About 19 years ago, at the age of 21, Mark Horowitz was very unhappy. He was studying to be a psychiatrist, but he felt his life was falling apart. Horowitz went to a family doctor and asked for a prescription for antidepressants. “She gave it to me in about 30 seconds,” he said. He cycled through a few different ones, each with its own side effects, before settling on escitalopram, known in the United States as Lexapro, a selective serotonin reuptake inhibitor, or SSRI.
In medical school in Sydney, Australia, Horowitz had heard the gospel about antidepressants and believed in the drugs. “I thought, ‘This is going to be the answer,’” he told me. “I read things on the Internet, in textbooks. I thought, ‘This is going to solve it for me.’”
The escitalopram seemed to work, but shortly after starting the drug, he became chronically tired and was diagnosed with narcolepsy. The drowsiness prevented Horowitz from working full-time. It affected his relationships. Slowly he noticed that his memory and concentration had become worse too. By 2016 he’d been prescribed Ritalin to stay awake and zolpidem (also known as Ambien) on the theory that it would help improve his sleep at night, so he’d be less drowsy during the day.
In 2010, Horowitz took a break from his career to work on a PhD at the Institute of Psychiatry, Psychology, and Neuroscience in London. He was still taking the antidepressants, and he was curious about how they were affecting him, so he focused his graduate school research on the neurobiology of depression. One day he came across an article on withdrawal from antidepressants. He’d never been told about withdrawal before, and as he continued studying antidepressants, his rosy view of them began to change.
So in 2015, Horowitz decided to come off the drugs, slowly. But when he got close to 0 milligrams, something in his brain chemistry shifted radically. “I’d get a couple of hours of very broken sleep,” he said, “and I would absolutely dread waking up, because I knew when I woke up, I would wake up to surging panic. I felt like I was on the edge of a cliff being chased by an animal.”
Horowitz felt so disabled by the withdrawal that he ended up moving back home to Sydney to live with his parents. “I was a 35-year-old male lying on my parents’ floor like a child, almost crying,” he recalled. He felt trapped. He knew getting back on the drugs would make his memory and sleep issues worse, but the withdrawal was unbearable.
Eventually he felt he had no choice but to resume their use. In 2018, he began tapering off the antidepressants again, this time even more slowly. He’s now in his third year of the taper. It’s tolerable, but still hard. Each time he reduces his dosage he has trouble concentrating, and sometimes he’s struck by the unsettling feeling that life isn’t real. “I find it very disturbing,” he said.
Horowitz’s experience sounded remarkably like that of others I talked to who’d been on antidepressants—and remarkably like my own. He went from being a true believer in the drugs to seeing the antidepressant industry as similar to Big Tobacco in its funding of research and academics who favor its profit margins. He’s more determined than ever to get off the drugs completely. “I hope whatever adverse effects I’ve had will fade away over time,” Horowitz said. “That’s what kind of drives me.”
It’s impossible to say how many people like Horowitz exist—people who once believed in SSRIs, or their close cousins the SNRIs (serotonin–norepinephrine reuptake inhibitors), as near-miracle-treatments for depression and have now changed their minds. But if you go by the hundreds of Facebook pages, Web forums, and subreddits dedicated to negative experiences with antidepressants, it becomes obvious that at least tens of thousands of people feel their lives have been negatively affected, or in some cases ruined, by the drugs. And those are just the ones who feel strongly enough to post about their experiences.
The field of psychiatry itself may be going through a similar reckoning. A wave of research suggests that the drugs are less effective and more dangerous than many previously believed. And some medical professionals are concluding what until recently felt too heretical to say out loud: Antidepressants may often cause more harm than good.
There’s the familiar story we’ve been told about depression: that it’s caused by a chemical imbalance in the brain, that it’s a disease like diabetes and thus requires medication. Would you tell a diabetic not to take insulin? And there’s the familiar story we’ve been told about SSRIs. The drugs do what their name implies: They prevent the reuptake of serotonin so that the imbalance—a lack of the neurotransmitter in your brain—can be alleviated. (SNRIs work by preventing the reuptake of both serotonin and norepinephrine.)
