Last winter was difficult for Clara, my nineteen-year-old daughter. Like me, she is a high-functioning neurotic—my phrase for what clinicians call people with Generalized Anxiety Disorder. Waves of dread often wash over us, but we keep our footing. There’s a quick-stepping alertness to our demeanour, like that of hurrying sandpipers on the tide’s edge. We take flight and settle.
For various reasons, Clara had been unhappy, and her unhappiness intensified her natural anxiety. Her GP prescribed the antidepressant Cipralex, but it didn’t calm her. Instead, it made her feel numb. Insomnia became a problem, as if she’d had too much caffeine. So that spring, the doctor suggested a new drug, Trintellix, which Health Canada approved in 2014 for the treatment of “major depressive disorder” in those over the age of eighteen.
I remember sitting in my Honda looking at the sample package, with Clara beside me, and noting the package’s black-box label, which warned of a heightened risk of suicidal thinking and behaviour in children, adolescents, and young adults.
“Why this drug?” I asked, “You don’t have major depressive disorder.”
“You’re not a doctor,” Clara said, with trademark teenage scorn.
Fair enough. Doctors know what they are doing. I had my doubts, but I was no longer her guardian. She swallowed her first capsule of Trintellix that evening.
One week later, while I was in a Kingston hotel room, she messaged me through Facebook. “Can I check myself into a hospital, is that a thing people do?”
“Why?” I typed. “What’s up?”
“I genuinely want to kill myself, and I’ve never had that thought before.” She was frightened by this turn in her mind.
I tossed aside my laptop, called my husband—who was working at home—and urged him to whisk her to the emergency department at the Centre for Addiction and Mental Health (CAMH) in Toronto. After interviewing Clara, the CAMH staff authorized immediate discontinuation of Trintellix. For two days, we had her sleep in our bed on a sort of impromptu suicide watch, until the alien pull to self-destruction receded.
Over the ensuing months, as Clara worked with a kind therapist to resolve what was troubling her, she resumed being a sandpiper. It’s not perfect, as those of us with anxiety would agree, but for the moment, it felt safer.
There was just one thing left to do now that my daughter had been prevented from killing herself: attempt to understand why she had wanted to do so in the first place.
Between 2010 and 2013, antidepressant prescriptions for Canadian children under eighteen jumped by 63 percent—in total, more than five million were issued in that time. The number comes from a recent study published in the Canadian Journal of Psychiatry in which researchers tallied drug-store transactions from across the country. The majority of those prescriptions were for selective serotonin reuptake inhibitors (SSRIs), a popular class of antidepressant medication designed to treat depression, anxiety, and obsessive-compulsive disorder by boosting the levels of serotonin in the brain. The basic theory has been that more serotonin—sometimes known as the feel-good chemical—can help strengthen the brain circuits that regulate mood.
“Our study could be a good-news story,” says Mina Tadrous, the study’s senior author and a research associate at the Ontario Drug Policy Research Network, “in that doctors are identifying more teens with mental health challenges. But the bad-news story would be that it’s much easier to give a pill than talk.” Tadrous worries that the uptick in prescriptions might be a phenomenon called “indication creep,” which occurs when doctors have the discretion to repurpose drugs indicated for other uses, leading to an increase in off-label—or unapproved—prescribing. Health Canada has never officially approved SSRIs for children under eighteen. Yet some doctors believe the benefits of off-label prescribing outweigh the risks. So what, in fact, do we know about the risk-benefit ratio of SSRI use in young people?
The path to understanding the efficacy of SSRIs in treating mood disorders in the general population has been a muddy one. Prozac was the first to come to market, in 1987. An article reporting that “intense suicidal preoccupation” might be one of the adverse reactions to the drug was published soon after. Pharmaceutical companies, however, had much to gain by touting SSRIs as personality-fixing miracle drugs. Several of those drugs—Paxil, Zoloft, Effexor—came down the pipeline in fairly rapid succession. From 1988 to 2008, the rate of antidepressant prescription in the United States increased by nearly 400 percent. Part of this increase came from the fact that sales reps, eager to maximize sales, persuaded doctors to prescribe off-label.
