Teen suicides have been growing more common, and no one really knows why — or how to stop it.
Here are some sobering statistics. Between 2009 and 2017, the number of high schoolers who contemplated suicide reportedly increased by 25 percent. Deaths by suicide among teens increased by 33 percent in that time period as well. Suicide is now the second leading cause of death among teens after accidents (traffic, poisoning, drownings, etc). But don’t be mistaken: Teen suicide is still rare. Just 10 out of 100,000 teens ages 15 through 19 die this way. But even a single death is one too many.
Researchers have been working for decades on interventions to decrease the rate of suicides among teens (as well as among adults). It’s hard. For one, many people who died by suicide never got mental health treatment at all. And the ones that did, well, there’s very limited evidence on what works.
In fact, according to the authors of a new paper in JAMA Psychiatry I read recently, “To our knowledge, no other intervention for suicidal adolescents has been associated with reduced mortality.” That line stopped me cold. There’s nothing that’s been scientifically proven to save lives when it comes to suicidal teens, except what they discovered in this paper?
I called up Cheryl King, who’s been studying youth suicide prevention for the past 30 years at the University of Michigan and was the lead author on the new JAMA Psychiatry paper.
King explained that the paper revisits a clinical trial she and colleagues conducted more than a decade ago. In the trial, half of 448 teens who were admitted to a psychiatric hospital for suicidality were asked to select up to four adults in their lives to receive continuing education and suicide prevention. Simply put: The adults were getting education and support, so they could better support the teen.
The results, which were first published in 2009, were modest. There were small, temporary reductions in suicidal thoughts among the teens in the treatment group, who were more likely to stick to their follow-up therapy. “It wasn’t a big effect, but just somewhat more likely,” King stresses.
But interest in the trial persisted. So King took advantage of a sabbatical, and she and her colleagues tried to see how many of their participants — those who received the treatment, and those who did not — died 11 to 14 years later.
There were 13 deaths among the control group participants (most died of drug overdoses — it’s unclear if they were intentional or not). But among those who elected adults to help them, there were just two. The most conservative interpretation of the data suggests a 50 percent reduction in death among the treatment group.
“If you can come up with a treatment where you had 50 percent less mortality with a treatment, that is actually huge, if that were to replicate,” King says.
But still: How can this small, preliminary study be it? This is the only intervention — or at least the only intervention known to its authors, or to the editors of JAMA Psychiatry — to show that more lives, post-hospitalization, can be saved?
The numbers in the study are just too small for a strong conclusion. It needs to be retested, with a larger sample and across more hospitals, she stressed. It’s not a miracle yet. But it is a promising glimmer of hope that a simple education program can save lives. But the background to it, I fear, is just as important. Why isn’t there good data on saving lives?
It’s not that psychiatry has no evidence-based treatments to offer these teens
Over time, psychiatrists have found various treatments, medicines, and therapies that have been shown to help reduce suicidal ideation, and even suicide attempts (including limiting access to lethal means). And these are hugely important outcomes.
But actual death outcomes aren’t often studied — and suicidal teens aren’t typically tracked into adulthood. “The issue is we haven’t studied mortality,” King says. “So I couldn’t say there is no intervention that doesn’t [save lives] — we just don’t know if any of them do.”
One reason why is that it takes a long time to study mortality.
Even among the highest-risk teens, “suicide is relatively rare,” as Kathryn Gordon, a clinical psychologist and researcher who recently left her academic job for a private practice, tells me in an email. In 2017, the Centers for Disease Control and Prevention reported there were 2,877 deaths by suicide among those ages 13 through 19 across the whole country.
“Often intervention research will instead focus on suicide attempts and suicidal desire — useful outcomes, but not the most crucial ones to establish that an intervention saves lives,” Gordon says. And it’s just hard to study something that is rare.
Making things harder: Studies aren’t typically funded long enough for mortality data to accrue in a statistically meaningful way.
In 2017, the National Institutes of Health spent $37 million on research grants for suicide prevention. That’s trivial when you compare it to the $6.6 billion it spent on cancer research. Out of 295 disease research areas the NIH funds, in 2018, suicide prevention ranked 206. Research on West Nile virus — which kills around 137 a year — is ranked higher.
“In mental health, our studies tend to be funded for studies with smaller sample sizes, and the funding generally goes for four to five years,” King says. That’s not enough time to assess mortality. She and her colleagues were able to do a reanalysis of the original study by looking at national death records and cross-referencing their participant files. But that’s not the same as tracking a group for 10 years and reevaluating them on a larger variety of outcomes.
Why this simple education program could save lives
So why were King’s most recent findings in JAMA Psychiatry so promising? It’s because they suggest that teens who elected adults to receive education and support were more likely to be alive 11 to 14 years later.
Gordon, who was not involved in the JAMA Psychiatry paper, says these findings are important, rare, and hopeful. It’s an intervention that’s not especially expensive and seems to have some effect on mortality, she noted.
But why might it work?
King developed the intervention after working with a lot of suicidal teens, and observing that they weren’t getting enough support when they transitioned out of the hospital. When they’re inpatients at the hospital, they get 24/7 care. “And suddenly,” when they are discharged, “they are supposed to go back to school and wait for their first weekly appointment,” King says. That transition is really hard, and can bring them back to a dark place. “I developed this out of wanting to build a supportive bridge from them.”
It’s key, King says, that the intervention targeted the adults around the teens — the ones providing support. She had the teens nominate up to four so it wasn’t just their parents charged with looking out for them. The teen were encouraged to nominate other family members, educators, or people in the community. They just had to be people that the teens knew cared about them.
The adults were educated in how to talk to suicidal teens and how to make sure they’re adhering to treatment. After an in-person training, the adults got support over the phone for a few months to help them work through the challenges of helping a teen in trouble.
King suspects what makes the intervention effective is that the kids were the ones to nominate the adults. Perhaps that makes them think about the connections they have with others — and opens a door to strengthening them.
The intervention also instigates the adults — not all of whom are the child’s parents — to be more proactive. “The truth is it’s not very easy for adults to go there, to reach out, to talk to and try to help suicidal teens,” King says. “We were always reassuring that their role was just to be a caring person, and they weren’t responsible for whatever choices the teen made. ”
The trial was conducted in the early 2000s through 2008, and the results published in 2009. Again, the results were modest. The intervention didn’t seem to cause harm, which is an important hurdle to clear in all clinical trial research. The teens were only tracked for a year. During the first six weeks, it appears the teens who nominated adults had fewer suicidal thoughts. But then that improvement disappeared by the end of the year.
But what happened from there on out, to make these kids less likely to die a decade or more later?
It could be the teens felt more affirmation, and felt understood. It could be that they learned to talk with a caring adult and ask for help. It could be that they stuck to their treatment better when adults in their lives were more actively involved.
King doesn’t know what the secret ingredient is. “Things can cascade” for teens, she says. Small choices about education, drug use, living situations, and romantic partners begin to accrue and set the course for our lives. And it’s hard to say how exactly this intervention could tip the scales.
But King wants to find out, by conducting a larger study across a few locations in the country. Applying for the grants is “a very long process,” she says. Longer still is the time spent waiting to enroll patients, the time needed to train the adults, the time needed for data to roll in, and the time to count the outcomes at the end.
Again, if it’s true that this intervention can save 50 percent more lives, it would be huge. Recall that 13 people in the control group in King’s reanalysis died.
“Think of it in terms of individual youths lives,” she says. “Six of them might have lived. This is just not that expensive of an intervention.”