Visit Your Local PBS Station PBS Home PBS Home Programs A-Z TV Schedules Watch Video Donate Shop PBS Search PBS
VIDEO ARCHIVESCHEDULETOPICSABOUT FRONTLINEBUY DVDSTEACHERS
The Medicated Child [home page]

Kiki Chang

photo of chang

Kiki Chang, M.D., is the director of the Pediatric Bipolar Disorders Program at Lucile Packard Children's Hospital in Palo Alto, Calif., and is an associate professor of child and adolescent psychiatry at Stanford University School of Medicine. He is a rising star in the field of juvenile bipolar disorder research. In this interview, he outlines his provocative new idea that early intervention -- identifying and treating children at high risk for bipolar disorder -- can prevent the development of full-blown bipolar. He also discusses the "uneasy partnership" between researchers and the pharmaceutical companies who fund clinical trials, the usefulness of brain imaging and why childhood bipolar diagnoses seem to be mainly an American problem. This is the edited transcript of an interview conducted on Oct. 8, 2007.

[How did you get into child psychiatry and specializing in bipolar disorder in children?]

I went into psychiatry in general because I found the interaction between the brain and behavior to be really fascinating. I also found there was a lot of biological research that needed to done, and we were behind a lot of other medical fields.

And I've always loved working with children. Childhood is such an important factor in our development. I believe psychiatric disorders really have their roots in childhood. So I got interested in looking at children -- this was in the mid-90s -- and seeing if they had early symptoms of bipolar, especially if bipolar disorder ran in their families.

During my psychiatric residency, I realized in working with families in which bipolar disorder ran that the children were really having early symptoms of bipolar -- maybe not the full disorders already, but they were having some early symptoms. And a lot of the parents said, "He reminds me a lot of what I was like when I was 12, when I was 10, when I was 15."

I was specifically interested in bipolar disorder because it was a very interesting and common disorder, affecting, depending on how you look at it, anywhere from 2 to 4 percent of the population. It's been around forever. So why is it still here? Why do we still have it? When somebody has a manic episode, it's very notable. It's not something that you can hide, like anxiety or depression even. It's very notable.

And you say that back in the mid-90s, it was not a diagnosis that was being given to kids.

Exactly. There was just some literature describing it as very rare in children. It's really in the last decade that the explosion has happened -- in recognition, in diagnosis and in research of bipolar disorder in children.

Why is there so much attention being focused on bipolar disorder? I believe, number one, it is a very common disorder, and number two, there has been a growth in the amount of kids with bipolar disorder in this century. I can't prove it, because we were never looking for it before, and we were diagnosing it in different ways and looking for it in the wrong way.

So I think a lot of the increase right now in diagnosis is due to better recognition of the disorder. But I also do believe, without proof, that the incidence of the disorder in childhood has increased. And there's a lot of models for that, too -- for example, genetic anticipation. This is the phenomenon in which a disorder in each successive generation starts appearing at an earlier age. So you go from Grandma having it at 45, and then Mother having it at 25, and then the child having the onset at 15. Usually, in these generations, there's increased severity, too. It sort of gets worse as you go on. So that may be one of the reasons why we're having more diagnoses now of bipolar.

You're saying, then, that not only are kids displaying bipolar at a younger age, but that their cases are more severe than their parents' and their grandparents'. What could explain that change?

There may be many reasons. One of the reasons that, again, I have difficulty finding evidence for would be the environment. Look at the huge growth of toxins in the environment over the last 100 years with our industry. What is that doing to our development as children?

Also, the advent of different technologies changing the way we live -- artificial lighting, for example. Some people think that's linked to an earlier onset of puberty. I don't know what's causing earlier onset, but certainly there has been a trend to have earlier and earlier puberty. So that might lead to earlier and earlier onset of the [bipolar] disorder as well.

And finally, there may be genetic causes for each generation having an earlier onset of bipolar disorder. In other disorders like Huntington's disease, they have found that the earlier children or adults have the disorder, it has been linked with sequences in the genome that expand with each generation. And the more these sequences are expanding in the chromosomes, the earlier that you get the disorder.

It's little bit controversial. Some studies have suggested that that's what's happening in bipolar disorder, and other studies have refuted it. But because we don't have one gene like we do, say, for Huntington's, it's hard to pin down what's happening genetically to cause this.

So bipolar wasn't something that was identified in kids when you were a resident in the 1990s. When and why did that change?

One very influential article was Barbara Geller's in 1997. It was a 10-year review in the Journal of the American Academy of Child and Adolescent Psychiatry recapping what was known about pediatric bipolar disorder. It generated a lot of interest among researchers.

But probably the biggest explosion might have been The Bipolar Child, a book [published] in 2000. The authors [Dr. Demitri Papolos and Janice Papolos] went on talk shows, and that publicity made people start taking a hard look at the disorder.

Plus, there was some research coming out suggesting that these adults that we've been diagnosing with bipolar disorder, many -- if not most of them -- had their onset of their illness during either childhood or adolescence. That was from a 1997 questionnaire to a support group for adults with bipolar disorder asking them when they had onset of their first symptoms. With the findings that two-thirds of them had it before age 18, this was a big smack in the face of what we used to think about bipolar disorder, and we realized that we're probably missing a lot of these early forms of bipolar disorder that may not look fully bipolar until later. That's when a lot of the research started coming out.

