Roger McIntyre, MD, FRCPC: Hello, I'm Roger McIntyre, professor of psychiatry and pharmacology at the University of Toronto. I'm also head of the Mood Disorders Psychopharmacology Unit of the University Health Network in Toronto, Canada. I'd like to welcome you to this program, "Bipolar I Depression Today: The Evolving Understanding of Bipolar Depression Neurobiology and the Relation to Diagnosis." It is part of a series on our current knowledge of bipolar I depression.

This program will include discussion of investigational drugs, mechanical devices, biologics or diagnostics not approved by the FDA for use in the United States. Specifically, there are no imaging tests approved for the diagnosis of bipolar disorder.

The goals of this program are to discuss the current understanding of the neurobiological pathways of mood disorders with depressive episodes in relationship to the characteristic end diagnosis of bipolar disorder, and to discuss factors that may contribute to the under-recognition of bipolar I depression in routine clinical practice.

Let's begin by looking at the clinical ecosystem of bipolar disorder. Bipolar disorder is one of several mood disorders, which include the unipolar disorders. There have been several organizational frameworks proposed to organize the bipolar and unipolar disorders. One is to think about these conditions along a spectrum. In other words, we know that by definition unipolar illness is the presence of depressive episodes, while bipolar I disorder includes the presence of manic or mixed episodes. Between these 2 opposite poles are phenotypic variants of bipolar disorder and unipolar disorder, comprising the so-called spectrum.

As a matter of fact, clinicians often encounter patients between these 2 poles, that is, along the spectrum, with most patients encountered in clinical practice exhibiting so-called subthreshold symptoms. Subthreshold symptoms are defined as symptoms falling short of the criteria for a hypomanic, manic, mixed, or depressive episode. So taken together mood disorders can be conceptualized along a spectrum.

As we look at the prevalence of the separate mood disorders along this spectrum, there's an important message that emerges. Historically, the estimated lifetime and/or 12-month prevalence of bipolar disorder have been cited in the range of around 1% to 2% for mania or hypomania. A more detailed look at the prevalence estimates, whether it's 12-month or lifetime, indicates that perhaps that may be an underestimate when factoring in subthreshold symptoms. In other words, if we look at bipolar disorder through the lens of the spectrum -- that is, individuals who have subthreshold presentations of biopolar disorder -- the estimated 12-month and lifetime prevalences are much higher than previously estimated.

For example, results from the National Comorbidity Survey Replication (NCS-R), one of several cross-national epidemiological studies, reported that the narrowly defined estimate of bipolar disorder, that is what we're most familiar with in the Diagnostic and Statistic Manual of Medical Disorders (DSM) manual, would estimate the prevalence at about 1% to 2%.^[1] If we broaden the criteria of bipolar disorder to allow for so-called subthreshold presentations, a much higher percentage of individuals meet the definition, closer to 5%, even in some cases closer to 10%. So taken together, the prevalence of bipolar spectrum disorders is simply a phenomenon dependent in part on how we're defining bipolar illness.

What we're witnessing in the field of psychiatry is a move away from the more narrow definition to a broader definition and there are reasons to believe that the next iteration of DSM, DSM-5, will try to capture more of these softer, or more subtle presentations of bipolar disorder. Clinicians have recognized for decades that in many cases individuals with bipolar disorder initially present as suffering from “depression”. In fact, we know that depressive episodes, and or, clinically significant depressive symptoms, are often the index presentation of bipolar disorder.

When we observe our patients prospectively over time, treated or untreated, what we've learned, and this has been also supported by empirical evidence, is that a significant portion of these individuals will later declare themselves as in fact having bipolar. The initial diagnosis of unipolar would be perhaps more accurately referred to as pseudo-unipolar; they were presenting as depressed, but the underlying diagnosis was bipolar disorder all along.

The results on this figure resonate with many other lines of evidence telling us several points. First, this is a group of 550 individuals with depression prospectively observed over many decades.^[2] What you can see is that there's a fairly consistent, almost linear, conversion from unipolar to bipolar disorder. Prior to the so-called psychopharmacological revolution in the 1950s, the estimated conversion rate per year was often around 1% to 2%. More recent estimates, which are reflected on this slide, suggest that conversion rate over time is around 2% to 4% per year.

