Bad logic, bad science, and sensationalism

This is the kind of news release that angers me.

Brian Quinn, LCSW, PhD., Author Of A New Book On Bipolar Disorder, Discusses The Risks Of Antidepressants In Angry, Agitated Teens

It’s ill-conceived, illogical, and dangerous. It starts with the statement that “Dr. Quinn’s new book reviews the most up-to-date research on the use of antidepressants in bipolar disorder” and then proceeds to demonstrate that the author has little grasp of the most basic principles of logic, let alone the principles of research.  His statements are based neither in research nor in logic. They are self-serving and misleading. I get that he wants to sell his new book. I also get that using sensationalism and duplicity as a means to increase sales is perfectly acceptable to him. And that, in my view, is reprehensible.

“Eric Harris, one of the shooters at Columbine, was on an antidepressant when he and Dylan Klebold killed 12 students and then themselves. Jeff Weise, who killed nine people and then himself in Red Lake, Minnesota, was also on an antidepressant, and Kip Kinkel was on an antidepressant when he fatally shot his parents, two students and wounded dozens at a high school in Springfield, Oregon.”

Okay. And so? This tells us what? This is selective bigotry and stigmatization of mental illness at its very worst. “One of the shooters” at Columbine was on an antidepressant? And the implication is that caused him to kill people? What about the other shooter? What caused him to kill people? What about all the other historical and emotional characteristics of Eric Harris? Were those factors unrelated to the shootings? What about all of the other people who commit crimes, including mass or serial murder, every year who are not on any medication at all? What about the millions of people world-wide who are on antidepressants and other medications who will live their whole lives without ever killing anyone?

“Antidepressants have been linked to worsening hostility, suicidal behavior and psychosis in depressed children and teens and some people with bipolar disorder. While no one can say that antidepressants caused Harris, Weise, and Kinkel to commit mass murder, all these cases raise the question of who should be given antidepressants and who should not.”

No. This is false. Antidepressants have not been linked to hostility, suicidal behavior, or psychosis in children, teens, or anyone else, except by anecdotal evidence which is increasingly being exposed as hysteria. However, it is the case that hostility, suicidal behavior, and sometimes psychosis are among the symptoms of the illnesses and conditions that various classes of psychotropic medications, including antidepressants, are used to treat.

“‘These drugs tend to be handed out routinely to anyone with symptoms of depression and without a great deal of thought given to accurate diagnosis or the risks involved in using them, especially in young people with agitated depression,’ said Brian Quinn, LCSW, PhD., author of a new book, Wiley Concise Guides to Mental Health: Bipolar Disorder (John Wiley & Sons, 2007). ‘We don’t know what the diagnoses of these kids were, but we do know that they all had angry, agitated depressions — an indication they may have had bipolar disorder or could respond to antidepressants with a worsening of their symptoms. Caution should have dictated that they not be given antidepressants initially.'”

What absolute balderdash. These drugs are certainly NOT “handed out routinely to anyone with symptoms of depression and without a great deal of thought given to accurate diagnosis or the risks involved in using them”. The author is a social worker. He does not and cannot prescribe medications. What evidence is presented to back up the claim that physicians are ignoring diagnostic considerations in prescribing these medications?

“‘Psychiatrists, psychologists and social workers frequently misdiagnose those in the depressed phase of bipolar illness as having simple depression. This often results in them mistakenly being given antidepressants alone,” Quinn said.”

Okay. So people with bipolar disorder are often initially misdiagnosed with depression. Why is that, Mr. Quinn? I can tell you that it’s not a result of incompetence and it’s not a result of antidepressant medication. It’s a result of the nature of bipolar disorder – a fact you should understand as a self-styled expert on the subject. The majority of bipolar patients first present with depression, not hypomania, and the only way currently to accurately diagnose bipolar disorder is on the basis of longitudinal evidence showing recurrent cycles of depression coupled with evidence of at least one hypomanic or manic episode, evidence which is frequently not available when a patient is first being seen by a mental health professional. Given that, physicians do what they should be doing, viz., treating the symptoms of depression. That doesn’t mean that the possibility of bipolar disorder is ruled out. It simply means that physicians, like other men of science, follow the principle of Occam’s Razor: The simplest explanation is often the correct one and one does not proceed to a more complex explanation (diagnosis) of medical-psychiatric symptoms without first eliminating the simpler explanations (diagnoses).

This is the Dr. Phil phenomenon – the confusion of entertainment and marketing with science and professional practice. It is blatantly anti-science and anti-logic. And it worries me because of the potential damage it does to vulnerable people who need help and may be denied it if sensationalist propaganda such as this is believed.

