Welcome. I'm Dr. Benjamin Drutt, Professor and Rosalyn Carter Chair in Mental Health at the Rollins School of Public Health at Emory University and health systems expert for SMI advisor. I'm pleased that you are joining us today for an SMI advisor webinar titled navigating controversies around antidepressants. Next slide. SMI advisor, also known as the clinical support system for serious mental illness is an APA and SAMHSA initiative devoted to helping clinicians implement evidence-based care for those living with serious mental illness. Working with experts from across the SMI clinician community, our interdisciplinary effort has been designed to help you get the answers you need to care for your patients. Today's webinar has been designated for one AMA PRA category, one credit for physicians, one nursing continuing professional development contact hour. Credit for participating in today's webinar will be available until July 31st, 2023. Next slide. Slides from the presentation today are available to download in the webinar chat. Select the link to view. Next slide. Please feel free to submit your questions throughout the presentation by typing them into the question area found in the lower portion of your control panel. We'll reserve 10 to 15 minutes at the end of presentation for Q&A. Next slide. Now, I'd like to introduce you to the faculty for today's webinar, Dr. Awais Aftab. Awais Aftab, MD, is clinical assistant professor of psychiatry, Case Western University. Dr. Aftab is a member of the community advisory board at Osmond. He is clinical assistant professor of psychiatry at Case Western University in Cleveland, Ohio. Dr. Aftab is recognized for his work across both critical and philosophical issues in psychiatry. He's also a psychiatrist in the Cleveland area working with individuals with serious mental illness. Thank you, Dr. Aftab, for leading today's webinar. I'm really excited to hear it. Thank you so much, Dr. Dress, for the invitation, and good afternoon to everyone. To begin with, I want to say that I don't have any financial or otherwise disclosures or conflicts of interest to report. These are some of the learning objectives for the session today. Fundamentally, we'll be talking about different sorts of controversies that exist around antidepressant medications. A big chunk of the talk is going to be focusing on controversies around efficacy and how to make sense of some of the research data that we have, but we will also touch on, to a lesser extent, controversies around side effects, adverse effects. There's been an ongoing discussion for the last several years, but it got enhanced quite a bit over the past year around the serotonin hypothesis and then how that might relate to how antidepressants work and what sort of effects they have. Dr. Dress already went over my bio. Something that I'll add is that for three years, I led the interview series, Conversations in Critical Psychiatry for Psychiatry Times. I was very involved with different sorts of conceptual, philosophical, and other scientific controversies and questions and debates that exist around psychiatry. I'm active in the philosophy of psychiatry circles. I'm a senior editor for the Journal of Philosophy, Psychiatry, and Psychology. I also write actively online on my sub-stack blog, Psychiatry at the Margins. So to kind of reiterate the outline that we'll be having, we'll begin with efficacy concerns around antidepressant medications. We'll talk about some of the debates around mechanisms and how do we make sense of that. We'll take a quick look at safety concerns and we'll end with a brief discussion of clinical implications and what are some of the takeaways for practicing clinicians because I imagine many people in the audience are clinicians or work directly with patients. This debate in some ways has been quite personal to me because I'm primarily a clinician. I work with patients. I use these medications on a regular basis, but I'm also quite involved in the academic literature and I've been following debates around antidepressant efficacy for many years now. I've personally felt quite conflicted because on one hand, I did see pretty robust effects of these medications on many patients in the clinics and different patients reported a wide variety of responses to antidepressant medications, but at the same time, research study after research study, at least over the last 10 to 15 years, seemed to be saying that the efficacy is pretty limited, that some would even go far and say that the efficacy is not clinically meaningful at all, that antidepressant medications are no better than placebo. These studies and often meta-analysis and relevant publications would get a lot of promotion in the press and they would generate headlines and a lot of debate online and a strong sentiment has emerged in sections of the critical community where they just strongly feel that these medications just are not doing much and they're clinically worthless for the most part. One of the goals of this presentation is let's look at the data, let's see what the research trials actually show. I think what they show is that it's a glass half full. We can't dismiss the efficacy of these medications, but at the same time, we have to recognize that there are limitations and that these medications may not be as effective as we originally might have thought. I'll begin with a relatively recent paper that came out last year and the reason I'm starting with this paper is because it is a meta-analysis of all of the trials that were submitted to FDA regarding antidepressants as part of the approval process. It's also not subject to publication bias since it's not looking at published literature, but it's looking at basically all the data that was submitted to the FDA. This analysis and this paper was published by individuals involved with FDA themselves. They had access to data, they were able to conduct the relevant analysis and make sure it was rigorous. The first author of the paper was Mark Stone, who is the Deputy Director for Safety for the FDA. It was a pretty large meta-analysis of all these trials. What they found was that on average, the difference between antidepressant and placebo group was fairly small. It was 1.75 points difference or what we might round it up as two point difference on Hamilton depression, 17 item rating scale. This is a scale that goes from zero to two and it's a very old scale. It was developed in around 1960, so it has been around for decades. Most clinical trials of antidepressants have been conducting using these scales. Some trials that were conducted using other scales such as Madras, the researchers converted those scores into