Managed Care: Public-Sector Managed Behavioral Health Care: IV. Integrated Versus "Carve-Out" Care

Behavioral health programs are changing at a dizzying pace. Organizations affiliate, merge, purchase one another, fail, and multiply with the unceasing turmoil produced by a market economy. Key personnel are here today and elsewhere tomorrow. How should those concerned with high-quality behavioral health services assess these changes? What can advocates advocate for?

We have argued that managed care organizations should be accountable for showing that their policies and practices reflect efforts to promote appropriate patient care under necessary resource constraints in a reasonable manner (1). Accountability for reasonableness requires organizations to demonstrate that the organizational changes they make are justified in terms of patient care objectives and responsible stewardship of resources. Administrative convenience and financial margin are not evil objectives (2), but on their own do not make policies, practices, and reorganization reasonable.

In 1998 we began a study designed to allow contrast between two forms of public-sector managed behavioral health care—the integrated, or "carve-in," approach and the "carve-out" approach. In a series of three columns in this journal we used the relationship between the Massachusetts Division of Medical Assistance, the state Medicaid agency, and its behavioral health care vendor to study the carve-out approach (3,4,5). We intended to use Harvard Pilgrim Health Care, a not-for-profit health maintenance organization (HMO) serving approximately 40,000 Medicaid recipients among its 1 million Massachusetts members, as our opportunity to study integrated (carve-in) behavioral health care provided to the same population.

Since the inception of the project, however, Harvard Pilgrim Health Care developed in 1998 an affiliation with Neighborhood Health Plan, a highly regarded organization that specializes in the care of vulnerable, low-income populations. As of October 1999, Harvard Pilgrim Health Care itself will no longer contract with the state Division of Medical Assistance. Harvard Pilgrim Health Care is effectively carving out its own Medicaid program by transferring its Medicaid members to Neighborhood Health Plan.

But that is not the only change. Although Neighborhood Health Plan believes in principle in integrated care, from its inception the plan, given the exigencies of running a complex organization, has elected to carve out behavioral health services. In other words, Harvard Pilgrim Health Care, our intended example of integrated or carve-in care, carved out its Medicaid program to an organization—Neighborhood Health Plan—that itself carved out behavioral health care to Beacon Health Strategies, a privately owned, for-profit company formed in 1996.

We were initially stunned by these changes and wondered if we could carry out the project we had contracted to do. But after a period of confusion we realized that our bewilderment reflected a common response to the tumult in the field. This fourth column in the series on public-sector care uses our struggle to interpret the story of Harvard Pilgrim Health Care, Neighborhood Health Plan, and Beacon to address two issues—how to assess the rampant organizational changes of the late 1990s and the long-standing debate about carve-in versus carve-out programs.

Organizational mission and financial margin

In 1995 Harvard Community Health Plan, a not-for-profit staff- and group-model HMO, and Pilgrim Health Care, a not-for-profit independent practice association (IPA), merged to form Harvard Pilgrim Health Care. Although both organizations served Medicaid members, Harvard Community Health Plan, formed in 1969 as an urban practice, had an especially strong commitment to low-income populations. In its original mission statement of 1970, it promised "to serve all socio-economic strata and age groups." The most recent, 1998 version of the mission— "to improve the health of the people we serve, and the health of society"—continues that commitment, defining "society" as "all segments."

However, Harvard Pilgrim Health Care projected a loss of up to $10 million for its Medicaid enrollees in 1999, and more if the Medicaid membership grew. The management of Harvard Pilgrim Health Care concluded that providing high-quality care within the Medicaid capitation was especially difficult for IPA practices that had few Medicaid enrollees and lacked the infrastructure and experience needed to care for an economically disadvantaged population efficiently.

Although many insurers and managed care organizations nationally have withdrawn from the Medicaid market because of losses or narrow profit margins, Harvard Pilgrim Health Care was reluctant to do so because of its commitment to serving "all segments" of society. At the same time, it regarded the projected losses as unsustainable.

In an effort to solve its conflict between mission and margin, in January 1998 Harvard Pilgrim Health Care developed an affiliation with Neighborhood Health Plan, a small, not-for-profit HMO founded in 1986 by the Massachusetts League of Community Health Centers and the Greater Boston Forum for Health Action. In 1997 Neighborhood Health Plan had 48,000 members, and it had developed considerable expertise at serving a membership predominantly enrolled through Medicaid via a network of community health centers and what its literature called "other community-responsive providers." Neighborhood Health Plan and Harvard Pilgrim Health Care management concluded that Neighborhood Health Plan would be able to serve a larger membership within the projected Medicaid capitation without incurring losses and possibly with a small surplus.

Under the affiliation Neighborhood Health Plan became a wholly owned subsidiary of Harvard Pilgrim Health Care. We asked the chief executive officer of Neighborhood Health Plan, "How do you get running room to pursue your mission?" He responded that "we make sure we are fiscally viable, and Harvard Pilgrim Health Care is happy for us to break even" (Hooley J, personal communication, July 1999).

Carve in and carve out

Neighborhood Health Plan, with its origin in the community health center movement, applies a strong psychosocial orientation and an expansive definition of medical necessity to health care. In principle, it strongly prefers integrated or carved-in behavioral care. However, as a small organization dealing with substantial management challenges, it elected to carve out behavioral care, initially to Mental Health Management of America, which had been the first statewide Medicaid carve-out vendor in 1992 (6).

