Brighter Futures

Improvements in Depression Care Pay for Themselves

By Michael Schoenbaum, Cathy Sherbourne, Lisa Rubenstein, and Kenneth Wells

Michael Schoenbaum is an economist. Cathy Sherbourne is a public health specialist. Lisa Rubenstein is director of the Veterans' Administration Greater Los Angeles/UCLA/RAND Center for the Study of Healthcare Provider Behavior. Kenneth Wells is a professor of psychiatry at the University of California, Los Angeles. All four authors are RAND researchers.

Impressive new evidence shows that general medical practitioners can substantially reduce the individual suffering and economic consequences of depression by making modestly improved efforts to identify depressed patients, help them get proper treatment, and monitor their progress. The clinical benefits for individuals last over an extended period of time, and the economic benefits for society appear likely to exceed the costs by a comfortable margin.

We conducted a randomized trial of an innovative program, called Partners in Care, which increased the availability of effective treatments for depression. Under the program, doctors and patients retained full control over their treatment decisions, but they also received training and resources to help them improve the quality of the treatments delivered.

As a result of the program, the average time that patients spent in a clinically depressed state dropped by well over a month over a two-year time period — even though many of the patients in the program chose not to receive any treatments at all. Meanwhile, the average time that patients spent employed grew by about a month. The added cost of the improved care over the standard care for depression was less than $500 per patient. Therefore, the program yielded substantial benefits — better care, less suffering, more employment — while only modestly increasing costs.

Considering the benefits, $500 over two years is not very much. The value of the extra month of employment alone would easily exceed the cost of treatment. Anyone making just $4 an hour, well below the minimum wage, would earn $640 a month. Even if society — in the form of health maintenance organizations, health insurance plans, or government health benefits — bore the entire cost, the value to society of just the increased productivity and tax revenues would justify the cost.

Yet the true value of mental health is, of course, much greater than increased productivity. Indeed, many patients going through a bout of serious depression would be glad to pay $500 for even a single day of relief. That kind of willingness to pay, moreover, would not begin to capture the value of the relief for family members of depressed individuals.

A Common — but Hidden — Malady

Depression is one of the most pervasive and debilitating of the major chronic diseases. It is one of the leading causes of disability worldwide. In the United States, from 10 to 20 percent of all patients in primary medical care practices — which include general physicians' offices and community clinics — show significant symptoms of depression. Nationwide, about 5 to 10 percent of all patients in primary medical care practices are clinically depressed, meaning that they suffer from intense and often continuous feelings of sadness and hopelessness, compounded by other symptoms.

There are two types of clinical depression: major depressive disorder and dysthymic disorder. Major depressive disorder is a severe episode of depressed mood accompanied by other cognitive or physical symptoms, such as suicidal thoughts or changes in weight and sleep patterns. Major depressive disorder lasts at least two weeks. Dysthymic disorder, or chronic depression, has fewer symptoms but persists at least two years with only brief periods of respite. Patients who exhibit depressive symptoms but who do not meet the formal criteria for either depressive disorder are said to suffer from subthreshold depression.

The personal and social costs of the disease are heavy. Table 1 illustrates that depressive symptoms are more debilitating than many chronic medical conditions, including hypertension, diabetes, arthritis, and lung disease. Only serious heart disease (either myocardial infarction in the prior year or current congestive heart failure) and arthritis are associated with greater morbidity in any of the domains of functioning listed in the table. Compared to all of the chronic ailments except heart disease, depression imposes greater physical, social, and work limitations; more days in bed; and worse perceptions of overall health. Depression also imposes greater bodily pain than all the chronic ailments except arthritis.

Despite the prevalence and heavy toll of depression, primary care professionals typically neither detect nor treat it appropriately. They usually detect only about half of the cases of clinical depression that they encounter. At best, only about a third of the patients suffering from clinical depression receive appropriate care, either medication or psychotherapy (see Table 2). Even though effective medication and psychotherapy treatments exist, much of the care received by patients is known to be poor.

Part of the difficulty in detecting and treating depression is that fatigue, hopelessness, and passivity are part of the illness. In addition, patients often feel shame for having a mental illness. Consequently, they are unlikely to demand medical attention for their depression. They also may not expect general practitioners to know how to care for mental illnesses.