In the mid-’90s, the England-based company SmithKline Beecham (now called GlaxoSmithKline) funded a landmark clinical trial, known as Study 329, which explored the safety and efficacy of Paxil for adolescents. Published in 2001, Study 329 reassured doctors that Paxil was an acceptable remedy for anguished teens. Sales soared. Skeptics, however, raised questions.
Some of the study participants, it emerged years later, entered the trial only mildly depressed, but wound up hospitalized once on Paxil. In 2004, Jane Garland, then a professor of psychiatry at UBC, wrote that “the authors dismissed this result by stating that these psychiatric adverse effects were not attributed to the medication—despite the fact that numerous reports of agitation and suicidal behaviour in young people treated with SSRIs have accumulated since the 1990s.” Indeed, enough evidence had accrued from clinical trials and lawsuits that both the US Food and Drug Administration and Health Canada mandated a warning for pediatric prescriptions of SSRIs—a warning they extended to cover eighteen-to-twenty-four-year-olds in the early 2000s.
Nevertheless, Study 329 became go-to proof of the benefits of such treatment—until, that is, researchers got hold of 77,000 pages of previously undisclosed data from the Paxil study. They reanalyzed the original trial findings and published their results in 2015. Paxil, they found, was not only ineffective in treating adolescent depression when compared to placebo—it was dangerous. Of the ninety-three young people who took Paxil, at least twelve became suicidal while on the drug. (Those already thinking about ending their lives had been precluded from joining the trial.)
A similar misinformation campaign seems to have surrounded another oft-cited study about SSRI safety and young people. In 2004, Forest Laboratories funded an adolescent trial of its drug citalopram (Celexa). Its aim was to win FDA approval for adolescent use, which it did on the basis of this study. But when a team of researchers reviewed the raw data last year, it found emails in which employees discussed how to massage the citalopram study’s findings. Pharmaceutical companies have been known to run drug trials, hire writers to carefully script the presentation of results, and only then attach a prominent psychiatrist as the ostensible study author so that they can publish in prestigious journals. According to the reanalysis, the massaged study “failed to mention” that five of the subjects on citalopram developed “potentially dangerous states of over-arousal.”
How has the industry so successfully maintained its narrative of safety? Lisa Cosgrove, a Boston-based psychologist and the co-author of Psychiatry under the Influence (2015), has pointed out that nearly 70 percent of the experts involved in producing the most recent edition of the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders (DSM-V) have financial ties to the pharmaceutical industry in the form of funded research or fees for speeches promoting certain drugs. What is especially striking about the DSM-V, wrote Cosgrove in an article for American Association of University Professors, is that “attention to adverse side effects of medications is virtually nonexistent.” A reference book consulted by busy GPs, who do the bulk of psychiatric prescribing, that does not address the risks associated with medications is, she writes, “unbalanced and raises questions about industry influence.”
The sidelining of side effects can produce the impression that they are too rare to offset the benefits of a drug. And the medical community is wary of depriving people of treatment that could be beneficial. “We don’t always do well in healthcare with communication of risk,” Tadrous says. “We don’t want to scare patients away.” Health care professionals want to heal, and the fear of dissuading young people from taking what are generally viewed as healing meds can be strong.
Yet, by relying on industry-funded research that can sometimes overplay the positive treatment impacts and obscure the potential harms, doctors can inadvertently become careless. A case in point is the story of John David Wood, an ambitious university student who found a psychiatrist in Toronto willing to prescribe an attention deficit disorder (ADD) drug to help him focus, even though he didn’t have ADD. Adderall was all the rage in college dorms when late-night essays were due. Wood powered his way toward an undergrad degree and then quit the drug cold turkey in 2004, not having been made aware of possible withdrawal effects by his doctor, who himself may not have fully understood those effects.