We’re already seeing evidence of rising suicide rates among adolescents and children as a result of the fundamentally unfounded restrictions on the use of antidepressant medications among these age groups. We don’t need more biased sensationalism to make the situation worse.

antidepressants, bad logic, bad research, bad science, bipolar disorder, depression, sensationalism

11 Replies to “Bad logic, bad science, and sensationalism”

  1. If I were asserting that antidepressants were the proximal cause for the school shooters’ behavior, the writer’s criticisms would be valid and my press release would most certainly be unscientific and reprehensible. Perhaps I did not explain myself clearly. But it may also be that the writer’s anger clouded his understanding. As he noted, I wrote that “no one can say that antidepressants caused Harris, Weise, and Kinkel to commit mass murder.” My point, as I clearly stated, was that “these cases raise the question of who should be given antidepressants and who should not.” I’ll get back to this critical issue later.

    For the moment, I would like to point out that the writer makes assertions that reflect an ignorance of the research and the work of some of the most distinguished clinicians and researchers in the field of bipolar disorder. He says, for example, that “antidepressants have not been linked to hostility, suicidal behavior, or psychosis in children, teens, or anyone else, except by anecdotal evidence. . . ”

    In fact, antidepressants have been linked to the development or aggravation of agitattion, irritability, mixed states (mixtures of depressive and manic symptoms), suicidality and even psychosis in children, adolescents and young adults in a number of studies. One study is: Faedda, G. et al. (2004) Treatment-emergent mania in pediatric bipolar disorder. Journal of Affective Disorder, 82(1), 149-158. I cite the rest of the research in my book on bipolar disorder. But the writer (and interested readers) can also find the evidence for my claim in a number of texts. One of them is “Bipolar Disorders: Mixed states, rapid cycling and atypical forms” by Andreas Marneros and Fred Goodwin (author of the classic text on bipolar disorder titled Manic-Depressive Illness). Another is “Bipolar Psychopharmacotherapy: Caring for the patient” by Tohen and Akiskal.

    The writer states, “[Antidepressants]. . . are certainly NOT ‘handed out routinely to anyone with symptoms of depression and without a great deal of thought given to accurate diagnosis or the risks involved in using them’.” The fact is, antidepressants are routinely and mistakenly used as the intial treatment even in patients with known bipolar disorder. This, in spite of the fact that the current APA standard of care calls for the use of mood stabilizers in patients with mild to moderate bipolar depression. The writer should read “Bipolar Depression: A comprehensive guide” by Rif El-Mallakh and Nassir Ghaemi (one of the foremost authorities in the world on bipolar disorder).

    Finally, the writer states “the only way currently to accurately diagnose bipolar disorder is on the basis of longitudinal evidence showing recurrent cycles of depression coupled with evidence of at least one hypomanic or manic episode.” This is the position taken by the APA in the DSM-IV.

    The writer is apparently unaware of the literature on identifying individuals who have a bipolar diathesis and who are at risk for developing antidepressant-induced agitation and mixed states, even in the absence of a history of hypomania. According to Maneros and Goodwin and Ghaemi (in the books I cited above), many primary care physicians and psychiatrists are not aware of this literature either. For instance, depressed individuals without a history of hypomania but who have a family history of bipolar illness, an early onset of depression, or atypical symptoms of depression such as oversleeping are more likely to have a bipolar than unipolar illness.

    Antidepressants have helped millions of people world-wide lead happier, more productive lives and may well have played a significant role in the overall decline in suicide rates we have seen among young people over the last 15 years. We should not abandon their use. But we should not be using them rountinely in anyone who meets diagnostic criteria for depression. There is ample evidence that their use in young people with clear-cut yet often over-looked markers of a bipolar diathesis (other than a history of hypomania) has caused harm. We can do a better job of identifying kids ahead of time who are at risk of antidepressant-induced agitation, psychosis and suicidality.

    Brian Quinn, Ph.D.

  2. Brian,
    Yes, your clarifications are helpful, but I agree completely with David that your assertions here in this reply don’t appear to match your apparent mission: no antidepressants for bipolar illness. Antidepressants do work for bipolar illness, are often indicated, and, I do agree, should not, however, be first line choices in a “bipolar” presentation.

    But then the next question, not having read your book, is, how does one make the diagnosis of that mysterious, multifaceted and deceptive creature: Bipolar?

    One of the main reasons all are confused about bipolar illness, including the luminaries such as Fred Goodwin, a distant DC colleague, is that few actually understand the science behind the bipolar presentation. Bipolar illness, no matter how you slice and dice it, is a phenotypic diagnosis, not endophenotypic, and therefore is not biologically based. But then, so is most of the DSM4.

    I suggest that all the “bipolar luminaries” read Thomas Kuhn: The Structure of Scientific Revolutions, or Alfred Korzybski: Science and Sanity. Reductionistic thinking is now going mainstream.

    Biology should drive pharmacology – not appearances.

