Do Antidepressants Actually Work? The Author of "Listening to Prozac" Is Back With the Answers

Aug 15 2016

Do Antidepressants Actually Work? The Author of “Listening to Prozac” Is Back With the Answers

August 15th, 2016

A couple of months ago, a study on the efficacy of antidepressants for children and adolescents was published in The Lancet. A flurry of headlines from major news outlets followed: “Most Antidepressants Aren’t Effective for Children, New Study Finds” (Wall Street Journal); “Most antidepressants ineffective in teens, study finds” (CBS News); “Study: Most antidepressants don’t work for young patients” (Business Insider). You get the picture.

The media is ever-hungry for the latest exposé of the pharmaceutical-medical industry. (And The Influence is certainly no exception.) But when it comes to antidepressants, do claims that they’re at best placebos, at worst dangerous, hold any weight?

Dr. Peter Kramer is a psychiatrist who made a name for himself by exploring the ethical and philosophical dimensions of using antidepressants to change personality in his 1993 book Listening to Prozac, a New York Times bestseller. Now he’s back with a new book, in which he argues that antidepressants get unjustly picked on, and in fact work “ordinarily well” (the book’s title). Media claims about the recent Lancet study, he says, were entirely wrong.

Kramer is a practicing psychiatrist, a faculty member at Brown Medical School, and the author of six other books. Ordinarily Well: The Case for Antidepressants has clearly struck a nerve, garnering three separate reviews in The New York Times alone.

Peter D. Kramer

It’s important to note the evidence that he is no Big Pharma shill. In 2008, as he recounts in Ordinarily Well, the British Medical Journal (BMJ) named him as one of 100 or so “independent medical experts” who did not take money from pharmaceutical companies. He was one of only six psychiatrists on the list.

Speaking with Kramer on the phone, I got the impression of a man both passionate about his subject and ever so slightly frustrated with a climate that he perceives as hostile towards Western medicine, and psychiatry in particular.

Still, it was great to talk to a psychiatrist that I didn’t have to pay. Dr. Kramer gave me the inside scoop on the best antidepressants, why depression is like tuberculosis, and the meaning of “the trout in the water.”


Sarah Beller: So, what was going on with that Lancet study?

Peter Kramer: The study found that Prozac [generic name: fluoxetine] did have an effect size that was statistically significant [compared to placebo]. It’s as good as the effect size for adults. The study tested something like 10 antidepressants, and it’s not that there was no effect picked up at all for the others. If you test 10 of anything, some will be higher and some will be lower.

The prevailing theory is that depression makes it harder for the brain to make new cells and new connections, and antidepressants seem to restore that capacity. You find it with lithium and ECT, too—lots of remedies for depression seem to “unstick the brain.”

Children and adolescents do that so well already—they are making new cells and new connections all the time—so it’s sort of hard to know if that mechanism is working for them. We don’t have as good a theoretical basis for prescribing for them … I don’t really work with children or younger adolescents, but what you want for your child is the same as what you want for an adult—a really good expert. Despite the average result [in his new book, Kramer writes: “My impression is that antidepressants work unreliably in children and carry serious risks”], you’ll try these things if the child is really suffering, not achieving developmental milestones.


Why do you think people are so distrustful of anti-depressants in general? 

That’s one topic of Against Depression [his 2005 book]. There’s such a long history of seeing melancholy as being somehow informative—a sign of sensitivity, a sign of discernment, an illness of superiority—the way at certain points tuberculosis was seen as a sign of refinement. There’s a reluctance to see it in medical terms. Some of it has to do more generally with medical psychiatry and reluctance to see phenomena that have psychological symptoms as syndromal and diagnosable. There’s a distrust of medication for things of the mind.

Then, under the heading of stigma, psychiatrists are not that well respected or trusted, in comparison to other doctors. Antidepressants are being used longer and longer and there is unease about seeing depression as something like diabetes where you might need to be in treatment indefinitely.


What pushed you over the line to the place of writing this book?

The idea of anti-depressants as glorified placebos came to the forefront in the last five or 10 years in the media. I started noticing that my patients, even people who owed so much of their lives to anti-depressants, were asking, “Am I just responding to this because I believe it? Is this just a placebo effect?”

Then a friend of mine had a stroke, a kind for which antidepressants are sometimes given to make the recovery more robust. Even though it’s not really related to depression, the neurologist had held back on prescribing antidepressants in this case because of the general drumbeat of doubt about antidepressants.

I thought this was really extraordinary. Here we have one of the few things you can do that is easy and almost instantaneous to help recovery from a stroke. My jaw dropped to think this remedy was being thrown away based on a what I thought was a cultural, rather than a scientific response.