And the skeptics might say, "Well, also the disorder started appearing when atypical antipsychotics were put out on the market." Do you think that that coincidence had any impact?

I really don't feel that there's a connection at all. I could see how one could make that link. But you could argue that prior to that, we've had lithium, for example, since the middle of the 20th century. We've also had anticonvulsants, carbamazepine and valproate used for many years before the mid-90s.

So I don't really think there's a connection precisely. The connection that one might think about is that there are studies coming out now showing that these treatments are effective. Not all these studies have been published yet. For example, the study with olanzapine just got published this month as far as a large trial with children with bipolar disorder comparing an active agent -- in this case, olanzapine -- to placebo.

So going back to your earlier question, the one connection would be then, well, these are pharmaceutical industry-generated studies, not the government. The government gave them incentive to do these studies and sometimes require these studies to be done, but without these companies, would these studies have been done? And then would people now be recognizing bipolar as much, because then there wouldn't be these studies?

So there might be an indirect link. But I think overall, the final effect is good. We are generating research with medications that now are showing efficacy, and also we're understanding more of the safety of these medications, because they're going to be used no matter what, because the families are desperate. The children are failing out of school; they're doing horribly at home; they're using substance; they're not able to function; they don't have any friends. They're completely derailed from a developmental standpoint, a psychological standpoint, a social standpoint, an academic standpoint. They're going to get medications, and we need to know what these medications are doing.

Bipolar is a spectrum disorder. How do you deal with that dilemma?

Yes, bipolar disorder overlaps with many different disorders. It's even more problematic with children, I believe, than for adults. For example, distractibility, a hallmark feature of ADHD [attention deficit hyperactivity disorder], can be also present during a manic [bipolar] episode. So how do you tell the difference?

What we try to do is stick with the criteria that are outlined in the DSM-IV [Diagnostic and Statistical Manual of Mental Disorders]. Now, these criteria were developed for adults with bipolar disorder. But we feel that if you use the criteria and apply it to children in a developmentally sensitive way, understanding that grandiosity in an adult is not going to look the same as grandiosity in a child, then we can make the diagnosis using the same criteria.

This all becomes a big argument over taxonomy, over our words and how to define it, because people say, "How do you know that's really bipolar disorder in a child?" Then I'll say, "Well, how do you know anything's really bipolar disorder?" You know, humans made up the term "bipolar disorder" to call this certain condition bipolar. It's really an artificial designation.

There is a big spectrum in bipolar disorder, because we're humans and we have so many variables and so many different brain circuits. So bipolar disorder is going to be expressed many different ways, depending on [what] your genetic and your neuronal makeup is.

It becomes almost artificial to try to put all these kids into boxes, yet that's how people operate. That's how these medications operate, because you get an indication for a box. So we still are trying to put people in the boxes, and that's understandable. It's just not great science, because the reality is no one fits in a box. We're very varied.

And that's why we need to understand, OK, if you have gene A, B and X, all right, and you have this kind of brain anatomy here, then you're going to be like this kind of person and that you may respond to this kind of treatment. But if you have gene C, D and Y, and you have this kind of brain over here, you're going to respond in a slightly different way. And that is all within bipolar, but it's getting a little bit more specific to individual differences.

Having said that, getting back to your original question, we don't have those biological tools yet, so clinically, we're still left to rely on these symptoms that appear, and there are symptom report[s] by the parents and by the child and by teachers. It's not perfect, you know. You won't see the child in the middle of a manic episode.

But what does classic bipolar look like in kids?

It does look like there are classic forms of bipolar in children. The children who have this type of episodic bipolarity, they'll be doing relatively well, and then they'll have an episode, kind of like an adult. You can see frank euphoria -- they are "up" more than you would expect, and other people notice, and their friends notice. They may not need as much sleep. They will be very energized. They may be singing when it's not appropriate to be. They may think that they're indestructible during those times and can go outside and jump from trees and do bicycle tricks and not get hurt.

They can be really involved in doing different kinds of projects. They may not be projects like [bipolar] adults do, such as spending a lot of money, because children may not have money, or starting new business ventures. But they'll do their equivalent, like trying to put on plays for the neighborhood and charging admission or working on the same project or just getting up in the middle of the night and writing words over and over on a page for some project. Those are things that have similarity to adult mania. And it exists in children. Now, how often? We're not really sure.

More often, you see variations on that; the younger you get, the less episodic it becomes. The variation would be that, OK, no longer do we have these nice, discrete episodes of a week to two weeks to three weeks of mania. Now there are short bursts within the day, where you have maybe three to four hours of this kind of behavior, and then maybe followed by an hour of sadness and crying. Does that count as a full day? Does that count as a week? Well, if you have those every day for a week, maybe that does represent something that's similar [to adult mania].

This is probably where the crux of the diagnostic problem lies: The duration of mood swings in children becomes a lot shorter as you get younger than they do in adults. And no one knows what that means. Does that mean that children have the same disorder as these adults do, or does it mean that we're seeing a different disorder? The only way we're going to know for sure is to do the long-term follow-up studies.

Now remember, adults with bipolar disorder also have a wide spectrum. There's classic adult bipolar disorder, and there's other kinds. Rapid cycling exists in adults, too. There's mixed states, where you have mania and depression at the same time. There's other disorders these adults can have, these comorbid disorders.

