Major changes need to occur if we are serious about improving mental health care. A recent national study by McGlynn and colleagues (1) found that only 58 percent of persons treated for depression and 11 percent of persons with substance use disorders received the recommended care. In response to this report, Steinberg (2) suggested four actions that will improve care: first, quality of care should routinely be measured and reported; second, clinicians must make greater use of information technology; third, consumers must take greater responsibility for their health care; and fourth, new incentives are needed to create a "business case" for quality improvement. It is this fourth action, making a business case for quality improvement, that is the focus of this column. Using data from a study of adherence to well-accepted best practices—practices that are in keeping with the strongest evidence about what works and why—we, a research group at Harvard Medical School, offer evidence that the use of financial incentives can promote best practices that will benefit patients and health systems.

How can anyone disagree with the implementation of best practices? Although it is easy to acclaim best practices, implementing best practices in clinical settings involves changing organizational or clinician behavior. Changing the status quo has proven to be far more difficult than first thought. In a literature review, Cabana and colleagues (3) identified a number of potential barriers to bringing practice in line with evidence-based recommendations: lack of familiarity with or awareness of recommended practice, lack of belief in or agreement with best-practice guidelines, inconsistency between best-practices recommendations and past training and current routines, and clinician-acknowledged inability to change their own behaviors.

Recently, the National Institute of Mental Health increased its emphasis on developing mechanisms for transferring research-based knowledge into practice settings. This change in focus has encouraged the testing of several different approaches to improving how care is delivered. Strategies for shifting to best practices have included reeducation through lectures and conference presentations, mailings, academic detailing, and electronic reminder systems (4,5,6,7,8), but there is little evidence that these strategies have succeeded in changing clinical behavior. One strategy that has shown some success is integrating professionals who are trained in specific educational and clinical practices into existing treatment teams (9,10).

One approach has not been studied systematically, but appears to hold promise. This strategy uses financial rewards and penalties for administrators of behavioral health carve-out plans. These administrators can then use their leverage with providers to bring about change. One example of this is the contractual relationship of the Medicaid agency in Massachusetts and its behavioral health managed care vendor. The terms of the contract between these two organizations detail specific financial rewards if the managed care vendor implements certain best practices that meet the standard set by the Medicaid agency.

In a recent study of guideline adherence (11) we examined whether financial incentives resulted in greater adherence to best practices for 420 adult high-risk, vulnerable patients with schizophrenia. The data come from medical records of Medicaid beneficiaries enrolled in a managed care plan (N=94) and in a fee-for-service plan (N=60). We chose four different best practices that had been listed in the managed care contract, each with financial incentives for performance.

We found statistically significant differences in practices between the two groups. Best practices were more likely to be documented in the charts of patients in the managed care group than in the charts of the fee-for-service group: aftercare plans before discharge from the hospital (98 percent of managed care patients compared with 90 percent of fee-for-service patients), a follow-up appointment scheduled with a physician after discharge (93 percent compared with 82 percent), contact with an outpatient clinician before discharge, (89 percent compared with 80 percent), and contact with the primary care clinician during the psychiatric inpatient episode (50 percent compared with 43 percent). Patients from both types of health plans were treated in the same hospitals; so we concluded that managed care contracting, not hospital-specific policies, was responsible for the differences that we found. We did not check to determine whether actual clinician behavior was consistent with the medical record data, so it is possible, but we believe unlikely, that these practices did not actually occur.

In addition to the managed care plan having financial incentives, the state Medicaid agency also set nonfinancial performance standards for network providers who provided acute treatment for inpatients, including therapeutic programming—for example, vocational assessment, individual and group psychotherapy, family evaluation and therapy, psychiatric and medical evaluation, pharmacological services, substance abuse evaluation, and other psychosocial services. Data from our study provided evidence that managed care inpatients were as likely or more likely than fee-for-service beneficiaries to have most of the specified services provided.

Performance standards are not new in contractual arrangements, but they are not often discussed in the scientific literature. There are a number of possible reasons. First, to clinicians these contractual standards are often seen only as an exchange between financial officers of the contracting organizations. Second, it appears that despite the vast changes in health care in the last few decades, talk about money remains unseemly, even unprofessional to clinicians. Third, clinicians are likely to see themselves as working in an entirely different world from the "suits"—in the clinicians' world, altruistic motives underlie every action. Admittedly, describing the clinicians' world view as a rather naive conception of human behavior is perhaps unfair. Certainly, we can all appreciate the slippery slope that begins with incentives to improve care through the implementation of certain acknowledged best practices and descends to financial rewards that only come when clinically sound treatment is compromised. One person's definition of justified financial rewards is another person's definition of greed, and the line between the two is not always clear. In addition, discussions about financial incentives to change clinician and organizational behavior may reinforce the belief that the health care system is driven by money and is undermining professional altruistic values.

However, concerns about implementing financial incentives need to be balanced with some understanding of how difficult it can be to overcome a lifetime of training that teaches clinicians particular methods of treatment and instills values of autonomous decision making. Furthermore, organizations have long-standing operating policies that are difficult to change, policies that make changing routines not just difficult, but often expensive as well. With responsible negotiations, opportunities exist for revenue enhancement to go hand in hand with improved care.

In the case we have provided, it was the managed care company, not the hospitals, that reaped the financial rewards. In the long range it will be in the best interests of everyone to find mechanisms whereby the reward system for managed care organizations to ensure best practices is converted into "carrots," not "sticks," for the hospitals. There is concern in practice settings as well as medical schools about the incompatibility of the invisible hand of the market and the caring hands of clinicians (12). The case example that we have offered illustrates how patients, providers, and managed care companies can all win and demonstrates the potential benefits of responsible managed care.

The obstacles to improving quality are not trivial—nor are they insurmountable. Brennan and Berwick (13) cautioned that "within medicine today, it still takes immense courage to insist that fundamental improvement is achievable, to demand it, and to accept the consequent changes." Clinical leaders must demand that care meets best practices, develop up-to-date information systems to support quality improvement efforts, and create financial support for these activities. However, at the end of the day neither government regulation nor financial rewards can ensure that patients receive the best care that we know how to give. This standard, that is, the best we know how to give, must come from within the values and culture of clinicians. The challenge is to ensure that this culture includes willingness to seek and employ that which works best.

This work was supported by grant RO-1 MH54076 from the National Institute of Mental Health.

Dr. Dickey is affiliated with Cambridge Health Alliance, 1495 Cambridge Street, Cambridge, Massachusetts 02139 (e-mail, barbara_dickey@hms.harvard.edu) and the department of psychiatry at Harvard Medical School in Boston. William M. Glazer, M.D., is editor of this column.