Monday, January 11, 2010

Antidepressants Ineffective? Don't Believe It! (Part 2)

Last week, I posted part 1 of my reaction to the JAMA study showing that antidepressants don't beat placebo for mild to moderate depression. The many comments I received were intelligent and informed--reading them will give anybody an instant education in the limitations of antidepressant research.

It has been clear for years that antidepressants barely outperform sugar pills
, at least in the artificial environment of the double blind, placebo controlled study. It is also clear that most psychiatrists in the trenches are perplexed by these results, because in our practices we appear to see people responding very robustly to antidepressants--including those with only mild to moderate symptoms. So what gives?

I presented a hypothetical patient who became depressed because of financial woes, who saw a psychiatrist, received a prescription for Zoloft and Klonopin, and got better. Some commenters wondered why the patient was given medication rather than therapy. My answer is simply that in modern psychiatry, such a patient is more likely to be given meds, even though therapy may be just as effective. The reasons for this are complex, but have to do with how psychiatrists are trained (primarily, they are taught how to prescribe medications), how insurance companies reimburse (they pay twice as much for an hour's worth of medication visits than for an hour of therapy), and what patients want (many prefer the quick fix of medications over multiple therapy visits). I personally believe that psychiatrists should incorporate more therapy into their practices, but the current state of the field is as outlined above.

Anyway, I then considered what would happen if the same patient got enrolled in a randomized, placebo-controlled trial of an antidepressant. Once accepted, he is told that he will be randomly assigned to either active medication or placebo.

Let's assume that our patient lucks out and gets assigned to the latest SSRI, "Newexa," and let's further assume that Newexa works as well as all other SSRIs (like Zoloft or Lexapro). One would assume that since our patient is assigned to a drug that works as well as Zoloft, he stands the same chance of improvement. But not so.

When patients are enrolled in preliminary "open label" studies, in which everybody is given the active drug, and there is no blinding, the response rates are in the 70-80% range. But when patients are given the same drug in the context of a placebo-controlled trial, the response rates plummet to the 40-50% range. The reason for this is that if you strongly believe that you are taking an effective treatment, this confidence in itself boosts the effects of the drug. On the other hand, if you know that there is a 50% chance that you are taking a sugar pill, you have less confidence in the chances of a recovery. Essentially, by not telling patients what they are taking, researchers are turning down the volume of the placebo effect. But when I see a patient in my office, I do just the opposite. I try to maximize the placebo effect. I say things like, "I've had great success with Zoloft in patients like you....I would say you have an 80-90% chance of feeling much better within the next 2 weeks....These drugs are remarkably effective...."

So this is the crux of the issue. Response rates are always lower in randomized controlled trials than they are in the real world of the office, where our patients know exactly what they are taking. Thus, when I read about the measly outcome results of antidepressants in these trials, I take it with a big grain of salt. I interpret it in light of many other studies that have indeed shown antidepressants to be more effective than placebo in conditions as diverse as severe depression, anxiety disorders, and eating disorders. Rather than concluding that these meds don't work, I conclude that they do work, but that there is a large placebo component.

In the best of all worlds, we would be allowed to prescribe sugar pills to patients, and we would presumably get a robust response without many side effects. This is considered unethical by the AMA.
Nonetheless, according to this recent survey of 679 U.S. physicians, about half reported prescribing a placebo on a regular basis, most of which were "active" placebos--things like vitamins and over the counter pain relievers such as Tylenol.

The bottom line is that while medicine could not survive without placebo-controlled trials, they are rarely generalizable to the real world. I view their results more as signals of efficacy rather than as definitive answers.


Joseph Arpaia, MD said...

I agree with your explanation regarding the varying placebo effect.

However, people in the studies can also break the blind. Estimates are that 70-80% do, and this skews the effect. I wrote an article for the layperson about that for Psychology Today ~ July 2000 which was cut to a sidebar because, as the managing editor told me, the advertising department complained that they would be able to sell their drug ads if the full article ran.

