My name is Vik Gupta and I'm the moderator for this upcoming exciting session and a proud member of the Clinical Updates Subcommittee being led by Ron Winchell and Catherine Krohn. Next session, the title is Real-World Strategies for Bipolar II Recognition and Management Under Conditions of Uncertainty. Few housekeeping points before we get started with the introduction. There'll be a talk by the speaker who's right here, Dr. Sachs, and right after his talk and the presentation, we'll have a Q&A. So while for the Q&A, you'll come up to the mic and speak your question in 30 seconds or less. It's very critical. There is also an online component to the questions. So to be fair to your colleagues who are joining us online, I request again 30 seconds or less. Let me get started with the introduction of our speaker. Dr. Sachs is an internationally recognized expert clinician and clinical trialist with extensive experience with bipolar and related disorders. He founded the MGH Bipolar Clinic and Research Program and was the principal investigator of the Systematic Treatment Enhancement Program for Bipolar Disorder, STEP-BD. He has been instrumental in developing practice guidelines, technology-based solutions for placebo response mitigation, and methods for integration of measure-based guidance into clinical practice. He's a clinical vice president at Signant Health and past president of ISCTM. And also, his mom is really proud of him and his grandkids love him. Please put your hands together to welcome Dr. Gary Sachs. Thank you very much. I didn't know my mom was going to be your writer for that introduction, but there we go. Thank you all for being here. It really is a pleasure to talk about this topic. I want you to understand what some of the goals are. One overarching goal for me is to get to all of your questions. And the other part of it, left out of the introduction, is that I'm a PGY-43, right? And I say that because I think we're all still learning. There's not received wisdom of absolute truth in much of what we do. And hence, you see that under conditions of uncertainty. We all work day in and day out to do the best we can for our patients. But there's always a lot of question marks. So just to lay out the way the plan is, I'm going to make a little introduction around the conceptual issues. We're going to talk some of the definitions around BP2, what it is, and why it's such a challenge. And then I'm going to emphasize how you can assess your diagnostic confidence. Again, not that you're going to make the one right diagnosis, but how you can kind of have a check on yourself and say, here's how sure I am as I tell a patient and their family what I think the diagnosis is. It's really good to have this kind of framework, I think, at your fingertips. And as you'll see, lots of reasons for modesty as we do that. Then we'll move on to the sort of systematic treatment approach that we used in Step BD. And I'll define for you some of the terms and work through how that iterative approach plays out, especially the so-called two-week rule. And then hopefully we'll still have roughly half an hour for Q&A. That sounds good. Let's go ahead. One other thing, I know people might want copies of the presentation. It will not be available through the APA website because I don't have a lawyer and I couldn't really read all of the authorization. But I do want to make it available to you. If you want, just write to me. You see my address. I'll show it again at the end. If you change your mind and decide you actually do want it, you'll have another chance at this. Okay? So the first important issue in terms of major point to get across, hotly debated, what in the world is this Bipolar II? Is this something that's real? You can see back in 2008, this paper, Bipolar II Disorder, arguments for and against it being a distinct diagnostic entity, really interesting kind of paper. It kind of came down on the side of saying, hey, this is a validated condition. But there's a lot of ways things can be, quote, validated. That's more of an academic idea. And now we move on to Jin Mali's paper, and that's 2019. You see the conclusion moves to Bipolar Disorder Type II is a myth. Now rather than taking a position on this debate, our purpose really is to consider how best to address the clinical needs of our patients. We're not here to score points. And I could show you chapter and verse in the statistics that led to each of those conclusions, and we're still not ready to say definitely one way or another. And I just want you to think a little bit about this. What Jin Mali said in 2019, there's really an episode definition that is important in distinguishing Bipolar II from MDD, Major Depressive Disorder, or Bipolar Type I. And this hangs on symptom severity, duration, and impairment. I'm going to say more about it, but as you see from the right-hand side of this slide, that space that defines the difference between the two, there are no nodal points. There are no places where there's a discontinuity. So it is like saying, you know, being tall, Type I is different than being tall, Type II. Well, you know, there's a lot of variation in height, and I don't think we're ready to say that height is bimodally distributed. It's really not. And we don't have any evidence showing a bimodal distribution among the mood disorders as DSM-5 has defined them. So now we are ready for the first question. And again, we have agreed, Dr. Griffin spoke before, we're going to go low-tech here. You don't have to worry about Slido. I'm just going to ask you to raise your hands. And I want you to think about your own beliefs and practice. Raise your hand if you believe Bipolar II is a clearly distinct disease entity separate from BP-1 and MDD. I have a feeling most of you would say that, because you probably wouldn't come to a Bipolar II talk if you didn't really think that that was the case. It only makes sense that you're here. So I hope this won't hurt anybody's feelings, but I want to introduce a little bit of data about the reliability of the diagnosis. Preface that by saying reliability and validity are two different things. It could well be that we have a valid, distinct condition. We just don't have a good way of making that distinction, right? Our yardstick is not so good. So here are the results from the DSM-5 field trials. The idea here is to measure the agreement between two clinicians evaluating the patient within a day or two, sometimes the same day, sometimes the next day, under conditions that really favored agreement. If you saw the details, it is better than most research studies actually do. A kappa of about 0.8 would really be good if we were talking about agreement in most any other area. You can see for BP1, it's fairly decent at 0.56. We could have a conference on whether we think MDD is a real entity, because look at how pathetic the reliability between two patients seeing a depressed person the same day. We've got a kappa of 0.28, and that's combining several different centers. I think there were five separate centers that did that, and some were like 0.2. That is like asking somebody, is it Tuesday or Wednesday today, and not being able to get agreement. It's surprising, but that's how it is. So you can see that MDD and generalized anxiety disorder are probably things that if you believe the patient had it today, and your colleague saw the patient tomorrow, don't get into an argument about it, because it's actually very hard. You yourself on a different day might come to a different conclusion. So again, reason for modesty. Now I show you this to set up the data for BP2. You'll see it really comes down to hypomania. Where does bipolar 2 come down? Well, not surprisingly, right in between MDD and bipolar 1. And if you really think about it, since one of the things that's required for a bipolar 2 diagnosis, as you'll see, is to have had a major depressive episode, that kind of low reliability is baked in the cake, isn't it? So it's not that surprising. And current hypomania, which is really what this number represents, is an extremely difficult thing to get people to agree on, both because some people will call it mania, some people will call it hypomania, and others will say, wow, that person's having a great day, right? And they'll call it normal. So it's actually inherently difficult to do. And so realizing that the reliability is relatively low, we needn't get into big arguments about who's right. And you'll see lots of papers about this person misdiagnosed as BP1 or BP2, and somebody else who's misdiagnosed as borderline, being called bipolar, none of that really, really helps our field. We just have to understand what