Major Depressive Disorder Podcast

Predictors of Suicidal Ideation, Suicide Attempt, and Suicide Death in Patients With Major Depressive Disorder

Madhukar H. Trivedi, MD; Christoph U. Correll, MD


June 08, 2023

This transcript has been edited for clarity. For more episodes, download the Medscape app or subscribe to the podcast on Apple Podcasts, Spotify, or your preferred podcast provider.

Madhukar H. Trivedi, MD: Hello. I'm Dr Madhukar Trivedi. Welcome to season two of Medscape's InDiscussion series on major depressive disorder. Today we'll be discussing suicidal risk in major depressive disorder, predictors of suicidal ideation, suicide attempt, and suicide death among people with major depressive disorder. I am so excited to talk with our guest today, Dr Correll, who is a professor of psychiatry and molecular medicine at the Zucker School of Medicine at Hofstra/Northwell, as well as professor and chair of the department of child and adolescent psychiatry at Charité University in Berlin, Germany. Welcome to InDiscussion and welcome, Christoph. It is a pleasure for me to invite you to this presentation. This is a very important topic, and we will be dividing it into two sections. One is about the preponderance of data we have, which you are clearly an expert on in terms of predictions, and then we will try to figure out clinically how one uses this information in practice.

Christoph U. Correll, MD: Absolutely. It's important for our audience to not just have data but know what to do with it. And thanks for having me.

Trivedi: Thank you very much. I like to ask our guests initially — and you started giving the answer — what is it [about this topic] that somebody listening to this podcast should be interested in? What is in it for them? What would you say?

Correll: I think life is what we have — and only one, as far as we know. So, protecting that for ourselves and for our patients is key. Also, there is a poem in Germany that says, "Suicide is the last door that you knock on, but you've never knocked on all before." Whoever has been in psychiatry most likely has been exposed to losing a patient by suicide. And that's devastating, obviously, for the patient and the family but also for us. This event is something we need to care about and try to identify, understand, and prevent for many, many reasons. That's why we should listen and understand a bit better what the pathways are into it and how we can potentially get people to stay alive longer and see some hope at the end of the tunnel.

Trivedi: Wonderful. And you've written about and paid attention to this. There are a whole lot of large-scale studies, small studies, etc., looking at the prediction of suicidal ideation, suicide attempt, and death. Maybe you can start off with a brief overview about how you see this ideation, behavior, and death itself. Then, you can tell us if you have any idea what the predictors are. Is there something to be gained from these large-scale meta-analyses?

Correll: First, we should look at the scale of this. Basically, the 12-month prevalence of suicidal ideation in general is about 2% in the population. Suicide attempts are maybe three out of a thousand. And suicide occurs in about half of two thirds of patients who have comorbid depression, whether that's major depression or persistent depressive disorder. But feeling really down and seeing no hope is very much linked to suicide attempts. You also mentioned suicidal ideation, suicide attempt, and death. These are the three levels. Even the ideation, we can, again, parse into passive ideation, which is, "Oh, I don't want to be alive anymore, but I don't do anything about it," to "Maybe I should hurt myself," and "I should hurt myself and have a plan, and now I execute it." These levels of suicidal ideation clinically are very important to understand. We should also see whether we can get a suicide contract so that people will not act on [their ideation]. Let's look at what the data show about ideation, attempt, and death. There are two risk factors that cut across each of those three domains — ideation, attempt, and death — and one is previous suicide behavior. The past predicts the future very much, so we should ask about this whenever we see a patient. Another is severe depression. The more severe the depression is, and maybe it is even psychotic depression, the bigger the risk factor. But there is a protective factor that cuts across ideation and attempt, and this is positive school factors. So, in adolescents, if the formation of your identity and social interaction is not affected by bullying and negative experiences, you have some resilience. But let's now look at some individual factors. For ideation, in addition to previous suicidal behavior and depression, there could be negative school factors, living alone, and clearly substance abuse, which makes you more disinhibited and, in the long run have more depressive symptoms. There are also negative life events, including negative family factors. For suicide attempts, anxiety disorders and alcohol use disorders are relevant, as are negative life events, stressful life events, feelings of hopelessness, and having an anxiety disorder as a comorbidity. And then finally, suicide death factors are being male, having an anxiety disorder, alcohol and substance use, and having anxiety symptoms. In that sense, we have a number of suicide risk factors we can identify and potentially even do something about.