equivalence of the Hamilton depression rating scale. On average, there was an eight point change in this rating scale for placebo group. There was about a 10 point change in the antidepressant group. This is fairly consistent with some of the previous meta-analysis that have been done over the last 10 to 15 years, which have also reported a two point difference on Hamilton depression rating scale between antidepressant and placebo. This in many ways is the fundamental crux of the efficacy because people feel that how can a two point difference be meaningful? How can that reflect anything clinically significant? My hope is to show that this average difference obscures meaningful differences in the antidepressant group and the placebo group that are relevant for clinicians for us to understand. In the past, NICE guidelines in the UK, they had specified that a three point difference in Hamilton depression rating scale should be considered as clinically significant, but it's widely recognized that that threshold was arbitrary. It was a clinical guess. It was not based on any sort of valid data. There's a lot of controversy around what difference between antidepressant and placebo when it comes to Hamilton depression rating scale should be considered significant. Some people think that two points should be sufficient to be considered significant. Other feel it should be much higher, like five points or more, but there's no clear consensus. This is because in some ways, it's not the right question to ask because the average score may not be the best thing to look at when we're looking at antidepressant clinical trials. I'll just go through some of the other additional data from this paper, looking at the FDA analysis. One thing you can see, this is divided between men and women, and it's showing placebo group and antidepressant group, and it's showing change from baseline. You can see that the red color represents about a 12 to 14 point change from baseline, and others are color-coded. You can see that there is a shift happening that compared to placebo, the response or the change gets shifted a little bit by about two points. We're seeing a lot more of the red. This is reflected in this overall graph of the observed antidepressant response and the observed placebo response as well. On the x-axis, we have change from baseline. At the end of the trial, by how many points has the depression improved? This line here is the antidepressant line, and this line here is the placebo one. You can see that there's a tremendous overlap between the two, but that there are also differences. In particular, we see a peak here with the antidepressant group versus we see a peak with the placebo group a little earlier. The researchers who were conducting the analysis, they broke it down into further subgroups. What they did was that they looked at trajectories of response. They conducted statistical analysis to see if this overall trend can be discriminated into further subgroups. They were able to identify three different trajectories of response. There was a group of people who experienced a large improvement in their depression scores with an average change from baseline of 16 points. This was a pretty significant change in this particular trajectory. This is represented here by this line. This is the line that is having the large change with the antidepressant, and this is having the large change with the placebo. You can see that 25 percent of people in the antidepressant group experienced a large...were in this trajectory of large response versus only 10 percent of people in the placebo group were in this trajectory. If you look at the minimal response, or by people who improved very little, 12 percent of people in the antidepressant group had a minimal response versus 20 percent of people in the placebo group had a minimal response. Then there's this middle category of non-specific response, which is pretty similar between antidepressant and placebo, which is represented here by the blue line. It's really when it comes to these large antidepressant response and minimal antidepressant response that we're beginning to see a separation between antidepressant and placebo, which shows that the average scores are obscuring a lot of heterogeneity. Different people have different trajectories, and some improve quite a lot, and some improve very little, and other people are in between. Just looking at the average score obscures that. It hides this heterogeneity. Just to reiterate, compared to placebo, antidepressants were more likely to show a trajectory of large response and less likely to show a minimal response. If you calculate the number needed to treat for a large response, it turns out to be 6.7. To remind everyone, in general, as a rule of thumb in medicine, a number needed to treat that is less than 10 is generally considered pretty decent. If it's less than 5, that's really great, but anything less than 10 is usually considered to be meaningful and therapeutic. Many medications that we standardly use in medicine often have a number needed to treat that are much larger than this. As far as the number needed to treat is concerned for the large response, it falls within the accepted threshold of clinical use in medicine. There's another paper that's really important when it comes to antidepressant efficacy. This is a meta-analysis of trials with 21 antidepressants for the acute treatment for major depression. It was systematically a network meta-analysis. It might be possibly the largest meta-analysis ever done for any substance or any intervention, but definitely one of the largest, if not the largest. They had 522 randomized controlled trials with 116,000 participants, so a huge sample size. They reported that on average, the effect size was 0.3, which converts into a two-point difference on Hamilton Depression Rating Scale. As I mentioned, this is a pretty consistent finding. All antidepressants were more effective than placebo. Even though the average difference was small, there is statistical separation. You can say that antidepressants are showing discrimination from placebo. The odds ratios were different for different antidepressant medications. The highest one was 2.13 for imitriptyline, which is a tricyclic antidepressant. The lowest one was 1.37 for ruboxetine, which is an antidepressant that is not available in the US, but is available in the UK. This figure just shows the differences in the odd ratios for many different medications. This is the list of all the antidepressants that were included in the meta-analysis. You can see that generally, they're all pretty similar. They're around the odds ratio for 1.5, 1.6. There are