Tuesday, April 10, 2018

Sensational Antidepressant Article from the New York Times

Take some quotes taken out of context, the suggestion that doctors know less about the problem than the New York Times does, and the suggestion that you may be "addicted to antidepressants" and what do you have - the latest article on antidepressants by the New York Times.  Although the New York Times has never been an impressive resource of psychiatric advice they continue to play one and the latest article  Many People Taking Antidepressants Discover They Cannot Quit is a great example.

The reader is presented with numbers that seem to make the case "Some 15.5 million Americans have been taking the medications for at least five years. The rate has almost doubled since 2010, and more than tripled since 2000." and "Nearly 25 million adults, like Ms. Toline, have been on antidepressants for at least two years, a 60 percent increase since 2010."  Guaranteed to shock the average reader, especially in a culture that systematically discriminates against the treatment of mental illness.

Adding just a little perspective those figures translates to 15.5M/254M = 6.1% and 25M/254M = 10% of the adult population in the US.  Looking at the most recent epidemiological estimates of depression in the US 1990 - 2003 shows one year prevalences of 3.4 - 10.3% of the adult population.  The lifetime prevalences from some of those studies 9.9-17.1%.  It seems that the claims of antidepressant utilization may be overblown relative to the epidemiology of depression and the number of people disabled by it.  The authors go on to quote a study on the overutilization of antidepressants on data obtained from the National Health and Nutrition Examination Survey (NHANES) study.  These same authors have quoted an increase of antidepressant use of 10.4%.  This same study estimated a lifetime prevalence of depression of 9.5%.

Depression alone is not the sole indication for antidepressants. Anxiety disorders is another FDA approved indication.  Anxiety disorders can add an additional 3% 1 year prevalence and 5-6% lifetime prevalence.  About 16.5% of the population has headaches and antidepressants are used to treat headaches.  Another 6.9-10% of the population have painful neuropathies that are also an indication for antidepressant treatment.  Over a hundred million Americans have chronic back pain another indication for a specific antidepressant.  The main reference points to a study (3) that suggests only about 7.5% of antidepressants are prescribed for nonpsychiatric conditions.  Only 65.3% of the prescriptions were for "mood disorders. A study looking at antidepressant drug prescribing in primary care settings in Quebec Canada (5) provides specific data and concludes that  29.4% of all antidepressant prescriptions were not for depression or anxiety but for insomnia, pain, migraine, menopause, attention-deficit/ hyperactivity disorder, and digestive system disorders. Those same authors go on in a subsequent paper to provide a detailed analysis of the off-label use of those antidepressants.

The number of antidepressant prescriptions is far less drastic when taken in that context.  I am not arguing that every person with an eligible condition should be on antidepressants.  I am definitely saying that given the large numbers of people who will potentially benefit - the number of antidepressant prescriptions is not as outrageous as portrayed in the article.

What follows is a brief descriptions of antidepressant discontinuation symptoms and the fact that the medical profession doesn't know what to do about it.  This is certainly not the case in any setting where I have practiced. Discontinuation symptoms are well know to occur with SSRI and SNRI medications.  I routinely describe them and their varying intensity as part of the informed consent procedure when I prescribe these medications. The reality is that 20% of people will stop taking antidepressants in the first month after getting a prescription. Many will just get the prescription and never start.  An additional 20-30% will stop in the next 3-4 months.  Stopping antidepressants without medical guidance is so common that I routinely ask patients if they have abruptly stopped at any point when I am making any changes in their medications.  The majority have stopped without getting any of the discontinuation symptoms.  I qualify that by the fact that I have not prescribed paroxetine in 30 years because I considered it to be a problematic medication and I have a very low threshold for stopping antidepressants if I don't believe they are tolerated.  Even in their referenced study (2) the authors state: "In one national study, for example, only about one-quarter of adults initiating antidepressants for new episodes of depression continued to take their medications for 90 days...".  Does that sound like it is a medication that is difficult to stop?

They don't stop there.  After making it seem like we are in the midst of an antidepressant epidemic and that people are unable to stop antidepressants they make an even more absurd argument - doctors are unable to help patients get off antidepressants.  Before I go into their details consider this.  I work at a facility where we routinely detox people off high doses of the most addictive drugs in the world.  If we are able to do that, why would a doctor not be able to figure out how to discontinue a non-addictive antidepressant?  This specific statement really had me rolling my eyes:

"Yet the medical profession has no good answer for people struggling to stop taking the drugs — no scientifically backed guidelines, no means to determine who’s at highest risk, no way to tailor appropriate strategies to individuals."