As far as prescribing placebos, really that is most of what works with chronic disease. I believe it was Osler who said that the most important aspect of medical care was the faith the patient had in the physician. The power of suggestion is huge, and since anxiety and depression are affected by expectation, suggestion is particularly powerful in those disorders.

Since the patient expects a medication I often prescribe very small doses of an antidepressant and make it clear to them that over time these small doses are likely to be as effective as larger doses. This is the truth, and it helps the patient have the patience to do the mental skills training that I take them through (CBT, DBT, ACT whatever).

What really amazes me is when someone who has learned to change their behaviors, automatic thoughts and schemas tells me after 3 months "You know you were right about that small dose of Zoloft working." They attribute the healing power to the pill rather than to themselves. Why? I always advise them that they are also making important changes in thinking and behavior and to maintain those or the medication will stop working.

OK. Back to work. Thanks for what you do here. Real life is so much more interesting and complex than a clinical trial.

bigvince said...

Your belief in the treatment no one really wants to admit it may be the placebo effect at work on the observer that allows you to see an improvement that studies just do not show may be transferred to the patient. There is an excellent study on ADHD which demonstrated this merely being told that a child was being treated improved observer rating of the child. In any case evidence exist that vitamin d may improve depression and fish oil may also do the same . So why nouse those agents turn up the placebo effect and create side benefit instead of side effects

Gallucci, MD said...

Very good comments above! Including the citation of Sir William Osler. The power of placebo and suggestion needs further investigations...
Daniel, congratulations for this important post!

KIatherine Stone said...

This is great. Thank you so much for enlightening us. I'm sharing both parts 1 and 2 with my readers today!

Antidepressants aren't the only answer. But the rush by the media to state that they don't work every time one of these studies comes out is truly annoying.

Anonymous said...

Dr. Carlat,

First of all, if there was a study that showed that fish oil capsules didn't work for depression but clinicians claimed it did in real life, they would be accused of claiming anecdotal evidence and not practicing evidenced based medicine. How is what you are doing any different?

Do you have links to studies that show that antidepressants work for at least 5 years? I don't doubt they can work in the short term but so far, I haven't seen any effective evidence that they work long term. But I am willing to be proven wrong.

Also, since a successful response is based on a 50% reduction of symptoms, that make the success rate not look as impressive. How do you judge whether ADs are effective in your patients or not? Do you also take into account the side effects which might prevent them from staying on the meds and could add to the depression such as weight gain?

Additionally, if ADs are so successful, why are we seeing ads for stuff like Abilify which are claiming the drugs aren't successful and a booster is needed? Somebody is lying.


Anonymous said...

I must disagree. I have seen far less than robust responses in 20 years with these meds in clinical practice. In the short term they may provide relief. I have no idea if they are better than sugar pills but I believe they are for some short term. In the long hall they fail over and over again to relieve people as far as I see for almost everyone.I think this is wishful thinking on your part. I wish you were right.

Eric said...

"But when I see a patient in my office, I do just the opposite. I try to maximize the placebo effect. I say things like, "I've had great success with Zoloft in patients like you....I would say you have an 80-90% chance of feeling much better within the next 2 weeks....These drugs are remarkably effective...."

Hrmmm...I'm not entirely certain that this is ethical. I might be misunderstanding this but it sounds dishonest. A patient should be able to trust his or her psychiatrist completely. I for one hope that my psychiatrist hasn't been "hyping up" my prescriptions in order to help induce a placebo effect (thankfully I doubt this to be the case).

Anonymous said...

Dr. Arpaia: If you wonder why your patient attributes healing to the pill, you should check out a fascinating article published today in the NY Times entitled "The Americanization of Mental Illness."

Anonymous said...

I may be in error but I do not understand why you say the JAMA study reported no efficacy versus placebo for "mild and moderate" depression. Didn't the study also include "severe" depression, with no benefit from antidepressants? It was only the "Very Severe" depressed group that showed, at best, a modest treatment effect for antidepressants. Correct me if I am wrong.

Anonymous said...

