Hello I Want to Die Please Fix Me

by Anna Mehler Paperny

  Crappy as we are at figuring out who has killed themselves, we’re far worse at figuring out who’s in danger of doing so. Much as I loathed being locked up against my will, I understand the excruciating nature of that calculation: Is she about to prance out of here and off herself? A 2016 study found clinicians evaluating suicide attempts in Winnipeg emergency rooms were laughably bad at guessing who was going to try again. Experience helped. Standardized assessment tools didn’t.

  Clinicians who went with their guts were 10 percent more likely to accurately predict a subsequent suicide attempt than clinicians who used a fancy-pants scale. People with more psychiatric experience were better at estimating reattempt risk. But not by much.2

  Yunqiao Wang, one of the study’s co-authors, expected accuracy rates to be low. As a psychologist with a practice of her own, she’s seen how challenging risk assessment is. But she didn’t expect them to be that low, she tells me by phone.3 She knows the horror stories of people who kill themselves within twenty-four hours of discharge. Keeping them just a bit longer, to assess them a bit more thoroughly, might make a difference. “It’s not enough to just, you know, ‘Check, check, check, you meet the criteria’ and then you’re released.”

  Further complicating the prognostication game are people who “use suicide almost as a gesture”—a dramatic cry for help. There were a few people like that in the psych-ward rooms beside mine. Some had checked themselves in because they needed a respite from the responsibility of staying alive. Others, especially in the longer-term ward, cycled through repeatedly because they had nowhere else or nowhere better to go. Wang can understand this perspective but it irks her. “Crisis intervention is not a hotel.”

  Aaron Beck, one of America’s most prominent psychiatrists and the father of cognitive therapy, formalized suicide risk factors into a scorecard: the seriousness of your attempt (Did you really think it would kill you? Did you really want it to?), the degree of premeditation (Did you put your affairs in order? Did you leave a note?), and the degree to which you tried to keep anyone from intervening (Were you alone, did you lock the door, did you wait for a moment when no one would notice your absence? Did you tell anyone?).4 “It’s a really complicated set of variables you have to factor in. But still, you’re not going to be 100 percent right,” says Paul Kurdyak at CAMH.5

  Trust him: he’s been there. One patient, in her mid-thirties or early forties, told him everything he needed to make him feel he could let her go. So he did. And she killed herself within hours. “Thankfully it’s incredibly rare. But it is devastating.” It can be brutal for staff psychiatrists to make discharge decisions knowing both how unreliable and how potentially deadly their best judgment is. “That’s why I have a deep sense of humility about the task at hand.”

  When the Toronto family doctor Javed Alloo faces the challenge of sussing out a suicidal patient’s likelihood of obliterating themselves before their next appointment, he’ll ask if they’re planning on killing themselves but knows that isn’t enough on its own. “If they have interests and if they’re feeling engaged with life, with hope of any kind, then I feel safer.”6 And he’ll look at non-verbal cues—obvious ones, like the consistency of their answers, and ones that would never occur to me: Which way are their feet pointing? Are they responding to questions in a way that’s different from the way they would normally respond?

  One of his longtime patients with a history of chronic, near-debilitating non-mental illness ended up in emerg after overdosing on her medication, and he knew she was bullshitting when she told emergency staff it was an accident. “I’m like, no, she did it on purpose. Because she’s not the type of person who makes mistakes.” The next time she was sitting across from him in his office, he didn’t mess around.

  “‘So, why’d you do it?’ It was a purposely provocative question. She kind of looked at me and started to get ready a glib answer. And I just said, ‘I know. So just tell me why you did it.’ And then she told me. She told me, ‘So I wouldn’t be sad.’”

  So he started her on depression treatment, and she stuck with it. His bold query made their doctor-patient relationship more open. “I felt more scared not asking the question than asking the question.”

  Of course it would be easier if there were a biomarker or a neurological test you could give someone to figure out how likely they are to try and kill themselves at some point in the future. “In cardiology,” the National Institute of Mental Health’s Sarah Lisanby points out, “we have EKGs, we have stress tests, we have ways of knowing if you are likely to die from your heart disease. We need the same for brain diseases. We need the same to know that this person in front of us, who is suffering from depression, is actually dying of it, is on their way to a future suicide.”7

  It sounds fanciful to me, or at least far away. But Maria Oquendo, past president of the American Psychiatric Association, does think it’s possible. Some people who attempt suicide, she says, produce inordinate amounts of the hormone cortisol when they’re stressed, and you can test for that in saliva. But even if you could disaggregate normal cortisol stress responses—if you’re late for something or unprepared or otherwise stressed out—from super-high ones, that test wouldn’t capture everyone. Some people’s desire to die may be tied to abnormal serotonin systems and weirdnesses around receptor density. That’s cumbersome to test for now, but she’s confident that in a decade, that will change.8

  Meanwhile, a group of researchers announced in 2017 that they’d designed a machine-learning algorithm that would read neural signatures of people’s fMRIs, which measure brain activity by tracking changes in your blood flow. This could differentiate not only people thinking of suicide from people who aren’t but people who’d attempted suicide from those who were just thinking about it.9 This sounds futuristic and super-cool, but as someone whose liberty would likely be compromised by such a device, I’m leery. Apart from anything else, this still wouldn’t tell you who is imminently going to try to kill themselves compared to those who just think about it all the time. If it worked it would no doubt flag me based on past actions and current ideation—but I’m not about to kill myself right now and I would absolutely object to being forcibly hospitalized as a result of such a scan.