So, what does that look like in children? Well, it's the same thing. In children, they could have rapid cycling. They more frequently have other comorbid disorders. Disorders happen at the same time, like ADHD, anxiety -- as they get older, substance use disorders. Then those are the kids in which it's not as clear, because there's all this other stuff going on at the same time. And those are the kids who are a little bit more what people think of atypical bipolar disorder. But they still met the DSM-IV criteria, so we're still including them in the same group right now.

So we're not even sure that the kids who are called bipolar now will be bipolar as grownups?

We're not 100 percent sure. Again, if you talk to adults who have the disorder and you ask them what they were like, a lot of times they were exactly like the kids that we're seeing now. So you can work backward and say, "Yeah, I'm pretty sure that these kids are going to grow up to be the adults that we're seeing today."

There's some problems with that kind of research. It relies a lot on people's memories. But what we really need are to follow these kids out over time.

Now, the few studies that have followed these children out over time do suggest that yes, they indeed continue to have this disorder. It doesn't go away. It's chronic, it comes back. The episodes happen again, so they relapse. And so far, into young adulthood, it does look like they are continuing to have what we call bipolar disorder.

How important is irritability as a symptom in order to diagnose bipolar?

Irritability is very important. The problem is, irritability is a very common symptom to have. Everyone gets irritable. Where is it coming from? That's a whole different question.

With kids who have big temper tantrums -- this happens a few times a week -- is that coming from bipolar disorder? Often in the community, because there's so much attention paid to the [bipolar] diagnosis, sometimes they will get that diagnosis, but it's not based on any symptom other than temper tantrums.

The most common cause of irritability in children is just being unhappy, right? If you have an unhappy child, it's possible you may have an irritable child, too, someone who snaps back, is oppositional, throws a temper tantrum. It doesn't mean they have mania or bipolar disorder.

But it can also come from full depression. It can come from anxiety. It can come from children with autistic spectrum disorders, who don't want their world to go outside this one prescribed box, and if it does, they blow up. So it can come from many different places. Post-traumatic stress disorder, kids can blow up because of that, too. So I think it's dangerous.

But when you take irritability to the extent that some of these children exhibited, that it's very severe, and you start weeding out some of the more common causes of unhappy child, and then when you add the other manic symptoms in there -- decreased need for sleep, increased goal-directed activity, racing thoughts, all those things together -- hypersexuality, for example -- if you have that together with extreme irritability, you're at least somewhere on the spectrum. And if you narrow down the rest of the symptoms, then you can get a good idea that this child may have full bipolar I or II disorder.

And if you're looking at the symptoms, you're talking about needing to find all of them together.

Right. I think that's the key thing. Say you have one or two symptoms of mania. You may not be having mania. If you have four or five during the same time, during the same general episode, whether it's during one day or whether it's during one week, then I think you're more on the right track, right to mania.

The problem is, again, this is a developmental disorder. So some kids only do have two symptoms, yet if you wait a year, they'll have four to five symptoms. The real question that we're trying to answer in our research is, well, how do we figure out which kid with two symptoms is going to develop the five symptoms, and how can we get in there and prevent that from happening?

So let's talk a little bit about delaying or even preventing a child developing full bipolar. Why is it so important to prevent it? And how can you prevent it?

We believe it's a disorder that kind of takes on a life of its own. There's this theory in the literature -- the kindling theory -- that was first presented by Bob Post [Dr. Robert M. Post], and the theory is that you have some genetic predisposition, right, some genes to get this disorder later on. But you also have to have the environment coming in. So certain stressful events or a stressful environment, together with these genes, are going to send you in a certain direction. If you have those bipolar genes, you're going to go toward a bipolar direction.

Eventually what happens is probably either before or after the first mood episode, you're not going to need as much of those stressors anymore, and the disorder is going to take on a life of its own. So your next episode is going to happen easier. Then more and more frequently you will get these episodes, and it will become harder to treat, too. So it's a phenomenon called kindling, where this happens naturalistically.

Now, we feel if you get in early, before the first full mood episode, before too much has kindled in the brain, then perhaps we can stop this whole kindling process and either delay the onset to full mania or make it so that it's not going to be so debilitating and have such high morbidity/mortality in this group with bipolar disorder.

How would you intervene before it's full-blown bipolar?

First we have to identify who those kids are, but the intervention -- that's a good question. Right now we're investigating multiple areas of intervention. Usually for the psychiatrist, the major tools are either therapy or medications, and number one is therapy. We're very hopeful that changing the environmental part of it, the stress part of it, could delay or prevent the onset of mania.

So first of all, educate the families so that they know what they're dealing with and how to prevent mood swings or mood symptoms. If they're already taking medications, [educate them about] what the medications are for, what the side effects are, why they're taking them and the need to take them. So that's the educational piece that we try to do in therapy.

Then there's other things. If we can decrease the amount of stress that's going on in their family on a daily basis, maybe we can also decrease this progression. If we can improve their coping skills and their problem-solving skills within the family, maybe we can delay this progression.

So therapy's number one. We've started a study with a University of Colorado team to study a family-based therapy intervention that's designed to be preventative before the child has bipolar disorder -- looking at high-risk kids.