This story has dominated the popular understanding of these drugs for more than three decades. Since the release of Prozac in 1986, nearly every year has seen an increase in the number of prescriptions for antidepressants. Between 2015 and 2019, the use of antidepressants in the United States jumped by 15 percent—and by nearly 40 percent in teenagers. In 2018, 13 percent of Americans were on some kind of antidepressant, and the anxiety and depression caused by the pandemic has created another surge. They constitute some of the world’s most popular drugs—there were more Prozac prescriptions in 2019 than prescriptions for the antibiotic amoxicillin.
And yet we don’t know much about how they work, or even if they work for many of the people who take them. The theory that antidepressants correct a chemical imbalance of neurotransmitters like serotonin and norepinephrine is not a proven fact. Their basic functioning has not been definitively established. The story we all know is more marketing than science. And the incentive to find out whether they are indeed the best way to treat depression does not exist. In a world where so much scientific research is conducted by pharmaceutical companies and the entities they back, if drugs are making a profit, there’s little reason to question them.
Much of the theoretical support for the use of antidepressants comes from the idea that we understand the causes of depression and the chemicals involved. But that narrative of scientific certainty obscures a murkier set of facts. Among scientists, it’s not controversial to admit that the brain is too complicated to be boiled down to a few chemicals and diagnostic categories. “I don’t think anybody can say with any honesty that they know exactly what happens at a synaptic level,” said Hugh Middleton, a British psychiatrist who has worked in the field for over 30 years and has come to view the chemical imbalance theory skeptically.
Experiments can show that there’s a correlation between SSRIs and increased levels of serotonin, but that’s far different from knowing definitively that a lack of serotonin is responsible for depression or that increasing the availability of serotonin ameliorates it. Bita Moghaddam, a professor of behavioral neuroscience and psychiatry at Oregon Health and Science University, suggests a metaphor: Let’s say you have a toothache, and you’re prescribed opiates to alleviate the pain. But by studying how opiates relieve the pain, we learn nothing about the cause of the toothache. Antidepressants are similar: We can know that they increase the availability of serotonin, but an entirely different chemical process might be responsible for depression.
Even psychiatrists who believe in the efficacy of SSRIs and SNRIs know that the idea that they correct a chemical imbalance is, at best, an oversimplification. “It’s mostly a discredited theory,” Michael Thase, a psychiatry professor at the University of Pennsylvania, told me. “There’s not any good evidence that depression in and of itself is caused by deficiencies or deficits” of serotonin or other neurotransmitters.
The true believers in chemical imbalance, it turns out, aren’t scientists, but an American public that has mostly gotten its information from advertisements. “While the cause [of depression] is unknown, depression may be related to an imbalance of natural chemicals between nerve cells in the brain,” one commercial asserts. “Prescription Zoloft works to correct this imbalance.”
A 2007 survey of hundreds of students at Cleveland State University found that 84.7 percent believed that a chemical imbalance was the cause of depression. Their main source for this belief was not a doctor or a therapist but rather television. And perhaps even more consequentially, the news media oversimplifies the research, glomming onto easy narratives, often against the intentions of the researchers they’re reporting on. The originator of the chemical imbalance hypothesis, a psychiatrist named Joseph Schildkraut, always said it was simply that—a hypothesis. But the media took his research as fact and disseminated it widely. One of Schildkraut’s early articles on chemical imbalance became the most cited article ever published by the American Journal of Psychiatry.
How, then, were we sold on drugs that promise to fix a problem—“chemical imbalance”—that scientists aren’t even sure exists? To the supporters of antidepressants, the answer is easy: Regardless of the limited information we have on the causes of depression, we know that antidepressants work to lessen its symptoms. Thus, despite their operational mystery, their use is worth it. “They’re indisputably a great pharmacological advance,” Jeffrey Lieberman, a professor of psychiatry at Columbia University, told me. “The problems that they’ve been criticized for have more to do [with] the way they’re used by doctors and the heterogeneity of the condition that they’re indicated for.” (In February, Columbia suspended Lieberman over a tweet that he acknowledged as “racist and sexist.”)
In other words, the issue is not the drugs themselves; the problem is that doctors are prescribing them to people who shouldn’t be on them. But as SSRIs and SNRIs have exploded in popularity, some researchers are questioning the basic premise of the drugs: Perhaps, in most cases, they’re not worth taking at all.