  We know more about what works and what doesn’t work when it comes to treating people who want to die. It mostly involves talking to people about why they want to die and trying to help them deal with those things in a way other than death. That may sound obvious but it’s a dramatic departure from the way clinicians—and just about everyone else—approach suicidality now.

  “Typically, suicide is something that makes everybody uncomfortable, including clinicians,” says Tom Ellis, the former senior staff psychologist at the Menninger Clinic. “So there’s a really strong emotional pull on both parties’ part to change the subject: ‘Let’s talk about your family; let’s talk about your job; let’s talk about various aspects of your life; but get off this business of wanting to die.’” 10 That strategy’s not super efficacious. So he seeks to attack suicidality directly. “Let’s prioritize, first and foremost, helping this person to survive.” To do that, “we really need to understand what is beyond the disorder.” He sits with patients and writes down what fuels their hopelessness and desire to die. “So it’s very, very concrete, and very explicit.”

  People often know what their triggers are, but either way the therapist will talk things through with them, maybe suggesting things they aren’t verbalizing even to themselves quite yet. He then develops a written safety plan with them “that says, you know, ‘Let’s see: What are your typical warning signs that you’re about to go into a suicidal crisis? What sort of things have you learned in therapy that you can apply?”

  But the evidence suggests that a short-term intervention won’t keep a chronic illness at bay for long. It’s not enough to make someone feel life is worth living for a week or so, only to have them try and kill themselves a few months later, as several people in both the test
and control groups did in a study Tom Ellis conducted. I’ve talked to people who felt way worse descending into despondency after a period of non-shittiness, because being reminded of the delta between hope and despair makes the latter that much more acute, the desperation to escape it that much more pressing.

  Thing is, Catholic University of America psychology professor Dave Jobes says, you don’t need that intensive inpatient environment. He designed a treatment model he calls Collaborative Assessment and Management of Suicidality that can be used in non-specialist outpatient settings—like your GP’s office. He argues that talking directly to patients about the reasons behind their desire to die would save health systems a massive amount of money. “We call it sharpening the driver….We ask the patient, ‘What are the things that make you feel like you should take your life?’”

  But asking if someone is thinking of killing themselves isn’t enough if you ask the question the wrong way.

  Wrong way: “You’re not thinking of suicide, are you?” That’s “leading them to a null answer,” says Brian Ahmedani, director of psychiatry research at the Henry Ford Health System. “Saying the question that way says, ‘I want you to say no.’ And, even further upstream, it tells the patient, ‘I don’t even want to know about it if you are.’ I mean, who’s going to say yes, you’re having that problem, when someone’s led you?” 11

  Yes, suicide is scary. But it’s okay: there are guides available that walk clinicians through, word for word, how to ask a patient about it. One is called the Patient Health Questionnaire (PHQ-9) and it’s just a check-the-box list of nine questions.12 (I’ve gotta say, though, that the suicide-related question is terribly worded: “Over the last two weeks, how often have you been bothered by thoughts that you would be better off dead or of hurting yourself in some way?” At first glance it sounds like you’re asking if the person has been bothered by thoughts she’d be better off dead OR be better off hurting herself. And every reporter knows double-barrelled questions are the devil: the interviewee is going to pick whichever’s easier to answer—in this case, I’d go for hurting myself as the easier option to cop to. But also, thinking you’d be better off dead is not the same as wanting to kill yourself or thinking about doing it—the latter is much more active and, I’d argue, far more lethal.)

  I’ve heard a lot of clinician variations on the suicide question. “Have you been having thoughts that life is not worth living?” was one of my psychiatrist’s favourites, along with, “Have you been thinking of ending your life?” Another psychiatrist favoured “Do you feel safe?” which sounds more like the kind of thing I’d ask myself before heading out on my bike at night, or that I’d ask a Muslim friend living in the United States in 2019. My bias favours compassionate bluntness: Have you been thinking about killing yourself? And, if the answer to that one is negative: Have you wanted to die at all in the past X days/weeks?

  Yes, it’s uncomfortable to discuss. But the stakes are too high not to.