Trivedi: You mentioned two important things. If somebody has significant symptoms within a major depressive episode or significant substance use, these are immediate risk factors, and this obviously would be a good enough explanation for us to try to treat [their major depressive disorder] aggressively. That is understood. But in somebody with a past suicide attempt or past suicide behavior who now has some symptoms, what other guidance can you provide in addition to knowing the person is still high risk? You mentioned something about hopelessness, family contact, and negative thoughts. Is there anything providers can think about in addition to knowing that somebody is male and has a history of major depressive disorder? What are other things you do in order to help patients stay in treatment?

Correll: The first thing is to identify or realize that suicidality is part of psychiatric disorders, particularly depression, and have it on your radar. Second is to assess it, but when it's there, try to mitigate it and also have a risk mitigation strategy so you know what level of care is needed. We know that poor family interaction is problematic, so good family interaction or having anyone who is supportive is important — first, to give the patient hope, but also to be there to monitor them when they leave the office. So, try to engage a significant other and understand if the patient has anyone they can trust and if they have talked to them about their symptoms. Do they have anyone they could call when they really feel overwhelmed by their symptoms? That's an important way to understand whether I can let a patient go. If they're alone and basically talk to the bottle instead of a friend when they feel bad, this might be a high-risk patient. Next, ask, "Do you have any hopelessness? Do you have any future plans?" If a patient says, "I'm suicidal, but I clearly want to finish school, and I want to go to work," or "I would never do it because I don't want to disappoint and hurt my parents or my friends," that's information that's very important to understanding the true risk over the long haul.

Trivedi: How do you use this information to educate people who are at risk to help them monitor themselves and add to their protective factors? How do you educate them on keeping track of their feelings, so that if they start feeling hopeless, it's time to contact their provider?

Correll: That's a very good point. Have some trigger points to identify when they feel more than passive suicidal ideas or feel very pessimistic, when they have very negative and repetitive thoughts and maybe also insomnia, when they are worn down and feel they can't take it anymore. These are risk factors where they should either call and contact a friend, a significant other, or go to their provider. Also identify whether when they feel that way, they skip medication doses. Do they feel that [the medications] can't really help? That's another red flag we need to educate patients about. Tell them that whatever they do, don't stop taking their medications. This can give them withdrawal symptoms that make them feel even worse and give them more anxiety. That's important in terms of education. It's also important to have a family member or significant other in the room to allow them to absorb this information, also psychoeducation about the risk factors and what a person can do about them; and that significant improvement in sleep, staying away from drugs, having social contact, and physical exercise are important. Being physically active and going outside into a green space can really improve mood and reduce suicidality. Sometimes the patient can't activate that skill, but if there is a friend who calls up and says, "How are you doing today? You feel down in the dumps? Let's go for a walk." That can go a long way.

Trivedi: It sounds like what you're saying, and I use it in my practice, is that you're creating a community for the patient so they can call on others to help when they are feeling their worst. And then obviously as a provider, you're available to them when they need it.

Correll: Yes. And they also need to know some emergency numbers because providers are not always available 24/7. Let them know if there is an 800-number or a number for the clinic where somebody after hours can attend to them. That's important.

Trivedi: One of the other things that unfortunately is still disheartening to me is that the myth remains that asking people — especially young people — about suicide puts the thought in them. It is often relieving and comforting for them when an adult asks them. Any thoughts on how parents or adults should address this issue with teenagers and younger people?

Correll: Yes. Thank you for making this very important point because there is this myth that you don't want to open up a can of worms that overwhelms the patient and they suddenly realize, "Oh, I'm suicidal." That's not the case. A patient knows they are suicidal far before you and anyone around them. And as you said, it's often so pent up. It's also stigmatized. Some people in religious circles feel very burdened by not being able to share this "evil" thought with anyone. It can be relieving to them if we normalize it somewhat by saying, "This can happen. We are here to help you. This is not new to us." And as you know, talking about things already reduces some of the burden. What we haven't understood we have to repeat, and what we haven't shared we have to carry. So in that sense, asking them about suicide is very important. Obviously, we have to ask in a caring manner and not just in terms of a checklist — then they might not answer or be afraid you'll lock them up if they've been suicidal before. We have to ask the patient in a very caring and kind manner, providing support and also hope.

Trivedi: Besides aggressively treating their basic psychiatric disorder like depression, bipolar disorder, substance use, or an anxiety disorder, we currently don't have true medications available for chronic suicidal patients except maybe clozapine and lithium. Why don't you talk a little about the evidence for targeted pharmacotherapy for people who are chronically suicidal.

