The Medical Model Doesn't Work for Mental Health

; Abraham Verghese, MD; Tom Insel, MD


March 25, 2022

This transcript has been edited for clarity.

Eric J. Topol, MD: This is Eric Topol with my co-host Abraham Verghese for Medscape's Medicine and the Machine. Today, we welcome Tom Insel, who led the National Institute of Mental Health (NIMH) for 13 years, from 2002 to 2015. He's also known as the nation's psychiatrist. He has just written a book called Healing: Our Path From Mental Illness to Mental Health. So Tom, welcome.

Thomas Insel, MD: Delighted to be here. Thanks for having me.

Topol: In Medicine and the Machine, we have not done justice to mental health. In this book and your work over decades, you certainly have taken this on.

Abraham Verghese, MD: I've just finished your book and I must confess that I was struck by how little I knew about mental health. I learned so much from your book; I was surprised by many things, but especially by the notion that diagnosis, which is so fundamental to what we do in internal medicine, for example, is a very different beast in mental health. In fact, one of your chapters says diagnosis doesn't matter. Would you talk about that?

Insel: The study of mental illness has been a difficult saga. It goes back to World War II, when the field began to realize that diagnosis was important and that we needed to come up with a way to distinguish the different forms of psychopathology because we suddenly had a choice of treatments and we needed to know who should get which therapy. The attempt to do that was taken over by the American Psychiatric Association, which eventually created the Diagnostic and Statistical Manual of Mental Disorders — the DSM, a much-needed innovation. It provided consensus definitions of the major syndromes based on symptoms. But the problem has always been that it never went beyond those kinds of consensus descriptions. There's value in that; it provided a common language. But what I believe has become a real problem for the field is that what was intended as a dictionary ultimately began to be used as an encyclopedia. People thought that somehow these labels had scientific and biological validity. They were never intended to do that. They actually were never intended to even identify which person would get which treatment beyond a sort of vague bias toward one kind of treatment or another.

This has been a real problem for the field — that we've never developed the kind of diagnostic rigor or validity that so much of medicine has enjoyed. One result is that treatments are often empirical — kind of hit-or-miss. We don't have the precision to say, "This individual will do best with this particular medication," or "...this particular psychological intervention," or perhaps "...this neuro therapeutic effort." We have lots of good treatments but we don't yet know how to match the treatment and the diagnosis.

Topol: One of the statistics I learned from your book is that 1 in 20 Americans have serious mental illness — schizophrenia, bipolar disorder, or severe depression. But out of those, only 15%-16% receive even minimal therapy. So, we fail not only with precise therapies but by not even providing the therapies that are needed. Is that a fair summary?

Insel: It's an important issue. When I started the book, I was trying to carry forward a project that I've worked on for 40 years, which is to convince the public that psychiatry is just another discipline in medicine and these illnesses are no different from endocrine or any other metabolic disease. But the more I worked on this, the more convinced I was that something is fundamentally different here.

One of those differences is around this issue of engagement — that more than half the people who should or could be in treatment are not in treatment. There are lots of reasons for that. Some of them have to do with the illnesses themselves. What is different is that the people who need treatment the most are least likely to seek it. This is the flipside of what you'd see in oncology, diabetes, or cardiovascular care, where generally, the people disabled by their symptoms are the people who are going to be driven into treatment, no matter what it takes to get there. When you find people who are truly psychotic and irrational, they don't think that they're sick and they'll do almost everything possible to avoid treatment. That makes this a really difficult problem to solve.

Verghese: What's remarkable is that the treatment may not depend on making an accurate diagnosis. In that chapter on diagnosis, you quote a psychologist, John Weiss, who talks about the commonalities of process that are helpful in almost all mental illness. Would you explain that a bit?

Insel: In terms of psychological therapies, we spent 20-30 years developing "bespoke" therapies, all of which have acronyms: CBT (cognitive-behavioral therapy), DPT (dialectical behavioral therapy), IPT (interpersonal therapy) — there's a long list. But in the past 5 years or so, we have been asking whether there are some common features. I see increasing recognition that good outcomes depend not just on the therapy, but the therapist — people who are good listeners, naturally empathic, and able to build rapport and a sense of trust. Give them fundamental skills, such as motivational interviewing so they know how to help people talk about what's really bothering them, and tools, such as behavioral activation that can help people who are shut in or inactive, to become more engaged and involved. A limited number of those skills cross almost all therapies. In the hands of a person who has an empathic ability, these seem to be quite effective. It may not depend on whether this patient has a particular Current Procedural Terminology (CPT) code or lands on a certain page of the DSM, but some of these interventions are helpful across the diagnostic structure.

