The good news about selective serotonin reuptake inhibitors (SSRIs) is that they are effective and well received by patients and clinicians. When properly used they can make an enormous contribution to well-being. At times they make the difference between life and death. The bad news is that in 2000 more than $7 billion was spent on SSRIs in the United States (1).
Management of SSRI use presents an important opportunity for increasing value—that is, the yield of quality per dollar spent. The concept of an unmanaged pharmacy benefit program is a chimera. The combination of marketing by drug companies to consumers and physicians, the absence of price controls, and the steady emergence of "me-too" drugs means that affordability requires management. As in other areas of managed care, the question is not whether management is necessary but how to manage wisely.
We recently proposed an "ethical template" as a decision guide for enhancing value fairly in managing pharmacy benefits (2). The template argues that for drugs that have comparable profiles of risk and benefit, it is ethically justifiable to favor the least costly alternative as long as individual patients are protected by a fair and efficient process for dealing with exceptions.
This column, the second in a series describing strategies for managing psychiatric drug use that promote access, high-quality care, and acceptable costs (3), draws lessons about enhancing value from Kaiser Permanente's approach to managing the use of SSRIs. Kaiser's experience suggests that the decision process recommended in our ethical template is not ethical pie in the sky but can be applied effectively in real-world settings.
Kaiser Permanente is a group-model health maintenance organization serving more than 8 million members nationally. Kaiser Foundation Health Plan, Inc., the insurance and facility management arm of the organization, contracts exclusively with regional Permanente medical groups, the physician-governed group practices that provide services to Kaiser members. The majority of members—a total of 6 million—are in California.
In 1997 Kaiser Permanente identified improvement of the diagnosis, treatment, and prevention of depression as a central goal and to that end launched a multipronged depression initiative (4). One component of the initiative is achieving optimal use of antidepressant medications from the dual perspectives of quality of care and cost-effectiveness.
In the early to mid-1990s, as evidence began to emerge that the higher cost of SSRIs compared with tricyclic antidepressants was offset by lower costs for outpatient and inpatient care (5), Kaiser Permanente designated SSRIs as the preferred antidepressant medications for initial prescriptions. The rationale for this policy was that comparable efficacy, fewer adverse effects, reduced lethality in overdose, physician satisfaction (6), and comparable overall costs made the SSRIs preferable.
Until fluoxetine became available in generic form in 2001, the branded forms of citalopram, fluoxetine, paroxetine, and sertraline battled for market share. As in the wider community, physicians and patients in the Permanente medical groups had individual preferences among the SSRIs. The medical groups asked themselves, "Should each of us make SSRI choices for ourselves, or should we establish priorities among the drugs? And, if we set priorities, on what basis should we do it?"
The medical groups conducted an internal study of medication switching and found that the rate of switching to other agents was the same for all four SSRIs. Given that finding and the absence of persuasive evidence that any of the four SSRIs systematically produced superior outcomes, the medical groups decided that cost was an appropriate basis for establishing preferences among the competing choices. Kaiser Permanente's national pharmacy purchasing program negotiated advantageous pricing arrangements for fluoxetine and paroxetine, which were made the agents of choice within the group practices.
In 2001 Eli Lilly's patent on fluoxetine expired and generic alternatives began to become available. In the same year, a nine-month randomized trial by Kroenke and colleagues (7) that compared fluoxetine, paroxetine, and sertraline in real-world primary care practice showed that the drugs did not differ significantly in effectiveness across a broad range of outcome measures. In the editorial that accompanied the study report, Gregory Simon (8), a distinguished psychiatric researcher at the Group Health Cooperative Center for Health Studies in Seattle, commented that because clinicians selecting an antidepressant for patients who have not previously taken an SSRI "cannot reliably predict that one drug will have greater effectiveness, fewer adverse effects, or fewer risks…an initial choice based on prescription costs is prudent, ethical, and clinically reasonable."
In light of the Kroenke study, the Simon editorial, and clinical data from the Permanente medical groups, Kaiser implemented a "fluoxetine first" campaign in 2001. When the program began, approximately 30 percent of first-time SSRI prescriptions were for fluoxetine. A year later, the figure was 80 percent.