The second part of it is medications, and I'm really excited about medications. People are so concerned about the long-term side effects that they have on children. And I agree, those are things we should be concerned about. But one thing that people often overlook are the possible beneficial side effects.

We know it acts acutely on symptoms to help them right now, but what about long term? These medications have effects on the brain directly -- that's how they work. But some of the effects on the brain could also be protective. In animal studies and cell-line studies, there's some evidence that suggests that these medications have what we call neuroprotective qualities. They actually protect the brain against injury and insult. They may even help with healthy neuronal growth in certain areas. If that's the case, perhaps finding the right medication early on can protect a brain against all these further insults of kindling later on and maybe even directly have some sort of neuroprotective effect, so that these children never do progress to full bipolar disorder.

But it also has enormous risks, because perhaps this person doesn't develop bipolar, and you're submitting them to the risk of tics and whatever you might develop down the line.

Right, and this is where you go into the whole risk-versus-benefit thing that, as physicians, we're always taught to consider. And we do consider that. Then it becomes a familial decision, and many of the families know this; they've seen it; they have it themselves. And they say, "I don't want my child to go through this." It's going be up to them to decide for the child, based on the facts that we have, whether or not they want to go through with this type of intervention.

Is there any evidence that medications are neuroprotective?

We have a little bit of human evidence to suggest that things such as lithium, or even some of the antipsychotics, things like valproate, may be neuroprotective. But as far as a real clinical evidence of how to get there early and prevent it, we don't know.

We have anecdotal data from an earlier study we did in which we used valproate, and some of these children who we thought were high risk are doing great now. They never did have their first full manic episode. They had strong genetic loading for it and early symptoms, but then they've never developed it, and they're doing fine now. Some of them even got taken off the medication.

So I think that's the great potential, because children's brains are still developing. And because they're still developing, it's possible that with the right environmental guidance, the right medication guidance, we can help them develop healthily so that eventually they come off of all medications. It was sort of unheard of before.

Even children with full bipolar disorder, it might be possible, because we're treating more kids now than we ever have in the past.

On the other hand, there are some children in which this was tried, too, and they did continue to progress to full bipolar disorder. Now, I don't know of any instances in which the medications caused so much problems that it became more of a problem that they were taking medications than their disorder. In all cases, it's either they're doing fine or their disorder did progress to the state where, despite the intervention, they were still developing bipolar disorder.

It's pretty fascinating and complex, though. What impact does the medication have on the brain? I know you did an early study -- it might have been 2003 -- where you were talking about Depakote and its impact on the brain.

We have some early indications of how these medications work. For example, we know that things like valproate, or Depakote, are generally GABAergic. In other words, they increase a neurochemical in the brain called GABA [gamma-aminobutyric acid], which is inhibitory. It kind of shuts down things.

Now, how it works is probably much more complex than that, but you can get the idea that if somebody is having overactivation in certain areas of the brain, they're not getting enough inhibition. By increasing the inhibition, you might be helping their mood out.

Valproate has also been studied, as I said before, in cell lines and in animals in which it can do things such as increase the amount of a neuroprotective protein that's known to be neuroprotective in the prefrontal cortex, in the front of the brain, that controls mood. So that's a direct effect saying, "Hey, it is doing good things to the brain." Now, that's in animals, but we haven't tested that in humans, and we probably won't, because it's very difficult to get that kind of information.

But we have an idea that at least that's how it's acting on the brain and that, perhaps, those beneficial effects could continue to be protective in some way from the stressors, from the mood episodes, because every time they have another mood episode we feel your disorder gets worse. And so if you just simply prevent the mood episode, that in itself can be neuroprotective.

When you have a manic episode or a huge depressive episode, what impact does that have on the brain?

We believe that the more episodes you have, the more the brain easily has a next episode. It becomes part of the brain's pathway. It realizes, "OK, now this is the way I go." ... Once you have that, there's no turning back. There's very, very few cases in which people who have manic episodes don't have a reoccurrence down the road.

So, as far as what happens on a neuronal basis during an episode, it's really kind of speculative. But there have been some studies showing that people who have more mood episodes have high atrophy to certain parts of the brain, or certain parts of the brain shrink.

Now, is it because when they have more mood episodes stress hormones go up, and stress hormones are neurotoxic? Possibly. There may be other mechanisms as well, but there have been those studies to suggest that.

You say "we feel" more than you say "we know." How frustrating is that?

It's frustrating, but I think when you're dealing with something as complex as the brain, it's hard to say "we know" about most things. But I'm pretty conservative. I think that we know now a lot more than we did 10 years ago. But this is the brain. It's super complex. It's the most complex organ in the body.

But we are understanding now brain anatomy. We know what the different regions are responsible for. We know how they act together to create mood, to create behavior, to create movement, speech, all of that. And we know what can go wrong in those things.

Who is Dr. Joseph Biederman, and what impact has he had in this field?

He is a very well-known child psychiatrist from Massachusetts General Hospital and Harvard who I'd say, along with Barbara Geller, who's at St. Louis -- the two of them together have done a lot of the groundbreaking research in childhood bipolar disorder since the mid-1990s until now.

And what has his contribution been?

He and his group, including Janet Wozniak and the other researchers there, Eric Mick, they have really taken a hard look at the kids with explosive irritability. I think that's been their big contribution. It's that they have been able to take these children and see what's going on and see if they have bipolar disorder or not and bring that idea that, hey, a lot of these kids who are explosively irritable may indeed have bipolar disorder. A lot of these kids with ADHD, in fact, may have bipolar disorder.