Topol: During your remarkable 13-year tenure at NIMH, you fostered basic research — including genomics — to understand mental illness. When Beth Stevens at Harvard discovered the complement factor underpinning of schizophrenia, I thought, Oh my gosh, this is amazing. This is going to transform schizophrenia, one of the most difficult conditions there is in all of medicine. Of course, that didn't happen. Is that emblematic of the problems with making great discoveries in basic research that take decades to improve outcomes? What do you think?

Insel: There's a piece of that; it's true across medicine. As Francis Collins used to say, "It's always a marathon; it's almost never a sprint." Getting those basic discoveries into new therapeutics is a slog. That's especially true with something like that complement factor discovery, which is essentially discovering a risk factor for schizophrenia in germline DNA. That's a very long way from finding anything actionable for the disease. It's interesting. But whether it will make a difference is unclear to me, and we now have hundreds of those kinds of findings. When you add them all together, you probably learn as much by taking a good family history. I believe it is important to understand something about the genetics of risk, but it's hard to see how you use that in any way that makes a difference.

When I wrote the book, I went from thinking about those issues to understanding that, as important as it was to get a deeper view of the pathophysiology of these illnesses, someone has to pay attention to the fact that so many people with schizophrenia are incarcerated or homeless with this disease, or that they are dying at age 55 of chronic pulmonary disease because they are smokers and they've been eating out of dumpsters and living a terrible life. That's really what began to bother me. That's what the book became... trying to understand how we can allow that to happen for an illness that's fundamentally treatable — not curable, but treatable. And while the C4 complement factor discovery is interesting, I don't see how that reduces the incarceration of hundreds of thousands of people.

Schizophrenia is not a rare condition. In 2022, ten times more people with serious mental illness are incarcerated compared with living in public institutions for mental illness. I find that egregious and unacceptable. I have to say that I'm not sure I understood that 5 or 10 years ago when I was at NIMH. I don't think I knew that was happening in the way I now understand. I've spent time in jails, prisons, and homeless shelters, and I got the sense, which I didn't have when I was so deeply involved in academic medicine and research on mental health, that this is as much a social justice problem as a scientific problem.

Topol: You made that point. You underscored how it was supposed to deinstitutionalize mental illness and instead, it transinstitutionalized imprisonment.

Insel: That's on us. We decided that we would put a tremendous amount of public funds into building jails and prisons. Bryan Stevenson, the founder of the Equal Justice Initiative, says a prison opened every 10 days through the 1990s. At the same time, we gutted our public mental health system and there were no beds and very few committed community centers. So there was no capacity, and people who needed care didn't have any options except to end up in the criminal justice system. That's not right.

Verghese: Your choice of the word "healing" as the title for your book was interesting, and I understood it better when I had finished reading it, in the sense that you're reflecting the feeling that there's so much broken with mental health — so much underdiagnosis and undertreatment. But you also brought in the importance of place and family. I love the passage where you quote Mandela's use of the South African term ubuntu, which roughly means "I am because of you." Would you expand on that? It's not what I expected to read from a book by you and it was so refreshing.

Insel: It was part of this evolution I went through. I have spent my whole career being focused on a kind of medical model. This is one of the ways I believe mental illness is quite exceptional. The medical model has worked so well for infectious diseases and other areas of medicine. It's working very well for cancer right now. But it hasn't worked for people with serious mental illness. It's led to a focus on symptom reduction rather than recovery. The ubuntu idea comes out of this need to try to understand both the problems and the solutions in a much broader context. I call that context recovery or healing. I don't think I fully understood what recovery was when I started writing the book.

I was talking about recovery to a very wise psychiatrist who works on Skid Row in Los Angeles. He said that if you want to understand this, it's just basically the three P's. Of course, where I was coming from, I was thinking, There's Paxil, and there's Prozac... but I couldn't think of what the third P was. I thought maybe psychotherapy. And he stopped me and put up his hand and said, "Look man, it's people, place, and purpose. Those are the three P's if you want to help people recover. You have to help them find a life. They need people, place, and purpose."