mood disorders are about. So let's begin with the easy things to agree about. Mood disorders are lifelong conditions. And this is a multiplex dysregulation syndrome that we're talking about, in which there are episodes of abnormal mood states. And in the mood disorders in DSM-5, those episodes typically start during adolescence, though they can happen at any time, and they alternate with periods of full or partial remission. Really pretty simple. We'll call those all mood disorders. And those kinds of mood disorders that are lifelong, they are built out of building blocks that are acute episodes. And acute depressive episodes are the most common manifestation of these dysregulated states. And the important thing to understand is now, as we have evolved from DSM-3 to 4 to 5, we begin to discern subtypes that impact prognosis and treatment outcome. That's why they're relevant to us. Some of these labels mean something about what treatment we're going to choose or how likely that treatment is to work. And that's really why we're talking about it. But again, we're always uncertain. And that is inherent to this. So don't feel bad if somebody says they disagree with your diagnosis. Depressive episodes accompanied by symptoms of anxiety and hypomania are common. We imagine that highs and lows are, quote, polar opposites. But if you have ever been to a wedding or a funeral, you realize that there's a lot about the opposite pole from what you expect that just comes out. And people crying at weddings are a perfect example of how our brains are really wired in the normal state, let alone the pathological state. So the other crazy thing about having a lifetime mood disorder diagnosis is that it's a lifetime. And that means in younger patients, the uncertainty is even higher. So if you're talking about a 17-year-old or a 27-year-old, your uncertainty has to be higher than a 47, 57, or 67-year-old. And we just have to accept that as part of the deal. The idea of diagnosing a patient as bipolar, they come to your office and they're terribly depressed or you see them on the ward and they're terribly depressed. And my colleague, Charlie Bowden, reviewed records in Texas and showed that the more severe the current depression, the more likely it was that the patient actually denied a prior manic episode that was documented on the chart. So asking a depressed patient, when did this start? I mean, I can't tell you how many people told me it's probably during my mother's second trimester. You know, it's been lifelong, doctor. That's all I've ever been. And then you go, well, I see in the chart, you know, you went to graduate school, you got married, you had kids, you did well until, oh, yeah, that's true, but I've always been depressed. So the mood state is actually the filter through which our whole assessment takes place. And we have to recognize that filter is often distorting the reality. So that's a big challenge. And then current and prior hypomanias are difficult for more or less the same kind of person. Ask somebody who's really wearing rose-colored glasses when they had difficulty, and it really can be never. So good reason to have collateral sources of information. And then I will tell you, every time I see a patient who has bipolar disorder and that's all, the first call I want to make is to the Smithsonian, because that is an extremely rare condition. We do not see many people who do not have at least one of the common co-occurring conditions. And I just want to remind you what the most common ones are, anxiety, substance abuse, personality disorders. You can see among patients, we're often dealing with people with high BMIs, ADHD, et cetera. And it's not limited to psychiatric comorbidity. In Roger McIntyre's work here, you can see, again, the metabolic syndrome that goes with obesity, cardiovascular disease, type 2 diabetes, et cetera. These are extremely common. So I've kind of set the stage that there are many reasons to remain modest and recognize our uncertainty. So let me shift now to the next part. What is it that's the definition of bipolar II? I know you don't need me to read through the DSM-5 criteria, but I do want to make a couple comments about it. There are really three core requirements to have a diagnosis of bipolar II. You must have at least one hypomanic episode and at least one major depressive episode. And then never have a full manic episode. I guess it's not on the slide, but that quote, full-blown manic episode, I don't know. That's not necessarily a technical term, but everybody seems to know what it means. And I'll run through this a little bit with you so we can be clear about our terms as we move on. A few other important definitions. So we have to have the definition for a major depressive episode. And again, you understand that there are requirements for an abnormal mood state and then the associated symptoms. So if you have at least five of these nine, including the abnormal mood state, you can have a major depressive episode. Under DSM-5, I want to emphasize the idea of assessing for these episode specifiers, particularly with mixed features and anxious distress. And I'll say more about that later. But when we talk about a depressive episode, it brings us to other important terms to be clear about, like we refer to so-called bipolar depression, a kind of blanket term that could be depression with BP1, BP2, or maybe BP not elsewhere classified, right? Somebody who has a major depressive lifetime diagnosis could also have a current major depressive episode, but they've never had any form of abnormal mood elevation. And then this one is my favorite, and I'm going to try to encourage you to adopt my biases and say, well, most of the time when I'm uncertain, I might want to use a different label, mood disorder not otherwise classified with a current major depressive episode. Why? Because I'm uncertain. And sometimes when we're uncertain, it's handy to have a good code for the diagnosis that you're going to apply. So F39, if you're thinking about it, is the code that I use the most often, particularly with a new patient who has a mood disorder when I really don't know. And we're going to come back to that in a bit. To keep going with the definitions, what's the difference between hypomania and mania? And you know, we could talk a long time about the difference between four days and two weeks and European data that showed it was just as valid if the mood elevation was two days. Could I really ignore two days of clearly abnormal mood elevation? No. But my reliability of diagnosis, the disagreement, gets much higher if we become more sensitive, we accept shorter periods, and that's a problem. So we've agreed under DSM-5 to keep the four-day rule. But I will tell you, just for being efficient, the real difference is impairment. The real difference is if you have hypomania, you do not have significant impairment in social or occupational function. You were not hospitalized. You were not psychotic. I can't tell you how many patients have come to me and said, I have bipolar 2, I was just hospitalized and finally the hallucinations are gone. If you were hospitalized with bipolar 2, either you didn't get a good assessment or I'm missing something here. So we need to talk about that. The important thing as a rule is assessing that impairment. And sometimes impairment is in the eye of the beholder. So you tell an off-color joke at your company's Christmas party that causes you to lose your job, that may have nothing to do with you and your brain condition, right? It's clearly bad judgment, perhaps, but it doesn't necessarily come out as a biomarker when we're talking about personalized medicine, that we can't rely on that. So I don't know what to make of that. There's a lot of difference of opinion about individual cases when we start to evaluate the kind of impairment that people with hypomania have. This is a good time for me to say at Yale there was a study done where they looked at, they did a questionnaire akin to the mood disorder questionnaire that you'll hear about, asking people during their four years at Yale, had they had these things? I think it was 93% endorsed at least one time. Now, I don't think 93% of Yale graduates have bipolar 1 or 2, but I am very concerned about the other 7%, right? Because think about it, they never fell in love, they never had a time of being overly exuberant, they went to sleep every night at 10 o'clock at night and never violated it. So there are just common sense things that tell you we can err on the side of excessive specificity or excessive sensitivity and we want to avoid that. One thing that helps me stay grounded is thinking about how the mood