Wood experienced a psychotic break. According to his mother, Julie, an ER doctor also missed the withdrawal effects and decided that Wood had a major mental illness. The doctor then prescribed Celexa, along with an antipsychotic, and two sedatives. Wood protested that the new drug cocktail produced serious side effects, but to no avail. He remained on the drugs for years but eventually got off them. During a particularly bad time in 2008, he went back on the stimulants, and this time, felt worse. “My son,” Julie says, “realizing he was getting suicidal thoughts, went to North York General. They told him to go sit down and wait like everyone else, and they did not watch him, and he walked out after two hours without anyone taking any notice and later jumped in front of a subway.”
Julie and her husband, Peter, now help run a website called rxisk.org, which provides information about adverse drug events. “Parents mostly never figure out that drugs were responsible,” Julie says. “The news stories always say, ‘Despite the fact that she was on medication,’ she took her own life. Or, ‘Even though he was being treated…’ The media can’t tease apart the story.”
According to rxisk.org, the FDA received 212 complaints of suicidal thoughts related to Trintellix—the drug prescribed to my daughter—between 2013 and 2016. The figure is slippery, because we don’t know how many people experience side effects they don’t report, or how many recognize that their medication may be affecting their behaviour.
David Healy is an Irish psychiatrist who began tracking the side effects of SSRIs since their first appearance over thirty-five years ago. In the late nineties, he recruited twenty psychologically healthy volunteers to examine the difference between SSRIs and older antidepressants. During the study, he unexpectedly witnessed two of the volunteers who were taking Zoloft develop suicidal fixations. Later, he was part of the team that analyzed the undisclosed documents from Study 329, and he has served as an expert witness at trials resulting from lawsuits filed by patients’ families in the aftermath of suicides. When I emailed him to ask why he thought SSRIs would trigger suicidality, he responded: “We know almost nothing about how psychotropic drugs interface with someone’s underlying problems.”
Neuropharmacology is a field in its infancy. Neurotransmitters were first identified less than a century ago, and new ones are still being discovered. An SSRI may lead to regulated levels of serotonin, but, according to Tadrous, “There’s also a chance it’s hitting neurotransmitters we don’t know about.”
Meanwhile, the evidence that SSRIs have a complex impact on developing brains keeps mounting. In 2015, Israeli psychiatrists published a paper describing “SSRI-induced activation syndrome” in children and adolescents—a host of psychiatric changes, including anger, panic, irritability, and mania, that appear to be activated by this class of drug. Among those changes is a condition called akathisia, which is marked by emotional agitation and, sometimes, physical restlessness. Think of ’roid rage mixed with dread: a revving effect that leaves a patient so supercharged that suicide becomes plausible as an escape route.
“Prescribing an antidepressant to someone who is suffering,” Kelly Brogan, a New York psychiatrist, recently observed, “may be like holding out a knife to someone who is falling off a cliff.”
One winter afternoon, I paid a visit to Elia Abi-Jaoude at Toronto’s Hospital for Sick Children. Like Healy, the young psychiatrist participated in the reanalysis of Study 329. Originally from Lebanon, he studied medicine in Winnipeg. “We’re not stopping to ask what this thing is we’re calling depression,” he said. “We’re conflating suffering with mental illness. You go into sub-Saharan Africa and people are broken, they’re in dire circumstances, and we say they have depression. Really? Are you serious? Before, depression was impossible to miss. When you saw the person, you thought, ‘There’s something wrong here.’ Today, you can have a person functioning reasonably well; they may be suffering or struggling, no question, but they’re functioning, and yet we tell them they have a mental illness.”
I thought of Clara as she’d been before she took Trintellix, which was originally intended to treat only those with major depressive disorder. She was working as a sales associate at a clothing store, alert and charming, going out for sushi after shifts, in constant touch with her friends, coming home with amusing anecdotes about customers.
“We are pathologizing adolescence,” says Abi-Jaoude, who is concerned by the number of young people referred to SickKids for depression who may just be struggling through difficult life experiences. “I’m not saying depression doesn’t exist,” he says, “only that depression isn’t always a justified label. Kids come in, and within days, they do well. They’re smiling; they engage with the team.” He points to research that suggests the therapeutic relationship between counsellor and patient accounts for 30 percent of treatment response. Peoples’ expectations of getting better account for an additional 15 percent. “The lion’s share,” he says, “is simply life circumstances changing.”