some differences at the tails, but in general, most of the modern antidepressants are pretty comparable when it comes to efficacy. In terms of tolerability, most of them are, in these trials, pretty comparable to placebo in terms of dropout rates. I focused on the average difference in depression rating scales for antidepressants. I've shown that although the average difference on these total scores is small, if you look at the heterogeneity of response, we can identify that there is a sub group that does really well. These are just one medication trial. As we all know, in practice, many people require two or three medication trials to find the medication that works well for them. With additional trials, we can expect that the response rate would be even higher. One of the publications that has shown why reliance on total scores can be misleading is this 2016 paper that was published in Molecular Psychiatry. It looked at multiple SSRI medications. They took 18 industry-sponsored randomized control trials. And they pooled all of them together. It was a large sample of 6,700 individuals with depression. Instead of looking at the total Hamilton Depression Rating Scores, it looked at the items. In particular, it examined the depressed mood item. There are 17 items, and they all look at different things. They were like, what if we just look at the depressed mood item? What if we use that as a measure? They conducted 32 drug placebo comparisons. In half of the comparisons in which they were looking at the total score, there was a failure to separate the antidepressant effect from the placebo effect. This is consistent with FDA trial results as well, where about half of the trials are negative, where you're unable to get a significant p-value. But if you look at just the depressed mood item, what you see is that across these trials, antidepressants show a more consistent effect. Only 9% of the trials were negative when we are looking at just the depressed mood item. And there's a table in that paper that breaks down what is happening with all of the different items. So these are the 17 items that constitute the Hamilton Depression Rating Scale. And you can see here that there is different change for each of these items. We see a pretty decent change here on the depressed mood item with an effect side of 0.4. So the overall effect side in this study is 0.27, which is considered small. So you get some decent effect on depressed mood itself. And then you can see with some of the items, you get very little change. So insomnia is being affected very little. If you look at anxiety here, psychic anxiety is improving pretty decently, but somatic anxiety is not. And some of the items are even worsening because they are related to adverse effects of antidepressants. So for example, gastrointestinal somatic symptoms, sexual symptoms, those are worsening. So this is another way in which an average score or sort of like of the total score can be misleading is because it is hiding all the ways in which different symptoms are responding differently to antidepressant medications, but reassuringly the core symptoms of depressed mood and psychic anxiety tend to respond much well. If you look at comparisons of antidepressants versus psychotherapy, this is a relatively recent meta-analysis from World Psychiatry from 2020. This compared psychotherapy versus antidepressant versus combination, and again, a large meta-analysis with a hundred studies, 12,000 patients. And what they found was that combined treatment with sort of like psychotherapy and medication was more effective than either alone with sort of like within a relative risk of about 1.3 sort of like, you know, approximately. But if you compared medication and psychotherapy directly, and for the most part, we're looking at short-term manualized psychotherapy. So we're looking for the most part at CBT trials, you know, to about six to eight weeks, maybe a little bit more, or other psychotherapy such as interpersonal psychotherapy, but also short, you know, manualized and limited. So if you look at short-term psychotherapy and you look at acute treatment with antidepressant medications, they have the same effect as sort of like, you know, one is not superior to the other, and similar results were found for remission. And if you looked at chronic depression or freedom-resistant depression, effects were similar for medication and psychotherapy at the same time. So this also, this provides additional evidence that whatever effect we are seeing with antidepressant medication is probably clinically meaningful because we are seeing the same sort of effect with sort of like psychotherapy as well. And if we want to maintain that the clinical effects of antidepressants are meaningless or sort of like, you know, not clinically worthwhile, then we are forced to conclude that psychotherapy also doesn't sort of like work, which I think would be a rather nihilistic conclusion to kind of at all. And if you look at comparisons of antidepressant medications and their efficacy with other medications that are standard and used across various other specialties in medicine, we also find that there's not that much difference. The meta-analysis that has looked at it is that it's almost a decade old now, but it's still a very popular reference. So this was published in British Journal of Psychiatry in 2012. And what they did was that they included 94 meta-analysis in which they used sort of like, they examined 48 drugs that are used in sort of like 20 medical sort of like, you know, diseases or conditions. And they took 16 medications that are used for eight psychiatric disorders. And they compared how sort of like, you know, how do their effect sizes compare? And generally they found that while there are some medications used in general medicine that have higher effect sizes, in general, psychiatric medications have similar effect sizes as other medicines used in, you know, in the field. And this is the figure from the same paper. And I drew a red line at the effect size of 0.3, because this is the effect size that we see with antidepressant medications. So you can see that there's nothing abnormal or unusual about a medication with an effect size of 0.3. And in fact, that is very common for us to rely on medications that have this sort of effect. So another thing to keep in mind is that we are comparing antidepressants to placebo. And it's important to ask what exactly happens in a placebo group? Because, you know, there's a lot of things going on. There is a group sort of like, you know, there's certain degree of natural improvement and natural history, you know, of the illness. People are getting better. There's