    We will never get the biology right if we only look at the surface symptoms, as many of the symptoms that are so “bipolar” swim downstream from a variety of brain biologic conditions such as gluten sensitivity, brain injury and estrogen dominance. It is interesting to me how many of these authorities eschew functional brain neuroimaging and consider it ineffective and insignificant [even though imaging findings fill most journals].

    It appears we all agree that evidence should prevail on the front burner. Why not look at brain evidence?

    I can testify directly to that point because I have asked some of those luminaries to visit our brain imaging center and look at our SPECT findings. -No visits, indeed no curious pokes.

    On a positive note, you do raise a good point, we should use antidepressants with more care, especially in volatile mood disordered individuals. But please refrain from the reductionisitic conclusion that antidepressants are bad medicine for those with bipolar illness.

    Good or bad is not the issue, as antidepressants are good for many. Proper sequencing of medications, is the priority.

    You do appear to add to the fervor, inappropriate in my view, that antidepressants are bad, that bipolar is under-diagnosed [when in truth it is the most common wastebasket diagnosis of the day], and that professionally we are wrong if we use antidepressants for those with mood disorder.

    Let’s get the polarized, indeed sensational, comments out of the discussion, look for proper sequencing of meds, and consider the fact that some with bipolar also suffer with ADD.

    If we can’t use antidepressants we most assuredly should not use stimulants with “bipolar.” Yet we all know that many do use both stimulants and antidepressants effectively and often. Interesting to me how different the “studies” are from reality in the office.

    David thanks for your comments, I am with you on this one.


  3. Charles,

    I did not state or imply that antidepressants should never be used in bipolar patients. What I said was that the young, angry school shooters who were put on antidepressants “raise the question of who should be given antidepressants and who should not.”

    I have two main points, backed by the research literature that I cite in my first reply, in my book, and that others have cited. The first is that patients with bipolar depression are routinely and frequently misdiagnosed with unipolar depression and then mistakenly given antidepressants. Ghaemi cites research showing that community psychiatrists misdiagnose bipolar depression as unipolar depression 60% of the time. He notes that the rate of misdiagnosis by primary care physicians is much higher. And the delay from the time of first symptoms to accurate diagnosis has been found to be an appalling 7 to 12 years.

    The second point is that even patients with clearly identified bipolar depression are given antidepressants much more often than mood stabilizers (study is cited in my book), even though the Amercian Psychiatric Association says that mood stabilizers are the standard of care for bipolar depression.

    A few other points: First,you and Dave are apparently not aware that the diagnosis of bipolar spectrum illness is not based on symptoms alone but on four lines of evidence first described by Robins and Guze in 1970 and elaborated on by clinician/researchers such as Akiskal and Ghaemi. These four lines of evidence are phenomenology (signs and symptoms), family history, course, and response to treatment.

    Second, yes antidepressants can be effective in acute bipolar depression (although not as frequently effective as in unipolar depression). Their use in bipolar depression, however, runs the risk of inducing agitated, dysphoric mania, even in patients on mood stabilizers. In addtion, antidepressants have never been shown to be effective or safe in long-term treatment. Two recent placebo controlled studies, for instance (one in the New England Journal of Medicine and one in the Archives of General Psychiatry) showed antidepressants were no more effective than mood stabilizers alone in preventing depressive relapse. In fact, antidepressants have been found to worsen the course of the illness in bipolar patients by hastening the onset of new depressive episodes (inducing rapid cycling).

    I must point out that neither you nor Dave cite any literature contradicting the points I have made. Dave’s reply, in particular, is a vile ad hominem argument.

    Brian Quinn, Ph.D.

  4. “Dave’s reply, in particular, is a vile ad hominem argument.”

    Actually, Brian, I approved your comment without a reply from me, deciding it was better to let my original post and your reply stand alone on their own merits.

  5. Brian,
    Thanks for your comprehensive and thoughtful reply. It does appear that we are very much on the same page in several respects:

    1. I completely agree that history, course, and response to treatment offer biologic implications, and share your concern and admonition that we should all think more frequently about these and any biologic markers on initial review. I am aware of those markers, and note, with you, how frequently they are overlooked. My only point in this context: we should more consistently begin to look at SPECT brain findings as significant and clinically useful for challenging diagnostic presentations – biology should drive pharmacology, I agree.

    2. I also agree with your note, and the findings of others, that many use antidepressants without careful overview of the biologic findings. -Very much with you on that point, and frequently write about that one myself. Antidepressants should not be chosen as first line agents in mood disorder. My only point: antidepressants do fit in as an augmentation strategy for mood disorder. This point, you will likely agree, is often derided by the many bipolar authors – sorry, I haven’t read your book! Again we agree. I have so often heard “experts” disagree with this augmentation point I have become a bit defensive.

