Toward a new understanding of mental illness - Thomas Insel

[Music] [Music] So let's start with some good news, and the good news has to do with what do we know based on biomedical research that actually has changed the outcomes for many very serious diseases.

Start with leukemia, acute lymphoblastic leukemia, the most common cancer of children. When I was a student, the mortality rate was about 95%. Today, some 25 to 30 years later, we're talking about a mortality rate that's reduced by 85%. 6,000 children each year who would have previously died of this disease are cured.

If you want the really big numbers, look at these numbers for heart disease. Heart disease used to be the biggest killer, particularly for men in their 40s. Today, we've seen a 63% reduction in mortality from heart disease. Remarkably, 1.1 million deaths averted every year.

AIDS, incredibly, has just been named in the past month a chronic disease, meaning that a 20-year-old who becomes infected with HIV is expected not to live weeks, months, or a couple of years, as we said only a decade ago, but is thought to live decades, probably to die in his 60s or 70s from other causes altogether.

These are just remarkable, remarkable changes in the outlook for some of the biggest killers. And one in particular that you probably wouldn't know about: stroke, which has been, along with heart disease, one of the biggest killers in this country.

It is a disease in which now we know that if you can get people into the emergency room within 3 hours of the onset, some 30% of them will be able to leave the hospital without any disability whatsoever. Remarkable stories, good news stories, all of which boil down to understanding something about the diseases that have allowed us to detect early and intervene early.

Early detection, early intervention—that's the story for these successes. Unfortunately, the news is not all good. Let's talk about one other story, which has to do with suicide. Now, this is, of course, not a disease per se; it's a condition or it's a situation that leads to mortality.

What you may not realize is just how prevalent it is. There are 38,000 suicides each year in the United States. That means one about every 15 minutes. It's the third most common cause of death amongst people between the ages of 15 and 25.

It's kind of an extraordinary story when you realize that this is twice as common as homicide and actually more common as a source of death than traffic fatalities in this country. Now, when we talk about suicide, there is also a medical contribution here because 90% of suicides are related to a mental illness: depression, bipolar disorder, schizophrenia, anorexia, borderline personality—there's a long list of disorders that contribute.

As I mentioned before, often early in life. But it's not just the mortality from these disorders; it's also morbidity. If you look at disability, as measured by the World Health Organization with something they call the disability-adjusted life years, it's kind of a metric that nobody would think of except an economist.

Except it's one way of trying to capture what is lost in terms of disability from medical causes. And as you can see, virtually 30% of all disability from all medical causes can be attributed to mental disorders, neuropsychiatric syndromes.

You're probably thinking that doesn't make any sense. I mean, cancer seems far more serious; heart disease seems far more serious. But you can see, actually, they're further down this list, and that's because we're talking here about disability.

What drives the disability for these disorders like schizophrenia and bipolar and depression? Why are they number one here? Well, there are probably three reasons. One is that they're highly prevalent—about one in five people will suffer from one of these disorders in the course of their lifetime.

The second, of course, is that for some people, these become truly disabling, and it's about 4% to 5%, perhaps one in 20. But what really drives these numbers, this high morbidity, and to some extent the high mortality, is the fact that these start very early in life.

50% will have onset by age 14; 75% by age 24—a picture that is very different than what one would see if you're talking about cancer or heart disease, diabetes, hypertension, most of the major illnesses that we think about as being sources of morbidity and mortality.

These are indeed the chronic disorders of young people. Now, I started by telling you that there were some good news stories; this is obviously not one of them. This is the part of it that is perhaps most difficult, and in a sense, this is a kind of confession for me.

My job is to actually make sure that we make progress on all of these disorders. I work for the federal government; actually, I work for you. You pay my salary, and maybe at this point, when you know what I do or maybe what I fail to do, you'll think that I probably ought to be fired, and I could certainly understand that.

But what I want to suggest, and the reason I'm here, is to tell you that I think we're about to be in a very different world as we think about these illnesses. What I've been talking to you about so far is mental disorders—diseases of the mind.

That's actually becoming a rather unpopular term these days, and people feel that for whatever reason, it's politically better to use the term behavioral disorders and to talk about these as disorders of behavior. Fair enough, they are disorders of behavior, and they are disorders of the mind.

But what I want to suggest to you is that both of those terms, which have been in play for a century or more, are actually now impediments to progress; that what we need conceptually to make progress here is to rethink these disorders as brain disorders.