expectancy effects. You know, you're expecting that you're receiving treatment. You know, that expectancy produces some benefit. They're in regression to the mean. You start off in a very acute state and sort of like, you know, with the passage of time, the scores normalize. There's a lot of non-specific therapeutic support and monitoring that is built in. You're meeting regularly with the research support staff. You're having lengthy detailed interviews. That itself has a therapeutic aspect. In many of the trials, you can use as needed anxiolytics or sedatives for acute anxiety control. And there are other aspects as well, such as many trials, subjects receive financial compensation. In some of the trials, the baseline symptoms might be inflated. So all of that means that in RCTs, especially RCTs that are industry sponsored, they are trying to have a very high placebo response. And that high placebo response makes it difficult for us to distinguish the effect of the antidepressant. And this is particularly sort of like the case when we ask the question, are placebo effects and antidepressant effects additive? Which is sort of like, you know, like, can you take, let's say, for example, you know, you have an expectancy effect, you know, that is two points. And then you have an antidepressant effects that is also two points. If you add them together, you get four points. Sort of like, you know, realistically, sort of like, you know, based on available research data, probably not. Because it is not something we can add together. We have seen a similar thing with psychotherapy and antidepressant. If you combine the two, it is superior, but it's not the case that you can add the antidepressant effect to the psychotherapy effect and get a combined total score. Which means that when we are subtracting the placebo effect from the antidepressant effect, we are missing a big chunk of efficacy of antidepressants. So subtracting the placebo score from the antidepressant sort of like score assumes that the two are additive and that is why we can subtract them. But in reality, they are overlapping. The antidepressant effect overlaps with placebo effect and there's no clear way for us to sort of like tell how much of it is because of placebo and how much of it is because of depression, antidepressant. So to recap so far regarding the efficacy issue, antidepressants on average have small effect size, 0.3 versus placebo. But if you look at patterns of response, they are pretty heterogeneous and patients are more likely to show a large response with antidepressants compared to placebo with a number needed to treat that is pretty respectable. Clinical effects of antidepressants are obscured by looking at total scores on rating scales. And if you look at just a depressed mood item, you see a more consistent and robust effect. Antidepressants and short-term psychotherapy have equivalent effect sizes on average compared to placebo. And the effect size of antidepressants is comparable to many other treatments in general medicine. So all of this suggests that efficacy is limited. It's not as good as we would like it to be, but at the same time, it's clinically meaningful and that we sort of like, some people respond very well, some people respond very poorly and some people get a non-specific sort of response. So with that, I'm gonna move into the second section, which looks at mechanisms. And a lot of the debate over the past year was kind of sort of like revolved around this umbrella review that was published in Molecular Psychiatry in July of last year. And it was a systematic umbrella review of evidence related to serotonin theory of depression. And what they kind of concluded was that that sort of like the overall body of evidence doesn't provide a consistent evidence that there is an association between serotonin, especially levels of serotonin or any sort of deficiency of serotonin and depression. And there's no support for the hypothesis that depression is caused by lower serotonin activity or concentrations. Now, there's still a lot of active discussion going on in scientific circle regarding the conclusions and the validity of these conclusions. So, but this was very widely reported in the media and newspapers. And this rapidly led to the question. So even though this meta-analysis was not looking at antidepressants, it was looking at just depression. The question emerged that if there is no serotonin deficiency, if there is no systematic serotonin differences in depression, then what exactly is it that the antidepressants are doing? How is it that they're working? And maybe if there is no serotonin deficiency or imbalance, then maybe the antidepressants don't work. So this debate rapidly got enmeshed with questions of antidepressant efficacy. And from my perspective, discussions of serotonin hypothesis become confusing very quickly because there's no precise articulation of what serotonin hypothesis of depression is. It's sort of like, it's a way buzzword that refers to many different sorts of possibilities. And so it's not clear at all that sort of like, what does it mean to disprove some versions of the serotonin hypothesis? And it's certainly sort of like, we cannot say that the results of this umbrella review prove that serotonin has nothing to do with depression. Sort of like here, I've outlined some of the different possibilities that we can point towards when we talk about serotonin hypothesis. So the traditional possibility that I think a lot of people in the general public and even many clinicians seem to have in mind is the first one, that the depression is caused by low levels of serotonin in the brain or low serotonergic activity. And this in general has not been supported by research. And this has been well known for two or three decades now that we don't consistently find any evidence of low serotonin levels in individuals with depression. But these other possibilities have various sorts of evidence to support them. So for example, the second point that depression generally or in some subset of patient involves alterations of the serotonin signaling system. For example, not the levels of serotonin, but sort of like we're looking at serotonin receptors and their distribution or their sensitivity. Or for example, sort of like serotonergic system mechanistically links depressive symptoms and neurobiological dysfunctions in other aspects of brain functioning. So for example, serotonin links things like neurogenesis or neuroplasticity. And it's these things that are more relevant, but serotonin is the one that's kind of like, mediating the relationship between antidepressants and depressive symptoms and these