disorder concept has evolved over time. And that means that we always go back and we start with Kreplin's great insight. So Emil Kreplin in Germany at the end of the 19th century, he was going on his rounds. And let's say he's seeing a bunch of people who are all between 65 and 70, they pretty much on rounds to him and his colleagues look all alike. And it dawns on him that some of these patients had been admitted in their teens for the first time and others had no problem until very recently. So he hits upon the idea of age of onset, early versus late. And being the head of the hospital, Kreplin could make his fellows propositions that they were not able to refuse. So he said to one of them, Alois, these people who have been well all their life until recently, I think they have a different disorder than those who got ill early. And he asked Alois Alzheimer to take care of those patients and when they died and they went to autopsy, the findings that you know about, the senile plaques and neurofibrillary tangles were recognized. And that was the first pathologically based diagnosis in psychiatry, if you will. And of course, Emil Kreplin was the first person to call that Alzheimer's disease. Kind of a cool thing. But poor Dr. Kreplin, fewer people know his name in the general public, he was left with the early onset patients and when they came to autopsy, he saw nothing. And he looked every way, he was really a very innovative kind of guy. But his next great insight was to take those early onset patients and divide them by whether they ever got better after their first admission. And so those that had no recovery, not all of them, not even half of them or a quarter of them, but maybe like 10, 15% had a kind of cognitive decline that reminded him of Alzheimer's patients. And they were really different than those that had full recoveries. But the ones who had no recovery, right, these ones with cognitive decline, he labeled that dementia praecox, right? Everybody knows the term we use now is schizophrenia, right? So that's really taking us to the beginning of the 20th century. That was 1921 when that term came in. But among those with full recovery, he noticed that some had only depression, some had depression and mood elevation, and some had only mood elevation. But he thought that the similarities greatly outweighed the differences. And so he called it manic depressive insanity. And of course, we have now unipolar, bipolar, or MDD and bipolar. And those distinctions have been made often based on family history. And I'll show you a little of the related data in a bit. But here we have kind of the chocolate, vanilla, and strawberry of psychiatric nomenclature, right? These are our big three groups. And it is great in the DSM-5 era that we have begun to subtype them in meaningful ways. That's the lifetime diagnosis part. Kreplin also had an idea that was akin to what we see about those episode subtypes that we talked about. He saw the fluctuations in mood, activity, and cognition as having peaks and valleys that then corresponded to pure mania, dysphoric mania, or what we might now call mixed states, manic stupor, agitated depression, and pure melancholic depression. You can see how that's the forerunner of what we have now in DSM-5. We have those current episodes, depression, hypomania, mania. We have the episode specifiers, and we have our lifetime mood disorders. Not really such a great departure from Kreplin's time. Adding in things like course specifiers, rapid cycling, seasonal, postpartum, others that are in there. It's really important, though, to remember that modest reliability. I believe, especially for research and using structured assessments, but when in doubt, which is almost always the case, especially at the beginning, I want to stay modest. And so that not elsewhere classified is my best friend. And so I have that idea of the lifetime mood disorder. This is there in DSM-5. I just think it's underutilized and underappreciated, so I push that. An important part of this is to recognize in this evolution, we're going from seeing mood disorders as kind of distinct episodes, as if there were nodal points between them, to things that are more like points on a continuum. And I think that trend will continue as we get better understanding of the underlying biology. So a little bit about what the mixed feature specifier is about and how it relates to these mood disorders. If we just think on the bipolar side, that Venn diagram idea you see here on the right, we have BP-1, 2, cyclothymia, BP not elsewhere specified, and maybe there's some overlap with unipolar. I think that's how many people would think of this. I like to think of it a little bit differently. And in part because that mixed episode idea is there and it doesn't necessarily make somebody fall into the bipolar universe. It's really more one big entity where there's a lot of not elsewhere classified going on and we still have our episodes. So I hope all those definitions have made you feel very confident in your confusion, maybe not less confused, right? It's still very hard to do the job. And what I want to do is now say, how do I cope with this? How can I maybe help you do a little better? We'll always stay modest, but I think we can focus our exam on what are the key points. And you'll see, there are going to be four major outcomes from this focused assessment that I think have a great deal of benefit to me in planning, not just the diagnosis, but the treatment plan with patients. So I just want to remind you that when we start off with an assessment, we can assess those current episodes. Is it a depressive episode, a manic or hypomanic episode? Are there mixed features, et cetera? Then we move on to the lifetime diagnosis. And that often means we're looking for a past hypomania or mania. And if there is one, it's bipolar. And once it is bipolar, it is not changing, right? So that's a so-called absorbing state as it's understood that you stay there once you have that diagnosis. That's at the first assessment. But after that, patients come back and it gets harder because you're going to do prospective monitoring. And at any point in the future, if the patient has hypomania or mania, if they were on the MDD side, we're going to change their diagnosis. And if they've already had bipolar, again, they're going to stay bipolar. So a really important thing, the practical implication is incredibly simple. Most patients who diagnose MDD, even if they have mixed features, over time, the majority do not, even with mixed features, the majority do not convert to bipolar I or II. But some do. And therefore, I think all patients who have had a major depressive episode should be told that they're at risk for that, right? This is a practical implication. And to say, oh, we can ignore this because you haven't had a bipolar episode, although your mother and your uncle did, that seems to me to be a disservice to patients. And here you see the famous Yogi Berra line. I have to admit to my patients that making predictions is difficult, especially about the future. So that is for sure. So now I have another question, and I'll ask you with show of hands about measuring things. I have a bias that it's always good to try to measure things because when you ask patients to do a measure, you have an opportunity not to say, I think this and you think that, but you can say, you know, the depression scale was this and the young man urating scale was that. What do you think? Right? It actually creates an opportunity for discussion that isn't there if I just tell the patient what I think their diagnosis is. So I'm going to ask you, have you used formal scales in your practice in the past 12 months? Raise your hand if you've used any formal scale. Oh, I'm so happy. Very good. Excellent. That's fantastic. I want to show you one formal scale that probably most of you have never used or probably heard of, but one that we used in Step BD called the Bipolarity Index. Now when we talked about Kreplin, you heard that, you know, he saw all the signs and symptoms on his patients when he went on his rounds, but then he used age of onset. He used that course of illness, recovery, no recovery. Well, in 1970, Robbins and Gouzet added two other dimensions and they said these five dimensions, they added family history and response to treatment. These other dimensions are the way to validate all psychiatric diagnosis and they were particularly concerned with schizophrenia. But what we did in Step BD is that we validated a quantitative approach and we took each of those domains and we simply assigned points. If you had the most convincing evidence, you got 20 points. So if you