In 1998, Irving Kirsch, then a psychology professor at the University of Connecticut, and Guy Sapirstein, then a graduate student working with Kirsch, published one of the most controversial papers on antidepressants. Analyzing 19 studies that included 2,318 participants who were given either an active medication or an inactive placebo, Kirsch found that 75 percent of the effect of antidepressants could be attributed to the placebo effect. He concluded that the act of taking pills was more effective than the active ingredients in the pills and that much of the remaining 25 percent of effect could be attributed to the very knowledge you were on something at all—given antidepressants’ side effects, it can be easy for participants to know which side of the clinical trial they’re on and set their expectations accordingly.
Kirsch is not a fringe scientist. He’s the associate director of the Program in Placebo Studies at Harvard University, and he’s published research on the placebo effect for many ailments, including back pain. The idea that the placebo effect is an important factor in pain relief (whether physical or mental) isn’t particularly controversial: A 2014 study, for example, found that a placebo was more than 50 percent as effective as real medication in reducing pain from a migraine.
But Kirsch’s research on antidepressants was attacked from every angle. People claimed the dataset he used was skewed and that his findings were warped to fit his biases. One response paper claimed Kirsch’s study used “a minuscule group of unrepresentative, inconsistently and erroneously selected articles arbitrarily analyzed by an obscure, misleading effect size” to reach its conclusions.
Kirsch responded at length to these arguments and kept researching antidepressants. He used the Freedom of Information Act to obtain the data from unpublished trials on antidepressants conducted by various drug companies. In a 2008 paper, he showed that nearly half of the studies the companies performed were never published and that the overwhelming majority of them found little to no effect for their SSRIs and SNRIs. In the resulting analysis, Kirsch revealed that when these studies were included, the placebo effect accounted for 82 percent of the response in trial participants.
Since then, some reanalyses have shown slightly better effects for antidepressants, and some agree with the conclusions of Kirsch’s original studies. But even the research that contradicts Kirsch shows nothing close to a cure for depression. A reanalysis of Kirsch’s data by two European researchers, for example, found a better efficacy for antidepressants, especially SNRIs, than he did. But several of the antidepressants still fell below the threshold of clinically significant improvement over a placebo. The question for some researchers is not whether antidepressants are an improvement over placebo but whether that improvement is large enough to clinically matter in patients’ lives. The strongest evidence for the beneficial effects of antidepressants applies to people with the most severe depression.
The main argument between pro-SSRI and anti-SSRI researchers is the cutoff for what is considered effective. Do a few points more on a scale that measures patients’ quality of life matter clinically? If so, what is the precise cutoff for “effective” versus “ineffective”? Researchers largely agree that antidepressants give a small bump on the standard Hamilton Depression Rating Scale, but the unanswered question is whether that bump means antidepressants are worth it given their possible side effects.
“My guess is that, 20 years from now, people will look back at prescribing antidepressants the way we now look at things like bloodletting,” Kirsch told me. “If they are to be used at all, it should be as a last resort when nothing else is working.”
For supporters of antidepressants, like Lieberman, Kirsch’s research obscures the fact that antidepressants work for many people but that they are misused by doctors and given to people with a wide variety of problems that all get lumped under the word “depression.” A study that includes, say, 50 people for whom antidepressants have no effect and 50 for whom they work will show a small overall effect, but it will conceal the fact that they worked well for half the people in the study.
Still, this calls for changing the mainstream narrative that “antidepressants work” to something more scientifically accurate: Antidepressants may work better than a placebo, especially in the most severely depressed patients, though we don’t know exactly how, and the data is conflicting—and even when the research is promising, it’s tainted by the fact that much of it is funded by drug companies.
This more critical view of antidepressants still goes largely unmentioned in the offices of primary care doctors and psychiatrists. Had I been given this counternarrative of all the possible positives and negatives and unknowns, I might never have started taking antidepressants.
In 2017, following a big move to a new city, I began having panic attacks with increased frequency. I started seeing a psychoanalyst and, fearing the meandering nature of therapy wouldn’t work fast enough to ameliorate my acute anxiety and dread, went to a psychiatrist, who, because of her vague European accent and stern nature, I trusted. She prescribed 75 milligrams of venlafaxine (brand name: Effexor), a relatively low dose of an SNRI I’d never heard of. I asked her about side effects—sexual dysfunction and withdrawal. Rare and manageable, she said.
