Skip to Main Content
Contribute Try STAT+ Today

About once a shift, I see a patient who has been unnecessarily — and often to his or her dismay — routed to my hospital’s locked psychiatric emergency unit in the process of simply trying to figure out where to go for routine psychiatric care. I’ve evaluated an attorney, a medical student, an internationally renowned musician, and many others who didn’t need to go through the unpleasant and potentially scary process of sitting through an extensive evaluation in an austere, security-monitored environment. Their journeys took them to this daunting unit not for lack of acumen or social or financial resources, but because finding the correct mental health care pathway can be bewildering and wrong turns are common.

Nonetheless, they’re among the fortunate minority who receive care. An estimated 40 million Americans experience mental illness in a given year. More than half of them do not, or cannot, get treatment.

For most Americans who seek treatment for issues like depression, anxiety, bipolar disorder, schizophrenia, and obsessive-compulsive disorder, navigating the way to mental health care is a challenging endeavor. Why? The dearth of providers, the uncoordinated panoply of practitioners with disparate and confusing titles and qualifications, and the costs of care have created a disorienting health care jungle where untreated mental illness is prevalent. It doesn’t have to be that way.


One of the biggest barriers to better mental health care is the lack of qualified providers. Some experts say we simply need to train more psychiatrists. That recommendation, often made by psychiatrists, is based on the idea that their medical background and specialized training are needed to improve mental health care.

Yet psychiatrists are the costliest care providers in the mental health care system. Just adding more of them to the current structure won’t fundamentally change the fact that care is expensive and confusing to access. And there isn’t solid evidence that psychiatrists alone provide better outcomes than other clinicians in treating many fundamental mental health issues like generalized anxiety and depression.


I believe we need three key innovations to improve the delivery of mental health care in the U.S.: shifting some tasks to physician extenders, integrating psychiatric care with primary care, and supplementing clinical capacity with software-based therapy tools.

Task shifting. Physician assistants (PAs) with specialized training in mental health could bolster the mental health workforce. Shifting the care of uncomplicated patients who require straightforward clinical decision-making to such PAs, working under the supervision of psychiatrists, could be a huge help where psychiatrists are few and their geographic distribution doesn’t match the need for them. Currently, there are too few psychiatric PAs to make a big difference — only 1.3 percent of the 125,000 certified PAs practicing in the U.S. work in psychiatry. Building more robust and well-structured PA training pathways in mental health would allow psychiatrists to spend more time working with people who have complex, unstructured clinical problems, which require greater expertise.

Integration with primary care. Another way to increase access to mental health care involves adding front-line practitioners and screening processes within primary care practices, the places where patients most often interact with the health care system. We can build capacity by having primary care doctors flag patients for further evaluation by psychiatric PAs and other mental health-focused physician extenders. These practitioners can screen more patients, implement straightforward first-line treatments, and maintain consultation and referral options to psychiatrists embedded in primary care offices. This structure, known as collaborative care, consistently demonstrates better mental health outcomes than standard management in primary care settings.

Technology-based tools. Novel technologies can help make care more efficient and expand the reach of practitioners. Using video conferencing to hold mental health “visits,” for example, can overcome some of the problems associated with the inequitable distribution of mental health practitioners. Software-based tools that deliver standard cognitive behavioral therapy have been shown to be as effective as in-person therapy. Although technology approaches are often oversold by early adopters as replacements for clinicians, and summarily dismissed by traditionalists, they have already demonstrated their value as a component of versatile and comprehensive mental health care.

Failing to expand access to mental health care comes with an enormous burden. First there is the corrosive effect of untreated mental illness on quality of life for millions of Americans. Then there is the cost — by one estimate, untreated mental illness costs the U.S. $225 billion a year, and that doesn’t include the cost of lost productivity. The University of Michigan’s Institute for Social Research once estimated that 217 million days of work are lost in the U.S. each year due to mental illness and substance abuse problems. That translates to as much as $50 billion in lost productivity each year.

Mental health care systems stand a better chance of meeting the needs of the 21st century not by just adding more psychiatrists, which would only augment how care has been provided in the past, but by fundamentally changing how care is provided. Psychiatrists will undoubtedly be essential for providing high-quality care, but they should do it mainly as clinical supervisors rather than as frontline workers. Well-trained physician extenders can reduce the cost of treating many cases of depression and anxiety.

Integrating mental health care in primary care settings represents the most effective way to seek out individuals in need of care and begin to address the disturbing fact that most people with mental illness don’t receive any care. Software-based video chat and therapy tools can help provide the new infrastructure to make care more efficient for all.

In the coming decades, the most valuable advance emerging from leaders in psychiatry and mental health care likely won’t be a breakthrough in drug development or neuroscience, or training more psychiatrists, but crafting sensible design decisions to create better mental health care systems.

