Letter/Observer Forum

A Response to James Coyne: The Depression of Primary Care Research and an Effort to Be Excited

Reading James Coyne’s article “Depression in Primary Care: Depressing News, Exciting Research Opportunities” [Observer, February 2001] left me, as he intended, both depressed and excited. For the last ten years, I have organized my practice to reflect the burgeoning literature that identifies non-psychiatric medicine, rather than the specialty mental health system, as the de facto mental health system (Regier, Goldberg, and Taube, 1978). I have affiliated with a primary care medical practice, I treat patients using evidence based procedures, who are exclusively referred by physicians. I primarily treat comorbid medical psychological disorders.

Over the last few years, with judicious use of the clinical research literature, I have influenced this rural medical center to develop a systematic approach to integrating primary care and psychological care. Currently, we have seven sites, where clinicians have been trained in medicine, and are co-located in practices that include family medicine, internal medicine, gynecology and obstetrics, and neurology. All the practices are busy, and today, as I write this, we have just been given our own research/administrative office, no easy feat in a rural under-funded medical center with limited space.

Despite this progress, all in all the experience continues to contain elements of depression, even though it is the most exciting opportunity I have ever had as a psychologist. Jim’s paper identified a number of contradictions and dilemmas that match my experience. The purpose of this response is to give a clinical practice-based corroboration to many of the issues he raised that confront psychological issues in the primary care setting.

Introducing psychological research and technology in a way that works in routine primary care is a daunting task that requires diligence, an incremental strategy, and a willingness to work with a much greater degree of flexibility than we might like. We are working on regular collection of SF-8 health status data as a screen, and SF-36 data as a change measure. (The SF measures are broadly-used health status measures in health care assessment and outcomes monitoring. They were selected because they allow patient self-report, are brief, and have scales that not only provide an overall assessment of health status, but also scales that assess physical health and mental health.) Prior to our efforts, there had never been any psychological data of any sort regularly collected in medical practices in our center. It has taken almost two years to move the system to support the SF data.

We also have begun collection of referral data, by use of a “behavioral health prescription pad,” which resides in the physician exam room, next to their regular prescription pad. It serves as rapid referral mechanism, and is taken to the front desk for immediate appointment with behavioral health practitioner. Also, data on demographics, reason for referral and patient release of information (to allow communication between physician and behavioral health provider), are regularly collected, both for ease of communication and to gather valuable data about practice and referral patterns.

Last week, I attended a meeting at an academic tertiary care center to discuss the potential for a multi-site trial of integrating care in non-academic primary care sites. Much of the discussion was interesting, methodologically important, and totally unlikely to be implemented in any primary care site that I work with, because the methodological demands and volume of data required is well beyond what is viable.

Attempts to utilize designs that assume randomization is just not going to happen in this medical world. It has not for medical research, and will not for psychological research. However, we continue to pursue a research strategy. We have responded to Katon’s observation that we really do not know what happens to patients with psychological problems in primary care (Katon and Schulberg, 1992). In all practices, we collect data on referrals within and outside of the practice, and the problems for which patients are referred.

With those data, we now can say that in one primary care practice over the course of one year, more than 80 referrals were made, most with depression diagnoses. Over 90 percent of those referrals were to a co-located clinician, and of those, over 90 percent of the patients scheduled and attended a first appointment. As this integration paradigm evolves, this sort of research is important, I think, and despite its methodological limitations still remarkably daunting.

As Jim’s article observed, if we are going to have an impact on health care systems, then services need to be population-based to optimize the potential effectiveness of psychological intervention. The interventions exist, the technology to implement exists, but the task is large and complicated, although not impossible.

For example, recently we collected data from one month of all new admissions to medical surgical beds. We demonstrated that psychologically-distressed medical patients stay in the hospital longer, and generate higher charges during their stay. As part of my activity within the medical center, I serve on the contracts committee of the Physician Hospital Organization, which negotiates contracts directly with insurers within Vermont. During one of these meetings, I shared the medical surgical data with the insurer, and the carve-out corporation that manages mental health services. As a result, we are developing a pilot with the insurer to regularly identify and assess psychologically distressed medical inpatients and outpatients, and fund the use of evidence-based clinical pathways for their medical and psychological care.

Where do you get the clinicians, you might ask? That is another difficult task. On the one hand, this work demands a shift in thinking and practice that virtually no one, no matter their discipline, has been sufficiently trained in. This means a substantial commitment to training, despite the fact that there is no funding source for the clinicians to be trained, or for the trainer doing the training.

We have been quite fortunate, in that we have recruited PhDs who just received their degrees and who need supervised hours for licensure. Also, a number of masters-level therapists live in our community who have a preexisting orientation to mind-body conceptualizations or are working in medicine. The problem here has been that medical offices want doctoral-level providers, so arrangements involving masters-level therapists require more intense front-end supervision and training.

What to make of all this? Jim Coyne is accurate. What has not worked, will not work. There is a movement toward increased physician detection, better assessment and treatment of depression and other mental health conditions using integrated models of care. If such efforts are to avoid the scrap heap of good ideas that have never gone anywhere, the key is the re-engineering, innovation and invention based upon the empirical foundations of our field but altered by the vastly different culture and operation of primary care.

Strategies need to mirror and mimic the practices of primary care even as it tries our patience and challenges our flexibility. Just this morning, I conducted a training session for two of the five physicians in my practice in interviewing skills for detecting problem alcohol use. In two days I will train the others. An inefficient use of my time, you might say. I say it is how primary care works and will continue to work.

There will be no revolution by primary care physicians, no arms flung wide to embrace us. They are fighting their own battles for survival. If we are willing to use our science and our methods in inadequate settings, do our work in physician’s exam rooms with our elbows on exam tables, then we can move past the depression and into the excitement. But the flexibility will be on our part, in our status, in our methods, in our treatments and in our research. That is what the situation is for us in primary care at this point, and it is there that we will find our opportunities.

Katon, W. & Schulberg, C. (1992). Epidemiology of depression. General Hospital Psychiatry, 14: 237-247.
Regier, D., Goldberg, I., & Taube, C. (1978). The de facto U.S. mental health services system: A public health perspective. Archives of General Psychiatry, 35: 685-693.

APS regularly opens certain online articles for discussion on our website. Effective February 2021, you must be a logged-in APS member to post comments. By posting a comment, you agree to our Community Guidelines and the display of your profile information, including your name and affiliation. Any opinions, findings, conclusions, or recommendations present in article comments are those of the writers and do not necessarily reflect the views of APS or the article’s author. For more information, please see our Community Guidelines.

Please login with your APS account to comment.