They really brought that up to the forefront and cast a big light on it, and I think that's generated a lot of discussion and a lot of thinking and research over the past decade.

But there have been those who say that he [Dr. Biederman] watered down the definition of bipolar, and ever since he did that many more children have been diagnosed.

Well, it may not be his fault entirely. I think it comes more from when people, or even professionals, who listen to these things and they say, "Oh, my child's very irritable; they must have bipolar disorder," or, "My child has big temper tantrums; they must have bipolar disorder."

It's when they don't really listen to the research or know all the details of the research that says, "No, that's just one aspect of it. You really have to have this, this and this." And usually it runs in families.

[Dr. Biederman] never said, "Hey, irritability means you have bipolar disorder." He never said that. But he brought irritability forth as a major player in pediatric bipolar disorder. So quite honestly -- no, I don't believe he watered down anything.

I mean, DSM-IV is DSM-IV. That's the problem with it: It has room for ambiguity in there. And he just took it and interpreted it in a certain way and went with it.

Now, has he pushed the envelope? Yeah, I think so. I think he has been someone to really push the envelope. I think in the interest of science in general you need those people. You really need those people to take a firm stand, even if it's a little controversial, and push the envelope.

Now, when we're talking about children's lives and well-being, that's a little more sensitive, and you've got to be careful about pushing the envelope. Again, it's how his research and his group's research has been interpreted.

I think Barbara Geller has pushed the envelope as well, but she hasn't taken as much [of] a controversial stand. It hasn't been on the forefront, because she tends to keep to herself and not do media interviews and things like that. She has really done her own research and has been just as much pushing things forward, but sort of [in] an under-the-radar type of way.

What's her contribution?

Well, her contribution is similar in letting people know that there are a lot of kids out there with bipolar disorder that we haven't been recognizing before. She takes a little bit more of a different slant on this. She takes a euphoria slant, or grandiosity. They have to have one of those symptoms, or else she doesn't call them bipolar. So just having irritability is not enough.

So you can see it's a slightly different patient group than Biederman's group, yet they overlap. And her studies do show that their samples do overlap. We're talking about similar samples, and we're talking about a spectrum, and we're talking about kids that are somewhere on that spectrum.

Now, maybe someone on her end of the spectrum over here is going to look very different from one on Joe Biederman's end of the spectrum over here. That's true, but we've now pushed the spectrum out. Now we can start to identify what the differences are.

We haven't even talked about this whole brouhaha of bipolar disorder -- the "not otherwise specified," or bipolar NOS. And that's a whole other ballgame, because bipolar NOS is a diagnosis that doesn't really mean anything. It just means that these children don't fit the strict criteria for bipolar disorder, yet we still feel they have it. These are kids who might be explosive and irritable and have a couple symptoms, and you might call them bipolar NOS, but you have no way [of] proving it, because we don't have firm criteria for NOS.

Yet probably the most common label that kids with bipolar disorder get is this NOS label. And there's somebody else who's trying to understand that, including ourselves and other places. There's also Ellen Leibenluft at the NIH [National Institutes for Health], and she's trying to understand how these more explosive kids without clear episodes and not full manic symptoms, how those kids differ from the narrow diagnosis, or bipolar I kids, who have very firm episodes and firm criteria for diagnosis.

But really, what you need to recognize is that, OK, you say bipolar -- it just puts them into spectrum. Let's take a look and see where they are, and let's study those, and let's figure out who's going to respond to what better.

Maybe kids over here don't need as much treatment. Maybe they're the kids who, with good psychosocial intervention, resolve over time and do better. And maybe these kids over here really need this kind of intervention. That's the kind of research we should be doing, instead of taking all our time arguing about whether this is bipolar or not.

But that's what a field does when it's just beginning to understand what something is.

But there should be more time and energy put into understanding the neurobiology, because that is really the future -- putting our resources into understanding genetics, neurobiology and any other biological facets of the disorder -- because clearly there is a biological interaction with the environment that causes this kind of disorder.

And in the meantime, kids are caught in this sort of conundrum where scientists who are trying to find answers don't have them quite. And yet we're medicating. How do you feel about that?

I feel like, again, when you're dealing with the brain, you have to take a certain degree of uncertainty and be able to live with it, because again, these children are sick. They're ill. They need help.

I've seen therapy do bad things, too, if you do the wrong kind of therapy. People don't often think about that. Any kind of intervention you do with a child is going to have repercussions, and we just hope that we're doing the right ones.

The data that we have so far indicate that yes, we are saving lives. These medications stop kids from committing suicide -- for example, depression and the SSRI [selective serotonin reuptake inhibitor] users. Now that SSRI usage has gone way down because of the black-box warning that they may cause suicidal behaviors and thoughts in children, suicides are now going up again for the first time in recent history.

Any kind of intervention, there are risks. But again, that's why we have physicians, and hopefully they're competent enough to be able to explain to the families what the risks are and what the benefits are. Right now, in my eyes, there's still such incredible benefits to treating truly bipolar kids with both medications and therapy that it's a no-brainer. We need to be doing this, but we just need to be studying it better for the future.

But if there's been a 40-fold increase in 10 years, what's up with that?