In fact, we all need that, right? But when it comes to people with serious mental illness, many of whom are incarcerated or homeless or just not thriving, we tend to forget this. We focus on their symptoms and whether they're getting the right medications and whether they have side effects. We don't think about their need for social support. Are they lonely? Are they cut off? Do they have someone who cares about them and someone who they care about? It's incredibly important for recovery. Do they have a safe place to live? A sanctuary? A place where they can get decent food? Where they don't have to worry about violence or being attacked? And do they have a reason to recover? Do they have something to live for?

I was at a clubhouse yesterday in Phoenix and somebody said, "What I need is for somebody not to ask me, 'What's the matter?' I need somebody to ask, 'What matters to me?'" And I realized that, in the medical model, we've missed all that. We don't do that. Yet for people to recover, this guy on Skid Row was right. Those are the basics. Ubuntu is very much a part of that idea. It's the connections we make in the sense that we have a relationship with others wherein they care about us, we care about them, and there's a community. That is just fundamental.

That's not rocket science. We can do that. A clubhouse does that. We have everything we need to make that happen and to help these people who today are our untouchables. They are outside our community. As President Kennedy said in 1963, they remain alien to our affections. We can fix that and provide community, sanctuary, and a sense of purpose so they can thrive. We're just not doing it. The book is a call to action to say, "Enough already." This must stop and we must begin to care for people. It's not going to be more genomics and more neuroimaging. Absolutely, we need more of those things. But the urgent need is to address this gross injustice. We shouldn't put that off any longer, and we certainly shouldn't put it off until we know more about the pathophysiology of schizophrenia.

Topol: In the last section on "The Way Forward," when you wrote the "measure of the soul of our nation," it seems that you encapsulated this idea. Our nation is hurting, and the pandemic didn't help. Nor is the war in Ukraine — watching people being massacred. What is the way forward? You're so thoughtful about where we've gone off the track, but what is the best way forward now, assuming an unlimited budget (which we don't have)? How could we fix this?

Insel: Most of what we need to do, we know how to do. So I don't think it's a mystery. A lot of it has to do with those three P's and creating that sense of community. I have been swept away by Ukraine; it's been hard for me to talk about the book because I've been so focused on what's happening there. It's infuriating and frightening but also inspiring to see people come together in a way our nation didn't come together during COVID. But we're seeing it now in another nation — perhaps a nation that's being born afresh, as people create this sense of common purpose and common sacrifice.

Where I'm hoping the book will lead is to a new social movement that is focused on those in greatest need, those who can't support themselves and have become destitute, homeless, or incarcerated because they have a brain disorder. We would never let this happen to someone with cardiovascular disease or asthma, so it's hard to imagine that we've permitted this to happen. First, we need the awareness that this is happening. I wrote the book with the idea that it would be a bit like An Inconvenient Truth, the documentary about climate change — that it would be a wake-up call for so many who, like me, haven't been aware of how serious this problem has become.

The problem has been made much worse by COVID. You're absolutely right about that. Now these issues are becoming more widespread. We're seeing enormous mental health issues among young people coming out of COVID. In California, at least in the Bay Area where I live, we lost far more young people to overdoses and suicide than we did to COVID. These deaths of despair have now become a more urgent issue, with more than 100,000 overdose deaths nationwide last year.

What's the solution? We need to name it. Just as we did with COVID, we've got to call it out. The Surgeon General has done that recently with youth suicide. I am delighted to see the administration come up with a fact sheet that lays out a whole game plan for how they want to address the mental health crisis — and call it a crisis. That's extraordinary. I think it's the first time in 41 years that the White House has even used the term mental illness or mental health. So leadership matters. Leadership will be helpful. But fundamentally, this is a bottom-up effort. It is creating community and creating community resources. Clubhouses are one way to do that — community clinics. The idea of using schools as a kind of a center for providing resilience and teaching kids about social and emotional learning — what they need to know to be able to compete in a brain economy. Giving them what we call brain capital by giving them a set of skills that allows them to talk about mental health issues and be able to develop resilience around that. We can do all this. It's on the agenda for the White House.

