, the enrollment specialist or ED staff member offers enrollment in the program. All systems use the same intake form and collect the same information from patients. During the grant period, enrollment specialists performed a health appraisal to identify those who have a history of and/or are at risk for frequent ED use. In most cases, the health system at which the patient first presents becomes the patient’s medical home. The patient receives a VODI identification card that notes the patient’s medical home. If a patient originally enrolled at one health system presents at another, the patient is treated and/or stabilized and then referred back to his or her medical home for followup.
    • Common services and out-of-pocket costs: All participating systems agreed to offer the same set of services to VODI patients assigned to them, including primary care and care management in a medical home, laboratory tests, pharmaceuticals, hospital-based diagnostics, subspecialty care, inpatient care, disease management, emergency care, and dental care. Each health system works with local FQHCs to provide primary care services. According to Federal law, FQHCs must offer these services based on a sliding scale fee schedule that applies to all patients. VODI patients receive other services (outside the FQHC) at minimal cost. To ensure that they did not become overburdened, the FQHCs decided to keep their fee structures similar so that patients do not choose a medical home based on cost.
    • Data sharing agreement: To facilitate care coordination, track progress towards enrollment goals, and ensure that patients return to their medical homes, the partner organizations agreed to share certain types of patient information. Originally, enrollment specialists recorded this information on paper forms. Each month, VODI staff entered the information into a common database shared by the partners. At each encounter, partner staff generated a charge slip similar to those used for insured patients with diagnosis and procedures codes. The information from these forms was transmitted electronically to the VODI central office, where staff compiled the information and reported each month to the executive committee on the number of patients enrolled along with aggregate data on utilization. Eventually, the program transitioned to an electronic health record that allows access to real-time enrollment data and up-to-date encounter reports at any time. The partners also agreed to share information with Wayne State University, which structured and oversaw program evaluation during the grant period.
    • Cooperative fundraising: Early in the process, the partners agreed to bring leads on possible outside funding sources for the VODI program to the executive committee. Whenever a promising lead arises, the partners jointly decide which health system should make the grant application. For example, when seeking a Federal Community Access Program grant in 2000, St. John’s Health applied, because it qualified for matching funds as an affiliate of Ascension Health (the nation’s largest nonprofit health system).
  • Individualized implementation: While all participating systems agreed to the policies and goals outlined above, each remains free to decide how to implement them and/or meet the agreed-to goals. For example, while all systems agreed to provide specialty care to VODI patients, each decided on its own who should provide this care. Henry Ford instructed its employed specialists to accept referrals for VODI patients, whereas St John’s Health developed a cadre of specialists who volunteer to see a certain number of VODI patients each month.
  • Centralized disease management support: During the grant period, the VODI central office provided disease management services to enrolled patients, using a team of nurse practitioners. The program provided an extra layer of care coordination to patients with five chronic conditions: asthma, diabetes, hypertension, cancer, and smoking. Eligible patients were identified through the aforementioned health appraisal conducted at enrollment. The nurse practitioners provided outreach, support, and assistance to help these patients get the services they needed, manage their conditions more effectively, and avoid costly ED visits. After the grant period ended, the primary care medical homes took over many of these disease management functions.

Context of the Innovation

The city of Detroit is currently served by three major health systems—Henry Ford Health System, Detroit Medical Center, and St. John's Health. (A fourth system, Mercy Hospital, closed in 2000). These systems serve residents in a city with very high unemployment and many uninsured residents. The center of the U.S. automobile industry and several other manufacturing industries hit hard by the recent economic downturn, Detroit has a higher proportion of unemployed and uninsured residents than the vast majority of cities in Michigan and around the nation.

The impetus for VODI goes back to 1995, when leaders of the City of Detroit Health Department, other local government officials, and the CEOs of the four major health systems came together with the City of Detroit Health Department to look for ways to address the rising rates of uninsured and uncompensated care in the city. With no local public hospital, these four health systems operated as the de facto safety net for the city, providing millions of dollars in uncompensated care each year. Even with these expenditures, many of the city’s uninsured went without needed care. Without enough primary care providers or a mechanism to pay for specialty care, the safety net was failing Detroit’s low-income residents. The situation had implications that went beyond the uninsured: the rate of uncompensated care threatened to bankrupt local health systems, which were among the largest employers in the city. To remain solvent, hospitals passed the costs of uncompensated care to insured patients, driving up health insurance premiums for local businesses and insured individuals.