were convincing on all five, you could score as many as a hundred points. Other convincing characteristics, maybe not the most convincing, you got 15. Other things that were kind of suggestive, 10, maybe a nonspecific kind of feature, five, and then zero if you had nothing associated with the disorder. So this was all based on the sort of classic conception of what the illness was about. And I'm proud to say not only did we use it in Step BD and find it helpful, but it's been validated in Russia, China, and the U.S. in clinical practice. It's been used in research studies, actually most of the ones I'm going to quote to you later when I show you the data for studies of bipolar depression in particular have used this scale. So it actually is something that can be extremely helpful. Yet I know it's kind of cumbersome. So here's the scale. And I don't show you this because I want you to memorize the scale. I just want you to get an idea that these five dimensions can be defined. You're going to get these. You don't need to worry about taking notes or pictures. There are definitions for these. But even if you just make your own, you don't have to agree to our definitions. Just make your own. Just say, what do I think are the most convincing? What else is convincing? What's not so convincing? I'm going to emphasize a couple because I want you to agree about them and they're actually pretty well replicated. That age of one set one, here is the age of one set data from Step BD where we enrolled a couple thousand patients. And you can see that in the 2000s, the pattern was pretty much the same as what Kreplan observed. You start to see a lot of people in the mid and late adolescent, early adult years and then it tapers off. Eighty-six percent of our patients had a first episode before age 30. So show me a patient who has been well until 57 and I'm going to say that's not a classic bipolar patient. They could have it. And especially if it is a female patient, you may find nothing with a medical workup. But workup particularly men because they have a high rate of having identifiable secondary medical causes for their mood states. So again, you can see what the deal is. You get 20 points if you're 15 to 19 in the onset. That's absolutely the prime time. If your first episode comes after age 45, not so much. Now another thing has to do with how many episodes do we see? And this comes from Roy Perlis and he got data from three Eli Lilly studies. Olanzapine, fluoxetine, the bipolar study, and then two duloxetine, unipolar, or MDD studies. And I point out to you the purple bars. Look under bipolar. More than half the bipolar patients who entered that study reported more than 25 episodes when they got into the study. The average age there was right around 40. That's a lot of episodes. And another 23% or so had 6 to 25 episodes. Now look at those two bars on the unipolar and you see that now we have a nodal point. There's not a lot of overlap here in the number of episodes. And so please assess how many episodes people have and think more like classic bipolar if we're talking in the teens and 20s as the number of episodes than if we're talking 1, 2, or 3 because that's pretty common. By the time you come to treatment, having a lot of episodes is clearly more associated with bipolar. Now along with this, they looked at data for family history. And same data set from Roy Perlis. And one of the intriguing things is when we look at family history of MDD, just unipolar depression, it is highest among bipolar probands. If the identified patient was bipolar, they have a higher frequency of just unipolar depression than the other two groups. However, what really is outstanding is the number of bipolar relatives that they have. The family history of bipolar is obviously nearly, it's between 5 and 10 times as much. So if you have a family history of bipolar and you haven't had a hypomanic or manic episode, your risk of converting to bipolar is probably a lot higher. And even if you don't convert, I would argue that your biology is probably much more similar to what classic bipolar is thought to be. So how do we use these tools? You can use the Bipolarity Index or we can just look at those domains individually. You can use your judgment to define it. Determine your own criteria for just these three things. What you think is really convincing, what's still convincing, and what is not so convincing. And apply them consistently. So if you do that, what you'll see is you can clearly score out on those Robbins and Gouzet domains, each of those three designations, and you can see where patients fall. And this will give you an idea of how likely it is the patient is a classic bipolar patient. And what I do with this is I share it with patients. And this idea of the clinical interpretation, what is all this going to lead to if I say, does this domain support a bipolar diagnosis? Yes or no. That's all I'm trying to do. If I think that it does, it's yes. If I think that it doesn't, it's not. And again, please do your own thing here. That will lead you to the four categories that I think are clinically useful. And it's very simple. High confidence diagnosis. That means that the episode characteristics, a full MATIC episode are there. Here, I'll make this a little simpler for you. High confidence, yes to the MATIC episode, and then at least two other domains are most or convincing evidence. So you have three domains, including a MATIC episode. Low confidence, you've got a MATIC episode, but nothing else is going for it. I have to concede that I have made a lot of diagnosis of bipolar I based on a classic, in my opinion, florid MATIC episode for patients going through divorce, other trauma, substance abuse, et cetera. So be very careful about that. Oh, they fulfilled every DSM-5 criteria for a MATIC episode. Three other people saw them. They all agreed. That may not be the same condition unless the other domains are in agreement. If we have nothing, we have no episode characteristic of mania or hypomania, and no other domains are associated with bipolar disorder, let's just say we don't really have to worry much about that person. And I would tell them they were at low risk, even if they've had a couple episodes of major depression. However, the group that I am most concerned about is the last group. These are the ones who do not have a MATIC episode or even a hypomanic episode, but they have three other domains that are consistent with bipolar disorder. And when they do, I will explain to them that they do not meet the DSM criteria for bipolar. I'm not going to call them that. I'm not going to diagnose them with any form of bipolar disorder, but I would recommend we think about treating them as if they have it, because it seems to me that they're at risk. So identifying this group of patients is how I can avoid, I told you so, I thought you were bipolar. Now I wait until they have a MATIC episode. They lose their job, their marriages, their place in the community. Really terrible outcome, something that I like. So identifying this group and saying here's why I don't want to give you venlafaxine for your depression, that's the point of all of this. So again, now we can move from that lifetime mood disorder and how we do a focused exam to a little bit about the episode specifiers. So now we're going to talk about having a current mood episode with either anxious distress or mixed features. And again, these are really common and they do impact treatment outcome in pretty negative ways. So what are we talking about? We're talking about these mixed features when the patient has predominantly one pole or the other, but has several of the non-overlapping criteria. So you have three or more that are unique to the opposite pole you can have with mixed features. For example, if you have a full depression and three of those that are on the right, that's what we're really talking about. How common is that? Well, Joe Goldberg in Step BD looked at people who one of our trained research clinicians was seeing and diagnosed that the clinical diagnosis was major depressive episode. Now one of the problems with the clinical diagnosis is that it creates a kind of expectation bias where you discount other symptoms. So we had the patients assessed independently. And when the clinician thought that it was a depressive episode, just under one-third of them were pure depressive episodes. You can see those subsyndromal manic features were present in over half the patients. And in almost 15%, a full manic episode was diagnosable. But because we were calling it depression, we simply ignored or discounted them, or our expectation bias led us astray. So a really