Kaiser Permanente bases its drug management process on three factors: the individual patient's clinical needs (as determined by the treating physician), evidence about comparative drug effectiveness, and cost (9). It conducts a careful, evidence-driven process to decide which drugs to include in its formulary. It has developed extensive educational programs, clinical guidelines, and decision tools about drug choices for physicians in the medical groups. But in its own view (9) "the most important formulary [management] element is the Permanente physician's ability to override and use a non-formulary medication if he or she feels that it is medically necessary for a specific patient."
For Permanente physicians, there is no prior authorization process for SSRIs. For example, if a physician wanted to prescribe sertraline as a first prescription for a patient whose twin had done poorly on fluoxetine but well on sertraline, the physician would simply have to write an exception code on the prescription. The patient would not have to make a higher copayment or wait for adjudication. The physician would experience no significant administrative burden. The process is hassle-free for both physician and patient.
However, the process is not one of unrestricted autonomy for the prescriber. Prescribing patterns are reviewed at clinical meetings. Discussion and debate occur regularly. As described by Robin Dea, M.D., who is chair of the chiefs of psychiatry for the Permanente Medical Group of Northern California, "We manage by education, peer pressure, and physician leadership, not rigid rules. We say, 'Here is how we want to do it,' but we understand that there will be exceptions. We use the power of influence rather than time-consuming procedures that take physicians away from patient care" (personal communication, August 2003).
Kaiser invests its managerial energy in increasing the use of generic fluoxetine in first-time prescriptions, but it does not seek to change the regimens of patients who are stable and doing well on nonpreferred agents. However, if a patient who was currently not taking medication but who had been successfully treated with one of the other SSRIs before the patent on fluoxetine expired developed a new episode of depression, a Kaiser Permanente physician would probably discuss the data on medication equivalence with the patient and ask whether he or she would be willing to use generic fluoxetine. If the patient was unwilling, the physician would be free to prescribe the original agent.
Experience at Kaiser Permanente suggests that evidence-driven deliberation about drug benefit policy within a clear framework of values is a crucial ingredient in improved pharmacy management. Kaiser has identified treatment of depression as a central objective. It uses managed care techniques to facilitate access to antidepressant medication, not to create barriers. However, it pays tremendous attention to selection of antidepressant drugs. When strong evidence suggested that increasing the use of generic fluoxetine could achieve significant savings without creating undue burdens or risks for patients, Kaiser initiated its robust "fluoxetine first" program. Within 12 months the proportion of first prescriptions for fluoxetine rose from 30 to 80 percent.
In accordance with a model prior-authorization process for drug benefit management that was recently proposed (10) and the ethical template for pharmacy benefits that we have described previously (2), Kaiser explicitly and openly includes cost as a factor in setting priorities for drug choices. It uses the full force of education and peer pressure to encourage adherence to its policies. Within a framework of open collegial scrutiny of prescribing practices, it provides a hassle-free process for making exceptions to its policies.
Improving psychiatric drug benefit management requires trust. Individuals with primary responsibility for stewardship of shared resources must trust frontline clinicians to use good judgment motivated by concern about each of their patients as an individual. However, this does not mean unbridled autonomy for the individual whose prescribing practices tap a common purse. Those with primary responsibility for frontline care of individual patients must trust the good intentions of those accountable for overseeing resource use for the wider population. However, this does not mean that costs and the needs of the population inevitably trump individual needs.
This formula builds in tension and ambiguity. Mental health professionals, however, understand better than most the psychic and organizational challenges posed by competing interests. They also understand the dangers of "splitting" and possible ways to avoid it. Kaiser Permanente's "fluoxetine first" program exemplifies the kind of clinically informed, ethically sensitive, fiscally prudent drug benefit management our health care system needs. Other programs—both public and private—should emulate it.
The authors thank David Campen, M.D., and Robin Dea, M.D.
Dr. Sabin, who is editor of this column, is clinical professor of psychiatry at Harvard Medical School and director of the ethics program at Harvard Pilgrim Health Care in Boston. Dr. Daniels is professor of ethics and population health in the department of population and international health at the Harvard School of Public Health. Send correspondence to Dr. Sabin at the Department of Ambulatory Care and Prevention, 133 Brookline Avenue, Sixth Floor, Boston, Massachusetts 02215 (e-mail, firstname.lastname@example.org).