In 1997, the Kellogg Foundation called for proposals for its Community Voices Initiative, a grant program to encourage community partnerships to improve access to care for the uninsured. The aims of the grant program meshed well with the goals of the Detroit community, and health care leaders in the city decided to come together to apply for the program. (See Planning and Development Process section for more details.)

Did It Work?

Back to Top


VODI has increased enrollment in public insurance, provided medical homes and specialty care for those without insurance, and reduced the costs of uncompensated care for local health systems.

  • Increased enrollment in public insurance: Between 1999 and 2010, VODI and health system staff assisted 59,244 individuals in applying for Medicaid or Adult Waiver Benefit programs, with 95 percent of those applying being accepted into the programs.
  • Enhanced access to primary and specialty care: Over the same time period, VODI has provided medical homes and specialty care to 78,781 Detroit residents who have no insurance and no regular source of care. VODI has also been a catalyst for several participating systems to develop and/or upgrade their ability to offer primary and specialty care within a medical home to these individuals.
    • Primary care: Before the program, St. John’s Health did not have a formal partner to provide medical homes to uninsured and Medicaid patients. After joining VODI, St. John’s developed a formal partnership with an FQHC, Advantage Health. In addition, the program brought increased attention to the need for additional primary care services in the city, leading to a dramatic expansion in the number of FQHCs or lookalike organizations in the city, from 5 such sites when the program began to 22 sites today.
    • Specialty care: Before the program, primary care providers routinely had difficulties procuring specialty care for Medicaid and uninsured patients. In response to VODI, participating systems developed mechanisms for providing this care. For example, St. John’s Health developed the Physicians Who Care program, which now consists of 400 specialty physicians who voluntarily see 1,300 patients a year. Henry Ford already had a program for providing specialty care, but the data sharing and tracking components of VODI helped strengthen communication between primary care and specialty providers, and helped to ensure patients got the care they needed.
  • Lower uncompensated care costs: During the 5-year grant period, VODI reduced the costs of uncompensated care for local health systems by an estimated $23 million. This figure includes $8.2 million in revenue enhancements from enrolling patients in public insurance and $13.3 million in cost savings (a 26-percent reduction) due to better utilization patterns (less ED care and more primary and preventive care).4

Evidence Rating (What is this?)

Suggestive: The evidence consists of post-implementation data and anecdotal information related to the following: the number of patients enrolled in the program and/or assisted in applying for Medicaid between 1999 and 2010, how participating systems have increased access to primary and specialty care as a result of the program, and estimates of the reduction in uncompensated care costs for local health systems during the first 5 years of the initiative.

How They Did It

Back to Top

Planning and Development Process

Key steps included the following:
  • Securing grant: After receiving the Community Voices grant application from the Kellogg Foundation, the mayor’s office called together the CEOs of the city’s four major health systems. The group decided to work together to develop a grant proposal. The effort succeeded, and Detroit became 1 of 13 cities nationwide that received $5 million in funding over 5 years.
  • Development of community vision: Senior administrators from the major health systems, local FQHCs, and the health department came together to discuss their vision for coordinating care. Instead of focusing on a small area of the city or a specific population, the group decided to develop a citywide collaboration among all major health entities, thus allowing them to meet the needs of as many uninsured Detroit residents as possible.
  • Reallocating patients after Mercy closure: When Mercy Hospital closed in 2000, the other participating systems agreed to absorb the 5,000 patients the hospital had committed to serving.
  • Securing additional funding after the initial grant period: The Executive Committee and VODI central office have continued to work together to secure additional funding after the Kellogg initiative ended and other funding ran out.

Resources Used and Skills Needed

  • Staffing: During the grant period, the VODI central office had between 15 and 20 full-time employees, including an executive director, data manager, administrative assistant, two peer navigators, and nurse practitioners for the disease management program. Each health system had one or two enrollment specialists on staff funded by the VODI grant. Since the initial grant ended, the VODI central office has five staff and some outside consultants.
  • Costs: The annual budget during the grant period ranged between $1 million and $2 million. Since 2005, the annual program budget has varied from $400,000 to $500,000.
begin fs

Funding Sources

Health Resources and Services Administration; Michigan Department of Community Health; W.K. Kellogg Foundation; American Legacy Foundation; Kresge Foundation; Ascension Health; Health Resources and Services Administration, Beacon Communities Program
end fs