important thing to realize, we're all subject to this. It's extremely common. This is in over 1,000 patients. So pretty strong finding. Other groups, joules-encs, lots of people have replicated this thing, this kind of finding. And mixed features are definitely associated with a more pernicious form of mood disorder. So a lot of people have it sometime during their life. And compared to those who only have pure depressions, there are more suicide attempts. There is more bipolar among the relatives. There's a lot more anxiety, impulse control disorder problems, and substance use. And even though it's a minority, most do not convert to bipolar. It's still a greater likelihood than those who don't have mixed features. So a little bit about the definition. Again, I'm not going to read through these because you know them and you don't need me to do that. But I will tell you that all you need is to have three of these during the majority of the days of a current depressive episode. So really important. That's the mixed feature. The criteria for the anxious distress is the specifier requires at least two of these symptoms while you're meeting criteria for major depressive episode. And those are being keyed up or tense, being unusually restless. Your concentration is paired due to worry. You fear something awful might happen or like you're losing control. And you can sort of rate out the severity of it from no anxious distress to mild to moderate, severe depending on how many symptoms are present. The real point to understand here is it's most likely if you have one that you have the other. You do have some people who have pure forms, only mixed features or only anxious distress, but a lot of people have both and not a good thing when you do have it. Why do I say that? Well, we can look at the difficulty caused, just this is a self-report during a depressive episode, how bad was it for you? And you can see in those green bars, those who do not have the anxious distress, they're always less than the purple bars who do have it in terms of the degree of difficulty. If you look at the treatment response, the anxious-depressed has not as good an outcome. In other words, the length of time of non-response is longer if you have anxious distress. And I'm going to come back to that in a minute. In STAR-D, this is now MDD, people who have MDD and met criteria for anxious distress, you're seeing the outcome with some standard antidepressants alone and in combination. And it's strikingly worse, right? In those anxious-depressed patients, the chances of remission are like a quarter to a third of what they are among those who don't have it. So this is something that I think needs to be recognized as a sign of poor prognosis, right? And it brings us to now move on to treatment and looking into what's an effective treatment. I would ask you what you think the best treatment is for bipolar depression. And I would like you just, I'm going to run through this list with you and again by show of hands. When you're seeing a patient, we'll say for the sake of argument and simplicity, for the first time you're seeing a patient and you think they have bipolar 2 depression, how many of you would say lithium is your first-line treatment? Okay, a few. How many would say I'll start with an SSRI, thank you very much? Wow, nobody. That's incredible because that is the most popular treatment when you see it. How about an SNRI? Wow, you guys didn't need to come to this talk. Ketiapine is your first-line, very good, evidence-based treatment. Olanzapine alone, okay. Olanzapine and ketiapine, it should have been olanzapine and fluoxetine. Well, I'll say olanzapine and fluoxetine, I mean, yeah, a few people. Lirazidone? Okay, excellent. Okay, another evidence-based treatment. Lumeteperon? A few? Okay. How about dextromethorphan and bupropion? A couple people have used that, I've used it. And then finally, the darling of treatment now, how many of you would start a BP2 patient with psilocybin? No brave souls out there. That's actually rather reassuring when you think about it. Okay, so just to lay out the approach to treatment, I'm a big believer in formal assessment at the start of treatment. So we do that lifetime diagnosis, we assign what we call a current clinical status. I really focus a lot on the therapeutic alliance. We have a so-called collaborative care approach, and a big aim of it is to begin with concordance. So I'm going to come back to that on the next slide. But this is an individualized schema that works for pretty much all chronic conditions. What I want to do, because I can't promise to get it right, I want to increase the rate that we make wise decisions amicably together, me and the patient and their family. So we're going to offer, as a rule, what we call a menu of reasonable choices. Not one thing that I think is best. I always slip in what I think is best, but I give at least two other things as a rule. Why? Because it's surprising to me, anything that a patient thinks is good works better than what I necessarily think is good, and my willingness to try it improves the therapeutic alliance. So unless there's a really good reason to not do what they want to do, as long as you see we build in that measure-based guidance, I'm okay with it. So this is a systematic kind of thing, and it really is an approach I commend to you, because it enables me to make progress even if the patient doesn't get well. Why? Because if I have a definitive outcome for the treatment, I might do one, two, five, or six treatments this year, I can cross them off the list, and nothing frustrates a patient more than being told, I want to go back to that treatment you had last year, I'm not really sure it didn't work, let's be sure it didn't work. And then we've at least improved the situation. So what do we mean by concordance? There's a difference between concordance and adherence, right? So adherence is, did you take what you were prescribed? Concordance is, do you agree that that was a reasonable plan? So what you see here, it's a very nice two-by-two. If the patient follows the treatment plan, yes or no, so they're either adherent or not, do they agree with it? When patients are concordant and adherent, we just say, what a good job they're doing. When they're non-adherent but concordant, they welcome external supports. So that's like me saying, I love telling people that I should not have dessert at dinner, and if I do, I owe them $50, okay? And that builds a lot of supports for me, they're great desserts, especially if you go to meetings like this and go to restaurants, right? And then there are people who are looking out to see if I've had dessert. Now if you want to see what happens when you have somebody who is discordant when losing weight, next time you're in a restaurant, look around at somebody having dessert who you think probably shouldn't, and go up and say, you really shouldn't be having dessert. Those are fighting terms. And it's what we often do with our patients. We're telling them what to do before we establish agreement about the goal. Establish agreement first, offer external supports. Everybody wants to do it on their own, almost nobody can. So let's start to build in those external supports. Let's find a trusted person, family member, friend, et cetera, who can do this, and establishing that therapeutic alliance is critical, okay? So once we have that, we can move on with our simple schema, and this is starting with this idea of defining a critical decision point. You're bipolar depressed, and you have bipolar type 2, for example. Then we look and try to formulate this menu of reasonable choices by taking into account two main things, the evidence supporting that treatment, and individual factors. I'll say more about that, but you get the idea that evidence has a grade. Category A evidence is higher grade, and F is not so high. And we will formulate that menu of reasonable choices, and we'll explain to the patient, we have to do some education and negotiation to do this, we'll explain to them the different options, and I say I try really hard to have at least three. We negotiate, and they pick a treatment. Fantastic. Once we have the treatment, we can now talk about an appropriate measure, and since most of you said you use measures, use the one you think is best for whatever you've agreed to treat. And here we have just the most incredible thing. I know many of you didn't go to Harvard, but we have figured this thing out at Harvard. It's called continue to do what works and stop doing what doesn't work. So once you have the measure, and you see that the patient is either getting better or not, you are really in a position