We Don’t Train the Physicians Who Provide Mental Health Care

Medical faculty could teach mental health care—if they were first trained.

A similar version was published earlier in Psychology Today.

You heard in my last post the encouraging news that some medical educators embrace the psychological and social aspects of medicine, reflecting a belief that they must integrate these features with medicine’s already strong disease focus. This also means, by definition, that they embrace George Engel’s systems-based biopsychosocial (BPS) model and the patient-centered interview (PCI) needed to operationalize the model.1

Given acceptance by some in medical education that patients’ psychological and social dimensions are as important as their physical diseases, we come face-to-face with this non sequitur: medicine does not train its graduates to be competent with mental disorders, the quintessential psychosocial domain in medicine. This training failure is the root problem leading to the national mental health crisis.

Meliacin Master  (–1312): Woman teaching geometry

To review, the ever-worsening shortfall of psychiatrists requires that 85% of all mental health care be provided by untrained primary care and other medical physicians. That mental disorders are the most common health problem physicians face in practice makes this a population problem and, therefore, a taxpayer problem. Over 80 million patients with formal mental disorders receive little or no care annually in the U.S.—and the little care they do receive is often outside guidelines. And it gets worse. The prevalence of mental illness has tripled or even quadrupled with the COVID-19 pandemic.2

So, if medicine is interested in biopsychosocial and patient-centered medicine, we must analyze why it fails to train its student and resident graduates to competence in mental health. The two should go hand-in-hand.

Before we can address this, however, we must first identify what we want our mental health teaching to accomplish, what’s the ideal situation? To be consistent with BPS principles, the aim for mental health training should be that graduates from medical schools and residencies be as competent in mental health care as they are in disease care. However, no more than we would train all students and residents to be as skilled as a cardiologist or a surgeon, the objective is not to create psychiatrists of everyone. The intent, rather, is that graduates be competent with common mental disorders—just as they now are skilled in handling common heart and surgical problems.

Nor is the intent to replace mental health professionals. But, given the severe shortages in psychology as well as psychiatry, we want to provide now unavailable basic mental care to the entire U.S. population. Only primary care providers have the necessary numbers and distribution to do this.

The prestigious Institute of Medicine (IOM)3 took the vanguard in describing how to achieve our training aim of mental health skills equal to graduates’ disease skills. The IOM advised, as had others since the 1970s, intensive training in mental health care in all years of medical school and residency.4

Training would focus on the most common problems clinicians face in medical settings (depression, anxiety, chronic pain/substance use), now mostly unrecognized and even less often treated effectively. For severe mental disorders and those refractory to primary care physicians’ ministrations, graduates will also be trained in referral to psychiatry, psychology, and for counselling, just as they now are trained to consult cardiology or surgery for problems outside their purview.

The present numbers of mental health professionals can operate far more successfully in this new plan where they would function as consultants—absolved of the impossible responsibility for all frontline mental health care. Nonetheless, to develop a large enough cadre of the consultants, we must increase training of psychologists skilled in primary care as well as psychiatrists, including much increased support for collaborative care. Sufficient increases in their numbers will allow them to function better as consultants to a trained primary care workforce.

Now that we know what needs to be done, we come to the real problem. To intensively train learners in all years of education means two things will need to change. First, now paltry teaching time in the curriculum needs to increase. Second, we need to identify faculty to teach a much-expanded curriculum. We have high hurdles we must somehow surmount.

With about one-half of all health care involving mental disorders, should curriculum time parallel this? It’s an open question right now, but we know instinctively that the curriculum time commitment is at least ten times more than the 5.1 weeks medical students now receive (even less in residencies), which is about 2.5% of total medical school and residency training time for U.S. graduates. This means we would need to increase our mental health curricula to at least 25% of total teaching time

Accepting a BPS orientation, which you now know is true for many educators, the issue of finding expanded curricular time is the lesser, although no less significant, of the two problems. Guided by their patient-centered orientation, educators will need to figure out the details of ensuring much increased training time in each year. In and of itself, this is not technically difficult, but we can expect considerable defensiveness about loss of teaching time (“turf battle”) from some disease disciplines. The answer, in my opinion, lies in many disease areas being willing to give up a bit of their own (disease) turf to provide the newly required time. Here’s what I think can convince many disease-based faculty to do this. We show them that their most difficult and refractory physical disease patients typically have associated mental disorders, and that the physical disease improves only when treating the co-occurring mental disorder. While not a simple problem, we can find the time, especially if we recall that medical education’s primary aim is providing the best care possible, especially for its most common problem.

Here’s the second, more difficult issue we must resolve—finding teachers. A look at the sobering data from the AAMC paints a grim landscape. We find that there were 176,732 full-time faculty in all 179 U.S. medical schools as of December 31, 2018.5 The following highlights the almost exclusive physical disease focus of U.S. medical faculty.

  • Basic sciences — 19,732 faculty teaching, for example, anatomy, histology, biochemistry, microbiology, pathology, pharmacology, and physiology.
  • Clinical sciences — 144,439 faculty teaching, for example, anesthesiology, dermatology, emergency medicine, family practice, internal medicine, neurology, obstetrics and gynecology, ophthalmology, orthopedics, otolaryngology, pediatrics, physical medicine and rehabilitation, public health, radiology, and surgery.

The number of mental health faculty (psychiatrists and other mental health professionals) was only 11,103, about 6% of the total.5 Have them do more teaching? They are already stretched thin to teach the miniscule amount of mental health training now rendered.

We need a massive increase in faculty capable of teaching mental health care, say, five times the present number of mental health faculty. That would be approximately 55,000, which still is less than one-third of all faculty. Guess what? That’s more than the total number of psychiatrists in the U.S., and only a minority of them function as faculty.6

Where then will the teachers come from? This is the deal-breaker. We must identify someone to teach. The mental health care for over 80 million U.S. patients hangs in the balance (much higher with COVID-19)—each year.

Freepik (https://www.flaticon.com/authors/freepik): Deutsch: Lehrer vor der Tafel als Icon

CC BY-SA 3.0

The only faculty group with numbers sufficient to support a marked increase in training of medical students and residents is the non-mental health disease-based faculty. My recommendation is that we train a select minority of them intensively in basic mental health care so that they can, in turn, train students and residents indefinitely. I think there would be 20-25 faculty at each school that would be interested in this.

Copyright Robert C. Smith, December 19, 2020


1. Smith R, Fortin AH, VI, Dwamena F, Frankel R. An Evidence-based Patient-Centered Method Makes the Biopsychosocial Model Scientific. Patient Educ Couns 2013;90:265-70.

2. Health Care Access and Mental Health–Anxiety and Depression. 2020. (Accessed November 25, 2020, at https://www.cdc.gov/nchs/covid19/pulse/mental-health.htm.)

3. Institute of Medicine. Improving Medical Education: Enhancing the Behavioral and Social Science Content of Medical School Curricula. Washington, DC: National Academy of Sciences; 2004.

4. Burns BJ, Scott JE, Burke JD, Jr., Kessler LG. Mental health training of primary care residents: a review of recent literature (1974-1981). Gen Hosp Psychiatry 1983;5:157-69.

5. U.S. Medical School Faculty by Department, 2018. 2019. (Accessed December 15, 2020, at https://www.aamc.org/download/494988/data/18table1.pdf.)

6. Number of Active Physicians in the U.S. in 2018, by Specialty Area. 2019. (Accessed May 10, 2019, at https://www.statista.com/statistics/209424/us-number-of-active-physicians-by-specialty-area/.)