I remember standing outside a CVS in South Philadelphia on a gray day, almost shaking with excitement as I called a friend and waited for my prescription to be filled. I was so happy to finally have a chemical explanation—and chemical solution—to my struggles.
I spent the first few weeks on the drug feeling like I was about to throw up. I felt very high, but not in a fun way. Eventually, things leveled out: I had fewer panic attacks, and the ones I did have were milder. It was as if a gong had been ringing in my head, and someone came in and placed foam padding around it. The gong wasn’t gone, but its noise was contained.
Then, a few months in, I did something I’ve done a thousand times: While on a drive, I blasted my favorite Korn song (“Here to Stay,” for those curious) at full volume, my car rattling to the crunchy bass.
And I felt nothing. No excitement. No adrenaline. No goosebumps at the chorus. I was terrified. The thing I’d enjoyed most in life was deadened, and I felt robbed. I started reflecting on my experience taking Effexor. I weighed myself—I’d gained 25 pounds in a matter of months. I realized that I had begun to have less sex, and the sex I did have was less enjoyable. It took forever to orgasm, and even when I did, that, too, was a kind of muted experience. That foam padding had tamed my harshest anxieties, but it had also tamed everything else.
I swiftly tapered off the drug and felt fine, surprised that I had experienced virtually no withdrawal except for a few “brain zaps”—the bizarre feeling that your head has been hit by a tiny jolt of electricity, combined with the sense that you’ve jumped ahead in time by a millisecond.
And then, one day about four months after my taper, I woke up shaking, with a sense of impending doom like nothing I’d ever experienced. The best I’ve heard it described is by Andrew Solomon in his book Noonday Demon: It’s like the adrenaline rush of terror you feel when you trip while walking and realize what’s happened just before you hit the ground. It was like that. Every second. Twenty-four hours a day. For months. I got close to making plans to end my life.
Through therapy, I could grasp parts of what was causing my pain—but not all of it. What I was feeling was not a return of my depression and anxiety. It was something much, much worse.
Desperate for answers, I began searching online for camaraderie, and I found it on forums like Surviving Antidepressants. In post after post, people wrote about their withdrawal from antidepressants, an experience that made them suicidal or caused their limbs to shake uncontrollably. They described not being able to leave their bed. Even in bed, they detailed a hell much worse than the one they had been in before starting antidepressants. I was relieved, in part, to know I wasn’t alone. But mostly I was horrified that so many thousands of people could be going through so much pain. “I am so lost and feeling so hopeless,” one typical poster wrote after being put on several different antidepressants over a few years. “I have a 3-year-old and a job, and I feel like I am losing everything.”
It turns out that the side effects of antidepressants are much more common than I was told. In one study, 60 percent of people reported sexual dysfunction while taking them. And that feeling of deadening, the feeling that ruined my favorite music for me, was common enough to have a name: tardive dysphoria.
The critics of SSRIs and SNRIs have theories about why the effects of withdrawal can be so serious and long-lasting, namely that your body down-regulates its serotonin production once it becomes accustomed to the drugs—and when you get off them, your brain doesn’t know how to recalibrate. The brain zaps, shaking limbs, and other physical symptoms suggest that a critical neurotransmitter like serotonin is not working as it’s supposed to. One study in rats treated with fluoxetine (brand name: Prozac) found that their serotonin receptors had become significantly less sensitive after a few weeks on the drug.
My mental breakdown was so severe and unrelenting that I gave up and went back on Effexor. Within a day of restarting, many of my withdrawal symptoms receded. I spent a few more months on the drug and then got off again, this time much more slowly, eventually switching to Prozac and then slowly tapering off that. It’s a process I learned not from doctors but from forum posters, who realized that Prozac takes longer than Effexor to leave your bloodstream and thus is easier to taper off from using. (The half-life of Effexor is just a few hours; the half-life of Prozac is several days.) The “Prozac bridge” worked, though I still sometimes feel not fully back to my old self, and I don’t know why.
I asked several scientists and psychiatrists I interviewed for this story whether my mental breakdown could be at least partially related to my antidepressant use. They said it was plausible, that they’d heard similar stories many times.
Studies show that over half of the people who quit taking SSRIs report withdrawal effects, and nearly half report them as severe. SSRIs have been shown to increase the risk of mania. Other studies have found an increased risk of heart conditions and death for those taking antidepressants.