to help that patient out. What you want to avoid is an inconclusive trial, right? For the reason I mentioned before, you go back and you say, I think we really need to try lithium in a higher dose. We did it two years ago, now we need to do it again. Probably not a great idea. Cross it off the list, fantastic, we're trying to have a definitive outcome that means sufficient dose and duration for you to conclude it either worked or it didn't work or they couldn't stand it. All three of those are definitive outcomes. Once you have that, you've actually come back to the next visit with the patient, having grown what you know, not about the literature, but about that patient. And it takes us into those individual factors. So here are a bunch of them, and I think there are a number that are really important. I've emphasized to you this idea of concordance. Eventually we'll have biomarkers, but prior treatment response is extremely helpful. I don't repeat things just because somebody else did the trial. I don't put patients through that. But you can see there are a lot of different things in here that might influence your choice of the initial menu of reasonable choices or subsequently as you learn more about the patient. So a little bit of guidance, construct your initial menu of reasonable choices. I think we have an obligation to patients to at least offer proven treatments first. And we'll come back to how you define that. But as we go along, we're going to revise the plan as we go forward based on the measured results. So we're aiming for those definitive outcomes. We're stopping things that don't work, continuing what works, et cetera. So how do we evaluate the quality of the evidence? What is a proven treatment? Well, it's pretty simple. Category A evidence is a double-blind placebo-controlled trial with an adequate sample. The asterisk is there for statisticians. And they will tell you that you have to have statistical power, 80% chance of detecting a difference at the 0.05 level. For anybody who's not a statistician, I'm going to make this super simple. If the trial has fewer than 100 subjects, you don't need to read it. I've just made your life easier. There are only a handful of studies you have to read ever. And in any one year, if there are two, that's a blessing. So it's not really burdensome to keep up with this rule if you just say, I'm going to keep it to that. And you can see how the evidence drops off from those category A. We have B, double-blind comparison with adequate samples. They're not placebo-controlled. Uncontrolled trials, no published evidence, but maybe there's a class effect that we're interested in. And then maybe at F, the available evidence is negative. Why you'd recommend those to people, I don't know, but that's in there. So with that context, let's move on to the controlled trials. And to say that there's a lot to learn from these and many, many well-defended beliefs, but relatively little data, in a way that's good because it makes it easier for us to remember. So what does the data say? Well, for acute bipolar depression, and this is mainly from BP1, but I'll show you in a minute what there is for BP2, at least one positive category A trial, lamotrigine, olanzapine, lorazine, cariprazine, lumateperone, and the combination of olanzapine and fluoxetine. That's what we've got. And then only negative, you see a bunch of standard antidepressants. I'm getting the signal that I have to go fast, so I will. You can see some of the early trials, Joe Calabrese published on lamotrigine. That initial trial was pretty promising. Mertzitoin, olanzapine, and olanzapine and fluoxetine, you can see the clear separation. Same thing with the early catiopine studies for both 600 and 300. When we look at the mixed feature design studies, this is really kind of compelling data for lorazidone versus placebo. You can see the effect size here of 0.80. That's pretty strong. That's overall. I can also show you for lorazidone, thinking about that anxious distress, you can see they had improvement in the anxiety rating scale there as well. What I want to emphasize here with lumateperone is not only do we have nice separation, but we have separate data for BP2. And you may notice that the difference between the placebo and the blue lumateperone arm is even greater for BP2. And interestingly enough, the effect size was clearly larger there. What about standard antidepressant medications? Well, they're commonly used, and again, I'm just going to have to tell you there's no Category A evidence for them. Nobody here endorsed using them, so I won't say a lot about it. But the role of standard antidepressants, you can look, and if we go back to those catiopine studies, you can see the all patients on the left, the BP1s in the middle. And what you see overall is there's separation for both the catiopine dose groups, but the paroxetine, the standard antidepressant, is not significantly different, and it's closer to placebo than it is to either of the active agents. And that's clearly the case in the BP1s. If you look at the BP2s, and I'll help you focus by making that a little bigger, it started to look like there was a difference through the first six weeks, but by week eight, paroxetine came out superimposable on the placebo outcome. So the case for standard antidepressants is not strong. In Step BD, we randomized 366 patients to get a mood stabilizer, lithium or valproate usually, and either bupropion or paroxetine versus one of those with a placebo. Durable recovery, staying well for at least eight weeks. You can see virtually the same slight edge for not getting the standard antidepressant. Switch rates were also nearly the same. There was no benefit nor harm from standard antidepressants that we observed in Step BD. And I didn't ask you about the use of standard antidepressants in mixed states, but this is from Joe Goldberg again. If you looked, you'd see that there was no benefit to having a standard antidepressant nor a harm in people in Step BD who had mixed states. For refractory bipolar depressions, suffice it to say, there's really no systematic data. There are no Category A studies. There are bunches of things here that are used, but no great data, so please don't feel like those have to be on your menu of reasonable choices. When you have mixed features, there's really nice data for lorazodone and lumateperone from post-hoc and exploratory analyses. We have to be careful about those. They weren't the aim of the study, but at least they're encouraging. For bipolar depression with DSM-5-defined anxious distress in particular, there's only an exploratory analysis available for lumateperone, but that was positive. That result has been presented. Overall, when we look at the Category B and C, we go from lumateperone to adding back catepine and lorazodone. And then you'll see case reports for a lot of other things, but be careful. Now we come to a closing idea here, and I just want to get back to these levels of evidence and how we do it. We always start out choosing that initial menu of reasonable choices from retrospective assessment, but when we go forward to prospective, we can really individualize the care. How we do this is kind of cool because we rely on the so-called two-week rule. What is that? It turns out that if you looked at all of the studies that had been submitted at the point where this analysis was done by Joe Calabrese's group to the FDA for bipolar depression, you had all the treatment arms, including placebo, interestingly enough, in both positive and failed studies. This is going to be a well-replicated finding. The two-week rule is you use a scale and you see if the patient is at least 20% better at two weeks. That's the positive predictive value. If they are, there is roughly a 50% to 60% chance that patient is going to be in remission at six to eight weeks. That's pretty good to know, but not nearly as important as knowing what does it mean if they are not even 20% better. Eighty to 90% chance that that patient will not be responding at six to eight weeks. The only humane thing I can think to do under those circumstances, when I see patients not have even 20% improvement, I raise the dose or change the treatment. I can make that call at the two-week mark, not the two-month or the six-month mark. At the end of the year, I am having more definitive outcomes. That's one way you make promise to the patient that we will have progress. Whether or not you're better, we won't be in the same position next year because I will definitively know what not to do again. Small benefit, but a benefit. Because I have five minutes left, I can dwell on this slide and say, if you would