So taken together, if we had 100 people in our practice today with so-called depression and we observed them over 10 to 20 years, it would not be unreasonable to expect as many as 20% to 30% of them would declare themselves as bipolar. And what's so important to emphasize is that the most common phenotypic expression of bipolar disorder, hypo/mania, is not mania but hypomania, a softer, more subtle presentation [although depressive symptoms are the most common presentation overall]. Taken together, hypomania and mania make up the bipolar spectrums, so a sizeable proportion in fact will have bipolar disorder.

As a clinician, it has been imperative for me to not only make the diagnosis of bipolar in a timely fashion, but to get that diagnosis correct. This has been a major deficiency for us in the clinical community. One of the ways we can begin to think about whether or not a person has bipolar disorder is to look at some of the features of the presentation of bipolar disorder early in the illness course. Towards that aim, Dr Nassir Ghaemi from Boston has provided some very helpful information in this regard, and the data are presented on this slide.^[3]

What Dr Ghaemi has done methodologically is look at three separate subgroups of individuals, those who have a diagnosis of unipolar disorder, bipolar disorder, and those who are initially diagnosed with unipolar disorder, only later to be diagnosed with bipolar. Dr Ghaemi looked at some relevant course of illness variables as well as treatment outcomes specifically; age at onset of the unipolar or bipolar illness, and the time where these individuals had their first manic episode. When do they first visit a healthcare provider or receive conventional unimodal antidepressants or mood stabilizers? And finally, when were they first diagnosed with bipolar disorder?

Several messages emerge that I think are very much teaching pearls for us. First, when you look at the features of people who have been diagnosed initially as unipolar, only later to declare themselves as bipolar, these individuals look a lot like those individuals who were initially diagnosed as bipolar: earlier age at onset, late teens, early 20s. We know this is an early-age-at-onset disorder, with 75% of individuals declaring their onset prior to age 25.^[4,5]

You can also see that the recurrent unipolar group, in fact, had exposure to antidepressants, perhaps not surprising given the fact that depressive symptoms and episodes dominate the longitudinal course of bipolar. What's concerning, however, is that a decade and a half goes by from the first onset of symptoms to the establishment of the diagnosis. That protracted delay not only results in suffering to the individual and inappropriate treatment, but in many cases I think it also inadvertently contributes to the progression of illness.

Let's now maybe look at a more summative approach to try to address this issue of bipolar. We're all looking for a blood test and biomarker not yet available. It's a clinical diagnosis to make the diagnosis of bipolar disorder. Towards that aim, we can never forget the classic paper by Robins and Guze from 1970 where they looked at how we validate mental illness and how to make the diagnosis.^[6] Remember some of their key features which include, but are not limited to: the phenomenology, the course of illness, and the family history.

What you see on this slide is an attempt to juxtapose bipolar disorder to unipolar disorder, taking what's known as a probabilistic approach, since we don't have a deterministic approach. In other words, we don't have any single symptom or sign that is pathognomonic of bipolar disorder. So what we're looking at is a probabilistic approach where we bring together several features that increase or decrease our confidence this person is bipolar.

Observationally what I've noticed for many years is that people with bipolar depression more often present with so-called atypical depression. That is, they have hyperphagia, hypersomnia, and so-called leaden paralysis — a significant terrible fatigue particularly accentuated in the winter, but not always. These individuals who have bipolar disorder very often report an early age of onset, often before 25 years of age; in fact, very often before 20 years of age.^[5]

What I've been particularly struck by, and what the evidence in fact supports, is that the individual with bipolar disorder is more likely to have a high frequency of episodes when compared to those with major depressive disorder. In other words, there's a cyclicity vulnerability that is particularly observed in the bipolar population. And finally, very high rates of psychopathology, particularly bipolar disorder, in the family of the individual who has bipolar, speaks to the heritability of the condition. So, taken together, this probabilistic approach increases or decreases our confidence that our patient has bipolar.

Let's now speak to the age at onset of bipolar disorder. This is a critical phenomenon. Seventy-five percent of all mental disorders begin prior to 25 years of age, and bipolar disorder is no exception to that.^[7] In fact there have been several age-at-onset studies that have determined that the modal age of onset for bipolar disorder is typically in the late teens or early 20s, again with depression being the index presentation for most of those affected. ^[8] In rare cases we sometimes can see prepubertal onset.

