In this issue, Olfson and Marcus document a sea change in the provision of mental health services during the decade spanning 1998 to 2007 (1). Using data from the nationally representative Medical Expenditure Panel Survey, the authors analyze trends in service use, the mix between psychotherapy and pharmacotherapy, and spending on psychotherapy services in the United States. The results point to a major shift in mental health services delivery away from psychotherapy and toward psychopharmacology.

Although the study focuses on trends in psychotherapy, these changes need to be understood in the context of the enormous expansion in use of psychiatric medications that occurred during the same years. The study reports that the number of people using any mental health care increased from 16.1 million to 23.2 million, continuing a trajectory of expanding use of outpatient mental health services that has been evident for at least 30 years (2). This increase was confined almost entirely to rising rates of pharmacotherapy, with the proportion of treated individuals who received only psychiatric medications increasing from 44.1% in 1998 to 57.1% by 2007.

A confluence of factors likely drove the growing use of pharmacological treatment during this decade. In specialty settings, managed care organizations had an incentive to substitute medications for psychotherapy, in part because they were not required to pay for drugs under capitated contracts (3). In primary care settings, availability of new medications, particularly selective serotonin reuptake inhibitors, made it easier for generalist physicians to treat common depressive and anxiety disorders. Across all settings, marketing efforts by the pharmaceutical industry helped drive both provider behavior and consumer demand. During that decade, spending on drug promotion grew at an annual rate of more than 10%; with new guidelines promulgated by the Food and Drug Administration in 1997, this spending for the first time included direct-to-consumer advertising of drugs on television and other broadcast media (4).

As Olfson and Marcus show, this expansion of psychopharmacology came in part at the expense of psychotherapy. While population-based rates of receipt of any pharmacotherapy stayed constant over time, the chance that a patient in mental health treatment would receive psychotherapy declined, along with a substantial reduction in the intensity of psychotherapy among those in treatment. Perhaps most telling are the figures on national costs for psychotherapy. During a decade in which national health costs increased by 88% (5), expenditures for psychotherapy declined by more than a third, from $10.94 billion to $7.17 billion (1). This change was driven by both a decrease in the average number of psychotherapy visits and a decline in average cost per psychotherapy visit.

What are the clinical implications of the growing emphasis on pharmacological treatments for mental disorders? The Medical Expenditure Panel Survey (6) defines psychotherapy very generally, and psychotherapy as delivered in routine practice rarely represents the evidence-based treatments demonstrated to improve care. Thus, it is safe to assume that rates of guideline-concordant therapy in both years are likely to be considerably lower than those reported in the study. This lack of data on the content of psychotherapy in this survey also makes it difficult to assess whether, or in what cases, declining intensity indicates problems with quality of care.

However, the growing emphasis on psychopharmacological treatment in both primary care and specialty mental health settings suggests that the range of options available to patients for mental health treatment is declining. Recent meta-analyses have raised questions about the benefits of antidepressant therapies used alone in treating less serious depressive disorders (7). Many patients prefer psychotherapy as a first-line treatment, and some of the most successful trials of collaborative approaches to care improvement have given patients a choice between psychotherapy and pharmacological treatments (8). Psychotherapy may have particular benefits when used in conjunction with pharmacotherapy, providing patients with long-term strategies for coping with symptoms and stressors, while also promoting adherence to medication treatments (9). Taken together, this research suggests the importance of preserving a range of both psychopharmacological and psychotherapeutic options for patients and providers.

How might these trends change in the coming years? Two new laws—health reform and mental health parity—could work synergistically to reduce financial barriers to receipt of psychotherapy and other specialty mental health services. The Patient Protection and Affordable Care Act, signed into law in March 2010, will decrease the pool of uninsured Americans through the development of health insurance exchanges and an expansion of eligibility for Medicaid to 133% of the federal poverty line. The Paul Wellstone and Pete Domenici Mental Health Parity and Addiction Equity Act of 2008, which took effect in July 2010, will expand the scope of previous state and federal laws and will extend protections to self-insured plans that were exempt from previous federal and state benefit regulations. It prohibits differences in coverage for mental disorders, including "nonquantitative" mechanisms such as prior authorization and utilization review, that may have presented particular barriers to use of psychotherapy services.

However, these two laws alone will likely be insufficient to address the underlying forces that have led to the declining emphasis on psychotherapy as a treatment for mental health disorders in the United States. For patients, the extensive marketing structure for pharmacotherapy has no counterpart for psychotherapies. Stigma, or a belief that these treatments are not effective, could keep patients from seeking out psychotherapy even in the face of broadened insurance coverage. Even if patient demand increases, the dearth of practitioners who can provide evidence-based psychotherapies is likely to remain a bottleneck for delivery of these treatments. This is likely to be a particular concern for primary care physicians, who receive little formal training in psychotherapy. And health and mental health systems are still far more oriented to the delivery of medications and procedures than to cognitive and psychosocial treatments.

In the U.K., recognition of the underuse of evidence-based psychotherapies led to a national initiative to increase availability of those treatments, the Improving Access to Psychological Therapies (IAPT) program (10). This program includes a range of strategies—including workforce expansion, standardized training in evidence-based psychotherapies, and use of novel treatments such as computer-based psychotherapy programs—to ensure that patients are able to receive psychological therapies in primary care settings. The IAPT could serve as a model for an effort to promote the more widespread use of appropriate psychotherapies in the United States. Whether or not the United States is willing to adopt this sort of comprehensive strategy, policymakers, advocacy groups, and clinical leaders should continue to monitor these trends and work to ensure that patients continue to have access to a range of both psychological and psychopharmacological therapies in the coming years.