  There are times when wanting your own death is seen not as pathological but as a rational decision, a choice to which you are entitled. It’s telling, though, that the kinds of pain North American society acknowledges as so unbearable as to make death an acceptable choice don’t include the pain caused by mental illness. In Canada and in some US states, a doctor can legally help you die if you have terminal cancer, but not if a mental illness is wrecking your life. That could change—there will likely be court challenges of the mental illness prohibition on medically assisted death—but a proper discussion of what that might look like, of how a doctor would distinguish between a desire to die driven by a disorder’s skewed thinking and a desire to die driven by a rational assessment of what a disorder is doing to your life, is beyond the scope of this book. It is no doubt a question society will have to answer: Why does the pain of people who are crazy carry less weight than the pain of those who are not?

  * * *

  —

  TREATING DEBILITATING DEPRESSION without addressing suicidality is “like treating a heart attack by saying someone needs to lose weight,” Ahmedani says. “If you only treat the weight loss…and you don’t also put them on cholesterol medication, give them nitro, give them other types of interventions for the heart attack…it may not be effective.”

  If changing the way you do things to drastically improve the answers to such questions as “Will patient kill herself?” and “Will patient cycle in and out of the emergency room dozens of times before killing herself or being indefinitely institutionalized?” isn’t your style, there are a few basic changes that could make a difference.

  For one thing, stop it with the stupid “no-suicide contracts.” Yes, this is a real thing: you make a sick person sign a paper saying they’ll stop doing the thing the illness keeps making them do. They’re about as effective as telling someone with cancer to stop letting their cells reproduce in a disorganized and uncontrollable way. Some institutions implement these kinds of contracts because it makes them feel better from a liability perspective, as though they’re less likely to be sued if patients kill themselves because they made them promise not to kill themselves. Spoiler: this ass-covering gambit does nothing to cover your ass. If someone can prove you knew or should have known a patient was a danger to themselves, you’re open to litigation. Especially in the US. It can also destroy any trust between a suicidal person and their clinician if the clinician gives the sense that any such behaviour will be met with punitive action. The prospect of being accused of breaching a contract, legally unbinding as it may be, certainly wouldn’t encourage me to open up. “Patients know that it’s only a piece of paper and that signing that is not going to change the fact if they’re in a desperate place,” Tom Ellis says. “It’s just so naive on the surface, it’s hard to imagine it took hold the way that it did.”

  If all this sounds like it should be a given, you haven’t talked to people who’ve been the loudest self-harm authorities in popular culture over the past half-century—those who think the mere mention of suicide is enough to send people plunging off cliffs like angsty lemmings. This queasiness is closely connected to contagion theory—the idea that exposure to the very concept of self-inflicted death will cause others to commit suicide when they wouldn’t have considered it before. That mentioning or publicizing suicide will lead impressionable individuals—kids and teens, especially—to just off themselves, has been dogma for decades. People refer to the “Werther effect,” after Goethe’s eighteenth-century novel The Sorrows of Young Werther in which the protagonist kills himself—and which, supposedly, gave rise to a slew of copycat suicides across Europe. Even academic researchers have had to confront ethics boards who have freaked out at the thought of talking to suicidal people about their suicidal thoughts for fear that would result in lots of suicides. “This is a common concern,” Tom Ellis tells me. “It doesn’t happen.” He maintains that people actually tend to be relieved to be able to talk about something that is so consuming them. (Yep: I can confirm.) “This is on their mind anyway. It’s not something we’re creating out of thin air.”

  The truth is, it’s complicated. A 2014 study of dozens of teen suicide clusters (a cluster being anywhere from three to eleven deaths) between 1988 and 1996 found that suicides that received prominent, detailed newspaper coverage—especially celebrity or teen suicides—were more likely to be followed by a “cluster” of three or more teen suicides.13

  The first thing that came to my mind on reading this study: whoah, teens read newspapers! (And it’s killing them in awful ways!) But seriously—even here, causation’s tough to suss out: only a quarter of the communities where the suicide clusters occurred had local coverage of that first suicide. That suggests that most clusters happen without prior precipitating coverage.

  It’s also important to keep in mind, I think, that the suicides in this study took place well before social media and the internet were a thing. It’s become easy to find websites and forums where suicide is glorified, where you’re giv
en a how-to, without ever having to pick up a newspaper. “It’s much harder [for us] to do research with social media. It’s so amorphous,” Madelyn Gould, an epidemiologist behind the suicide cluster study, tells me over the phone. “There are a lot of causes and the media definitely has an impact but so does just hearing about [suicide], being exposed to it if you’re vulnerable.”14

  News coverage still matters but not always in the way you might think. There was a decrease in suicides immediately after Kurt Cobain’s death.15 The weeks following a pair of highly publicized teen suicides in the Ottawa region in 2010 and 2011 saw a spike in the number of young people presenting to a children’s hospital emergency department for mental health reasons. The seriousness of their conditions hadn’t worsened, though, suggesting extensive but sensitive media coverage that emphasized the role mental illness plays in suicide may have spurred kids and their families to seek help, enabling more youths to receive needed services.16