Adoption Considerations

Back to Top

Getting Started with This Innovation

  • Set manageable goals: Although the partners agreed to a citywide approach, they did not want to be overwhelmed with new patients. Consequently, rather than launching an ambitious enrollment drive through community organizations and primary care clinics, the partners decided to enroll only patients who presented at EDs. They also agreed to first determine eligibility for other insurance programs before enrolling patients in VODI. This realistic approach made it easier for partners to agree on the scope of the program and facilitated more complete data tracking and enrollment processes.
  • Make sure system has capacity to serve patients: An insurance or enrollment card does not guarantee access to care. Rather, the system must have the capacity to serve patients across the full continuum of care. By strengthening the collaboration among health care systems, VODI extended the reach of already existing services and encouraged the development of new ones to meet the needs of enrolled patients.
  • Expect to spend time and money: Successful community building and collaboration takes time and resources, with few "short cuts" available to hasten the process. Each partner paid a modest amount of dues to participate in the collaboration.
  • Get right people at table: Ensure that those creating the committee structure and serving on the various committees are senior enough within their home organizations to make decisions and have the support—and the ear—of their CEOs.
  • Prepare for influx of patients to government insurance plans: In the first few years of the program, nearly 60 percent of uninsured patients screened in the EDs ended up being eligible for and enrolled in government-sponsored insurance programs, such as Medicaid or Wayne County's Adult Benefit Waiver Program (called PlusCare). As a result, PlusCare quickly reached capacity and had to turn away eligible patients. For the next few years, VODI enrolled nearly 90 percent of screened patients.4

Sustaining This Innovation

  • Add partners as necessary: Initially, the FQHCs did not have representation on the executive or oversight committees. Roughly 6 months into the grant, the partners decided to invite the FQHCs to the table. As the number of FQHCs and similar sites expanded, these organizations formed their own workgroup that reports monthly to the executive committee.
  • Take up slack as necessary: As noted, when Mercy Hospital closed, the other hospitals absorbed the 5,000 VODI enrollees the hospital had committed to serving.
  • Promote continuity as much as possible: During the 13 years since VODI began, multiple changes have occurred in the senior management of participating health systems. However, most of those involved in the initial grant-writing process still participate in the program in some capacity. This continuity—as well as the regular reports to CEOs and other senior management—ensures ongoing commitment to the program.

More Information

Back to Top

Contact the Innovator

Lucille Smith, M.Ed.
Executive Director
Voices of Detroit
4201 St. Antoine
University Health Center- 9D #324
Detroit, MI 48201
Phone: 313-832-4246

Innovator Disclosures

In addition to funding for its programmatic offerings, Ms. Smith reports that the Voices of Detroit Initiative has received additional funds from the Kellogg Foundation to cover travel expenses and expenses from marketing a book about the program.

References/Related Articles

Chesney JD, Smitherman HC, Taueg C, et al. Taking Care of the Uninsured: A Path to Reform. Wayne State University, 2008.


1 U.S. Census Bureau, 2010 American Community Survey.
2 Institute of Medicine. America’s Uninsured Crisis: Consequences for Health and Health Care. Washington, DC: The National Academies Press, 2009.
3 American Hospital Association. Uncompensated Hospital Care Cost Fact Sheet. January 2012. Available at: (If you don't have the software to open this PDF, download free Adobe Acrobat ReaderĀ® software External Web Site Policy.).
4 Minich-Pourshadi. Five Healthcare Leaders Weigh-In on Rising Charity Care. HealthLeaders Media, February 15, 2010. Available at:
5 Chesney JD, Smitherman HC, Taueg C, et al. Taking Care of the Uninsured: A Path to Reform. Wayne State University, 2008.
Comment on this Innovation

Disclaimer: The inclusion of an innovation in the Innovations Exchange does not constitute or imply an endorsement by the U.S. Department of Health and Human Services, the Agency for Healthcare Research and Quality, or Westat of the innovation or of the submitter or developer of the innovation. Read more.

Original publication: September 26, 2012.
Original publication indicates the date the profile was first posted to the Innovations Exchange.

Last updated: July 02, 2014.
Last updated indicates the date the most recent changes to the profile were posted to the Innovations Exchange.

Date verified by innovator: June 20, 2014.
Date verified by innovator indicates the most recent date the innovator provided feedback during the annual review process. The innovator is invited to review, update, and verify the profile annually.