Correll: I think psychotherapy is the building block. People need to feel connected in order to stay connected with life and their own life and their hopes. That is the basis. If there are reasons for being depressed, such as trauma that hasn't been addressed, trauma-focused therapy is very important. We also obviously need to address the comorbidities. And then, as you said, aggressively treating the underlying disease and depressive symptoms is very important. In addition, we have, as you mentioned, lithium and clozapine, which are not used as much for chronic suicidality. But we now also have esketamine approved for suicidality in the context of this psychiatric emergency, at least acutely, so that is something to consider. We have some other agents that might speed up the response, like dextromethorphan plus bupropion. Getting people out of the depression early is very important. Sometimes what we are using — although not approved — in order to take the edge off, is acute benzodiazepine just to relax patients a little bit. That is obviously not something we would want to give long term. We also shouldn't forget about sleep because if you don't have appropriate sleep, you're worn down every day, you feel like you can't make it, and you're exhausted to the bone, it's hard to stay alive. So we also need to look at that. It's one of our vital signs — the fifth vital sign. The sixth vital sign is pain. Physical pain and sleep are often not assessed and addressed and can all fuel into depression and suicidality.

Trivedi: Maybe that tees us up to what we also talked about earlier, Dr Correll, which is being much more nuanced and careful about measurement. You mentioned measurement of the basic psychiatric problem patients are coming in for, such as depression, bipolar disorder, posttraumatic stress disorder, etc. Using rating instruments and monitoring patients using measurement-based care may be the first thing. But in addition, we need to keep in mind the other triggers like sleep, pain, and also talking about hopelessness and interpersonal difficulties as you mentioned earlier. Those measurements and taking a careful history on an ongoing basis are things worth thinking about. Any thoughts on measurements in the service of treating people with measurement-based care?

Correll: Yes. The minimum is to think of ruminations, negative thoughts, feelings of guilt, and hopelessness, and then also sleep, pain, depression, and specific suicidality levels. You need to ask about those in the clinical setting, but there is often not time. Having something you can hand out to people in the waiting room is helpful. It saves you time. You go over the ratings that were selected as being present. Sometimes patients feel more comfortable admitting these things on paper than to someone who looks them in the face when they also know the doctor or prescriber is rushed — it doesn't have to be a doctor only. In that sense, measurement-based care is enormously important. There are also some tools we have. We've also been using the Sheehan Suicidality Tracking Scale, for example. You could also use the Beck Depression Inventory. These are some self-ratings that are easy to use. But you have worked on a more comprehensive tool you can share with our audience.

Trivedi: Right. We have developed a nine-item suicide self-rating instrument called Concise Health Risk Tracking (CHRT). I'm glad you brought it up because this idea of suicide and medications has been adopted by the FDA for a different purpose. They were looking at whether or not medications are causing suicide. In the process, they recommended measurements, but they are really just trying to be more regulatory. We are interested in it clinically. What we've done is develop the CHRT. We recently published several papers on it to give metrics. Six of the first nine items are tracking exactly what you were saying — hopelessness, social support systems, and so forth that serve as a propensity for suicide. The last three items are suicide items themselves. We have been able to use that rating instrument to be able to predict higher risk vs lower risk. There cannot be a 100% accuracy, as you all know, but this one is getting us to 80%-85% accuracy in terms of risk for suicide events. It is called CHRT-9, and you can find it online. This is an instrument for which we've published the psychometrics in the last few years quite extensively. In general, what I tell providers when they ask about the instrument is that they should track [symptoms]. They should educate patients about self-monitoring their own symptoms, not only when they are thinking about death or suicide but also their propensity for the risks you mentioned — hopelessness, sleep, and social support systems. These are the things that really help. One additional domain we have been interested in that I'd be curious to hear your thoughts on is irritability. Irritability often gets excluded from diagnostic criteria. And in order to do that, as you know, I'm much more of a fan of developing the right instruments to measure with — we developed measurement-based care for the STAR(*)D trial 30 years back — we have developed a rating instrument called Concise Associated Symptoms Tracking (CAST-SR) that also measures irritability. And we've shown that irritability is independently associated with suicidal ideation and behavior, over and above depression. The CAST rating instrument is also a 10-item rating instrument. Both of these are self-rating instruments, so patients can measure themselves.