other phenomena. And sort of like, more generally speaking, serotonin system is generally involved in the regulation of moods and emotions. And there may be no specific abnormality in the serotonin system in depression by and large, but it still provides us a target for intervention with serotonergic antidepressant. So even if we accept that there is nothing wrong in the serotonin system by itself, sort of like, the serotonin system is functioning as it should be functioning, that doesn't mean that we cannot target that system and use medications effectively. For example, if someone is experiencing fever, they're sort of like prostaglandin pathways and all of the other pathways, they're functioning exactly as they're supposed to. There's nothing wrong with them. They're designed to respond with fever when an infection is present, but we can still act on those pathways through sort of like, to acetaminophen and other antipyretic medications, and we can achieve symptom control. So the idea that just because we're not finding consistent serotonin abnormalities doesn't mean that the medications are not producing an effect through the serotonergic pathways. So the basics are like, well, I'm going to convince you that the scientific debate is far from over. It has not been definitively shown that serotonin has nothing to do with depression. In fact, there are a lot of sort of like areas of research which seem to suggest that it does have something to do. For example, this is an older paper from 2007. This was a meta-analysis of monoamine depletion studies, and this seemed to suggest that if you deplete the levels of monoamines, that seems to produce an effect in mood in certain individuals who have a vulnerability for depression. For example, they have a family history of depression or they themselves have a history of depression. So it seems to suggest that monoamine depletion has some kind of a vulnerability, plays sort of like the role of some kind of a vulnerability. If there was another meta-analysis from a few years ago, from 2015, looked at serotonin transporter and sort of like imaging and sort of like other findings that were both sort of like in vivo, as well as post-mortem, and they found reduced serotonin transporter binding in different areas of the brain and also reduced serotonin binding in post-mortem brains, again, suggesting that there are these complex relationships going on. There was a small study of 17 patients that came out just last year in October, and this was the first direct assessment of serotonin release capacity. Most of the other studies have used some kind of indirect measure, but this tried to look at serotonin release more directly, and they did find some evidence of reduced serotonin release capacity in patients experiencing major depressive episode. So it means that small study, preliminary evidence, the difference is sort of like modest, but it suggests that we still have a lot of empirical work to do in identifying what exactly is going on. And in general, the involvement of serotonin system in the general regulation of mood and emotions is backed by a large body of literature from animal studies as well as all sorts of clinical trials and other trials with humans. This is a really excellent review paper from 2020 looking at animal data as well as human data and trying to examine what is the role of serotonergic modulation when it comes to mood and other aspects of behavior. And so even if we set that aside, even if we say that there is no dysfunction, the link between serotonergic mechanisms and aspects of mood behavior allow us for the possibility of effective intervention. And serotonergic pathways appear to be mechanistically involved in the regulation of mood and emotions. And sort of like, you know, there may not necessarily be anything dysfunctional or imbalanced in these pathways, but we can still act on them to produce desired effects. So no one should think that just because a study shows that there is no direct involvement of serotonin, that it means that the medications are not producing their effect. At the same time, I want to say that none of this justifies this common narrative of chemical imbalance, of depression as a chemical imbalance. You know, depression is highly heterogeneous and multifactorial, and it involves a range of sort of like, you know, factors across multiple levels. And, you know, this is the message that we should be focusing on sort of like, you know, that depression is heterogeneous, multifactorial, and we have to give up these simplistic narratives of depression being a chemical imbalance. And when it comes to sort of like, you know, current understanding of how antidepressants work, here's, for example, a passage from Peter Kramer from his book, Ordinarily Well, where he focuses on neuroplasticity and neurogenesis. And he writes, the prevailing understanding is not that antidepressants reverse a fundamental deficiency of serotonin or a related chemical, but rather that medications restore resilience in the mind and brain, allowing the growth of new nerve cells and the elaboration of new connections between cells. The drugs permit depression to diminish by allowing repair and new learning to proceed in the brains and persons previously left stuck in depression. So this is sort of like the hypothesis that increasingly we're focusing on. All right, now I'm gonna quickly sort of like talk about some of the serious adverse effects and concerns that have emerged. And one of them are issues with withdrawal, especially complicated and protracted withdrawal. There was a lot of controversy around this, especially in the UK a few years back to a point where the Royal College of Psychiatrists had to issue official guidance on the topic of stopping antidepressants. And if you look at that guidance, they acknowledge that between a third and a half of people who take an antidepressant will experience withdrawal symptoms to some extent. And a portion of them will experience a withdrawal symptoms to a more severe degree. And we don't yet sort of like, you know, know who exactly will experience it. But if you have been taking a high dose of antidepressant, or if you have been taking it for a long time, you're more likely to experience withdrawal symptoms. And they also recommended that one way of mitigating withdrawal is hyperbolic tapering. Instead of reducing the dose by a fixed amount, say 10 milligram per day, we reduce the dose by a certain percentage of the last dose. So 50%, for example, you know, from 40 milligram to 20 milligram, then 50% 10 milligram, 50% 5 milligram. And then we switched to sort of like even lower doses, sort of like, you know, until we