It's a huge issue here in California, where we have a governor who has committed $4.4 billion for youth mental health. Our governor's idea is that you're never going to cure your way out of this, you're never going to have enough clinics. You have to move upstream and begin to talk about prevention and population approaches to helping people deal with these issues. So there is a game plan. When I finished the book about a year and a half ago, I didn't realize that we would be moving so quickly from the top down, that there would be so much going on in Congress, so much in the White House; that the leadership in places like California would take this problem and run with it. A lot is happening — more than I have seen in my four decades in the field.

Topol: That's reassuring.

Insel: It's a great time to be taking on these issues. That was the president's comment in his State of the Union message. He said, "Let's take on mental health." I believe there's already a momentum to do that. But I want to caution that it's not just about money and policy and programs. There is a need for building up this community that we had more of when I was a kid, when I was growing up, and when I first had my kids 40 years ago. We had a more supportive community than we have today. It's harder for parents. It's harder for kids. We can get back to that. We can fix that, but it's got to be both bottom-up and top-down.

Verghese: The statistics you had in your book about prisons were a revelation to me. I was struck by one anecdote of the prisoner who was anxious about being released because paradoxically, he had community in prison and he had medication and he had a purpose and a process. But once he was released, he would be back out there swimming in the world without any assistance. It made me wonder, given the numbers of incarcerated — not that it's ideal — if there isn't more opportunity to deliver the kind of place, people, and process that are needed?

Insel: It's a great question. There are places that are trying to provide better care in the criminal justice system, but it's a Band-Aid. The criminal justice system is essentially set up for punishment, not for therapy or even for rehabilitation. We have to hit reset on this and begin the process of rebuilding the infrastructure, the network of services within the healthcare system. That also requires us to reframe what we mean by healthcare. In the book I tried to get us to talk about health as something more than what we call healthcare. It does require this recovery focus — the three P's.

Just to make it more concrete, why not have caretakers be able to write prescriptions for housing or food, or for Medicaid or private insurance to pay for a clubhouse? None of that happens today, but it is not impossible to imagine. I'm inviting us to rethink what we mean by healthcare, in the context of mental health anyway, and to decide that, while the criminal justice system is where that's been happening, we can't expect the criminal justice system to become an ideal place for someone to recover. These are not criminals; these are people with an illness, and we ought to accept that.

There's a quote in the book from Abraham Lincoln in which he reminded someone that melancholia (the old name for depression) is a misfortune, not a fault. We've forgotten that. We've treated it as a fault or a crime. We've criminalized the very thing that we should be treating. That's something we have to start to turn around. There's an attempt to do that with a reformed crisis service, called 9-8-8, that Congress is requiring every county in America to put in place by July 16 of this year. This will mean that if you're in a mental health crisis, you don't call 9-1-1 and get cops with guns or an ambulance set up for a medical-surgical emergency. You call 9-8-8 and get a social worker, perhaps a nurse, or a peer. They come in a van and, if needed, they can spend hours with you to help you through that crisis. If you do need to be hospitalized, you don't have to go to a medical-surgical emergency room or to a jail. You go to a psychiatric emergency room or a crisis stabilization unit, and then there's a whole system that goes from there.

What we have to do is build out that pathway. This is happening around the country, but not fast enough. Often what we're seeing is that counties are building out one part of it, like the 9-8-8 phone line, but they're not building out the mobile crisis vans. They don't have the workforce. They haven't thought about where people get dropped off. There's a lot to fix, but we know what needs to be done and how to do it. Now it's a matter of making the commitment to make it happen.

Topol: I want to touch on a lesser but more common form of mental illness. Most of what we've been talking about throughout this conversation is the serious stuff. But what the pandemic brought out is a lesser form of depression and anxiety that's pervasive because it's reactive. There are not enough people out there to help these folks. What is your sense about the apps that are marketed for a monthly fee to support people at this very difficult time? It doesn't seem like the pandemic is over yet, and people are still suffering and they still need help.