Some of the researchers I spoke with pointed out that we have a much more balanced view of most other drugs. We know that benzodiazepines, for example, can be helpful for panic attacks, but we also know that they can worsen symptoms and that if people use them chronically, their brains adapt to them, leading to a terrible withdrawal. Why, it’s reasonable to ask, would antidepressants be any different?
Jessica Gonzales is stable and does not take antidepressants. This is, in many ways, a novel mental reality for the 36-year-old. In 2004, Gonzales went to college at the University of Southern California. She had just left a terrible relationship. Her parents had recently begun divorce proceedings. She was, by the most basic diagnostic criteria, depressed: She cried all the time; she couldn’t get out of bed.
She went to the school’s mental health center and saw a therapist and then a psychiatrist. Soon enough, she became one of the tens of millions of Americans to take an antidepressant. “I had no trepidations about going on them whatsoever,” Gonzales told me. “I was just like, ‘I feel like shit. Give them to me. I’ll do anything.’”
She tried several medications, including Zoloft. Then she switched to the SNRI duloxetine (brand name: Cymbalta). And for a while, she might have made a good candidate for an antidepressant commercial: The medication worked; she felt less depressed. She was majoring in psychology and learning the traditional chemical imbalance model of depression. These medications, she believed, helped correct that imbalance, and her life was proof of that. Gonzales stayed on the drugs for years.
Slowly, a slew of symptoms of something—though for the longest time she couldn’t figure out what—began creeping in. Years after starting antidepressants, Gonzales could no longer function without eight to 10 hours of sleep, plus a nap in the middle of the day. Her memory became unreliable—she’d blank on basic facts, like the street she lived on or her best friend’s name. She wanted to be an actor, but she couldn’t remember her lines.
Confused and frustrated, Gonzales saw doctor after doctor. They prescribed her other drugs—Wellbutrin (an atypical antidepressant that prevents the reuptake of dopamine instead of serotonin) and Adderall (a combination of amphetamines used to treat ADHD) to help with her constant tiredness. But by 2017, Gonzales felt inert—not like herself onstage or anywhere else. She wanted to get off the drugs, because she had nothing to lose. She wanted to remove any confounding variables and figure out what was wrong with her.
So in 2018, Gonzales began tapering off from her antidepressants. She would take two months to do so, at the advice of her doctor, who told her to drop her SNRI dosage by 25 percent every two weeks.
She soon began to suffer from terrible anxiety. Her stomach hurt. She experienced brain zaps. She began searching the Internet for answers and found thousands upon thousands of people who had experienced something similar. Most of them also said the only solution was to taper off the drugs, only much more slowly. So she slowed her taper. But things still got worse.
One night she woke up with stomach pain so intense she felt like she was being stabbed. She stood up and fainted. She came to, crawled to the kitchen, stood up to get water, and fainted again. Then she dragged herself to the bathroom and fainted a third time. She lay on the floor sobbing. Gonzales had never experienced an episode like this until attempting to get off her medication. Her story mirrors many others heard by psychiatrists who specialize in antidepressant withdrawal.
It would take years for Gonzales to feel normal again. Even after the most acute pain went away, she felt overly sensitive to all stimuli. Her stomach hurt constantly. She still sometimes doesn’t feel fully like herself—mostly during bouts of insomnia and anxiety. “I think what I experienced during withdrawal is outside the scope of what a human is capable of feeling without chemical manipulation,” she told me. “Our language doesn’t have the words to describe it.”
Gonzales choked up as we talked. She said she used to feel angry—angry that no one was there to address the underlying reasons for her sadness when she was put on antidepressants, like her past relationships or her struggles during college. And she was angry because she believes that the drugs she thought were helping her had incapacitated her for many years instead.
“It’s like you’re in a car, and you have this ‘check engine’ light, because your car is about to fall apart,” she said. “And someone comes up and is like, ‘I have this great solution.’ And they just turn off the light. And meanwhile your car is smoking and on fire.”