[Thanks to Kramer’s urging, the patient’s physician put him on a course of antidepressants. Soon, Kramer’s friend had regained movement and speech.]

And that was before I myself had really looked at the research to see what the flaws were. And it’s really bad. After having written the book, and speaking to readers, it seemed that I even underplayed it. When you see some of what’s wrong with some of these studies, there’s been an enormous amount of tolerance for flawed research by health and medicine professionals.


What’s an example of this egregiousness?

There’s one that Donald Klein called “the trout in the milk.” The phrase comes from when dairy farmers went on strike in 1849, and deliverymen were accused of watering down the milk. The saying originates because if you find a trout in the milk, it may be circumstantial evidence, but you can be pretty sure someone has watered it down.

In 1998, Irving Kirsch conducted a meta-analysis of different drug trials. He argued that studies that showed that antidepressants worked were probably due to the fact that the placebos [in the control groups] weren’t strong enough [to convince people they were on real medications]. He wanted to find studies that used placebos—non-antidepressant medications—with active side effects. So he used a study of medications, like a thyroid hormone and lithium, that had been used to augment antidepressants in people who hadn’t responded to antidepressants alone…But most people did respond to the combination of the two. If you divide the data that way, you end up having people who are actually on anti-depressants [plus the non-antidepressant “catalyst” medication] in the “placebo” group! That was the trout in the milk—if there are anti-depressants in the placebo group, you can be pretty sure the author of the study is up to something.

This author used that same study in three or four other papers. He has a new paper in 2016—almost 20 years later—based on the same misunderstanding of the study.


How has the reception of Ordinarily Well been so far, compared to Listening to Prozac?

Listening to Prozac was a cultural phenomenon. It hit the public at the right moment. This book is much more narrowly focused, so it hasn’t been that same type of response. But I have to say, when I wrote this book, I was really worried about bad reviews. There’s such willingness to oppose medicine on any basis. I’ve been just gratified that it hasn’t been the case. Even when people don’t agree, they’ve been very respectful. I’ve been worrying about how much I’ve had to go into technical issues—I had to be absolutely technical about what the evidence does and doesn’t say. But it’s full of narrative illustrations. It’s less a defense of anti-depressants than a defense of psychiatry as a humane undertaking.


How do you as a psychiatrist choose what anti-depressant to prescribe first? Are there ones that are your go-tos?

Every doctor has an opinion, and we don’t really know that one is better than the other. I do give a little rationale in the book for using one kind of anti-depressant for people who seem more slowed down and another for those who seem more agitated. But I think it’s very hard to name medicines.

Andrea Cipriani [the author of the Lancet article] has tried to rank order the anti-depressants in terms of their efficacy. For adolescents, Prozac came out the best by far, and Lexapro and Zoloft came out the best for adults. Those are all SSRIs. There’s certainly a case to be made too for non-SSRIS, like Effexor, and Wellbutrin certainly has its uses. Those five—Prozac, Lexapro, Zoloft, Effexor, Wellbutrin—at least some of those would probably make the list for any doctor. Prozac is probably not in the top for adults though.


Have you ever been on anti-depressants?



I find your chapter on the end-stage depression you witnessed in Boston 20 years ago really interesting. You write: “But if I am honest, there are reasons, beyond the numbers, that I would find it hard to believe that antidepressants do not work. Among them is an apparent decrease in the end-of-the-line depression that was familiar in my medical school years.” 

It seems to me that the nature of depression has changed—I’m talking about depression not accompanied by psychosis. On the wards in good hospitals where the diagnosis was presumably good—you would see people thin, no energy, repeating depressive sentiments whenever you spoke to them. Just suffering. Nowadays, anyone integrated into the medical system, there’s a chance they would be treated with medication [antidepressants] before they got to that stage. It’s been such an interest of mine—I thought, where are these people? I asked young medical residents and they said they’re really not seeing many of them. Lots of illnesses change form and frequency, so it’s not really knowable—it’s very hard to study—but I think some of it is due to the fact that we’re less reluctant to prescribe.

Sarah Beller is an associate editor of The Influence. You can follow her on Twitter: @SarahLBeller.

  • Sarah, why don’t you interview antidepressant critics as well, since your headline suggests an evenhanded approach? Or at least read up on it? Let’s say, David Healy or Robert Whittaker. Your questions are very soft. You don’t ask about iatrogenic harm, which can be hellish. (See for thousands of horror stories. could also give some insight.) I hope The Influence will someday deal with the problem of overprescribing these overhyped substances. To readers on SSRIs: How’s your sex life?