Well, that study that you're referring to was a study of just community diagnoses. So we're not talking about kids who were subject to a strict, research-based diagnosis that most people feel better about when you're talking about diagnosing somebody with bipolar I disorder. We're talking about what was written down on an insurance form, a bill.

So what I think it means is that it's become a more commonly accepted form of diagnosis to use for a child, rather than there's actually more children with this diagnosis. So people are using it. I don't really feel that there's truly been a 40-times increase in the actual amount of kids with bipolar disorder.

I do believe there's a lot of children being overdiagnosed with that. I see that in my practice. We never used to think that kids had this. We used to think they were bad kids, or they got thrown in the juve hall, etc. Now we're realizing, "Hey, they can have bipolar," so we're diagnosing it. So increased recognition, better way of diagnosing it, I think have led to all of these things, and overdiagnosis as well, too.

Overdiagnosis. You say you see this with those who come into your clinic. How is this happening?

It's usually the children who are having the big anger outbursts and they have a diagnosis in the community of either bipolar disorder or bipolar disorder NOS, and I see them for a consultation. About half the time, I remove the label of bipolar disorder. These children instead are having things such as autism spectrum disorders; they're having anxiety disorders or depressive disorders leading to irritability.

But for a lot of these children there's a lot of mood dysregulation going on, kids who at age 9 should be controlling their moods better but can't. And where is that coming from? Is that coming from bipolar? Not in a lot of cases.

But if we wait another five years, would these kids look more bipolar? Possibly. Or could they look more depressed? Possibly. Or could they go on to have other kinds of problems? Possibly. Or could they just get better?

So I do believe the term "bipolar disorder" in certain areas of the country, including the Bay Area, are being overused. I would argue that, in some ways, that's a good thing. It's kind of like appendicitis. You have to catch some normal appendices when you're doing those surgeries in order to know that you're being careful enough in calling the diagnosis, OK? But just because we think of bipolar and we're bringing these kids in and we're not missing it, we also don't want to treat them erroneously with the wrong kinds of medications.

Now, that's more of a problem if you think they have ADHD or depression, because you're giving them medications that may not be in their best interest if they really have bipolar.

Probably 50 percent of kids who have bipolar disorder are really sensitive to these kinds of medications. If you give them an antidepressant and you make them worse, that's bad. With bipolar disorder, at least most of the time, the medications we use for that are not going to make people psychiatrically worse, like a stimulant or antidepressant could.

The main concern is about side effects and other physical adverse effects of those medications. But one of the side effects they don't have, usually, is making the kids worse. So in some ways, overcalling it can be safe. But you want to make sure you have the proper diagnosis as soon as possible.

Let's talk a bit about the medications and how much we know about them. We are using in these children medications that are tested in grownups, and we're dosing for kids' size and weight. But do we even know if they're working?

In some cases yes, in some cases no. For example, anticonvulsants: A lot of them have been studied in children, and they do have dosing guidelines based on weight, which is different than for adults, which is just a fixed dose kind of guideline. The problem is they've been studied for epilepsy and not behavioral disorders, so we don't really know how that works for behavioral disorders, or psychiatric disorders in general.

For things like lithium, again, it's been studied mostly in adults. There's never really been a good study in kids for lithium. So we do need more of that, and there is a study going on right now, eight different sites in the country investigating lithium's use in children with bipolar disorder.

We are really at the forefront of an explosion of understanding these medications in children. Antipsychotics, same thing -- again, originally developed for adult populations and tested for adults, but because of governmental regulations, financial incentives for the companies and just general science, we're getting a lot of information. This month alone, we're getting an explosion of data from hundreds and hundreds of children who were in these studies to see whether or not these atypical antipsychotics work better than a placebo. What we don't have [are] good head-to-head studies between different medications. And adults don't have that either.

... [And] unfortunately, most kids with bipolar disorder will need more than one medication. That's because we haven't developed the one magic medication that's going to do everything. So right now, very often they're on two or three medications. Once you start getting above that, then you start getting a little concerned about, what are the different interactions of these different types of medications, and we have to be very careful about that.

How confused are doctors out there about how to use the medication?

Unfortunately, just like any field in medicine, there is a varying level of competency. You would hope that the standards of medical school would really weed that out, but it doesn't.

So you just have to pick your doctor carefully and find out who has had good results and references and credentials, because in psychiatry, just like every other field, there's a variation. Some people aren't doing a good job in dosing and combining medications and not trying to do what we call empirically based treatments, or basing it on evidence. They're just kind of winging it out there.

And not all of it is their fault. Some of it is because they don't have access to all this information that's coming out now, but that's why I think we need to bring access.

So who's responsible for the overmedicating? Is it pediatricians unsure of what they're doing?

I don't know really. There's some studies looking at that. Overall, I'd have to say it is probably folks who don't have access to good psychosocial interventions or good family-based programs in their communities. There's not enough money in poorer areas, so they have to rely on medications more.

I would say it is possibly folks who don't have a child psychiatry background. In a lot of places general psychiatrists who have no child psychiatry training are forced to treat kids because there's no one else to do it. Pediatricians too, but I don't see that as much as general psychiatrists. So again, if they had more proper training, I think they could do a pretty good job with it.

What needs to change?

We need more funds. We need to develop more child psychiatrists. It's amazing how few child psychiatrists really are out there.