In the early 2000s, Abi-Jaoude was fascinated by the findings of Irving Kirsch, an American psychologist. Kirsch was interested in studying the placebo effect, but became frustrated with the drug companies, which refused to make all the data from their clinical trials public. After the FDA released both published and unpublished studies in response to a petition he filed, Kirsch discovered that SSRIs weren’t significantly outperforming placebos. Indeed, the difference in responses to placebos and antidepressants was negligible. Kirsch reported his findings in an influential 2008 article, and the following year wrote a book, The Emperor’s New Drugs, in which he argued that instead of being a first-line treatment, antidepressants should be a last resort. After 60 Minutes aired an episode on these findings in 2012, Jeffrey Lieberman, the former president of the American Psychiatric Association, claimed Kirsch’s work was “misleading to people and potentially harmful.”
“Everyone circled the wagons,” recalls Abi-Jaoude. “‘This is a psychologist with an agenda!’” But after reviewing Kirsch’s evidence, Abi-Jaoude became convinced he was right and wrote his own paper in 2011. In it, he detailed the aggressive marketing strategies for SSRIs and considered the ways in which SSRI studies were manipulated. Kirsch’s findings have since been replicated in other studies, including a 2016 review of thirty-four clinical trials of SSRIs for the pediatric population that concluded that the drugs “do not seem to offer a clear advantage.”
“I’m very aware that this is a hard message to sell,” Abi-Jaoude said. “A lot of my colleagues have used antidepressants and seen their patients get better. It’s hard to see there may be another reason they got better.” For instance, many regress toward their mean, in that they eventually return to their own emotional baseline. Decades ago, grief resulting from a traumatic life event such as a divorce or job loss wouldn’t have been linked to a biological illness. Symptoms had to be unduly severe or prolonged before clinical depression was decreed. But the DSM now stipulates a two-week threshold. Walk into the doctor’s office, proclaim your distress, and receive your pill.
By the same token, as a culture, we tend to underestimate the power of placebo. But there is ample evidence that when patients believe they are getting treatment—even when that treatment involves pills containing only inert substances—they can also basically heal themselves. According to ongoing research at Harvard, placebos prompt the brain to release endorphins. Other elements of the brain’s reward system, like dopamine, are activated as well, resulting in a reduction of symptoms of depression, chronic pain, migraine, and other conditions. Regardless of what SSRIs can or cannot accomplish as chemical agents, the very act of taking them tends to trigger this placebo effect. In other words, both the placebo and the drug can make people feel better on the basis of expectation of recovery. What this means is that antidepressants aren’t just failing to do better than placebo in clinical trials: in the treatment of depression, they might in fact be acting as placebos.
“We have to be careful” when discussing SSRIs and placebos, says Abi-Jaoude. “There’s a whole socio-cultural aspect to this.” When people believe medications are going to work, he says, “they are more likely to work. They have been advertised so much that people expect that they’re going to get better. I want to be mindful of that when I critique them. This is part of our culture. It is how we’ve conceptualized suffering.”
The day after I saw Abi-Jaoude, I went out for lunch with a young woman who then worked as a peer counsellor at Stella’s Place, a Toronto non-profit that offers mental health services for teens and young adults. In Canada, access to such services is extremely limited. The average wait time is over 100 days, and, as Lindsay Ranger told me over eggs and hash browns, “When you’re younger, it’s the first time your mental health crisis has happened, and the worst time, because you’ve never experienced it before.”
Stella’s Place is a service for young people with a range of mental health challenges; there, they can engage in creative activities and one-on-one counselling while they wait to see a psychiatrist. But many are unable to wait. “For young people, the narrative has been, ‘If you need help, you need to go to the ER,’” Ranger explains. “The waiting lists are so long to see psychiatrists that, sometimes, it is the only way you can be treated in under a year.” This, then, is how teens are funnelled into the pharmaceutical maw. “When I was young,” she says, “I didn’t know you could be given something other than SSRIs.”
