Depression in Primary Care: Depressing News, Exciting Research Opportunities

When psychologists think of research to improve treatment outcomes for depression, they typically think of efficacy studies: randomized clinical trials evaluating psychotherapies or comparing psychotherapy to medication. As important as such studies are, there is a disconnect with the needs of depressed persons in the community, since the majority of these people will not be treated in a specialized mental health setting. Depressed persons in the community are much more likely to visit a primary care physician than be treated by a mental health professional. Unfortunately, it is likely that their depression will remain undetected by their physician. Even if their depression is detected, a growing number of studies demonstrate that patients diagnosed with depression in primary care fare no better than those who are missed. Patients receiving treatment for depression in primary care improve at approximately the same rate as patients assigned to placebo conditions in clinical trials conducted in tertiary care settings.

One irony is that so much is made of marginal differences in the efficacy for various treatments found in clinical trials, while the gap is so large between their efficacy in clinical trials and the effectiveness of the same treatments when they are offered in routine primary care.

A few years ago, it could have been said with confidence that most depression in primary care remained undetected, with physicians not inquiring and patients not volunteering that they were depressed. Awareness that depression was going untreated and promotion of the newer antidepressants led to a tremendous increase in the diagnosing of depression and in rates of prescriptions for antidepressants.

The rate of prescriptions for antidepressants now exceeds the prevalence of depression in the community. For example, over 10 percent of the elderly received a prescription for antidepressants in the past year. This practice does not necessarily translate into improved outcomes for depression. Physicians often are responding to nonspecific distress, rather than systematically applying the criteria needed for a diagnosis of depression. This also means that many depressed individuals remain undetected.

Similarly, once depression is diagnosed and treatment begins, patients are often dismissed without the follow-up needed to determine the effectiveness of the treatment, to address patients’ questions or compliance problems, or to make necessary adjustments in treatment. Given the costs associated with the prescription of antidepressants, reducing inappropriate prescription and improving the follow-up of patients being given antidepressants ought to be seen as priorities.

Results from over a decade of efforts to improve the outcome of depression treated in primary care are, in the words of a recent National Institute of Mental Health (NIMH) program announcement, “largely disappointing” (NIMH PA-99-073). We now have sufficient data to conclude that educating physicians or patients, screening and facilitating detection of depression, and providing recommendations about treatment are not sufficient to improve outcomes either for individual patients or on a population basis.

Results from the newest generation of studies are now accumulating. These studies typically involve examining both practice-level interventions and outcome monitoring. For instance, at the University of Pennsylvania, studies are currently evaluating the introduction of depression specialists into primary care. These Masters-level professionals maintain follow-up contact with patients, address their questions and concerns, monitor outcomes, and provide feedback to both the patients and their physicians. We are also evaluating a system for facilitating referrals to mental health professionals.

There are great challenges to mounting such studies, and, even if successful, there will be greater challenges to the implementation of similar programs in routine care. In the short term, such interventions seem to have a clinically significant advantage over routine care, but the difference may decrease over time, along with patient satisfaction. Furthermore, although designed to be cost-effective, such programs consume scarce resources, and may prove difficult to integrate into routine primary care, given the competing demands for resources.

It must be kept in mind that the typical visit to a primary care physician centers on the patient’s presenting complaints, follow-ups about pre-existing health conditions, and inquiries and education about preventive health practices. All this is expected to occur in the span of no more than 12 to 14 minutes. Given this, primary care may not seem like the optimal setting in which to treat depression, but to paraphrase Willie Sutton, it is where depression is likely to be found.

Even as we discover how difficult it is to improve the outcome of depression in primary care, the standards for success are being raised. Traditional efficacy trials evaluate success in terms of both the number of patients completing a study who benefit and the number originally enrolled (i.e., intent to treat analyses). However, effectiveness studies of the outcome of depression treated in primary care may adopt as their denominator the number of depressed patients identified through screening. Given that 15-50 percent of such patients will likely refuse randomization or drop out of treatment, the ability to demonstrate an effect is attenuated.

