What we wouldn’t do to be able to classify suicidal risk over someone’s life. The ability to see when a person is – and isn’t – suicidal would be a great boon to our work to prevent needless deaths. This is the grand vision to which Suicide Over the Life Cycle: Risk Factors, Assessment, and Treatment of Suicidal Patients aspires. We’re not there now – and we may never get there – but there’s value in continuing to attempt to understand suicide before it claims even one more life.
Most of Those who Died by Suicide Were Mentally Ill
It’s hard to study whether someone who died by suicide was – at the time – afflicted with a diagnoseable mental illness. The person is no longer around to discuss the situation, and therefore the psychological autopsy process – with all its limitations – must be used. (See Review of Suicidology, 2000.) The primary problem with the psychological autopsy approach is that it is subject to the biases of the investigators – and if they believe there should be mental illnesses, they’ll look for it.
DSM-5, the current Diagnostic and Statistical Manual of Mental Disorders published by the American Psychological Association, is frequently referred to as “the book of woe” and is further characterized as leading clinicians to over-pathologize normal responses. If you’re supposed to find something, and you look in the DSM, you’ll probably find something.
So, the retrospective interviews with family members, friends, and colleagues often led to identified mental illness. Perhaps the most telling aspect of the results that found mental illness is that the primary finding was alcoholism – what would now be called alcohol use disorder (AUD). This is telling, because in 2019, 25.8% of people in a general population survey reported having indulged in binge drinking in the preceding month. This means they’d likely qualify to be diagnosed with at least mild AUD. In the studies referred to from Suicide Over the Life Cycle, the percentage of patients diagnosed with a primary diagnosis of alcoholism is substantially similar to that number.
Certainly, alcohol use could be considered a mental illness, and it is also correlated with suicide deaths. However, few would characterize it as a severe mental illness – what people often think of when they believe someone is impacted by a mental illness. If we lower the bar to any kind of mental impairment, we could perceivably make most people who die by suicide have a mental illness. It would, however, necessarily include nearly every adult.
The second primary diagnosis in the referenced studies was depression, something that many people in the United States struggle with today. Again, it’s something that most people would not characterize as a serious mental illness but something that is included when categorizing most people who die by suicide as having a mental illness.
And that’s not all. The other challenge is in the identification of those who are suicidal. The approach used was coroner determination, which is known to be quite inaccurate. It’s entirely possible that a coroner will decide to mark something as a suicide only when they suspect mental illness, thereby biasing the samples.
So, is there research that says most people who died by suicide had a mental illness? Yes, that’s true. However, the research is fundamentally flawed in at least two dimensions, and the conclusion that someone had a mental illness might include more of us in the general population than anyone would like.
Let me return to the problem of alcoholism as a factor for suicide risk. On average a person who has alcoholism who dies by suicide have been alcoholics for 20 years and die at age 47. It’s hard to separate the impacts of alcoholism from the disorder itself. We know that alcoholics often have unstable home, professional, and social lives. Their disordered drinking leads them to lose their jobs, their homes, and their relationships. These are all substantial factors that lead towards suicide – with or without the introduction of alcohol.
To be clear, it’s not that there isn’t a relationship between alcohol and suicide – there definitively is. The challenge is that alcohol can induce alcohol myopia whether or not the person is an alcoholic, and it’s difficult to separate the work, home, and social losses from the alcohol use when determining how correlated they are.
Roles and Responsibilities
Durkheim’s assertion that suicide seemed to increase during periods of economic downturn has been well replicated – for men. Men’s expectations are shaped by society such that their worth is driven by their ability to work and provide for themselves and their families. Economic downturns obviously make that harder, and it’s easy to accept that men will choose suicide rather than face and address their inability to find work – presuming there is a solution.
Women, on the other hand, are often shown to be more distraught over relational or family-relational issues. They’re more likely to be influenced by divorce or estrangement than moderate fluctuations in the business cycle. This seems to be driven by acculturation. We expect that women will be more focused on family and relationships and disruptions are more impactful.