get to a point where we can easily taper it off. And in this hyperbolic tapering, instead of linear tapering, seems to be associated with better effects, and better tolerability for withdrawal issues. But there has been increasing attention sort of like paid to that. This was sort of like a quantitative analysis of an internet research forum, where a lot of people were experiencing protracted withdrawal symptoms. And in this, for example, in this small sample of 69 individuals, there were people who reported protracted withdrawal symptoms ranging from five months to 166 months, you know, with a median of 26 months. So a lot of complicated stuff happening with a subset of patients with a wide mix of symptoms, including suicidality, ecstasy, and agitation. So something to take seriously. And this is something that we haven't quite taken seriously in the same sort of way as we should have, especially in the US. So for example, the New York Times did a story in 2018 on problems with withdrawal with antidepressants. And they reported that many people who have been on antidepressants for a long time, when they try to go off it, they experience withdrawal symptoms and they realize that they cannot quit, or they're having difficulty kind of quitting. When this story was published, there was a response letter published by many prominent psychiatrists from Columbia and Bill Cornell and affiliated institutions. And in general, they basically, they tried to reassure the public. They said, we write as clinicians and researchers with experience treating antidepressant withdrawal to practicing psychiatrists and vast majority of patients, withdrawal is low on the list of priorities. And I felt like that gives the wrong message because even though a relatively small percentage of people will experience complicated symptoms, but the people who are experiencing these withdrawal symptoms for them, it's not a low priority. For them, it means a lot. And this is something we should be taking seriously. And because we haven't, people have been forced to find other methods. So for example, this is a paper which talks about experience of an individual who themselves were experiencing antidepressant withdrawal and they started an internet forum to try to help and guide other people. And the note in this paper, medical information about how to safely go off the drugs has been lacking. To fill this gap over the last 25 years, patients have developed a robust internet-based subculture of peer support for tapering off psychiatric drugs and recovering from withdrawal syndrome. So the individuals who are experiencing these complicated withdrawal problems found out that they were not getting much help from clinicians because clinicians in general were not aware of the magnitude of the problem, they were not aware of the strategies of dealing with them, and they were minimizing these issues. So these patients were forced to try to figure out means of helping each other and figuring out hyperbolic tapering strategies on their own. And I feel like that is a collective shortcoming of the profession, it's something that we should have addressed. And this sort of minimization of other severe and poorly understood adverse effects of antidepressants is still going on. For example, there's an active debate going on around post-SSRI sexual dysfunction. Some individuals experience sexual problems with SSRIs, and in some cases, uncommon cases, they linger on even after the antidepressant has been discontinued, and they might even be permanent. And there are also issues around SSRI-induced akathisia and suicidality that are being examined. So my general takeaway from this section is that while in clinical trials, antidepressants are generally well-tolerated, there's a subset of people who do experience serious adverse effects, and these adverse effects have been poorly recognized, poorly acknowledged, and poorly studied. And this is something that we need to be more mindful of, and a lot more research and study needs to be done. I'm going to skip this in interest of time, but this was one of the more recent trials looking at maintenance of antidepressant medications, and I'm going to jump to clinical implications to make use of the last couple of minutes of formal presentation that I have. So let's start by, how do we begin to apply all of this discussion to the clinical setting? So how do we evaluate the risk-benefit ratio in the clinic, for example? So one thing to recognize is that antidepressants are not benign, you know, they do have significant side effects for a certain subset of patients, they might even have more serious adverse effects. And at the same time, you have to balance that with efficacy considerations, which on average are modest, but there's a subgroup that responds much more favorably. So in general, the idea is that the more chronic the depression is or more severe the depression is, the risk-benefit ratio is more favorable. And if the depression is milder or if it is time limited, then we might want to start with some of the other strategies in priority over antidepressant medications. So when to start antidepressant medications? I think before doing that, we have to consider and discuss alternatives, such as psychotherapy, exercise, social support, cross-screnal magnetic stimulation, and even for like mild situational depression, which tends to be quite common in primary care settings, watchful waiting and close observation can be helpful as well. And once these strategies and alternatives have been discussed, and either the patient doesn't prefer them or they can't access them, or they have failed these strategies, then sort of like, you know, starting antidepressant makes more sense. And combining antidepressants with these other strategies, such as psychotherapy, tends to produce even more favorable outcomes. When to maintain antidepressants? You know, we have to take into account how many episodes of depression they've had. You know, in general, the recommendation is that if they've had three or more episodes, that's when maintenance is recommended. If they've had very severe episodes, you know, they've had psychotic symptoms, you know, if they've had a short time in remission, if they still have ongoing subthreshold symptoms, those would sort of like suggest the need, a possible need for maintenance treatment. But at the same time, this is an active discussion where we have to take into account how well the patient is tolerating these medications, what sort of side effects they're having, and what, you know, whether they're aware of the risks of, such as withdrawal and sexual side effects, and what sort of preferences they have. And