Indeed, another part of the problem is that general practitioners often lack the time, the training, or the access to specialists that are necessary to diagnose and treat depression effectively. As a result, many seriously depressed patients who do receive medical care are treated for a problem other than depression or receive the wrong treatment for depression. For instance, some depressed patients are prescribed sedatives, which are ineffective for depression.

A third difficulty is the mismatch between the unique needs of depressed patients and the typical processes of primary care practices. Depression, unlike other chronic illnesses, waxes and wanes. The needs of depressed patients change over time, depending on their unique clinical histories and life circumstances. Standard medical records, however, often capture limited information about patients' experiences with depression or mental health treatments. What primary care practices need are active, ongoing strategies to detect current major depression and to prescribe appropriate treatments.

A Noninvasive Approach

We designed our trial to confront the unique challenges of treating depression. Most clinical trials test a new drug or surgical procedure. Our trial, in contrast, tested new detection and management procedures. Most clinical trials randomly assign patients to experimental and control groups to judge the efficaciousness of individual experimental treatments. Our trial, in contrast, randomly assigned clinics to intervention and control groups to judge the effectiveness of experimental improvements in clinical care. Our trial involved 46 primary care clinics in six managed care organizations across the country. Control group clinics (one third of all clinics in the study) were observed without any changes to their care.

The intervention clinics received both clinical and financial support, including lectures for clinicians, meetings with patients to educate them about their options, follow-up meetings with patients concerning medication management, and reduced copayments for psychotherapy. Each participating clinic nominated a team of leaders — a doctor, a nursing supervisor, and a mental health specialist — to attend a two-day workshop and then to educate the other staff about how to implement the quality improvement programs. Nurses in the practices learned, for example, how to assess and educate patients by using patient checklists and brochures. We provided clinicians and patients with pamphlets, videos, manuals, and tracking forms in both English and Spanish and in formats suited to ethnically and socioeconomically diverse populations.

All the intervention clinics received the same training about effective treatments for depression. However, half of the intervention clinics received supplementary nursing resources specifically for medication management. The other half of the clinics received supplementary resources specifically for short-term, structured psychotherapy of 8 to 12 sessions. Not surprisingly, the patients in clinics with enhanced resources for medications were more likely to take medications, and the patients in clinics with enhanced resources for psychotherapy were more likely to undergo therapy.

Once we supplied the training and resources, however, we allowed the patients and clinicians to make their own treatment decisions. Our goal was to help the clinics help themselves by increasing the overall rate of effective care, regardless of the methods used.

About 1,350 patients enrolled in the study. We monitored them for two years. After the first year, they were 10 percent less likely to be clinically depressed and had better quality of life than patients in comparable clinics. Both treatments — medication and psychotherapy — appeared to have similar positive effects. After the second year, however, the benefits tapered off for patients in the programs that had emphasized medication, whereas the benefits persisted among patients in the programs that had emphasized psychotherapy. In both cases, the interventions were only short term, yet the benefits endured for at least a year, despite the recurrent nature of depression.

From a policy perspective, it was striking that 5 percent more of the patients in the quality improvement programs remained employed after 12 months as compared with their counterparts in the care-as-usual settings. Since depression reduces employment by about five percent, the Partners in Care program virtually negated the detrimental effect of depression on employment for at least a year. No other quality-improvement evaluation for any condition in primary care has shown that kind of positive employment boost.

In sum, we learned that it does not take expensive, elaborate, and mandatory treatment protocols or highly trained specialists operating in academic settings to make a big difference in the mental health, daily functioning, and job performance of depressed patients. Major progress along these lines can be made in everyday clinics when managed care practices implement modest, practical programs to improve the quality of care for depression.

Our study offers several hopeful messages. It shows patients that they can hope to improve their lives. It shows medical practices that they can help their patients once again contribute to society. It shows providers, employers, managed care companies, and insurers that they can hope for improved outcomes in functioning and employment given reasonable efforts to steer patients into appropriate treatments. And it shows that our findings may have broad applicability, given that the patients who participated were highly diverse and the practices in which they were treated were very typical.