There also seems to be a larger issue in this study (like many others with similar findings). Biological psychiatry seems predicated on a "bottom up" view of mental illness. In other words, from biology to psychology as the flow of causation for mental illness. It is the stuff of the widely disseminated (but hardly proven) "chemical imbalance" theory of depression (or anxiety, etc.)And the source of endless drug company ads on TV for pills. But when placebo is shown to be a powerful "treatment effect" the bottom-up view looks weak. In other words, the flow of causation seems to be equally strong for a reversal of causation -- psychology trumps biology. Just trying to add to the discussion about the one-sided view of biological psychiatry and the medical model.

Gallucci, MD said...

Dear Eric,
is not unethical neither dishonest to encourage your patient to take the correct steps to get better. The use of suggestion and information about your experience with medications is what we call ""therapeutic alliance". It is basically used to improve adherence to treatment and to instill hope. During a clinical interview psychiatrists use this strategies for a better rapport.

Joseph Arpaia, MD said...

Anonymous - thanks for the lead on the NYT article. It is fascinating. My impression of the DSM is that it is not really different from the texts on demonic possesion in the middle ages. Everything in it is phenomenological and it was written by adherents of one faith.

Eric - If my patients have responded to Zoloft then I am telling the truth if I tell that to my patient, and that is ethical. If they have not responded to Zoloft and I make that up for my patient, then I am lying, and that is not ethical.

Note the self-reinforcing cycle here. Belief enhances treatment. And once the physician believes in a treatment, then the belief in that treatment is conveyed to the patient, making the treatment more effective.

As a physician I try to find treatments I can believe in that cause as little disruption to the patient's life as possible.

Note that this is not restricted to medications. If a therapist believes that once a week therapy for years is needed, then that belief will be communicated to the patient, who will then come to therapy weekly for years.

We can also foster beliefs without doing so consciously. For example, by scheduling someone for weekly therapy appointments early in treatment we may foster the belief that therapy needs to be weekly. I learned that and now routinely tell my patients that we will meet less frequently as they learn how to manage their condition (my belief system there).

Kudos to you Dr. Carlat for getting such a discussion going. The questions are so often more stimulating than the answers.

Anonymous said...

Wow, interesting how a doctor on here thinks that its ethical to convince his patients that a drug is solving his problems ("maximize the placebo effect"), when in fact, he has worked them out on his own. Wouldn't it make more sense to explain the limited efficacy of antidepressant, and tell him that he is largely in control of his psychological well being, rather than making him feel that he is powerless, and that there is something fundamentally wrong with his brain chemistry?

Sarah said...

"My answer is simply that in modern psychiatry, such a patient is more likely to be given meds, even though therapy may be just as effective."

Ouch! If we assume that the patient's depression was a maladaptive response to a life trigger, then... Meds mediate the response in this one situation. I would hope that therapy would teach the patient to mediate his own response to bad situations, should a situation like this ever occur again. To me, then, therapy should be _more_ effective, not _just as effective_. This doesn't say much for therapy.

"But when I see a patient in my office... I try to maximize the placebo effect."

Well, this doesn't really say anything about the efficacy of SSRIs versus any other drug. Just the efficacy of drugs in general. I mean, you could also take an MAOI and say, "In your situation, i really think this would help you."

You're arguing that these drugs are effective based on your ability to maximize the placebo effect. I would be more comfortable with this if you were prescribing an actual placebo, rather than drugs with physiological effects that are sometimes detrimental. (Echoing what bigvince said, in a glass-half-empty sort of way.)

On another note, your explanation is fascinating to me on a personal level as it offers a possible explanation for why, ten years ago as a young adult, i was switched from one SSRI/SNRI to another and only ever put on a mild to moderate dose of anything (I feel like the doctors should have figured out that strategy wasn't working by the fourth antidepressant, though). It was a disheartening two-year experience that convinced me, at the time, that nothing could help me. Just to illustrate the flip side of what you've described.

I'm not sure whether this blog is targeted primarily at mental health professionals; if so, well, i'm just some schmoe and you can take my comments for what they are! =)

Anonymous said...

From today's NYT

SteveM said...