Not rare, but less common, would be to see an age at onset of bipolar in the 40s, 50s, and in some cases, over the age of 60 or 65. But taken together the vast majority of individual bipolar disorder will declare this affective vulnerability in their late teens and early 20s. Obviously, this has a host of implications, not only for diagnostics, but also for the implications on their educational attainment, characterological development, and functional trajectory.

When we make the diagnosis of bipolar disorder and we establish rapport, we initiate guideline-informed, measurement-based care for our patients; we clearly want to provide for patients the most favorable illness trajectory. And what we've learned from experience, again, and this has been mirrored by results from the STEP-BD initiative,^[9] is that there are features of the bipolar patient that would indicate to us this individual is at a higher risk of developing not only a relapse into depression but also mania. And that risk prediction seems to be most robust in those individuals with bipolar who have so-called residual symptoms or, again, subsyndromal symptoms.

In other words, not only are subsyndromal depressive symptoms the most common presentation of bipolar admixed with some hypomanic symptoms, but we've also learned that symptoms predispose future relapse and recurrence of illness — again, underscoring their clinical relevance, the importance of using measurement based care, and also reminding us of the perniciousness of these low-grade symptoms.

What I'd like to do now, if I can, is turn to you with a question. We've been spending a lot of time talking about the diagnosis of bipolar, its phenotypic approximation and overlap with unipolar disorder. We've got to make a correct and timely diagnosis.

Let's say a few words about the use of screening tools, diagnostic tools, and symptom measurement tools. It is important that we disambiguate these separate categories of tools; screening tools from diagnostic tools from a symptom measurement tool. Perhaps one of the better-known screening tools for bipolar disorder is the MDQ. It has sufficient sensitivity and specificity to warrant its designation as a screening tool. A key message, however, I want to convey is that it is not a diagnostic tool; it only alerts the practitioner that this individual is a screen positive or screen negative. That then invites the need for us to conduct a clinical assessment, a comprehensive assessment to assure that in the case of a screen positive, this person is truly bipolar, or in the case of a screen negative, that this is not a false negative. We should never confuse screening tools with diagnostic tools.

In academic centers, academic studies and clinical trials, we often use diagnostic tools to assist us. Clinicians have heard of these tools; the Structured Clinical Interview for DSM_IV, (SCID), the Mini-International Neuropsychiatric Interview (MINI), or the Mini-Mental State Exam (MMSE), but these are not really practical for busy clinical practice.

The diagnosis of bipolar is made with a comprehensive assessment. It's my strong viewpoint that once the diagnosis is made through a probabilistic approach that we discussed earlier, that we should embrace measurement-based care. Come up with either your own, or embrace some type of symptom checklist that allows the clinician and the patient to determine: are symptoms present or absent, if so, what are their severity and whether my intervention is effective or not.

We need to make sure we don't misdiagnose bipolar, but we also want to make sure we don't have false positives. I've been hearing a lot about the rising prevalence of bipolar. I think many of us have been as concerned about false negatives, those who have it who are missed, as we are about false positives, those who don’t have it who've been told that they do.

Let’s talk about the issue of the underlying vulnerability to bipolar. We've been talking about the phenotypic complexity of bipolar disorder, a timely and correct diagnosis, and the unfortunate reality that many people have a protracted delay from onset of symptoms to the establishment of the diagnosis. So what we're speaking about is a condition wherein there clearly is vulnerability. And this is where there's been some really exciting research conducted looking at what are the causative factors, maybe the protective factors, that lead to bipolar and the underlying pathophysiological changes.

We know that bipolar disorder is a highly heritable condition. In fact, the estimated liability that is heritable is in the range of around 80% to 85%,^[11] similar to schizophrenia, and much higher than major depressive disorder, which is in the order of around 30% to 35%. And we know it's heritable based on several lines of evidence including, but not limited to, twin adoption and family studies. A commonly asked question from patients who have bipolar disorder is, what is the risk to my child? What is the risk to my son or daughter?

We've learned from the literature that the risk to the offspring is increased — it's variable, but most estimates are between 10- to 20-fold. So taken together, a conservative or more narrow definition of bipolar disorder would say that the risk would be in the order of 10% to 20% for offspring of an affected individual. It’s important to highlight that we don't have any specific gene or genes for bipolar disorder, and the evidence leads us to believe that there are a host of vulnerability and perhaps protective genes that conspire to increase risk to this phenotype of bipolar.