Do I really need a study to do something that I have been doing successfully for 30 years?  Tapering people off of medications is something that every physician has to do.  Successfully using antidepressants means being able to taper and discontinue one and start another or taper and discontinue one while starting another or starting another and eventually tapering and discontinuing the original antidepressant.  That is not innovation - that is standard psychiatric practice.

I can only hope that the quotes from family physicians that follow were totally out of context.  Statements about "parking people on these drugs for convenience sake." and that the "state of the science is absolutely inadequate" are ludicrous.  I would say if you have to park somebody on a psychiatric drug or have questions about how it is used - it is time to send that patient to see a psychiatrist.  Nobody should ever be "parked" on a drug.

These physicians seem to have lost sight of the fact that they do not have similar problems prescribing equal amounts of antihypertensive medications and leaving people on them indefinitely.  There is no rhetoric about "parking" somebody on an antihypertensive medication or a cholesterol lowering drug or a medication for diabetes.  The fact that depression is the leading cause of disability in the world seems to be ignored.  The fact that up to 15% of people with depression die by suicide is not mentioned.  The suggestion is that this disabling and potentially fatal condition should not be addressed as rigorously as other chronic illnesses.

In the midst of all of the confusion created in this article, the authors fail to point out the likely cause of increased antidepressant prescriptions but they quote one psychiatrist who comes close.  He points out that the increase in antidepressants is due to primary care physicians prescribing them after brief appointments and (probably) not being able to follow the patient up as closely as a psychiatrist.  This was one of the main findings in the paper by Mojtabi and Olfson (2).  The specific quote "...the increase in long-term use (of antidepressants) was most evident among patients treated by general medical providers."

What is really going on here?  This blog has repeatedly pointed out that mental health care and treatment by psychiatrists has been rationed for about 30 years.  The result of that rationing is that there are few reasonable resources to treat all kinds of mental illnesses.  With that end result, the argument is now being made that we really don't have to build the infrastructure back up - we just need to shift the burden to primary care clinics.  In order to make it more simple for them we can just screen people with a rating scale for depression (PHQ-9) or anxiety (GAD-7) and treat either symptoms with a medication.  That way we can not only ration psychiatrists, but we can also ration psychologists and social workers who could possibly treat many of these patients with psychotherapy alone and no medication. For that matter, we could treat a lot of these patients with computerized psychotherapy - but managed care organizations will not.  State governments and managed care organizations will screen people, make a diagnosis based on a rating scale, and put that person on an antidepressant medication as fast as possible.

That is a recipe for high volume and very low quality work.  A significant number of those patients will not benefit from a medication because they do not have a compatible diagnosis.  A significant number will not benefit from the medication because it is not correctly prescribed.  In order to compensate for that inadequacy, a model of collaborative care exists that provides a psychiatric consultant to the primary care clinic.  That psychiatrist never has to directly see the patient.  The collaborative care model depends on putting patients on antidepressants as soon as possible and even more classes of psychiatric medication.

That is the real reason for increased antidepressant prescriptions and people taking them.  It is not because nobody knows how to prescribe them or stop them.  It is not because they are "addictive". It is because there is a lack of quality in the approach to diagnosing and treating depression in primary care settings and that is a direct result of federal and state governments and managed care organizations.

To be perfectly clear I will add a series of rules that will not question the current business and political rationing of mental health resources but will address the problem of antidepressant over prescribing and antidepressant discontinuation:

1.  Stop screening everyone in primary care clinics with rating scales - there is no evidence at a public health level that this approach is effective and it clearly exposes too many people to antidepressants and other medications.  I am actually more concerned about the addition of atypical antipsychotics to antidepressants for augmentation purposes when nobody is certain of the diagnosis or reason for an apparent lack of response and nobody knows how to diagnose the side effects of these medications.

2.  Provide any prospective antidepressant candidate with detailed information on antidepressant discontinuation syndrome - including the worse possible symptoms. While you are at it give them another sheet on serotonin syndrome as another complication of antidepressants.  It is called informed consent.  I encourage the New York Times not to write another article about serotonin syndrome.

3.  Triage depressed and anxious patients with therapists rather than rating scales - brief, focused counseling, CBTi for insomnia, and computerized psychotherapy all have demonstrated efficacy in addressing crisis situations and adjustment reactions that do not require medical treatment.