we have to sort of like be open and transparent about the possibility of mitogenic harm. We have to sort of like consider that this might happen. We have to make it a part of the informed consent process. We have to recognize that complications such as withdrawal and post-SSRI sexual dysfunction are understudied. And if patients complain of them, we shouldn't dismiss these complaints, but rather we should look up the latest guidance and see sort of like how understanding of these adverse effects has changed over the years. This slide is from a paper that compares the British guidelines with the New Zealand and Australian guidelines, and in general shows that there's a lot of congruence among them, that both these guidelines recognize a variety of interventions as being effective, with medications being one of them, and any of these can be first-line depending on patient history and sort of like their patient preference. And then if these fail, then we have additional strategies to address treatment resistance and other aspects. And last year, I did an interactive opinion piece for the New York Times, which is a choose your own adventure style piece where you're in the psychiatric seat and a patient comes to you with depression, and you make choices that are presented to you, and depending on what you choose, it shows you different probability outcomes, and then you take further steps. And if you're not very well familiar with outpatient management of depression, then this interactive experience can give you a sense of what happens in the real world and what are some of the difficulty and frustrations clinicians face in the treatment of depression. So this is a list of references that goes on for many slides, and this will be in the handout, which you can access. So with this, I'm going to stop the formal part of the presentation, and I'm going to open this up for discussion and questions and comments that will be moderated by Dr. Dress. Thanks so much. Really wonderful presentation, Dr. Aftab. So in a minute, we're going to shift into Q&A. Remember you can put your questions into the chat. We do have maybe 10 to 12 minutes for questions for Dr. Aftab, but before we do that, I want to share a moment to let you know that SMI Advisor is accessible from your mobile device. Use the SMI Advisor app to access resources, education, and upcoming events, complete mental health rating scales, and even submit questions directly to our team of SMI experts. Thank you. Download the app now at smiadvisor.org forward slash app. So now we can move on to some of the questions. I'm going to jump around a bit. We had a question that I think tied into some of the latter part of your talk, Dr. Aftab, about kind of a gap between the fact that most of the studies that have been done are done both efficacy, side effects, are short-term, six-week studies, but in practice, it seems like the bulk of patients are on them chronically. So how do you think about that when you're understanding both efficacy and kind of risk benefit profiles at a population level? Yeah. And I think this is very important consideration to take into account that acute efficacy RCDs are all sort of like several weeks, six to eight weeks, sometimes 10 weeks. We have very little studies that go beyond that. And it's difficult. There are many challenges to that. It's very difficult to conduct randomized control trials that last like months or years. And also because the trial design that people have followed in the field of depression for many years is this idea that people have an acute episode of depression, they get treatment, then they get better, and now we are trying to prevent future episodes of depression. So the trial designs after a couple of weeks, even when they have continued these studies and many studies have looked at that, they switched to a relapse prevention model where they're trying to see whether staying on that antidepressant prevents the next relapse or not. And what these artifacts show is that there is a reduced risk of relapse. In the most recent study, for example, that I skipped over, there was about a difference of about 16% between the patients who stayed on antidepressants over the course of a year versus patients who went off and switched to placebo. So it was about sort of like close to like 46%, 56% people who went off antidepressant experienced a relapse of depression versus 40% on antidepressants who did that. So there's a reduction of that. But I think there's also sort of like I think an issue that many people don't follow this pattern of acute treatment, relapse, and prevention, and they have continuous symptoms. They have a mix of depression and anxiety all the time, and they derive some sort of a constant benefit from antidepressants. They feel like the sort of like, you know, as they take antidepressants, they're using it for active symptom controls because their symptoms haven't really gone away. And that sort of clinical practice has not really, to my knowledge, been examined well in research trials, you know, whether, you know, RCTs or otherwise. So we have very little data to guide that. I think one of the issues that comes up is whether there is some degree of tolerance that develops in people over time. Many people respond initially, and then over time, they lose that response, and they might require switching to a different medication or adding on these strategies. But it's clear that we desperately need more evidence-based guidance regarding long-term management of depression since the vast majority of clinical trials are short-term, and that's a big gap in evidence. Yeah, I mean, if you're just looking at, say, a typical first-time depression case, you know, how commonly then would you see this being a time-limited course of antidepressants, you know, versus kind of keep – most guidelines that I'm aware of don't really have guidelines on the length of the course of that initial treatment. Most of them talk more about starting than about, you know, having a delimited course of treatment. So for the first episode of depression, like if someone experiences major depression for the first time in their life, the recommendation from many different guidelines is once they experience benefit from an antidepressant, to continue it for another 6 to 12 months. And after they have been stable for about 6 to 12 months on the medication, then the antidepressant can be tapered off because most people who have one episode of depression are not going to experience another episode of depression, you know, in a reasonable time period or even in their life. It's only when we have multiple episodes, and typically guidelines say about three