One of the areas that has been really exciting for us is the refinement of the disease model in bipolar disorder. The contemporary view that has been promoted and supported by many lines of evidence is that, in the individual who has bipolar disorder, the principal disturbance within the brain that is subserving the phenotype that we observe -- disturbances in affect regulation, neurocognition, disturbances in some cases in social cognition, as well as disturbances in emotional processing -- can be narrowed down to discrete brain circuits as well as distributed networks throughout the brain. These circuits, broadly cortical and subcortical circuits, are really exhibiting what appears to be a disconnectivity syndrome. The emotional generation centers deep in the brain are disconnected from those parts of the brain that are the emotional regulators. And taken together this model is changing the way we think about not only the underlying causes but also, in fact, how we can treat bipolar.

Let's say a few words about one of these circuits, the cortico-striatal-thalamo-cortical (CSTC) circuit. The message is not how many syllables we can put together in terms of defining a circuit, but the message really is that we no longer think that the biomarker or the biological lesion in bipolar is something wrong with the hippocampus, the amygdala, or the cingulate cortices (anterior and posterior). These nodal structures are clearly implicated, but we now talk in the language of circuits.

These nodal structures, the ones that I've mentioned and many others, comprise circuits that connect: the limbic, the emotional center, in many cases the autonomic center, the endocrine centers, and the deep down centers of the brain to the cortical regions of the brain. The normalonnectivity, the regulation, the top down to the lower structures, seems to be abnormal in bipolar disorder. Importantly, what we've learned from neuroimaging studies is that with effective treatment the disconnectivity seen in bipolar disorder can be normalized in many cases.

Brain circuits are abnormal in structure and function in bipolar. These circuits communicate through messengers. The well-known messengers are monoaminergic, such as serotonin, norepinephrine, and dopamine. There are many other chemistries that communicate within the brain, including, but not limited to, amino acid neurotransmitters such as glutamate and gamma aminobutyric acid (GABA). These observations have been made for many decades and it has provided the basis for employing so-called monoamine-based treatments for bipolar disorder. This is the rationale, for example, for considering an atypical antipsychotic in treating many of the features of a depressive episode or a bipolar episode, targeting these and many other systems.

There are three lines of evidence that would really support the notion that there are disturbances in the so-called catecholamine or indolamine system; that would be evidence from genetic studies showing polymorphisms, putting people at risk for mood disorders.^[12] Second, we have so-called monoamine depletion studies resulting in a recrudescence of affective symptoms in those with bipolar. Third, imaging or ligand studies, which I'll come back to in a few moments. What we've been able to do, through many animal studies, as well as supported by a host of interventional studies in humans, is to trace the constellation of features we see in bipolar to some of the neurotransmitters.

We have evidence that serotonergic dysregulation may be more critical to sleep or impulsivity, which leads in many cases to suicide. We know in some cases that norepinephrine is implicated in cognition or anergia or apathy. Dopamine is so important from a reward perspective, and hedonism. And we're learning that other chemistries, such as GABA, glutamate and histamine, have different, but overlapping roles in subserving affective and cognitive presentations of bipolar.

Again, these observations are not only academic but, in form, treatment interventions. What we're trying to do as we go forward in treating depression is use this information to come up with more tailored approaches for how we can target the phenomenology of bipolar, whether that be the affective domain, the cognitive domain, or some of the emotional-processing domain.

Let’s say a few words about the binding abnormalities. This is an important issue. You may be familiar with some of the radioligand studies which have looked at the monoamines within the brain in vivo in a human subject, whether a healthy control or a patient with unipolar or bipolar disorder. And we were hoping that we had a signature; in other words, we had a discreet pattern of binding in bipolar when compared to unipolar or other mental disorders in healthy controls.^[13,14] So far, we have signals that are promising, but for today we don't yet have a biomarker or a compellation of biomarkers, which is referred to as a biosignature, that is indicative that this person is absolutely bipolar or they're not. And that is clearly the clarion call for us for the future to try and identify such a way to make the diagnosis. And until we do, we're still in fact going to be relying on phenomenology.