Insel: I'm glad you brought that up because most of the people listening to this podcast are probably thinking about the anxiety in their kids or depression in their families. We have gone through this transformation, accelerated by COVID, from the brick-and-mortar world of therapy with your 50-minute or hour once a week, to a world in which people can access psychotherapy and even medication online with a click. It's a bit like what we saw with Amazon and Uber. It's a transformation that provides convenience and access and timeliness, and I believe all of that has been extraordinarily helpful for a lot of people.

The "Yes, but" part of this — the caveat — is that for the most part, we haven't seen a change in the quality of care or services. It concerns me that we have psychological treatments that work quite well and require learning a set of skills, but it's not what much of the workforce is delivering. We would not accept this in the treatment of diabetes or hypertension. In the medical sciences, people are pretty much required to do what the science tells them works. But in the world of psychotherapy, it's been much more free form; people give what they're comfortable giving. What you get depends on who you access.

I think this shift to teletherapy is interesting and potentially transformative. But this is the first act of a five-act play and this first act has just been about access. It's not that interesting to me in terms of changing outcomes overall. That will come in the second act as we start to focus on quality and begin to use the tools we understand. By giving feedback not just to the patient but also to the clinician, we can improve care. And as we improve care, we will improve outcomes. I have no doubt about that. But that's act two. We haven't gotten there yet. That's the work to be done in the next 3-4 years,

Verghese: I want to come back to diagnosis for just a moment, recognizing that it's not as vital as it is in my specialty of infectious diseases, for example. You recognize that genomics did not hold to the potential we once thought it might have for mental illness. What about functional MRI (fMRI)? Is that going to become the way we classify disorders? If not, what is the next thing? As a scientist, I'd like to think that at some point we will find the substrate that allows us to parse these things out. Or is that a naive idea? Have you dismissed that from your thought processes?

Insel: I have not, Abraham. I would push back a little that diagnosis doesn't count. I believe it does. Even today, though the categories we have are not perfect, they're a start. I just believe, as you're suggesting and as we would do in infectious disease or in oncology, we have to get past symptoms. We have to begin to think about other measures. fMRI is difficult to scale and difficult to do. There's a paper out in Nature this week on the use of MRI, showing that there's so much variation in the human brain, in the connectivity and the structure, that you would need literally thousands of patients to be able to get reproducible results when you're looking at depressed vs healthy brains. So while there are thousands of papers out there, most of them are not reproducible because of the inherent variation and the fact that the effects were looking at are rather small. This is the same problem we had with candidate genes and genomics when we started 20 years ago.

I do believe there are other ways to get at this. fMRI may be useful academically to demonstrate the circuitry differences or the biology for, let's say, the three or four different forms of depression or the two or three different forms of obsessive-compulsive disorder. I believe there will be some value in fMRI, but I don't think it will scale as well as EEG. Using something like EEG ultimately could become like using EKG in cardiology, where it's in every primary care doc's office. What's happened in the last 3 or 4 years is the advent of machine learning and artificial intelligence (AI) approaches to EEG, and these have transformed the way we use those signals. In fact, they have given us much better precision around the diagnosis of posttraumatic stress disorder and depression. But to be clear, there's no magic bullet here. It's probably not going to be just the EEG with good AI. It's going to be EEG, along with cognitive testing and with data from a wearable that tells you about sleep and activity, and maybe even social interaction. Symptoms need to be included as well, and subjective experience. We've seen that if you put those together, you can begin to define subtypes of the major mental disorders that are predictive of treatment response. They are beginning to get us to precision medicine. But unlike in oncology, it's not a single genomic defect. It will be three or four tools that we put together, and it has to be easy enough to do in a primary care setting because more than 80% of depression and anxiety is treated in primary care, not in specialty care.

Topol: This has been a terrific conversation, Tom. We're indebted to you for many reasons. What you've done in your career has been momentous, not just your leadership at the NIMH, but also this book. Perhaps the work that you put into the book led to this epiphany and reinvention. I hope the people who listen to this podcast learn about the three P's, the wisdom you've imparted to us, and the hope that we finally get mental health to a better place than it's been for many decades. I wish you were running the whole program, but maybe your ability to advise the key people in government leadership will get us there. Thanks so much for joining us.

Insel: Thanks for having me. I hope that, if nothing else, the book now transmits that hope because we do need hope at this moment, to figure out how to fix so many of these problems. We can do it. It's nothing more than just making a commitment.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.