Seven thousand, they say.

OK. I said 7,000 10 years ago. I was hoping it's higher now.... Think about that. Is there 7,000 of any medical specialty out there? It's such a low number.

Yet if you haven't had child training, how are you supposed to know what a 6-year-old who is acting out -- what's going on with them? You may have no idea.

And you're at the forefront of a lot of this research today. Where do you fit in that puzzle?

My own personal interest is in prevention and early identification, understanding biological markers and ways to intervene early with the child so that hopefully in the future, we won't have to use medications in children. We won't have to worry about the long-term side effects, because the amount of medications they'll be on will be much reduced or zero, because they won't have full expression of the disorder.

I'm trying, in a way, to put adult psychiatrists out of business, because I want to get to the root of this. And I think many child psychiatrists [and] psychiatrists feel the same way: These things begin early on; let's develop early-intervention prevention programs.

Do you feel there's enough funding from the government to do the research that you need?

There is funding, but they have been cutting it the last few years. There's a lot of good research that we've tried to do that probably would have gotten funded during other decades that, because of our current funding crisis, did not get funded.

We've had to try to turn to private sources, and that's very difficult given the current stigma of psychiatric disorders. But we're trying very hard to see who is interested in funding this kind of research, either foundations or just private donors to fund directly.

And you need to partner with the pharmaceutical companies to get the research done. How much of a dilemma is that?

I think it is an uneasy partnership because they have the money and they're going to do the research in some ways, because either the government requires them or because they want to do the research anyway. It's got to be done right, and there needs to be people who are advising them how to do it correctly.

The question is conflict of interest, really. Do these people then, including myself, feel a pressure to report only positive results for the company?

And the problem is you can say "no" all you want, but yet there are unconscious motivators, and there are other kinds of stressors on you to produce positive results. This doesn't just happen for Pharma; anyone who is an academic psychiatrist is under pressure to have positive results.

But people just see it as big business being evil. And maybe big businesses are, but I can tell you that I've worked with plenty of big businesses that are not evil, and they really do have this kind of balanced agenda. They have to meet their bottom line, but they also care about what they're doing and want to have good products for the community.

You know, they've done such an incredible amount of work that without them, where would we be now?

And yet you call it an uneasy relationship.

It is uneasy, because there is this potential conflict of interest and definitely that perceived conflict, even if there isn't a conflict of interest. ... So that is a problem. We can solve that by having, again, the government fund everything, but there's not enough funds.

But aren't the companies driving the research, because they decide to figure out what questions will be answered and which ones won't?

Not really. We have the questions. It's up to them to say whether or not they want to fund the study. Now, in certain cases they say, "Well, we won't fund the study unless you do it this way." And then it becomes an ethical question: OK, I wanted to do it this way originally; they want it this way.

In some cases we've just not done the study, because it's not going to work.

When you're dealing with pharmaceutical companies, doing research trials on their medications, have you ever felt that your research has been compromised in any way?

No. Never felt that has been a problem at all. No one has ever come in and said, "You know, those results, maybe you could present it this way," or, "Why don't you say this?" For me, it's never been an issue at all.

But can you do the research without the help of the pharmaceutical industry?

We can. We don't need the pharmaceutical industries to do the research. We need money, though. It costs a lot of money to do this kind of research. And so we do need the NIH budget to be increased.

When I was researching FRONTLINE's earlier report on ADHD seven years ago, it was thought that 20 percent of the people with ADHD would outgrow it, and now I think it's reversed. One doctor told me it could be as much as 80 percent of kids with ADHD may outgrow it. Is there a danger that we believe one thing now, and then we do more research, and this totally changes? And how to deal with that?

I think there is some danger in that. But again, we have to live right now. You have to live where you are. And the problem is, these kids are not doing well. They're doing horribly at school and socially, and horrible at home. They're developing bad habits, bad self-esteems. They're having suicidal thoughts. You have to do something now.

Maybe 10 years from now we'll say, "Yeah, we weren't doing exactly the right thing." But throughout medicine, that's always happened, right? I guess those leeches weren't such a good idea after all. Maybe those bumps on the head really didn't mean I could diagnose you. But you have to do what you feel is best right now with the evidence that you have, and we're trying to base it on what we feel is pretty good science.

The funny thing is, people don't realize that psychiatry and psychology papers are, in general, the most scrutinized and scientifically rigorous types of med studies, because we have to be. We have to be that good to survive the scrutiny because we're dealing with a very murky area.

[What are your thoughts on the usefulness of brain imaging?]

I do feel that brain imaging will have a very important role in the near future for helping with not only diagnosis, but figuring out what kinds of treatment are going to work better for certain individuals, and also for early identification and prevention of disorders like bipolar disorder. But we're at least five to 10 years, if not more, off from that being a really reliable clinical tool. Right now, it's strictly a research tool.

There are places that will kind of insinuate that they can make diagnoses, and I think that may be misleading. We have to be very careful, knowing that information gotten from brain imaging right now at the very best can give you some sort of idea about what ballpark you're in. Or someone who has done a lot of use of clinical imaging may have a better idea [that] maybe you might respond to this medication better, maybe not.

But it's really not too much different from your own clinical judgment of being a doctor for 30 years and seeing patients of all different kinds and coming up with a judgment. As long as people realize that when they're going in for these brain scans, then I think at least they're making an informed decision on it.