Moments ago we were talking about alterations in brain circuits and distributed networks. This is the current disease model in bipolar disorder, and it's well supported. What we've learned also from animal models of bipolar, which are not as robust as we would hope, and from postmortem work, is that within the brain of an individual who's bipolar there appears to be a neurotoxic or degenerative process. And that's evidenced by some replicated observations.^[14,15]

As a matter of fact, what we've seen in the brain of bipolar individuals is they have a loss of cell count. Now, in the brain there is white matter and gray matter, and there is reduction in cell counts seen in both, with more replicated evidence, frankly, in loss of white matter.^[16] So in other words, bipolar could be conceptualized as a white matter disease. And we also see a loss of what's called neuropil, which is what I refer to as the connective tissue within the brain. Taken together, this loss of brain tissue may be more likely observed in those with more progressive illness. Speaking to the neurotoxicity of bipolar disorder, this is not only a disquieting observation, but it's also really alerted us that at the brain level something's changing. This may subserve the phenomenology we see, — the more recalcitrant bipolar over time, as well as observations of patients who may not be responding as well to the treatment after 10 or 15 episodes compared to those who've had 1 or 2 episodes.

My own clinical experience has been very clear that the more episodes they have, the more cognitive impairment they have and the more difficult it is for them to get back to work and fulfill their roles and responsibilities. Not all treatments we have in bipolar are targeting monoamines. Some available treatments, such as lithium and some of the newer treatments being developed, are targeting other systems. One other system that's been implicated is the so-called signal transduction cascades. And simply stated, within the cell, the neurons and glia, there are a host of proteins that not only form scaffolding within the cell, but serve important functional roles for that cell. Lithium is a well-known mood stabilizer that targets some of these downstream intracellular systems, and that has implications again for not only understanding what's happening from the point of view of pathogenesis, but also may inform future treatment directions.

I've been obliquely speaking to the issue of progression throughout this presentation and the idea that bipolar disorder is this condition characterized by vulnerability to cyclicity. I want to spend more time really pressing the point of progression. This is a relatively newer concept for us. We as clinicians, and the literature certainly supports what we've observed, have noticed that many patients with bipolar disorder pursue what appears to be a progressive course. In other words, episodes become more frequent over time, and become longer in duration. The well intervals become shorter and shorter, and the symptoms become more severe. As I was saying moments ago, patients often complain of, or evince cognitive deficits. And the treatment interventions, whether pharmacotherapy, psychosocial interventions, and maybe even neuromodulator treatments such as electroconvulsive therapy (ECT), may not be as robust after 15 or 20 episodes when compared to the first 2 or 3 episodes.

This observation has really fostered this conception these are progressive conditions and Robert Post has put this forward for us.^[17] He actually introduced us to the idea of kindling in bipolar disorder, and this has been further refined in this figure, where people with bipolar have a genetic vulnerability. This genetic vulnerability conspires with adverse life experiences during a latency period or a vulnerability period leading to a prodromal phase, and that prodromal phase is characterized by nonspecific symptoms, followed by a syndromal phase where the person has depressive and manic episodes, followed by a more progressive illness.

There is a whole host of opportunities here in terms of what we can do for our patients if we can intervene earlier and arrest the progression. It’s a tantalizing thought: can we ever prevent the onset of symptoms and intervene much earlier? That's certainly not empirically established but a question we're all very excited to ask and maybe address in the future.

We have evidence, not just speculation, this is something we really think is happening when you look at either monoamines or their metabolites, or some of the biomarkers of mitochondrial activity; the mitochondria is the battery of the cell. Whether you're looking at neurotrophins, such as nerve growth factor, brain-derived neurotrophic factor, insulin growth factor, inflammatory peptides, or so-called markers of oxidative or nitrosative stress, a whole variety of molecular processes are implicated in disease pathophysiology, maybe even resiliency, in bipolar disorder. ^[18] What we're walking away with from these observations is that the molecular portrait of bipolar disorder at 15 and 20 years of age is very different than how it looks at 35, 50, and certainly 60 years of age. In other words, we're seeing a composite changing and a progression.

One of the more replicated examples is that early in the illness course, for one of the neurotrophins, brain-derived neurotrophic factor (BDNF), the circulating levels are normal. But later in the illness course, the BDNF levels start to go down, and that's interesting because it corresponds with some of the observations at the volumetric level that we see loss of brain tissue.^[19-22] We're observing this and there's a need to refine this further. It's not yet something we use clinically to diagnose or monitor disease, but the aim for the future is maybe to characterize our patients with these biosignatures as a way of not only treating but also preventing downstream consequences.