Scott Breitinger, M.D., is a resident in psychiatry at NewYork-Presbyterian Hospital|Weill Cornell Medical College in New York City.

  • Although I’m sure you did not intend to leave out Psychiatric Mental Health Nurse Practitioners and other Advanced Practice NP’s with additional psychiatric certification, it is disappointing that they were not referenced in your article.
    I am a Pediatric Nurse Practitioner with Pediatric Mental Health Certification.

  • Nice article. Most of what you suggest is already in place in many areas. I am in central Florida and lead a large community mental health center. We use physician extenders, we offer telehealth and teletherapy, we operate a medical clinic, and we have embedded therapists in primary care providers across the community. Come visit! Bill

  • 100,000 psychologists 600000 social workers over 13000 psychiatric APN along with many MFT and LPC. But more psychiatry PA are the solution. ?

  • Psychiatric nurse practitioners have practiced in this role for a long time.
    I am surprised that you did not mention this

  • Wow….I realize that Dr. Breitinger, as a resident, is still in training to hopefully be a board certified psychiatrist some day, but he seems woefully oblivious the some of the more pressing and obvious issues that impact his thesis. As a practicing psychiatric nurse practitioner in a solo private practice I can tell you that the root of the problem isn’t a lack of “technology based tools” or “integration with primary care”…it’s reimbursement, stupid. Psychiatrists and psychiatric NP’s are among the lowest paid specialists. I make %75 of what I did as an RN. The insurance industry and government just don’t view mental health with the same level importance as other specialties. Also I’ve never heard of a single psychiatrist of psychiatric NP “embedded” within a primary care office. What planet does this guy live on ? Furthermore his characterization of psychiatric NP’s as “physician extenders” is demeaning and misleading. I practice on my own and am not in any way shape or form anyone’s “extension”. Perhaps when Dr. Breitinger completes his residency and gains some valuable experience in the outpatient setting, his views on this critically important topic will be more….shall we say…. reality based.

    • You are correct regarding both the author’s glaring omission of the role of PNP’s in the care of mental health patients and the issues surrounding reimbursement. I have, though, heard of the concept of embedding a PNP in a PCP office and hear it works quite well, but, again, not enough providers are available and reimbursement sucks, so we all know this isn’t going to happen in 99.9% of the world.

      The sad thing is that if people get the mental health care they need, it improves quality of life and helps many people return to society as productive individuals. We can’t afford not to address the mental care crisis.

    • I have to agree, Dr. Breitinger’s views need to be “more….shall we say…. reality based.” He’s apparently unaware of the fact that all of his DSM “mental illnesses” were declared to be “invalid” by the head of the National Institute of Mental Health in 2013.

      And it sounds like he’s unaware of Robert Whitaker’s work, which pointed out that the ADHD drugs and antidepressants create the “bipolar” symptoms, resulting in over a million misdiagnosed American children.

      And he probably doesn’t know that his “schizophrenia” treatments (which are now being give to the “depressed” and “bipolar” labeled too), the antipsychotics/neuroleptics, can create both the negative and positive symptoms of “schizophrenia,” via NIDS and anticholinergic toxidrome.

      Expanding an “invalid,” iatrogenic illness creating, “mental health system” is not actually a wise idea.

  • Psychiatry is poorly reimbursed, if better reimbursed there would be more. Many practicing psychiatrists do not take insurances because of the reimbursement rates and then charge high amounts.

  • How odd. An author supposedly looking for solutions to the lack of qualified mental health providers completely ignores the role of psychiatric nurse practitioners? I am a psychiatric nurse practitioner in Washington State. I evaluate, diagnose and treat (through both prescriptions and therapy) those with mental health issues. I do absolutely everything a psychiatrist does. Only difference? I was trained in the more holistic nursing model of care, rather than the medical model (a fact about which I am quite proud). This complete disregard, complete lack of acknowledgement of psychiatrically-specialized nurse practitioners reveals to me the inherent ego in this author. Without knowing him, and solely based on this article, I would presume that he loves calling physician assistants “physician extenders”, seemingly to emphasize the power that the MDs ultimately have over this important role. Another oft used and offending term is “mid-level” practitioner. The truth is that most medical students don’t want to go into psychiatry. Physician assistants and nurse practitioners are already filling those shoes (at least in the states where nurse practitioners are allowed to practice to the fullest extent of their license). Instead of using belittling terms, this author should have taken a more expansive view of the health care system, because the truth is, we need people in the trenches. More NPs and PAs are willing to go there. Most MDs are not. Show some respect.