like the slides, do feel free to contact me and now we can move on to your questions. Fred Goodwin told me that if he were to give another name for bipolar disorder, he would call it a sleep disorder. I'm happy to hear you. Fred Goodwin told me that if he was to give another name for bipolar disorder, he would call it a sleep disorder. There you're not sleeping at all for four days. I mean, that's obviously bipolar one, but I find that two and a half days of an upswing mood is more likely to be a bipolar two. They spend almost nine months in the depressive phase if you do a yearly exam. And usually, rarely people come and talk about hypomania. Usually it's like an irritability. You call it dysphoria. We may call it dysphoria, but they feel that they are irritable for two days or two and a half days. And I always like to call somebody accompanying the patient. And once in a way, I would say, can I call home and speak to your daughter? And I find it's so easy if you have another objective person involved. I mean, I remember somebody, I just called home and the daughter picked up. I said, I'm Dr. So-and-so. And your mother is here. Doctor, she's bipolar. So if you involve somebody from the family, it's... I think you're right. You made two important points. I think you were talking about, in the days Fred Goodwin wrote about this, he had data with Tom Weir showing sleep disruption as the final common pathway to mania. That was a small study, but still valid. And the other point is the benefit of collateral information cannot be overestimated. So thank you for making this point. And the last thing is, if I have doubts, I think in 2005, when seroquelcortiopine was approved, this is almost an objective test. 50, 100, 200, 300, four days later, come see me next week, and wow. You get that wow response, you know? Yeah. And so again, that is another interesting point. A lot of times people get better very quickly. And I don't know if that's a true cotyping antidepressant effect or the benefit on the anxiolytic side, that people really do feel much better very quickly. Improving sleep doesn't hurt either. The other question is about anxiety disorders. The most common thing is both bipolar and panic disorder involve the amygdala. Panic disorder with agoraphobia, very common comorbidity with bipolar. Absolutely. Thank you very much. Bend it down here for a second. Any practical tips on differentiating between borderline patients with significant impulsivity and affective liability that worsens episodically due to psychosocial stressors versus someone with possible bipolar II diagnosis? Well, I'll tell you, the interesting thing for us is that the rate of diagnosable comorbid borderline in the bipolar clinic was a little bit higher than 30%. So it's a very common thing. And again, getting back to the start of the talk, we're not interested in debating. We're interested in helping. So instead of either or, I'm looking at both. And the only thing I have against borderline as a diagnosis is that it's often used as a pejorative. And if we can avoid that, explain to patients that there are these issues. And one of the reasons that their lives are so bumpy is the way they're interacting with family and friends and the high level of drama. Has that been lifelong? Is that only when they're depressed or manic? Once we agree that, yeah, that's the way they always are, and they've made a lot of suicide attempts or other kinds of self-harm, their interperson relationships are unstable, even when well, I definitely entertain the borderline. Thanks for the talk. Do you have any sense of why certain atypical antipsychotics, which you would think would be effective for bipolar depression, nevertheless have turned up negative in trials, as you mentioned, like Abilify? I feel like commonly my residents use that. I love that question. And so I'll put in my signet health hat where we really do a lot of quality review and oversight of clinical trials to try to explain why trials fail. So I don't think the Abilify trial is an example of a trial that came to a definitive negative conclusion. I think it's one that had an interesting swoosh effect. I showed you how, in that cataepine trial, paroxetine went from looking pretty good at week six to the same as placebo at week eight. I don't know if that's because paroxetine really didn't work or, at the end of a trial, a lot of patients are feeling like, oh, I'm going to be abandoned. All of a sudden, they're worse. And those kinds of things influence trials. Trial design and execution is critical to detecting a signal that's really there. We often have these negative trials that are really failed trials. So I think that's part of it. It is also true that this term, antipsychotic, that we take as a class effect is a marketing term. And it is not a science term. The ratio of the different kinds of receptor interactions that you have could well turn out to be a decisive factor in whether something has an antidepressant effect. You could please comment on medication-induced hypomania and what's the implications for treatment. The DSM has gone back and forth on whether that counts toward calling somebody bipolar or not. And again, the debate is fascinating. But the important thing is that happened. So I'm happy to call. I don't require people that I treat to say they have bipolar disorder. If they want to say that their disorder is peanut butter, I'm just trying to make it smoother. That's all I need to do. And that's where the concordance comes in. So if they have had what looks like treatment emergent mania on a steroid, a standard antidepressant, cocaine, et cetera, we see that they are liable to having that kind of treatment emergent manic response. And we can realize that that's a vulnerability. What do you want to do given that? I don't require a change in label. If the patient says, does that mean I should never take an antidepressant again, I'd say certainly not without a counteracting medication present. And why do we want to do that this week when we have these other medications that don't have that liability? So down the line, again, that iterative approach, I might someday put it back on the menu of reasonable choices. But I want the safest things first. Hi, excellent presentation. Thank you. I'd like to ask a couple of things. One is regarding your slide on medications in the beginning, you didn't mention lamotrigine. It wasn't even one of the options for us to raise our hands. So I'd like to hear you on that, and what's your experience, and how you feel about it. Also, there was a slide where TMS was included in the negative studies. And there are some studies for TMS for treating bipolar. And the last thing is, if you could comment on bipolar spectrum and your idea on that, because you mentioned Akisco before as well. Yeah, well, I'm glad you asked the lamotrigine question. Let me show you. This slide, for some reason, was hidden. I meant to show it. I showed you that first Calibri study that was positive. And here, five studies were done. That was study one. Study two included BP1 and BP2 patients. So pay attention to these two bars on the Hamdi, where there's no effect in study two. I don't think it's not showing. Sorry. It's not showing. It's really a great slide. I believe you. What am I doing? Oh, maybe I need to be in presentation mode. OK. Technical help is so good. So this is the study one, the Calibri study I showed you. They had only BP1s. In study two, there were BP1s and 2s. And you see that the difference that we saw between lamotrigine in the blue and placebo in the gray has gone away, not statistically significant. And we go through here on the Madras. In the Calibri study, it was. Most of the others did not find a statistically significant difference. But interestingly enough, on the CGI, look at this study here. Study four, which was BP2 alone, that was actually positive for lamotrigine. And so while overall, you see four out of five studies failed, these are hard studies to do. And there are tiny advantages for lamotrigine. So John Geddes did a meta-analysis of those five trials. And he found that, indeed, overall, when you did a meta-analysis, there was evidence supporting lamotrigine as effective. But let's admit, modestly effective. So that was your first one. Remind me of your second one. TMS for bipolar. No good evidence there. I don't know. For TMS and ketamine, when I see refractory bipolar patients and I run through my menu of reasonable choices, I think those are really reasonable options. But I have to tell you, my personal experience with them has not been great. And the last thing was bipolar spectrum, if you could comment on. Yes, so the bipolar