An even more stringent standard is that of the public health perspective, which asks the question: “What proportion of the depressed patients in primary care would improve if an intervention program was established?” Meeting this standard involves coming to terms with the additional 15 percent or more of primary care patients who can be expected to refuse screening. These standards may seem on the surface to be unrealistic and arbitrary. However, once we acknowledge that the prevalence of depression and its associated social costs make it a major public health problem, it is incumbent upon us to assess whether our efforts are making inroads into this problem on a population basis.

The elements of success in an effectiveness or public health study are likely quite different than the determinants of efficacy in a randomized clinical trial. Regardless of a treatment’s efficacy under well-controlled conditions, its impact will remain unrealized if targeted patients refuse to accept diagnosis or fail to follow through with treatment.

Depressed primary care patients typically differ from the patients who are enrolled in clinical trials on the basis of referrals or responses to newspaper advertisements. At the time of visit, depressed primary care patients are likely to be experiencing less severe depressive symptoms than depressed individuals actively seeking treatment. Because of the mildness of their depression, they may be more difficult to enlist into a treatment study. Even if these patients do enroll, there is a decreased likelihood that they will comply with treatments, be willing to tolerate the adverse effects of medication, or attend therapy regularly.

If these depressed patients are not already in treatment with a mental health professional, they may have attitudinal barriers to accepting treatment; notably, they may reject the label of depression or believe that they should handle their problems without the intervention of professionals. Furthermore, given the rates at which physicians dispense antidepressants, it is likely that the depressed patients most likely to accept treatment are already receiving it. The depressed primary care patients who are not already receiving medication may be particularly unwilling to volunteer symptoms or accept diagnosis and treatment.

Although the literature concerning depression in primary care continues to grow, psychologists are not the major contributors to it and have had little influence on the setting of public policy. The Depression Guidelines promulgated by the Agency for Health Services Research have been widely interpreted as promoting the prescription of medication, and most efforts to improve the outcome of depression in primary care have had a similar bias. It could be that some of the shortcomings of current efforts are due to their emphasis on medication and requirement that patients accept a biomedical concept of depression.

Researching this issue is just one of the pressing priorities to which psychologists could have an important and distinctive contribution. But it would be unrealistic of us to assume that there will be a great deal of interest in addressing the problem of depression in primary care by stationing doctoral-level psychologists in these settings.

Schulberg and his colleagues demonstrated that interpersonal psychotherapy (IPT) was equivalent to enhanced management of antidepressants, and that both were superior to routine primary care. The importance of these findings should not be missed; this therapy is acceptable and efficacious for some primary care patients. However, interventions of this sort are difficult to implement and require a large investment relative to their impact. IPT requires highly skilled therapists, and in this study only 42% of the patients receiving IPT completed the trial.

Having witnessed the difficulties involved in enrolling primary care patients in studies that provide free psychotherapy, I remain convinced that substantial resistance to accepting psychotherapy exists among some segments of the public. Aside from the issue of patients not accepting therapy, there is the practical issue of funding adequate numbers of skilled therapists in the primary care setting: There are not enough well trained, highly skilled IPT (or cognitive) therapists to go around, nor is there funding to staff many primary care settings.

Psychologists can make a significant contribution to research aimed at improving the outcome of depression in primary care. The modest success achieved by efforts focusing on physician- and practice-level interventions necessitate that we take a fresh look at the neglected patient factors which affect acceptance and outcome of treatment.

Once we acknowledge the attitudinal barriers and practical obstacles to offering conventional psychotherapy in the primary care setting, we can begin to explore how to re-engineer treatment. Specifically, we need to offer treatments that are acceptable to those depressed persons not currently receiving or benefiting from treatment; that can be delivered successfully without reliance on highly trained therapists; that allow for a large margin of error in their delivery; and that can be delivered in primary care settings without requiring an unlikely revolution in routine care.

Rather than being discouraged by the results of past efforts to improve the outcome of depression treatment, we need to respond to the challenge and the opportunity to develop more creative solutions.

This article is based on a presentation to the NIMH director and staff during a workshop on reducing barriers between behavioral science and schools of public health. The workshop was part of an ongoing initiative at NIMH that originated out of discussions with APS (see the December 1999 issue of the Observer for background).

Observer Vol.14, No.2 February, 2001

Leave a comment below and continue the conversation.


Leave a comment.

Comments go live after a short delay. Thank you for contributing.