Correll: This makes a lot of sense to me because irritability is so closely related to impulsivity. A person gets hit by normal stressors or situations in life and they can go off. And then if they have depressive symptoms and hopelessness paired with the impulsivity, either endogenously or also facilitated by ongoing substance use, it creates the perfect storm. And the thought becomes a plan, a plan becomes an action, or there's even no long-term plan. It's just spontaneously there, and then they can act and regret it afterward.

Trivedi: This has been a wonderful discussion, but I would like to end with a couple of important things. One is, you deal with this all the time, and you mentioned that ketamine/esketamine has potential benefits for suicidality going forward. Any other thoughts on what you see coming down the pipeline that might be of value for our listeners to keep track of?

Correll: Any treatment developed for depression that has a very fast onset will be of value — most likely, some NMDA (N-methyl-d-aspartate) antagonists, or glutamatergic antagonists, seem to be in that realm if they come down the pike and don't have any psychotomimetic effects. Remember there are plastogens, neuroplastogens, and psychoneuroplastogens. The psychoneuroplastogens also have psychotomimetic effects. We're now trying to harness drugs that are neuroplastogens that help the brain reestablish some of the arborization and connectivity very quickly — quicker than we ever thought — but without having some of the other effects and addiction associated with it. So whatever comes down the pike, that will be something to look out for. Also, we should keep an eye out for combination treatments with our currently approved antidepressants because suicidality is quite common in patients with resistant depression. Do you have any other medications or treatments you recommend people look out for?

Trivedi: Newer antidepressants like esketamine and the combination you mentioned with dextromethorphan and bupropion that lead to rapid onset of response are likely going to change the treatment landscape. This is because one of the problems our patients run into is that their depression or bipolar disorder is unremitting, and then if they have a life event or something that induces hopelessness or helplessness, they are at high risk. If we can, we should reduce the duration they have to go through the suffering.

Correll: Another agent that comes to mind that hasn't been approved that would also potentially be able to speed up depression response is zuranolone, which is another neurosteroid that has been tested and obviously is already approved as an intravenous formulation for postpartum depression. That agent seems to speed up the response quickly, too.

Trivedi: GABAA receptor modulation is another domain where very exciting molecules are being studied, so we will see over the next 5-10 years what happens with that. Thank you so very much, and I appreciate all that you've shared with us. One of the other things Dr Correll shared is that we should all continue to keep in mind — especially for suicide but also otherwise — the importance of conjoint psychotherapy for these patients. Thank you for bringing that up, and I appreciate you taking the time to be with us. Thank you all for tuning in. If you have not done so already, take a moment to download the Medscape app to listen and subscribe to this podcast series on major depressive disorder. This is Dr Madhukar Trivedi for InDiscussion. And thank you again, Dr Correll.

Listen to additional seasons of this podcast.



Prevalence of Suicidality in Major Depressive Disorder: A Systematic Review and Meta-Analysis of Comparative Studies

Twelve-Month Prevalence of and Risk Factors for Suicide Attempts in the World Health Organization World Mental Health Surveys

Persistent Depressive Disorder

Risk and Protective Factors for Suicide and Suicidal Behavior

Prevention of Suicide by Clozapine in Mental Disorders: Systematic Review

Lithium Suicide Prevention: A Brief Review and Reminder

Trauma-Focused Psychotherapies for Post-Traumatic Stress Disorder: A Systematic Review and Network Meta-analysis

Intranasal Esketamine and Current Suicidal Ideation With Intent in Major Depression Disorder: Beat the Clock, Save a Life, Start a Strategy

Dextromethorphan/Bupropion (Rx)

Status Update on the Sheehan-Suicidality Tracking Scale (S-STS) 2014

Psychometric Properties of the Beck Depression Inventory-II: A Comprehensive Review

The 9-Item Concise Health Risk Tracking – Self-Report (CHRT-SR9) Measure of Suicidal Risk: Performance in Adult Primary Care Patients

Maximizing the Adequacy of Medication Treatment in Controlled Trials and Clinical Practice: STAR(*)D Measurement-Based Care

Psychometrics of the Self-Report Concise Associated Symptoms Tracking Scale (CAST-SR): Results From the STRIDE (CTN-0037) Study

Mechanisms of Action and Clinical Efficacy of NMDA Receptor Modulators in Mood Disorders

Psychoplastogens: A Promising Class of Plasticity-Promoting Neurotherapeutics

Zuranolone in Major Depressive Disorder: Results From MOUNTAIN-A Phase 3, Multicenter, Double-Blind, Randomized, Placebo-Controlled Trial

GABA Receptor Positive Allosteric Modulators

Conjoint Therapy

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.