spectrum idea goes back to the very beginning, when I said there were no nodal points between them. And so the idea of having a spectrum, it really makes sense to me, unless and until we start to say, this is so unreliable that people can't agree about it. So you had those ideas of bipolar 1, bipolar 2, 3, 4, 5, and 11. And it turns out that bipolar 1, you saw, had pretty reasonable reliability. Bipolar 2, not so much. And when we get beyond that, it's really hard to talk about it, because the reliability is negligible. And therefore, it doesn't help me to start to think of other forms of bipolar when I can't even really agree on bipolar 2. Great, thank you. Thank you. So are there, how about, any comments? Two pieces of it. In the concept of rapid cycling and treatment implications, and including in younger patients. Yeah, rapid cycling is also a phenomenon that's often associated with BP2. It's not as dramatic. In the early cases that described very frequent cycling in BP1s aside, the phenomena of BP2s with rapid cycling can be hard to distinguish from borderline personality disorder. So the first question is, how do we define that? And if we define it based on a full requirement of a full depressive or mood elevation, that, in a way, constrains how rapid the cycling can be. Because if you need to have two weeks in one phase and at least one week in the other, with a period of recovery in between or not, you still can't have but so many episodes. In Europe, they see these people, BP2s, with one or two days of highs. And they see ultra rapid cycling. In my hands, what I have found there is the single most helpful thing has been tapering off standard antidepressants. There's no controlled data for that. But in our hands, stopping ineffective drugs is one of the most effective things that we do. The next thing is, I get polysomnography in those patients. And fairly often, you will see sleep apnea. And so CPAP has turned out to be a fairly good option for those kinds of patients, as well as evaluating their pituitary thyroid axis. Go ahead. So let's say you have a patient with bipolar 2 depression. And you treat them, and they go into remission. What is your recommendation in terms of preventing becoming hypomanic or recurrent to the depression going forward? So I've emphasized, because of the nature of the talk, medications. But self-management strategies, those are key. Establishing, first thing is, we try to establish a consistent sleep and wake up time. We try to do as many of the other things with a set schedule. We also, it's just so sad when you see this happen. Every time I go to Thanksgiving with the family, somebody calls the police because we get into a fight. Kind of destabilizes your mood. How about not going next Thanksgiving? And one of the things that happens, we saw that co-occurring alcohol substance abuse is very high. Again, one of the most effective things is to say, we've tried everything else. I know you don't think you're drinking excessively. Give me a month of abstinence, and let's see how you're doing. I can't tell you how often that turns out to be the trick. So those things, I didn't talk about them enough. Those things really turn out to make the maintenance phase of the treatment have a much better outcome. Getting somebody well is wonderful. Keeping them well then gets to be the whole art of this. And it is a matter of understanding the stressors, the habits, and the extraneous substances in a person's life. OK, I mean, that sounds great. But what do you do about the medicine? You mean, what do I do about it if they're drinking? No, no, no. OK, so you have them on medicine. You get them better. OK, so they're in remission. What do you do about the medicine going forward? So again, in this idea of collaborative care and talking about the risks and benefits, there are a lot of patients who want to go off medicine for the sake of it. And if you're not having side effects and you want to go off medicine, realize that you're taking a risk. The best I can do is I follow the measures at least once a month on my patients going forward. And if we see something, we can restart you. But I would rather, if we're tapering things down, let's get to a point where you don't have any side effects that are hard for you to tolerate. And you tell me if the benefit of being off meds is enough to justify the risk of a relapse. Because I will tell you the risk of relapse is pretty high. And even when I say that the first one, two, or three times, patients will want to go off their meds. But having it well-documented for them with real measures is extremely helpful. And so we don't have to repeat that cycle six, seven, eight times. If we just do it once or twice and then say, I'd like you to have a whole year of doing well. Let's come back to that question once a year and see what we can decide. So you would basically keep them on the same medicine, ideally. That was the continue-what-works theory, yes. You've been at the time to get the name. Oh, you know, you raise a good point. That indicated, meaning they have an FDA approval. I didn't get into this. But that's a different thing. That is for advertisements. We're doctors, right? And that's why I showed you category A is a good study, not FDA approval. FDA doesn't regulate what you do. You regulate what you do. And if you see a high-quality study that says this is a proven treatment, I don't care if it's off patent, one patent, or never approved. If it has category A evidence, I've got to tell the patient about it. And I'm happy to keep it in place, even if it's only approved in the short term. Again, depending on what the, you know, TART, tardive asconesia, metabolic syndrome. If it's not there, I'm not worried. But if it is, then it's reason to taper. Thank you. Sure. I'm curious whether your list of reasonable options would look different for a patient who you were convinced met the bipolar 2 criteria versus somebody who you think, maybe, but I can't make that diagnosis. Does a list look any different? Yeah. So that's why I said those four categories. That person who has a high index of suspicion, I will suggest to them that we leave their diagnosis what the DSM-5 would say it is, but that we consider their personal risk. And yes, I would treat them, if they have a family history of bipolar 1 or bipolar 2, I would treat them accordingly, as if they are at risk for that. Because these are really serious conditions. It is amazing how forgiving the world is of whatever you did when you were depressed, but not when you're jovial and high. So real difference. And I lay that out for patients. Again, they don't necessarily listen the first one, two, or three times, but they've heard it. And eventually, that therapeutic alliance becomes serviceable to them after they've heard it for a while. Thank you. So we've reached almost the very end of the time. And I'm going to have one more from the online audience. Asking about antidepressants. And what are your thoughts on SSRI use in addition to a mood stabilizer in a patient with severe anxiety in addition to bipolar 2? And also, someone else was asking about similar, if you have any particular antidepressants that you particularly like or find success with. Well, again, antidepressant, as I would define it, is a treatment that makes depression better. And therefore, medications like we saw chiroprazine, lumateparone, catiopine, lorazodone, I consider them atypical antidepressants. And I'm happy to recommend them. When patients say, well, I've never been psychotic doctor. So why are you giving me antipsychotic? I try to explain. They don't like that. OK. We can try an SSRI. And again, people will often do quite well on these. My expectation was always that SSRIs were going to turn out to be effective but have a higher rate of switch. You saw in step BD, neither of those turned out to be true. And so I'm OK with the bipolar 2 having a trial of an SSRI or SSNRI. And by the way, I didn't say this, but the only evidence for any one standard antidepressant to cause worsening, to cause mania, is for venlafaxine. And even there, with bipolar 2, with careful monitoring, could it be done? Yes, it could be done. But you need to explain that to patients. Patients need to be brought into the decision-making so they can decide which risk they want to bear. We've reached the end of our time. And we'd all like to thank you. Thank you very much. I appreciate you being here.

bipolar II disorder

recognition

management strategies

diagnostic challenges

episode subtypes

mixed features

anxious distress

individualizing treatment

lumateperone

quetiapine

collaborative care

comorbidities

treatment outcomes