Now, for some of you, you're going to say, "Oh my goodness, here we go again. We're going to hear about a biochemical imbalance, or we're going to hear about drugs, or we're going to hear about some very simplistic notion that will take our subjective experience and turn it into molecules, or maybe into some sort of a very flat, unidimensional understanding of what it is to have depression or schizophrenia."

When we talk about the brain, it is anything but unidimensional, or simplistic, or reductionistic. It depends, of course, on what scale or what scope you want to think about, but this is an organ of surreal complexity, and we are just beginning to understand how to even study it.

Whether you're thinking about the 100 billion neurons that are in the cortex, or the 100 trillion synapses that make up all the connections, we have just begun to try to figure out how do we take this very complex machine that does extraordinary kinds of information processing and use our own minds to understand this very complex brain that supports our own minds.

It's actually a kind of cruel trick of evolution that we simply don't have a brain that seems to be wired well enough to understand itself. In a sense, it actually makes you feel that when you're in the safe zone of studying behavior or cognition—something you can observe—that in a way feels more simplistic and reductionistic than trying to engage this very complex, mysterious organ that we're beginning to try to understand.

Now, already in the case of the brain disorders that I've been talking to you about—depression, obsessive-compulsive disorder, post-traumatic stress disorder—while we don't have an in-depth understanding of how they are abnormally processed or what the brain is doing in these illnesses, we have been able to already identify some of the connectional differences or some of the ways in which the circuitry is different for people who have these disorders.

We call this the human connectome, and you can think about the connectome sort of as the wiring diagram of the brain. You'll hear more about this in a few minutes. The important piece here is that as you begin to look at people who have these disorders, the one in five of us who struggle in some way, you find that there's a lot of variation in the way that the brain is wired.

But there are some predictable patterns, and those patterns are risk factors for developing one of these disorders. It's a little different than the way we think about brain disorders like Huntington's or Parkinson's or Alzheimer's disease, where you have a bombed-out part of your cortex.

Here, we're talking about traffic jams, or sometimes detours, or sometimes problems with just the way that things are connected and the way that the brain functions. You could, if you want, compare this to, on the one hand, a myocardial infarction—a heart attack—where you have dead tissue in the heart, versus an arrhythmia, where the organ simply isn't functioning because of the communication problems within it.

Either one would kill you; then only one of them will you find a major lesion. As we think about this, probably it's better to actually go a little deeper into one particular disorder, and that would be schizophrenia, because I think that's a good case for helping to understand why thinking of this as a brain disorder matters.

These are scans from Judy Rapaport and her colleagues at the National Institute of Mental Health, in which they studied children with very early onset schizophrenia. You can see already in the top there areas that are red or orange or yellow are places where there's less gray matter.

As they follow them over 5 years, comparing them to age-matched controls, you can see that particularly in areas like the dorsolateral prefrontal cortex or the superior temporal gyrus, there's a profound loss of gray matter.

It's important, if you try to model this, you can think about normal development as a loss of cortical mass—loss of cortical gray matter—and what's happening in schizophrenia is that you overshoot that mark. At some point when you overshoot, you cross a threshold, and it's that threshold where we talk—where we say this is a person who has this disease because they have the behavioral symptoms of hallucinations and delusions.

That's something we can observe, but look at this closely, and you can see that actually they've crossed a different threshold. They've crossed a brain threshold much earlier, that perhaps not at age 22 or 20, but even by age 15 or 16. You can begin to see the trajectory for development is quite different at the level of the brain, not at the level of behavior.

Why does this matter? Well, first, because for brain disorders, behavior is the last thing to change. We know that for Alzheimer's, for Parkinson's, for Huntington's, there are changes in the brain a decade or more before you see the first signs of a behavioral change.

The tools that we have now allow us to detect these brain changes much earlier, long before the symptoms emerge. But most importantly, go back to where we started—the good news stories in medicine are early detection, early intervention.

If we waited until the heart attack, we would be sacrificing 1.1 million lives every year in this country to heart disease. That is precisely what we do today when we decide that everybody with one of these brain disorders, brain circuit disorders, has a behavioral disorder.

We wait until the behavior becomes manifest; that's not early detection, that's not early intervention. Now, to be clear, we're not quite ready to do this. We don't have all the facts; we don't actually even know what the tools will be, nor what to precisely look for in every case to be able to get there before the behavior emerges as different.

But this tells us how we need to think about it and where we need to go. Are we going to be there soon? I think that this is something that will happen over the course of the next few years.

But I'd like to finish with a quote about trying to predict how this will happen by somebody who's thought a lot about changes in concepts and changes in technology: "We always overestimate the change that will occur in the next two years and underestimate the change that will occur in the next 10."

Bill Gates, thanks very much.