A Clinincal Bargain

The program with extra resources for psychotherapy proved to be slightly more expensive than the program with extra resources for medication management, but the outcomes of the former program were also better. Compared to usual care, the average cost increase for clinics with the extra psychotherapy resources was $485 per patient over two years. The average cost increase for clinics with extra resources for medication management was $419. The costs included medications, facility charges, professional fees, and the forgone wages of patients.

On average, the patients in clinics that emphasized psychotherapy worked an additional 21 days more than patients in usual care, while patients in clinics that emphasized medication worked about 18 more days. Regardless of employment, the patients in clinics that emphasized psychotherapy suffered 47 fewer days burdened by depression over the two-year period. The patients in clinics that emphasized medication suffered 25 fewer such days (see Table 3).

The apparently greater effectiveness of the program with increased access to effective psychotherapy has important implications for public policy. Currently, insurance benefits for psychotherapy are often relatively restrictive, because it is considered more expensive than medication-based treatment. However, if programs offering greater access to psychotherapy are more effective, and if their benefits last longer — as our evidence indicates — then such programs may have similar or even greater cost-effectiveness than medication-based treatments. This finding seems generally inconsistent with the recent trends to reduce psychotherapy coverage.

Overall, our study found that modest interventions to improve the quality of care for depression can substantially increase individual and societal welfare, even when initiated and implemented under realistic conditions that promote the treatment choices of patients and clinicians. Because providers and insurers would normally incur the costs for these improvements, their widespread implementation may require either increased consumer demand or public policy incentives. Improved medical care can improve the quality of life for depressed patients and their families and communities — if we can create the conditions to put such programs in place.

Related Reading

"Cost-Effectiveness of Practice-Initiated Quality Improvement for Depression: Results of a Randomized Controlled Trial," Journal of the American Medical Association, Vol. 286, No. 11, Sept. 19, 2001, pp. 1325 - 1330, Michael Schoenbaum, Jurgen Unutzer, Cathy Sherbourne, Naihua Duan, Lisa V. Rubenstein, Jeanne Miranda, Lisa S. Meredith, Maureen F. Carney, Kenneth B. Wells.

"The Design of Partners in Care: Evaluating the Cost-Effectiveness of Improving Care for Depression in Primary Care," Social Psychiatry and Psychiatric Epidemiology, Vol. 34, 1999, pp. 20 - 29, Kenneth B. Wells. Also available as RAND/RP-761, no charge.

"Evidence-Based Care for Depression in Managed Primary Care Practices," Health Affairs, Vol. 18, No. 5, 1999, pp. 89 - 105, Lisa V. Rubenstein, Maga Jackson-Triche, Jurgen Unutzer, Jeanne Miranda, Katy Minnium, Marjorie L. Pearson, Kenneth B. Wells. Also available as RAND/RP-841, no charge.

"Impact of Disseminating Quality Improvement Programs for Depression in Managed Primary Care: A Randomized Controlled Trial," Journal of the American Medical Association, Vol. 283, No. 2, Jan. 12, 2000, pp. 212 - 220, Kenneth B. Wells, Cathy Sherbourne, Michael Schoenbaum, Naihua Duan, Lisa S. Meredith, Jurgen Unutzer, Jeanne Miranda, Maureen F. Carney, Lisa V. Rubenstein. Also available as RAND/RP-889, no charge.

Improving Depression Outcomes in Primary Care: A User's Guide to Implementing the Partners in Care Approach, Lisa Rubenstein, RAND/MR-1198/15-AHRQ, 2000, $12.00.

Improving the Quality and Cost-Effectiveness of Treatment for Depression, RAND/RB-4500-1, 1998, 4 pp., no charge.

Partners in Care: Hope for Those Who Struggle with Hope, RAND/RB-4528, 2000, 5 pp., no charge.

"The Quality of Care for Depressive and Anxiety Disorders in the United States," Archives of General Psychiatry, Vol. 58, No. 1, January 2001, pp. 55 - 61, Alexander S. Young, Ruth Klap, Cathy D. Sherbourne, Kenneth B. Wells.