Re: Anonymous and the NY Times:

See this incisive rebuttal to Friedman by psychiatrist "Mark from Providence" (comment #28)

Apart from indication and efficacy, a huge problem with the debate is the evasion by the strongest proponents of expansive psychopharm use of the very real and sometimes very wicked side effect profiles of anti-depressants.

Perhaps the biggest travesty of current psychiatric practice was the publication of Peter Kramer's Listening to Prozac. and the advent of DTC advertising. Because Kramer presented psychopharm as benign pixie dust with the potential for "cosmetic psycho-pharmacology". And the general public bought into his biased mantra.

But Kramer's ego prevents him from backing off that premature and evasive oversell of what these drugs actually do. He still waxes rhapsodic about psycho-pharm as some kind of panacea and never addresses the issues of lousy side effects and debilitating withdrawal syndrome. Kramer is smug as a bug in rug in Providence collecting royalties and acting illegitimately wizened for the New York Times.

Unfortunately, the Kramer's and the Friedman's of psychiatry have the ears of the elitist cognoscenti who don't ask the right questions. And that is exacerbated by the old boy psychiatry network (excepting a few) that will not call a spade a spade.

Anonymous said...

I love your blog. But regarding the placebo effect, and why you feel placebo-controlled studies do not translate well into "real world" practice, and your telling patients that they have an "80 to 90 percent" chance of responding to medication: Didn't the STAR*D study not contain a placebo or control group? And weren't response rates to medication about 30 percent?????

Anonymous said...

By the way, if antidepressants are so effective, how is it that Astra Zeneca gets away with its Abilify commercials, which basically state that "2/3 of depressed patients taking antidepressants" don't respond??? Just curious.

Daniel Carlat, M.D. said...

There are too many great comments for me to touch on all of them. But I will clarify something about the STAR-D trial, which continues to get a lot of undeserved airplay, probably because it was the largest clinical trials ever done in psychiatry. You often hear that only a "third" of patients got better with Celexa, and this is thought to be further evidence that existing antidepressants don't work very well.

Over 4000 patients with depression were enrolled and all of them were put on Celexa, open label (no blinding, no placebo). 47% of these patients responded, and 30% achieved remission. So it is more accurate to say that half of these patients responded to Celexa.

This is still lower than the 70-80% response rate seen in some other open trials of antidepressants, and the reason is that the STAR-D patients were unusually ill people. The average patient in the trial had 6 prior episodes of depression and had been depressed continuously for an average of 2 years. These are the hardest to treat patients, and I was amazed that they actually did as well as they did.

As always, one can use these figures in various ways to advance one's own interest. Thus, if BMS is pushing Abilify by claiming that 2/3rds of patients don't respond to existing antidepressants (I'm not certain of this, since I haven't seen the ad), then the company is being deceptive by picking out one isolated piece of data and failing to put it into context.

Jai said...

You seem to think it is comical to give patients drugs that don't really work, which i guess it would be if they didn't have such devastating side effects.

When you only see a patient for an hour a week, in an controlled environment, you can't see just how much damage the antidepressants are doing. All you can do is trust what the patient tells you. The problem is the patient now has flies in their eyes, which is why they can't see them.

I have seen the lives of my closest friends have been ruined by antidepressant medication, while their doctors and they themselves refuse to accept they have made things worse. As they spiralled out of control of their own lives the answer given was always to simply increase the dosage or try something stronger. A vicious circle!

Only after they gave up the drug were they able to have the motivation to changes things for the better.

I urge all doctors to consider this when prescribing these drugs, the psychological side effects are very difficult to quantify in drug trials and far from well understood.

Anonymous said...

Thanks Dr.C for your comments. BTW, I stand corrected -- I meant to say BMS and not AZ for the Abilify ads. But I can't believe you haven't seen the ads though. Don't you have cable TV? The ads run about 100 times a day! LOL.

bigvince said...