The history of modern psychiatric medications—pills that are mass-produced and mass-marketed for mental health ailments—is a short one, and it’s largely a history of cycles of overenthusiasm followed by reckonings. As detailed in Anne Harrington’s Mind Fixers, in the early 1900s, the industrial chemical industry found that some synthetic fabric dyes could affect people’s bodies and minds. After a flurry of mergers between industrial companies and small-scale pharmacies, these new corporations began experimenting with psychiatric medications. In 1954, the FDA approved chlorpromazine, aka Thorazine, the first ever antipsychotic medication. Thorazine was a commercial hit, and new sedatives followed. By the end of the 1950s, another tranquilizer, meprobamate (marketed as Miltown), accounted for one in every three prescriptions in the US. The media hailed it as a wonder drug, capable of curing anxiety in anyone and everyone.
But then the backtracking began. The tranquilizers caused weight gain, involuntary limb movements, and facial tics. In 1961, it was estimated that nearly 40 percent of schizophrenia patients put on the tranquilizers suffered from these side effects. But mostly the drugs went out of fashion (though not completely; they’re still used today) because they were replaced with new drugs that were once again hailed as miracles. Benzodiazepines like Valium quickly outstripped sales of the older tranquilizers. These new drugs followed a similar path: Soon the media had sounded the alarm that benzos were highly addictive. By 1975, Valium was regulated by the FDA as a Schedule IV narcotic.
And that provided the opening for antidepressants—drugs that could treat the same ailments as benzos but without, at least theoretically, the side effects and physical dependence. The idea that depression and a host of other ailments were caused by a chemical imbalance that could be corrected by the drugs was in some ways a response to the PR crisis caused by the widespread addiction issues of benzodiazepines.
“The chemical imbalance theory is the pharmaceutical industry’s way of trying to revive the reputation of the prescription of these sorts of drugs,” said Joanna Moncrieff, a British psychiatrist and a founder of the Critical Psychiatry Network, a group of psychiatrists and trainees who are skeptical of the mainstream narrative of chemical imbalance and chemical cure. “We have a disease-centered model of antidepressants, so we have this idea that they’re rectifying an underlying abnormality, and we don’t pay any attention to the psychoactive effects that they might have.”
Josef Witt-Doerring, a 31-year-old psychiatrist in Utah, was taught the same theory about antidepressants as most other med school students. He also worked for a year at the FDA as a medical officer, where he saw how psychiatric medications were approved and the challenges regulators faced in detecting any adverse drug reactions. “Most people think that medications that are FDA-approved [are] free from life-altering adverse reactions. That’s not true,” he told me. “Serious problems can be missed for years, especially if they’re subtle or could conceivably be attributed to the condition the drug treats.”
Witt-Doerring received virtually no training in medical school on the possible adverse reactions to psychiatric drugs, and so he began doing his own research on SSRIs and other medications after stumbling upon Anatomy of an Epidemic, a popular book critical of SSRIs. He came to believe that many of the problems he was seeing in patients—everything from suicidality to sexual dysfunction to tinnitus—were actually caused by the drugs used to treat them. “So many of the patients I’d see would be on four or five psychiatric medications,” he told me. “Once you start being prepared to say, ‘Maybe one of these drugs is actually making people worse,’ you start asking a lot of questions…and then you begin to notice patterns.”
Witt-Doerring also had more and more patients who wanted to get off psychiatric medications, including antidepressants. Helping them to do so and then manage the adverse effects now accounts for about 75 percent of his practice. He still prescribes SSRIs, but only after carefully informing his patients of the possible benefits and drawbacks. “I find it very difficult to prescribe them,” he said. “It’s a very serious decision for me.”
Witt-Doerring and Mark Horowitz might be indicative of ripples of change within psychiatry. But the change is still mostly relegated to Internet forums and small pockets of the medical establishment. Antidepressant prescriptions are as high as they’ve ever been. Drug companies still do research on them. And most people struggling with depression don’t see a psychiatrist—nearly 80 percent of antidepressant prescriptions come from primary care doctors. Studies suggest that therapy is just as effective as medication over the long term for treating depression, but it’s more expensive and often not covered by insurance. For many people in the throes of depression, SSRIs are the only option.
Barring systemic change to how we deal with mental health in this country (and in most other Western capitalist nations), many of us may be stuck taking antidepressants while knowing the drugs sold to us are not some kind of magic bullet, but rather flawed tools that, like benzos and tranquilizers before them, could suddenly fall out of favor and be replaced by something else. Until then, the least we deserve is a full picture, one that has space for all the conflicting science and the thousands of personal stories of triumph and harm.