Competition and the Mental Health System

The role of managed care in behavioral health care has increased substantially, as it has in general health care. By the end of the 1990s, almost 80% of the privately—and publicly—insured received their behavioral health care through a managed care system (3) . By 2004, 80% of state Medicaid programs continued to use managed care for behavioral health (4) .

A unique feature of mental health is that managed care plans often separate mental health benefits from general health care benefits and subcontract to specialized managed behavioral health organizations (MBHOs), called carve-out plans. Somewhat more than half of managed care plans contract externally for the management of mental health benefits (5) . MBHOs have established specialized networks of psychiatrists, psychologists, master’s-level therapists, and other mental health specialists whose treatment philosophies are consistent with those of the MBHOs (6) . In Medicaid, MBHOs have more comprehensive services than general managed care plans (7) and in some cases are operated by county or local mental health authorities (4) . One-third of state Medicaid plans used MBHOs in 2000 (7), but this number appears to have declined, with 10 states reporting using MBHOs in 2004 (4) .

While many plans and Medicaid programs use MBHOs, they do not appear to enroll most of their clients in them. Sixteen percent of workers who have health insurance coverage were in plans that carved out mental health benefits in 2002 (8), while MBHOs covered roughly 15% of the Medicaid population (9) .

To select MBHOs, private and public purchasers use competitive procurements. The procurement process involves MBHOs competing for contracts on the basis of their reputations, fees, provider networks, and other factors. Having only a few MBHOs available can create challenges for purchasers. The number of national MBHOs active in Medicaid procurements declined to only three in 2000, as a result of mergers and of more stringent procurement provisions and perceived “underfunding” (10) . In fact, state and county purchasers often avoid competitive bidding for MBHOs altogether by setting capitation rates administratively rather than soliciting bids.

One reason purchasers consider MBHOs rather than integrated managed care plans is the presence of “adverse selection.” Adverse selection exists when individuals who use behavioral health services select health plans with more generous benefits (6) . Consequently, plans that offer generous benefits enroll a population that is sicker on average. Research has found this to be of particular concern for certain services, including mental health. Adverse selection gives plans the incentive to structure benefits, create limited specialty provider networks, or otherwise limit access to avoid high-risk consumers (11) . Thus, adverse selection leads competing plans to focus on avoiding high-cost enrollees rather than on improving efficiency.

One way for purchasers to address the adverse selection problem is to limit enrollees to a single MBHO. Alternatively, purchasers can contract with health plans for management services, but not on a risk or capitation basis. Furthermore, several Medicaid programs have contracted with managed care for general health care but included only limited mental health benefits, leaving remaining mental health benefits covered through fee-for-service (4) . Each approach implies different incentives to control costs and enroll patients. The evidence on either private or public purchasers’ experience with these strategies is limited and dated (12, 13) .

Managed care plans in the general health sector successfully reduce prices and use by increasing competition among providers (14) . Managed care plans introduced “selective contracting,” i.e., the formation of limited provider networks, including only those hospitals and physicians willing to negotiate lower prices. If managed care plans can make credible threats to exclude certain providers from their networks, providers have the incentive to negotiate lower prices in return for inclusion in the network and, thus, a higher volume of patients.

In contrast to integrated managed care plans, MBHO networks tend to be large, in order to meet the “best practice/performance standards” established by the industry, e.g., in urban areas only 10 miles to practitioners and 25 miles to facilities (15) . There is little research on the network-formation strategies used by MBHOs, but anecdotal evidence suggests that these plans do not rely on limiting provider networks to induce price competition among individual providers. While MBHOs negotiate reimbursement rates with hospitals and substance abuse treatment programs, they set fixed reimbursement rates by discipline (psychiatrist, psychologist, master’s-level social worker, nurse, etc.) in a given geographic area (16) .

These differences in provider network formation may attenuate the degree of price competition among providers of mental health services relative to providers in general health care. Nonetheless, between 1993 and 1999, MBHOs decreased their reimbursement rates for 50 minutes of psychotherapy by approximately 14% ($104.43 to $90) to psychiatrists, 14% ($82.27 to $71) to psychologists, and 9% ($67.07 to $61) to master’s-level providers (15) .

One impact of MBHOs has been on the mix of mental health services. A mental health network introduced by a national employer in 1996 increased the probability of initiating mental health treatment but shifted care from psychiatrists to nonpsychiatrist specialists (17) . Similarly, a study of a Medicaid MBHO found increases in access but lower intensity of services (18) . Overall, MBHOs have used multiple techniques to ration care (such as care management and provider profiling) and have been credited with reducing behavioral health costs by 30% to 48% percent in the private sector and by 17% to 33% in the Medicaid program (6, 19) .

The role of specialty versus general or primary care providers in the delivery of behavior health services is changing, with important implications for competition policy toward inpatient and outpatient care.