Results from the OMEGA-3D a interesting piece on omega 3 oils and depression ...,b6hFvj8q

"Fish Oil Rivals Antidepressants in Clinical Trial
Clinical findings affirm earlier indications and further justify strong omega-3 support from the American Psychiatric Association "
The results of the largest-ever clinical trial found that omega-3 fish oil may significantly benefit half of all people diagnosed with depression.

Specifically, fish oil seemed to help the 50 percent of depression patients who are free from diagnosed anxiety disorders.......'"

It is an interesting study which brings us to the first question why not opt for safer alternatives; that do exist instead of the using drugs that have a much less desireble risk benefit profile. From the above link
...But among those diagnosed with major depression ─ but not anxiety ─ the patients who took omega-3 EPA had significantly better scores than the placebo group.

As lead author Francois Lesperance, M.D., told Medscape Psychiatry. "... the level of improvement we saw in this subgroup is on a par with what has typically been reported with pharmacologic treatments." .....

Eric said...


Call it what you will, be it "therapeutic alliance" or some other name, but I am not convinced that this is acceptable practice.

Of course, I agree that there is definitely a practical benefit to the "placebo effect"--if you think that what you are doing is making you better your brain effectively tells you that you ARE doing better, which in itself isn't a bad thing. Whatever works, right?

Except--this could be disastorous for people with "severe" mental illness types--be it depression or OCD or any other example.

Admittedly, this post details the fact that severely depressed people benefited from antidepressants the most. Still, whatever one's condition, I believe a psychiatrist owes his patient to be completely straight-forward in order to establish an important bond of trust.

Neuroskeptic said...

"You seem to think it is comical to give patients drugs that don't really work, which i guess it would be if they didn't have such devastating side effects."

But they don't. Antipsychotics have devastating side effects, SSRIs by comparison are gentleness itself, don't abuse the term "devastating side effects" by applying it to antidepressants, please.

Anonymous said...

You started by saying that someone who's lost much of their savings, who would understandably feel anxious and depressed and likely be processing stages of grief over the event, could go to a "psychiatrist" and be told to take zoloft and klonopin and feel better.

(Klonopin is not intended for use longer than 14 days. Do you ever see this short term use actually happen? Of course not, it becomes a long term burden.)

Yes, perhaps he would temporarily feel better if given drugs. But ...

A very real event ~ losing security ~ has happened to someone. He would be abnormal if he had no reaction to that event, correct? There are "diagnoses" that can be applied to people who have no emotional responses to things. But this person is normal and does respond normally, with unpleasant feelings. So, right off the bat, ANYONE is damned if they do and damned if they don't. Do you feel? Take these drugs so you won't. Do you not feel? Take these drugs so maybe you will. THAT is crazy.

So, he takes the drugs and initially feels a little better ... because it impairs his ability to feel anything at all. NOT because of any genuine healing process. And what happens a few months or years into that disaster of a treatment plan?

That person would have had their normal and very appropriate grief process curtailed and avoided. In addition, they would be suffering the effects of numbed emotions and impaired cognitive function in the other parts of their lives, not just over the original issue of a very real loss.

And finally, there will come a day that the person must either get off the drugs or continue dying slowly from them. And that's when a "dr" such as yourself will pronounce them an "addict" (at least with the klonopin) and "prescribe" a very fast taper schedule. Or even worse, order them to a "rehab" so the "professionals" can do the job. When this goes wrong ~ and it will ~ another, stronger drug will be given. Perhaps a bit of Seroquel so the withdrawal insomnia isn't too bad (prompting another cycle of drugging and loss and withdrawal). Maybe the person will be very lucky and will survive this with only a year or so of recovery to get through to get back to normal ... the normal he would have been had he never been drugged in the first place. If not, they must taper more slowly ... sometimes for years, to withdraw from these drugs. Or worse, be given more and more "diagnoses" to justify more and more brain damaging drugs.

If the person ever does get away from these drugs, when they go through the fire of withdrawal from these poisons and come out the other side, they may just be grateful for having made it out alive. The loss of their retirement now means nothing because they almost truly lost their life to psychiatric drugging.

But the original loss is still there, compounded by years of more losses caused by the "treatment."