My daughter, unable to find a psychiatrist after her run-in with Trintellix, had gone to her GP. As she was now leery of drugs, CAMH suggested a group therapy session, but the waiting list was about eight months long. She ended up paying to see a private psychologist, but many visitors to Stella’s Place don’t have that option, which is why most people seeking mental health treatment wind up at their family doctor or in the ER. Indeed, a 2012 survey by Statistics Canada found that only 65 percent of patients who sought out therapy felt their needs had been met. In contrast, among those who wanted medication, 91 percent reported getting the drugs they needed. In other words, if you’re a patient, it’s cheaper and faster to get pills, and if you’re a doctor, it’s easier just to prescribe them. And this despite research that suggests it might actually be more cost-effective to cover mental health services: in some cases, therapy works just as well as medication, and it can prevent relapses and reduce the number of doctors’ appointments and ER visits.
Globally, there are a couple of developments that point toward healthier approaches to the treatment of mental illness in young people. One is personalized medicine based on genetic testing. As part of its research into psychotropic drugs, CAMH offers a cheek-swab test that can swiftly determine which, if any, agents a patient can metabolize. Its findings indicate that the presence of certain liver enzymes may affect how different people absorb psychiatric drugs . After the cheek swab has been analyzed, a report is provided that lists the antidepressants and antipsychotics that might work best for the individual—and, importantly, which medication to avoid. Clara now knows about the test, and is considering taking it in the near future.
Also intriguing is the prevention strategy, cognitive behavioural therapy (CBT), which teaches people how to break the feedback loop between negative thoughts and feelings by showing them where their thinking leads them into trouble. First developed by American psychologist Aaron Beck in the 1960s, the approach’s effectiveness has been borne out in dozens of studies. A 2011 review of fifty-three studies involving more than 14,000 Australian youth at risk for anxiety and depression showed promising results for CBT.
How does it work? When I was Clara’s age, I worried about serial killers. On any given day, I might find myself wondering, “What if there’s a man outside my window?” This thought would create a surge of adrenaline, a sudden physical tension. That reaction would then further alarm my mind, which would add a new catastrophic thought: “What if I can’t reach my parents?” Cue the shallow breathing. And around in a circle I’d go, feeling anxious feelings and thinking anxious thoughts. CBT teaches you to recognize what you’re doing before it spirals out of control. It’s now considered a first-line treatment in the United Kingdom: more than 500,000 patients receive the therapy each year, and 43 percent of them recover from their anxiety and depression.
This could be the most effective—and least toxic—intervention ever devised. If children were taught to regulate their catastrophic thoughts and physical sensations, many could self-regulate before their mood intensified and became “diagnosable.” This would take us back to the time when we dealt, medically, with only serious clinical depression and other severe mental illnesses.
The day that Clara first showed me her packet of Trintellix, we were driving to the country, to the simple A-frame cottage built more than a century ago by my great-grandfather. Here there is always healing in the quiet, even if we’re only staying overnight. And what do we return to? Twenty-four-hour news cycles, incessant social-media use, job insecurity, global warming. What’s not to dread?
On World Health Day this year, Dainius Pūras, a doctor and a representative for the Office of the United Nations High Commissioner for Human Rights, made a powerful statement. By focusing so intently on the biomedical model of treating depression, he argued, we are diverting resources from the “social and underlying determinants” that may have a larger and more lasting effect on mental health: economic injustice, political oppression, social violence. “Mental health services, policymakers, medical students, and medical doctors have been misinformed,” he declared. While there’s an important role antidepressants can play in addressing severe depression, the use of psychotropic medication as first-line treatment, he said, is “unsupported by the evidence.” Before off-label SSRI prescription rates rise any further, perhaps it’s time to set aside the pill bottles.
Correction: An earlier version of this story stated that the commercial name for the drug citalopram, for which Forest Laboratories funded an adolescent trial in 2004, is Cipralex. In fact, the name is Celexa. The same earlier version stated the author’s daughter, Clara, took citalopram. In fact, she took Cipralex, which is escitalopram. The Walrus regrets these errors.