This slide is speaking at the biological level about some of the aspects around progression we've discussed. Bipolar disorder clearly is a heterogeneous group of disorders subserved by heterogeneous genetics -- so-called epigenetics -- which is the modification at the postgene level. Once the illness begins to declare itself it then takes on almost a galloping progression over time, with observations of structural changes in the brain and a variety of molecular changes observed peripherally. We think these observations speak to a disease process that we could in fact arrest and reverse. We think it is relevant to how the illness presents itself.

Too often, we've been guilty of conceptualizing bipolar disorder only from the neck up. This is clearly a brain and body disorder. When persons suffer from bipolar disorder they are differentially affected by disparate medical conditions. Over half of individuals with bipolar disorder meet the National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP) criteria for metabolic syndrome — that is, a constellation of biochemical and clinical risk factors for diabetes, heart disease, and premature mortality. ^[23] They have very high rates of visceral adiposity, dyslipidemia, dysglycemia, and hypertension.

Intuitively, for many years, we thought this was only a function of weight-gain—promoting medicine, and there's no question that weight-gain—promoting medicine contributes dangerously to this significant comorbidity. But it's not the only variable that contributes. There are a host of variables ranging from health systems access and delivery, all the way down to the neurobiology of bipolar. Having bipolar disorder is fundamentally, a metabolic disorder. There is something wrong with the cellular metabolism in these individuals that leaves them at risk for obesity and diabetes, and that then contributes not only to its own morbidity and mortality, but may also conspire with the bipolar illness to progress the illness further. And we've seen this so many times where patients who have bipolar disorder who are overweight or obese, don’t respond as well to treatments as those who are normal weight; speaking to some type of differential biology and also the importance of preventing metabolic complications.

In summary, bipolar disorders are a group of disorders, heterogeneous in phenomenology, pathophysiology, and, certainly, underlying genetics. We know that bipolar disorders exist along a spectrum ranging from the most severe mood excursion, which we know is mania and mixed states, all the way down to more subtle presentations — people with subsyndromal depressive symptoms and hypomanic symptoms. I like to focus on the subsyndromal because, in clinical practice, that's what we see most often: subsyndromal depressive symptoms, often confused for a unipolar disorder. Confusion with unipolar disorder unfortunately leads to a delay in too many people from the onset of symptoms to the correct establishment of the diagnosis, and, unfortunately for many, delayed diagnosis leads to the initiation and continuation of inappropriate therapy.

The underlying biology of bipolar disorder is still being parsed, and we know that there are features more likely to be observed in the biology of bipolar, but we don't yet have a compelling body of evidence that says that it's entirely discreet from unipolar illness. In fact, the phenotypic approximation of bipolar and unipolar disorder is also mirrored by some data supporting a genetic and underlying neurobiological approximation of these two conditions.

Clearly, on the phenotypic surface, bipolar disorder and unipolar disorder have many features in common, such as the predominance of depression, but there are differences: the earlier age at onset of bipolar disorder, the vulnerability to cyclicity, the progressivity of bipolar and the pattern of comorbidity, including higher rates of obesity and metabolic syndrome. What I've also observed in clinical practice is a much more pronounced neurocognitive impairment in bipolar disorder when compared to unipolar. And certainly the family histories are much more loaded with psychopathology in bipolar than unipolar.

These observations are absolutely essential to keep in mind when making the diagnosis. It's important to screen patients for bipolar, and in the office, we need to offer patients a comprehensive assessment to rule in or rule out bipolar. We are learning that this is a progressive illness, not only phenotypically, but also at the neurobiological level. Obviously, a disconcerting observation may help us understand why some interventions are not as effective later in the illness course as they might be earlier in the illness course, but it raises the tantalizing question: Can we intervene even earlier; that is, during a prodromal or during even a latency period before the illness declares itself? That's obviously something we need to establish in the future.

The takeaway message also needs to include a strong statement about the relevance of physical health conditions. Bipolar exists above and below the neck, and in fact, it is the metabolic complications of bipolar disorder, in part related to medicine and in part to biology and other factors, that leads to not only a decreased likelihood of recovery and a more unfavorable course of illness, but also accelerates premature mortality, with heart disease the most common cause of mortality in this population.

So taken together, the message has to be an emphasis on timely and correct diagnosis and, once that's made, the incorporation of guideline-informed, evidence-based, personalized, and measurement-based care for our patients.

Thank you for participating in this activity. You may now take the CME posttest by clicking on the Earn CME Credit link. Please also take a moment to complete the program evaluation that follows.