To me, it seems like "psychiatry" is the same as stomping on someone's foot when they say they have a headache ... "There, that will make you forget about the headache!"

We injure people more so they forget about their original injury. And we heap on more and more injury to cover the previous ones ... And we keep doing it, over and over. It's definitely a vicious circle ... done by those who believe they are doing something good and helpful and necessary.

Stuart Kelter, Psy.D. said...

Putting aside for a moment the question of how to ethically harness the placebo effect in clinical practice, I applaud Dr. Carlat's and other prescribers' honesty in admitting to themselves that this is what they are doing. As a prescribing psychologist I am keenly aware that it is not so easy to substitute just any placebo as an antidepressant, given that marketing and publicity about antidepressants as a class of drugs are among the strongest contributors to the (placebo) response. In no small measure, the pharmaceutical industry has created their own efficacy, and not in the laboratory. For a wonderful review of this topic and of the history of the placebo response, see "Placebos Are Getting More Effective. Drugmakers Are Desperate to Know Why,"
which recently appeared in Wired Magazine:

Stuart Kelter, Psy.D.
Prescribing Psychologist
Las Cruces, New Mexico

bigvince said...

If it is true that marketing creates the placebo effect in the patient than the effectiveness should be much greater in the the United States than in most other countries. Why because of DTC advertising which prohibited in most other countries. I do not believe the data would support that. That leaves us with the placebo effect being created in the eye of the practitioner where marketing to providers is more on an even footing.

Anonymous said...

I truly enjoy your blog and believe that you are an ethical person. However, I agree with another person who commented here that if you are using deception to increase placebo effect, you likely are doing something unethical. I already don't trust my psychiatrist because he is on several speaker's bureaus. If I found out he was inflating medication success rates in hopes of triggering a placebo response, I would fire him so fast, he wouldn't know what hit him. And then I'd be sure to use word-of-mouth to let others know about my experience (both in-person and online). I expect accurate information from my physicians. Honest and accurate. No lying. No inflating. No dodging of questions. EVER. If your stuff doesn't work, don't prescribe it. The profession ought to consider going back to a time when the psychosocial was an important component of the psychiatric interview. My psychiatrist doesn't even have tissues in his office. Use your sleeve if you need to cry, but avoiding the crying part is preferable. It slows down the assembly line. And I cannot remember when he last displayed a shred of interest in, or asked about, any life-related occurrences that may have exacerbated what he just medicated. This is a problem for your profession. Your stuff doesn't work, but nobody knows how to do psychotherapy anymore. It seems to me that's a great way to drive yourselves right out of business. I don't know anyone who sees a psychiatrist who actually WANTS to be on medication and I know that all of the people I know who admit to seeing a psychiatrist would appreciate it if their doctors showed at least a remote interest in what is going on in their lives. We can dream...

Anonymous said...

My simple wish is for the psychiatric community to STOP prescribing antidepressants to fragile people who may have a short term bout of the blues. You've NO IDEA how many very capable people can't withdraw from these antidepressants. They have lost confidence in themselves and fear withdrawl. Open your eyes and see the truth: YOU TREAT THEM LIKE CONSUMERS, BUT SOMETIMES THEY JUST NEED SOMEONE TO LISTEN. YOU HAVE FAILED MANY, JUST LOOK BACK AT YOUR PATIENTS AND YOU WILL SEE...PLEASE, HAVE A HEART.

Ana said...

I agree with Anonymous above and I'm amazed that after all the evidence of SSRIs hideous side effects, and facts like people being unable to get out of the drug do to withdrawal symptoms; the use of SSRIs to "treat" sex offenders; birth defects caused by SSRIs and the simple fact that this theory is WRONG!
But the American government is doing clinical trials in people 7 - 17 years old to test Cymbalta and Prozac. They started in 2009 and will only end in 2011.
You can keep on prescribing.
Now I know why psychiatrists say:
"It's all in your head!" "My other patients don't have this problem."
I would love to see how this chapter of the Psychiatry History will be told in 3.034.
Take a good look at what you are doing.