These are, obviously, stereotypes. However, both fall into the key category of missed expectations. We’ll find that people are more likely to die by – or attempt – suicide when their expectations aren’t matched with the results that they’re getting – and that applies to men or women.
On the Same Team and No Suicide Contracts
It’s subtle. When you insist on a contract with a person, you’re acknowledging the potentially adversarial direction of the relationship. Contracts are used as instruments to document an agreement – but more frequently, they’re the basis for determination of right and wrong. When we pressure someone into a contract, we’re acknowledging the very kind of adversarial relationship that we should be avoiding with a suicidal person. On the surface, asking for a contract that says the other person won’t attempt suicide is pointless, since they’re not likely to think of the contract during a suicidal crisis. More importantly, what consequences can the contract extract from a dead person? (The answer is none.)
Rather than a focus on no-suicide contracts – which don’t work – we can do something that will potentially improve our outcomes. We can find ways to signal that we’re on their team. We’re there to support them. We don’t think that suicide is the right answer, but we want to better understand them and help them solve the problems that may make them believe that death is a better option.
Blame Seeking Messages
Often after a suicide death, there’s a rush to figure out who is to blame. It’s a bad outcome, so someone must have done something wrong. The problem with this is a belief that, for something bad to happen, then someone must have done something wrong. We don’t expect there is someone to blame if a tsunami wipes out a village, so why do we believe that there is always someone to blame when someone dies by suicide? I’m not saying there are never people to be held accountable for malpractice, but this is much rarer than we seem to give credit for.
Seeking the answer to “why” is an unfortunate artifact of our evolution and our desire to predict the future. Rare events, whatever their cause, are met with skepticism and confusion, since our prediction engines have failed. (See The Black Swan, The Signal and the Noise, Superforecasting, and Noise for handling rare events.) We look for someone to blame, so we can incorporate their malfeasance as a part of our models.
Suicide Over the Life Cycle oversimplifies the response to the hopeless person and says, “The clinician should not expect to dissuade patients of their hopelessness; rather the clinician must win the patient’s cooperation to undergo, and stick with, treatment.” Certainly, there’s no point in developing a direct conflict with a patient. However, there’s a path between directly disagreeing with the hopelessness that someone feels and basically ignoring it. In fact, both cognitive behavioral therapy (CBT) and dialectical behavioral therapy (DBT), the most common treatments, encourage patients to challenge their own thinking and to remove cognitive distortions – with the assistance of the therapist.
Rather than dissuading patients from hopelessness, it may be enough to just try to understand it. As the research around Motivational Interviewing shows, sometimes listening is all that is needed.
Sometimes, the cognitive distortions facing a patient can be identified and addressed. Habitual errors in thinking are termed “cognitive distortions” by Aaron T. Beck. They include the following:
- arbitrary inference, drawing a conclusion based on insufficient or even contradictory evidence;
- selective abstraction, attending to only a portion of relevant information;
- overgeneralization, abstracting a general rule from a single event and applying it to both related and unrelated events;
- magnification and minimization, exaggerating or underestimating the magnitude and importance of events;
- personalization, attributing causality to oneself when several factors contributed to an outcome; and
- dichotomous thinking, categorizing people and events in absolutistic, black-and-white terms (e.g., good versus bad).
What to Do When a Client Dies by Suicide
When a client dies by suicide, the counselor, therapist, social worker, or coach will feel the loss themselves and need to process these feelings – but they’ll also need to consider how they will engage with the family. Some will try to minimize contact and pretend that nothing happened even to the point of failing to address the next of kin’s questions and requests. Suicide Over the Life Cycle makes it clear that this is a bad strategy.
In our own situation, Alex’s social worker elected to not be responsive when we reached out. That failure to respond was very problematic – enough that there was a cross-agency escalation. Even if you don’t feel comfortable answering questions, it’s a good idea to be responsive as possible to the family, because you want to avoid becoming adversarial with them.
While we cannot foresee suicide as an outcome in most cases, it’s important that we begin to see Suicide Over the Life Cycle as best we can.