4.  Refer the difficult cases of discontinuation symptoms to psychiatrists who are used to treating it.

5.  Don't prescribe paroxetine or immediate release venlafaxine - both medications are well know to cause discontinuation symptoms and they are no longer necessary.

6.  Every physician who starts an antidepressant needs to have a plan to discontinue it - the idea that a patient needs to be on a medication "for the rest of their life" in a primary care setting is unrealistic.  If that determination is to be made - it should be made by an expert in maintenance antidepressant medications and not in a primary care clinic.

7.  Every patient should be encouraged to ask to see an expert if either their medication prescribing or treatment of depression is not satisfactory.  The standard for treating depression is complete remission of symptoms - not taking an antidepressant.  If you are still depressed - tell the primary care clinic that you want to see an expert.

In an ideal world, people with severe depression would be seen in specialty clinics for mood disorders, by psychiatric experts who could address every aspect of what they need.  That used to happen not so long ago.  It still happens in every other field of medicine.

But quality care like that is no longer an option if you have depression.

George Dawson, MD, DFAPA


1: Carey B, Gebeloff R. Many People Taking Antidepressants Discover They Cannot Quit. New York Times April 7, 2018.

2: Mojtabai R, Olfson M. National trends in long-term use of antidepressant medications: results from the U.S. National Health and Nutrition Examination Survey. J Clin Psychiatry. 2014 Feb;75(2):169-77. doi: 10.4088/JCP.13m08443. PubMed PMID: 24345349.

3: Mark TL. For what diagnoses are psychotropic medications being prescribed?: a nationally representative survey of physicians. CNS Drugs. 2010 Apr;24(4):319-26. doi: 10.2165/11533120-000000000-00000. PubMed PMID: 20297856.

4: van Hecke O, Austin SK, Khan RA, Smith BH, Torrance N. Neuropathic pain in the general population: a systematic review of epidemiological studies. Pain. 2014 Apr;155(4):654-62. doi: 10.1016/j.pain.2013.11.013. Epub 2013 Nov 26. Review. Erratum in: Pain. 2014 Sep;155(9):1907. PubMed PMID: 24291734.

5: Wong J, Motulsky A, Eguale T, Buckeridge DL, Abrahamowicz M, Tamblyn R.Treatment Indications for Antidepressants Prescribed in Primary Care in Quebec, Canada, 2006-2015. JAMA. 2016 May 24-31;315(20):2230-2. doi: 10.1001/jama.2016.3445. PubMed PMID: 27218634.

6: Wong J, Motulsky A, Abrahamowicz M, Eguale T, Buckeridge DL, Tamblyn R.Off-label indications for antidepressants in primary care: descriptive study of prescriptions from an indication based electronic prescribing system. BMJ. 2017 Feb 21;356:j603. doi: 10.1136/bmj.j603. PubMed PMID: 28228380.


  1. It's apparent that what has become the "community standard" is in fact the HMO model of psychiatry. Consider this article from Psychiatric Times that lists psychotherapy as option 7 when antidepressants don't work:


    I am also reluctant to give SSRIs to anyone under 25. The NNH/NNT ratios in that group are not good and that's a group that really could benefit from therapy first. Or at least getting the hell off all the nonsensical med combos they started in college.

  2. Agree - in Minnesota you are required to report PHQ-9 scores if you treat depression. They don't know what to do with that but it certainly reinforces the idea that a PHQ-9 is a diagnosis and treatment monitor all rolled into one.

    The majority of the mildly to moderately depressed would probably do better seeing a therapist or counselor first. We know from DSM-5 field trials that they will morph back and forth from depression to an anxiety disorder and a better pattern and persistentce may declare itslef (or not) at a distance from the crisis. But the HMO model also rations counselors and therapists - PHQ-9s are a lot cheaper.

  3. Who do you have to report it to? Medicaid?

    How reductionistic. I remember when psychiatry this dumbed down would have been universally ridiculed. But make no mistake, psychologists and social workers look at us (collectively, not nuanced thinkers like you) like we are dolts with one hammer always looking for a nail. Collabo-care is worse, the idea of not seeing a patient in consult and suggesting treatments would have gotten you thrown out of residency.

    There is also the insulting implication that the primary care doctor's MSE is as good as ours. This is not an ego thing, it's reality. My auscultation exam is not as good as theirs.

  4. Several years ago Benedict Carey wrote a wildly misleading article on an early intervention trial for psychotic disorder. I wrote a letter to the editor and the first author, and several weeks later noticed a multi paragraph correction buried in the back pages.