In 1996 Mental Health Management of America lost the Massachusetts Medicaid contract and ceased operations in the state. Neighborhood Health Plan put out a new request for proposals for its behavioral care. It selected Beacon Health Strategies, a newly formed company whose three top executives had been the leaders of Mental Health Management of America and consequently were well known to Neighborhood Health Plan. Based on trust that the new company was aligned in values and would provide reliable, high-quality services, Neighborhood Health Plan signed a three-year contract with Beacon on October 1, 1996.

When pressed to explain how they could reconcile their commitment to integrated care with the fact of a carve-out contract, the leadership of Neighborhood Health Plan cited six factors in addition to administrative expediency. First, because the Beacon leadership shares a similar background of community-oriented care, Neighborhood Health Plan was fully confident about congruence of vision and values. Second, the Beacon leadership was based in Massachusetts and brought a network of local knowledge and personal relationships to the work. Third, the contract shared financial risk and opportunities for savings—which constituted surplus for Neighborhood Health Plan and profit for Beacon—50-50 between the two organizations. Because unmet psychosocial needs are likely to increase medical and surgical costs and create additional demands for community providers, the shared-risk arrangement avoids any incentive for underservice.

Fourth, given the strong commitment to functionally integrated care, Neighborhood Health Plan and Beacon designed systems to encourage a high degree of collaboration between behavioral and general health clinicians and to monitor the results. Fifth, the two organizations have created a high degree of functional integration at the level of the care management process, as, for example, by having behavioral and general care managers meet weekly.

Finally, three years into their carve-out contract, Neighborhood Health Plan and Beacon are exploring how to carve back in the care of the 200 highest utilizers among the 7,000 members who received behavioral interventions through Beacon in the last 12 months. Neighborhood Health Plan has pioneered integrated care of members with AIDS and severe physical disability (7) and plans to work with Beacon to apply the same disease management approach it has used for those conditions to members with severe and persisting behavioral disorders.

Conclusions

We believe that the story about Harvard Pilgrim Health Care, Neighborhood Health Plan, Beacon, and Medicaid contains two important general lessons.

First, although the concept of "mainstreaming" care for low-income populations arises from an admirable commitment to equity and quality, in practice dealing effectively with the health impacts associated with low income require specialized skills, attitudes, and infrastructure. For Neighborhood Health Plan and Beacon, Medicaid is a mission, not a "product line." Both have created networks of clinicians skilled at and committed to caring for low-income populations.

Harvard Pilgrim Health Care, operating through a wider provider network, could not provide care within the funding provided by the state without incurring high and ultimately unacceptable financial losses. Its commitment to serving all segments of the population precluded simply withdrawing from Medicaid, as so many managed care organizations have done in recent years. Its decision to affiliate with Neighborhood Health Plan and carve out Medicaid to its wholly owned affiliate appears to meet the criteria for reasonableness. However, managed care organizations that evaluate Medicaid as a "business opportunity" and not as a crucial part of their mission, would almost certainly drop their Medicaid "product line" in similar financial circumstances.

Second, the time has come to put to rest the polarized debate about the relative merits of integrated or carved-in care as opposed to specialty or carved-out programs (8). This debate has never made sense to experienced clinicians. Anyone who has practiced in a solo fee-for-service setting, as the first author did from 1970 to 1980, knows that with a modicum of effort, it is possible to collaborate with general health clinicians, the social service sector, and other key participants in the care process. Anyone who has practiced in an integrated setting, as the first author has since 1975, knows that it is possible for clinicians in adjoining offices to collaborate poorly.

In the absence of outcome data showing that one organizational structure reliably produces better outcomes than another, we will continue to see the emergence of hybrid forms of the kind developed by Harvard Pilgrim Health Care, Neighborhood Health Plan, and Beacon. We should hold these new hybrids accountable for justifying their actions in terms of patient care objectives and responsible stewardship of resources and not judge them by preconceived notions about the right format for providing care.

Acknowledgments

The authors thank Christine Degan, Steven Friedman, James Hooley, Robert Master, Michael McGreal, Deborah Nelson, Michael Norton, Elizabeth Pattullo, Audrey Shelto, James Spink, and Donna Zeh for helping them learn about Harvard Pilgrim Health Care, Neighborhood Health Plan, and Beacon; Karen Brodsky, Eric Goplerud, and Stephen A. Somers for helping them learn about national policy issues; and the Greenwall Foundation, the Robert Wood Johnson Foundation Medicaid Managed Care Program, and the Substance Abuse and Mental Health Services Administration for their support.

Dr. Sabin, editor of this column, is associate clinical professor of psychiatry at Harvard Medical School and codirector of the Center for Ethics in Managed Care of Harvard Pilgrim Health Care and Harvard Medical School. Dr. Daniels is Goldthwaite professor in the department of philosophy and professor of medical ethics in the department of social medicine at Tufts University in Boston. Address correspondence to Dr. Sabin at Teaching Programs, Harvard Pilgrim Health Care, 126 Brookline Avenue, Suite 200, Boston, Massachusetts 02215 (e-mail, [email protected]).