episodes of depression, that we consider the idea of indefinite maintenance where we just sort of like, you know, continue them on an antidepressant medication and recommend that they stay on them. But there's a lot of gray area, and I think in practice what we're seeing is that people get started on these medications, even for the first episode, without any clear discussion of when they're going to stop it and what that might look like. So this is a big kind of gap in clinical practice that even though guidelines are pretty clear on this, that for first episode, maintenance is not recommended for the vast majority of patients. But in practice, many people get started on them, and then it's never discussed about sort of like when to go off them or not. Yeah, great point. Thank you. Well, there are a lot of questions. There's one that a couple of, one person asked and another person kind of plussed one. So I want to ask that, and that's about your views on esketamine and where that fits in this picture in terms of, you know, potential efficacy vis-a-vis traditional antidepressants. Yeah, so I mean, there's esketamine and esketamine is a very active area of investigation and debate. In fact, just in the last, you know, month, there have been like two high profile trials that have come out, one comparing ketamine versus a rigorous placebo and in one comparing ketamine to ECT. In general, right now, the sentiment is that ketamine and esketamine do have powerful acute effects. So they have powerful effects that, you know, within hours and days, they seem to improve depressed mood. The question is that how well does it last beyond that initial period? And if you look at sort of like, you know, three weeks down the road, four weeks down the road, a lot of that effect seems to sort of like, you know, dissipate and the difference from placebo tends to become quite small. So my own kind of assessment is that it's a very, it appears to be an effective acute treatment. I think the issue is going to be how long can you continue someone on esketamine and ketamine? And there's no clear guidance on that. If someone does well on that, do you just keep on giving them for weeks and months? And unlike, you know, the first antidepressants, ketamine, esketamine has a much higher risk of dependence, much higher risk of abuse. It's a sort of like controlled substance. So there are these additional considerations to take into account, take into account as well. So right now it's recommended only for people who have treatment resistance. They have not responded to several trials of first-line antidepressants. And I would encourage sort of like, you know, I think it's a good option to consider, but it's poorly regulated. Different providers are doing sort of like things in their own way. And it's important to take into account what the long-term plan is going to be if someone is starting ketamine. Great. Thanks. There are also several questions that are about, you know, a little bit related to your kind of heterogeneity comments, you know, how to think, since these studies are generally looking at averages, how are you thinking about it, say, and kind of weighing risks and benefits when you get into, say, older populations or groups with particular comorbidities such as autism or ADHD? How do you kind of map that more general literature that's looking at averages to these more heterogeneous subpopulations? Yeah, that's a very important question. So I think that's a general problem with antidepressant literature. There's kind of, you know, a vision of that, but there are relevant differences. For example, like, you know, in children and adolescents, the effect sizes are lower. So antidepressants are not as effective in children and adolescents as they are compared to adults, something to keep in mind. They are effective in the older population. So they seem to work just as well as adults, except that the time to response seems to be slower. So, for example, where an adult might experience a response in eight weeks, it might take up to 12 weeks for an older individual to experience response. Research has shown that the more psychiatric comorbidities a patient has, you know, if they have a comorbid, let's say, anxiety disorder, substance use disorder, they have autism, or they have other comorbidities, personality disorder, for example, the response rates go down. The more complex the presentation is, the less likely the individual is to respond to first-line antidepressants, and we might have to consider additional augmentation strategies in those cases. Thanks. Well, there were a ton of wonderful questions. I wish we had time to get to all of them, but I want to be respectful of everyone's time. So if we go to the next slide, if there are topics, including topics related to what we've been discussing today from this webinar that you'd like to discuss with colleagues in the mental health field, you can post a question or comment on SMI Advisor's webinar roundtable topics discussion board. This is an easy way to network and share ideas with other clinicians who participated in the webinar. If you have questions about this webinar or any other topic related to evidence-based care for SMI, you can get an answer within one business day from one of SMI Advisor's national experts on SMI. This service is available to all mental health clinicians, peer support specialist administrators, and anyone else in the mental health field who works with individuals who have SMI. It is a completely free and confidential service. Next slide. SMI Advisor offers more evidence-based guidance on medication, such as the psychopharmacology myth versus fact infographic, which offers information on common myths around SMI and psychopharmacology. Access the infographic by clicking on the link in the chat or by downloading this slide. Next slide. To claim credit for participating in today's webinar, you'll need to have met the requisite attendance threshold for your profession. Verification of attendance may take up to five minutes. You'll then be able to select next to advance and complete the program evaluation before claiming your credit. Next slide. Please join us on June 8th as Megan Arett presents Long-Acting Injectable Antipsychotic Medications in the Elderly Population. Again, this free webinar will be June 8th from 3 to 4 p.m. Thursday. Thanks so much for joining us. Until next time, take care.

Navigating Controversies Around Antidepressants

Dr. Awais Aftab

SMI Advisor

evidence-based care

efficacy of antidepressants

mechanism of action

side effects of antidepressants

individual responses to antidepressants

withdrawal symptoms

long-term use of antidepressants