    • Unfortunately, I see too many mistakes made by nurse practitioners in psychiatry. Psychotropic drugs are very complex. NPs get a watered down version of the sciences, and even pharmacology compared to physicians. Any patient receiving these meds, especially children, should seek a second opinion with a board-certified psychiatrist.

    • I agree with Robert Stone. Working in an inpatient medical setting exclusively, I see far too many messy and sometimes downright dangerous medication regimens from NPs and PAs in the community. Not to say this can’t happen with MDs but most of the MDs I speak to (and I call every patient’s psychiatrist or other provider) typically inherited the messy regimens and are in process of cleaning it up.

    • It’s insulting to make a blanket statement about the abilities of NP’s and the mistakes they make. I assure you that I am constantly adjusting errors do inpatient psychiatry who make medication changes without really k owing my patients. As an aside – I also speak for a pharmaceutical company and am regularly concerned when a room of MDs can not explain how a partial agonist works – yet they write the meds daily. Yes, the medications are very complex, but it is the responsibility of the provider to learn how to safely use the medications- including the psychiatrists. A general statement is insulting. Bad jusdements are not specific to NPs and as an NP that works diligently in the field – it’s just ignorant.

  • Scott,
    Great article. Thank you.
    I am curious as to why you don’t include NPs with PAs as Physician Extenders. What do you believe is the first few steps in order to set these wheels in motion.
    I look forward to your response.

    • Hi Lori,
      Thanks for your message. You’re absolutely right. This article was by no means exhaustive and I definitely could have spent more time discussing NPs and LCSWs (amongst many others who contribute greatly to mental healthcare provision). I focused on PAs in part to counter the argument that we need to focus simply on training more psychiatrists. I dont believe this is as cost effective as fostering the development of more coordinated team structures of which psychiatrists play a smaller direct role. I see NPs, LCSWs, psychologists, and psychiatrists as having a good degree of overlapping job responsibilities, and this inter-professional interface warrants its own consideration from a healthcare systems design standpoint.

      Regarding first steps, I’d look towards the locus of change:
      1. Task-shifting- greater focus on augmenting pathways towards careers in mental healthcare at training institutions (whether that’s nursing school, Social work school, medical school, etc.). Also will be impacted by reimbursement mechanisms described in greater detail in #2, below.
      2. Primary care integration- legislative and policy changes which control incentive structure (i.e., reimbursement). When healthcare uses strategies to measure more of what we do *for* a patient rather than what we do *to* a patient, the incentive will be greater for the expansion of Collaborative Care. Bundled payments and capitation models are beginning to accomplish this. Hopefully, more progress on the way.
      3. The private sector is creating these tools now, and there are many more on the way, I’m sure. I’m a fan of companies like Pacifica, Joyable, Lantern, and Headspace, amongst many others.

      Would love to hear your thoughts, especially about how you see (or would like to see) the interface of NPs with other mental health practitioners changing in the coming years.

      Kind regards,

  • “Yet psychiatrists are the costliest care providers in the mental health care system. Just adding more of them to the current structure won’t fundamentally change the fact that care is expensive and confusing to access.”
    To the contrary, if there were more psychiatrists, they would be less expensive, and they would make themselves more accessible. It is the law of supply and demand. People in health care love to think that basic economic principles do not apply to them, but in fact health care is just another product that people can purchase. Health care and pharmaceuticals in the United States are overpriced because there is insufficient supply for the demand.

    • Hey what about the nurses?!? Nurses’ are the Number 1 trusted healthcare providers for years now and Nurses week is just around the corner (May 6-12 2018). Hey Dr Scott Breitinger, resident in psychiatry at NewYork-Presbyterian Hospital|Weill Cornell Medical College in New York City and Stat News, can you please print a revision to correct this!?! This article is a farce and inaccurate without the mention of nurses. I thought we were trying to be “evidenced-based” these days. Even AI and a computer could figure this one out. Besides what is a “physician extender?” How about at least being respectful to the PAs you are describing. Never mind not even mentioning the role of the Advanced Practice Nurse Practitioner, the RN, the Nurse Care Manager . . . the list goes on and on. Nurses week is just around the corner so shame on you Stat News Editors for not addressing this prior to publication. We should probably just call this article fake news. Stat News is being removed from my bookmarks.

    • The statement you quoted was regarding relative position as “costliest care providers.” Because of the nature of the investment in training both from a societal perspective and from the individual’s perspective (in this case, psychiatrist’s), there is somewhat of a floor for the price of psychiatrists’ services. It is highly improbable that their time would ever reach the point of costing less than that of a psychologist, NP, PA, or LCSW.

      In addition to the laws of supply and demand–which I agree, do apply–there is the need to consider further still the economic principle of price elasticity of supply.

Comments are closed.