    1. That's the irony of trying to keep big platforms like the NYTimes or some of the larger antipsychiatry platforms honest.

      Once the distortions are out there - they are taken as the truth because of the reach of the platform and people ACT on this information.

      I learned today that a person I know decided to stop taking maintenance antidepressant medication that has worked for him for years based on "the New York Times said too many people were taking antidepressants so I am going to stop mine."

      If readers of this article were polled - my speculation is that many would have similar ideas and many would think that antidepressants were addictive.

      That makes these articles potentially more harmful than good.

    2. Keep up the good work by the way!

      What we lack in market and penetration we have to make up in individual effort.

    3. This article is most likely written by someone who is characterological, and uses characterological folk as example of all these allegedly heinous responses to meds. After 25 years of treating people, the ones who are the most dramatic and most demanding to allegedly be tapered off antidepressants for months if not over a year are Personality Disordered folk who just need the spotlight and have the tolerance for problems like a rabid dog with a thorn in its paw!

      I do like your comment about Paxil should not be a treatment option as of today. Between the withdrawal issues and the sexual dysfunction pervasive in men alone, what the hell is it good for until every other drug has been exhausted? And also on the mark, that physicians should NEVER say to a patient they will need to be on meds the rest of their lives. A moment of hideous candor from those who say that, maybe have the chutzpah to say "you'll need to see me the rest of your life, until I retire or die!"

      It is so sad how psychiatry has degraded into a profession that has abandoned most of the principles and beliefs that once made it something of value and importance.

  5. The fact that Serzone (and Merital, I am dating myself) was pulled and Paxil is still around and popular shows the abject failure of the FDA and the psych establishment.

    BTW, I've already seen enough ketamine patients to know the effect doesn't last. Yet that's the latest thing we are supposed to be excited about.

    1. Merital (nomifensine) was heavily promoted when I was a resident (PGY4). We had the largest boxes of free samples that I have ever seen.

      And then it just disappeared - apparently due to cases of hemolytic anemia and was subsequently withdrawn from other world markets suggesting it was significant. Would be interesting to go back and see what criteria were used considering that a drug in its class with seizurogenic potential (bupropion) was still approved and of course clozapine is monitored for leukopenia/agranulocytosis.

  6. I remember I had three or four patients on it and after it was pulled nothing else they tried worked as well. And let's not forget that clozapine was pulled then brought back.

    I remember thinking at the time of the Merital warning (1985?) that we were being treated as second rate MDs as if we incompetent to order labs. Even though we followed LFTs on other drugs and got Lithium levels routinely. And back then EKG screening was routine with TCAs.

    1. 1985 sounds about right.

      That same year I tried to get clozapine on a compassionate use protocol for one of my patients. I had to argue with a PhD from the pharmaceutical company about whether I could safely prescribe it or not. Of course that was before there was any standard protocol or pharmacies set up - but the attitude I got was not matter how much you know about the drug we are not going to give it to you on a compassionate protocol.

  7. The whole SSRI/SNRI movement has had the unintended side effect of making us psychiatrists (and generalists prescribing same meds) collectively lazy about fundamental principles of internal medicine. Late in my clinical career I have been much more persistent about measurement and biomarkers, including not only routine bloodwork EKGs etc but also psychometrics. My own n=1 biohacking has reinforced the necessity of measurement especially when everything is so subjective. I am also very interested in following vitamin D, glucose, insulin and CRP in patients. German psychiatric research is now focused more and more on inflammation/gut health/carbs/glutamate/mitochondria than trimonoamine hypotheses and this movement is just starting to get headway in the American publications.

    1. Minnesota has always been big on it - probably because on the inpatient side we had to do a lot of medicine. There is also a lot of collegiality in hosptials and frequent discussion with consultants. For decades I was able to discuss imaging studies, EEGs, labs, and specialty procedures. Very rich learning environment for medicine and many of them wanted to learn something about psychiatry. Still recall my first Vit D related case now about 25 years ago. Young patient with schizophrenia who claimed he drank milk, but came in with a hip fracture profoundly D deficient. Gave a grand rounds on inflammation about 15 years ago - very interested in cytokines since.

      Another element that added to the learning environemnt was MD lounge for lunches. MCO did away with it for a while. I always suspected their motive was to give the message that doctors were just employees like everyone else. Paying for the meals makes it equal but allows discussion with colleagues who you will only see in that environment and who you definitely learn from.

      Would not have missed that for the world.