But blood flow means squat, doesn't it?

Yeah. It's just blood going to a certain part of your brain, right. Right now it's really a tool to figure out more about different neurocircuitry. I don't think things like functional MRI will ever be great for really diagnosing, but maybe for predicting certain kinds of treatment outcomes and for determining whether or not you're depressed or not.

And, for example, we have a graduate student, Nancy Adleman, who's doing a study right now looking at depressed teenagers. This is a very big clinical question, which is, if you have someone who's depressed as a teenager, how do you know that they have what we call unipolar depression, or major depression? In other words, they're not going to have mania; they're just going to have recurrent depressive episodes. How can we tell that person apart from somebody who's bipolar, but yet this is their first mood episode? Two or three years from now, they'll have a full manic episode.

Why do we care? The treatment is radically different. If you treat this person with antidepressants, there's a good chance they may then develop a manic episode because of the antidepressant. And who knows what that's doing to their brain? We think nothing good, and then you've caused this patient harm. That happens time and time again, because we can't tell the difference between these two teenagers.

Well, she's doing a study where she's looking at teenagers with unipolar depression and teenagers with bipolar depression and trying to understand what the differences are between them from a functional activation, or neuronal, brain standpoint. We don't have the answers yet, but the early data suggests that there are differences between these two girls.

And maybe eventually, clinically, we could use this to tell the difference between who's going to react to this medication poorly and who's going to react better to medication. So it's a proxy. Looking just at blood flow itself, it's not going to tell you a whole lot, but being able to identify subgroups of patients, that could be very helpful.

Let's talk about CME [Continuing Medical Education; e.g., lectures, workshops for doctors] and Pharma's funding of this and whether it's become a marketing tool.

Well, it could very well be used that way. I'm sure some pharmaceutical companies probably believe that they'll get a lot of publicity and good marketing from CME events.

What I see instead is a great opportunity to get the information out to people. Who else is going to spend the money to do this kind of education? Psychiatrists are not going to go out and spend a ton of money to get information.

I think the kind of programs that Pharma can really produce are really quite remarkable. And it does take a lot of money to put on a good program and do innovative and good teaching. Unfortunately, no one else is paying for that. So in many cases, it's really been a great help.

Does it matter that the pharmaceutical companies are paying for it?

You know, doctors are going to use the medications anyway, and they're going to use them in ways that they think is right to help the patient. If we can at least teach them how to use them correctly based on the evidence, and give them enough information to make better decisions, then I think what we're doing is actually helping the overall outcome of the patient.

Now, that's where our goals are in line with pharmaceutical companies, because they do want good outcomes. They don't want patients having horrible side effects or being overdosed, because it's bad for the patient and certainly it looks bad for the company and will hurt sales, too.

Why is bipolar seemingly such a problem for American children? Why isn't it happening internationally?

You're right. It doesn't seem to be as much of a problem internationally.

We use medications in general a little bit more in the U.S., and people think that maybe that's why we have more kids with bipolar disorder. For example, if you have a depressed kid in Europe and you don't use an antidepressant, maybe they won't develop mania; maybe they'll develop bipolar disorder later when they're adults, and then they become an adult with bipolar disorder, because the adult instance of bipolar is no different. That's true worldwide. But what about the childhood incidence? Maybe in some ways we're creating this disorder earlier in the U.S. That's possible.

I also think it is really underrecognized in other countries, because, again, they're kind of like where we were 10 years ago. And now they're starting to recognize it as well. In certain countries, they're not where we are at all, psychologically and psychiatrically, at looking at this as a child disorder. They're still looking at it as behavior problems and things like that.

So there's cultural differences, stigmas. There's even some diagnostic difference. There's medication-use differences. And there's just the general bias against differences that may eventually change and ease up.

So I don't believe that it is a U.S. disorder only. I believe that maybe we do have more childhood onset. There are studies now showing that the onset of the disorder looks to be a little bit later in European countries.

If that's true, it makes sense that if the onset is later, then there's less children who have bipolar disorder in those countries than here. Maybe they're a little bit behind the curve and then eventually their onset is going to drop, drop, drop, like it has been for us, and then they'll have more kids with bipolar disorder. It's possible.

For example, when we think about America and who came to America -- people who were novelty seekers or not afraid, or people who had a little more adventuresome qualities -- there may be some links between bipolar disorder and novelty seeking, and perhaps people who have more tendencies toward mania may be more prone to make those brash, impulsive decisions.

Those are the people who came to the U.S. and truly determined a lot of our culture. And maybe those people are intermarrying more. So maybe we're getting a different strain, if you will, of bipolar disorder that looks a bit worse and happens earlier.

But it's due to genetics, and it's due to genetic combinations that are different here in the U.S. With the world intermingling more and more and more, that may cease to be the case, and everyone's going to have the same type of bipolar disorder and will all have early onset.

home . introduction . watch online . frequently asked questions . interviews . a parents' guide . join the discussion
producer's chat . site map . dvd & transcript . press reaction
credits . privacy policy . journalistic guidelines . FRONTLINE series home . wgbh . pbs

posted january 8, 2008

FRONTLINE is a registered trademark of wgbh educational foundation.
main photograph © corbis, all rights reserved
web site copyright 1995-2014 WGBH educational foundation