Another important tendency is the increase in outpatient behavioral health care provided by master’s-level therapists and primary care physicians rather than psychiatrists. In 2003, individuals with mental disorders were more likely to have been seen by a general medicine practitioner (22.8%) than a nonphysician, mental health specialist (16.0%) or a psychiatrist (12.0%) (25) . The shift away from care by psychiatrists has been observed for adults with serious mental illness, for persons with less severe mental conditions, and especially for children and youth (26, 27) .

According to recent national data, these sectors differ in treatment quality. Close to half of patients in the specialty mental health sector received at least minimally adequate treatment, compared to only 13% in the general medical sector, while differences between psychiatrists and nonphysician mental health specialists were very small (25) . Consumers may prefer nonpsychiatrists owing to their geographic proximity. Patients face shorter travel times to general physicians and nonphysician specialists than psychiatrists, and distance affects which provider type they choose (17) . One policy issue is whether consumers knowingly make trade-offs between price (which would include their travel time) and quality of care or whether lack of information affects their decision making, leading them to weigh price more heavily than quality.

The entry of more nonpsychiatrists puts pressure on psychiatrists’ salaries. Recent data show that median incomes of adult psychiatrists rose only 2% from 2001 to 2007, compared to 10% for other specialist physicians (28). Over the same time, median salaries for entry-level doctoral-level psychologists rose 23% (29, 30). Even among psychologists, recent increases in salary have slowed relative to earlier years.

Psychiatrists prescribe only one-third of psychotropic medications; primary care physicians and other providers prescribe the remaining two-thirds (31). Nonetheless, current policies continue to create practice barriers for nonphysicians, including state laws, restraining who may prescribe medication other than physicians. As of 2006, only 14 states permitted nurse practitioners to prescribe independent of physician involvement (32). The Bureau of Labor Statistics projects that the number of nonphysician, mental health providers will continue to grow by 23% over the next 10 years, which may put further pressure to redefine the relative roles of psychiatrists and other mental health specialists, including who can prescribe medications.

The evidence suggests that competition spurred the development of new forms of mental health care financing and changes in the delivery of mental health services. MBHOs surpassed general managed care plans in improving access and reducing costs. They also helped to address important adverse selection challenges. More recently, Medicaid programs have moved away from MBHOs toward integrated plans. As Medicaid moves away from MBHOs, policy and research should pay attention to the renewed possibility of adverse selection and whether competing managed care plans are limiting access to mental health services in order to avoid high-risk consumers.

It is difficult to generalize from the Medicaid experience to the private insurance market because Medicaid differs from private insurance in important ways, including the health status of the eligible population and its historically low payment rates. But no recent studies document whether MBHOs’ role in the private sector has waned or not. Do they continue to meet the private market test?

The evidence, primarily from the 1990s and frequently anecdotal, suggests that MBHOs played a significant role in reshaping mental health, in part by shifting care away from inpatient settings, increasing access to outpatient care, and promoting alternatives to psychiatrist’s care that are less costly. Selective contracting does not appear to be the primary mechanism through which price competition was generated. Carve-out plans, relative to fee-for-service or integrated managed care plans, achieved greater cost savings primarily through utilization management and shifting care to lower-cost outpatient settings where possible. Where do we stand today? How do integrated managed care plans contract for mental health care? Have they have been able to replicate efficiencies created by MBHOs? What is the nature of price and quality competition? These questions suggest a need for a new generation of studies by researchers interested in understanding how to control costs and enhance quality in behavioral health care.

Inpatient care, whether in general health care or in psychiatric care, is costly. Consequently, policymakers should pay particular attention to inpatient care changes, including consolidation and shifts in care across hospital type. Research should address the impact of such tendencies and whether they are leading to socially desirable outcomes in behavioral health.

There are tremendous opportunities to learn how changes in the price and quality of behaviorial health have been affected by policy environments. To what extent does providing quality information to consumers or introducing pay-for-performance programs change the competitive landscape for mental health providers? Some of these policies have been studied in the context of general health care, but only sporadic studies exist with respect to mental health.

From Columbia University and RAND Corporation and Smith College. Address correspondence and reprint requests to Alison Evans Cuellar, Department of Health Policy and Management, Mailman School of Public Health, 600 West 168th Street, 6th Floor, New York, NY 10032; [email protected] (e-mail). Commentary accepted for publication October 2008 (doi: 10.1176/appi.ajp.2008.08091383).

The authors report no competing interests.

The authors thank Saul Feldman, Howard Goldman, Sherry Glied, Margarita Alegria, and Thomas McGuire for their comments on an earlier draft and Rachael Wojnowicz for excellent research assistance. All errors remain our own. This work was supported by the John D. and Catherine T. MacArthur Foundation and Robert Wood Johnson Foundation grants to the Network on Mental Health Policy Research as well as grant K01 MH067086 from the National Institute of Mental Health (Dr. Cuellar, principal investigator).