    3. A final point, upon which we will likely have some differences, and not found in the literature, is the quite common finding of ADD with moods comorbid with depression. This oversight is pervasive and relevant to this conversation, and underlies my sensitivity to some of your careful considerations. Interestingly, ADD will also provide an atypical “cycling” picture with inappropriate, first line antidepressants, as the serotonin downregulates dopamine, aggravating impulsivity and cognitive abundance. ADD often shares many of the “biologic overtones” of bipolar – including multivariant thinking, cognitive anxiety and unmanageable cognitive abundance to name a few – and if ADD is properly identified in the first place, it will turn quickly around – stopping cold the cycling process. Unfortunately, the current diagnostic politic moves against ADD recognition, treats the mood rather than the underlying biologic PFC dysfunction, with mercurial results. Mood stabilizers don’t correct ADD, as you know. So many make the diagnosis of “Bipolar” simply by noting a paradoxical reaction to antidepressants – a singularly counterproductive and shallow approach. I am not asserting that you do so.

    On final review, it seems that you and I are not so polarized on bipolar as it first appeared. I regret and apologize for any misunderstandings in my comments without having read your book.

    I appreciate your points of clarification, and don’t see any contradiction in any of these remarks, only an opportunity to think more carefully about these complex presentations.


  6. Charles,

    A few points and then some questions.

    I say in my talks on bipolar disorder that the clinical diagnostic method in psychiatry, which relies on symptoms, family history, course and treatment response to make diagnoses, is kind of where general medicine was at in the late 1800s. In the absence of definitive biological tests, the best diagnosticians back then were the ones who knew symptoms, family history and course characteristics cold (i.e. diagnosis was based on more than just phenomenology or appearances). I then say that neuroimaging and the discovery of biological markers may make diagnosis more accurate. I’ll get back to this in a bit.

    The bipolar experts I know advocate the avoidance routine use of antidepressants in acute bipolar depression (not the absolute avoidance of them over the long haul), especially in those with agitated, mixed depressive states. This is based on the research evidence showing that antidepressant do not prevent depressive relapse in such patients better than mood stabilizers and anticonvulsants alone. In addition, antidepressants typically cause what these experts have termed a long-term “roughening” of the course of bipolar illness: more frequent depressions and treatment-refractory dysphoric mood states.

    Re ADHD: A 6-year prospective study showed that a little over a quarter of kids with ADHD later developed mania or hypomania (Tillman R and Geller B (2006) Controlled study of switching from attention-deficit hyperactivity disorder to a prepubertal and early adolescent bipolar disorder phenotype. J of Amer Acad Child Adol Psych, 42(12), 1486-1493). Mood stabilizers can work in ADHD-looking kids whose distractibility, hyperactivity and irritability are, in fact, due to a mood disorder.

    Questions: Does your experience with neuroimaging suggest that what we now label as ADHD combined with depression is more commonly at the root of your patient’s problems than bipolar disorder? Have we already arrived at the point where we can use neuroimaging studies to make valid and reliable diagnoses and guide treatment? I know little about neuroimaging but have the (perhaps mistaken) impression that we are still a ways away from that. Finally, antidepressant response is kind of non-specific, but in my experience, a pretty good indicator (although not diagnostic) of which adults may be bipolar. What kind of responses to antidepressants, other than jitteriness, and insomnia have you seen labelled as indicative of a bipolar diathesis that are not (in adults)?

    Brian Quinn

  7. Bryan,
    Yes, I am familiar with the literature and your several references – and have been interested in “symptoms and history: bipolar” since the mid 70’s. You appear to regularly assert that I am unfamiliar with the literature.

    I am only responding to all of this to add a dimension from the office trenches, informed experience, for this readership – the larger audience.

    I am not making the assertion of “more or less common,” regarding the ADD and depression diagnosis. I am only bringing this observation to the table as it is real, not imagined, and can support informed treatment. ADD, PFC dysregulation, can have significant anxiety and mood dysregulation that does “look” bipolar, but responds to stimulant meds – if the context of the history is compelling for ADD.

    We can see these comorbid conditions with SPECT scans, they are reliable, predictable, and useful – but only reveal brain function. Brain function is often different than DSM4 diagnosis – see the posts below.

    The last observation you made regarding “pretty good indicator” is the one that I posted the previous comments for. My opinion: we too often conclude that “bipolar” finding subsequent to amplified dysregulation, and miss several other comorbid diagnoses listed in the posts below.

    We agree:

    A different comorbid overlooked diagnosis visible on SPECT – TBI:

    SPECT in diagnosis:

    Beyond Bipolar – from a year ago:

    Something to think about –

  8. Charles,

    Well, the assertion is based on not having heard you (and especially Dave) cite the literature. Of course, that doesn’t necessarily mean you are not aware of it, so I apologize for making that assumption.

    In any case, I truly appreciate the info you provided and your perspective. I agree that we agree on the need to be more careful about jumping to conclusions on the basis of phenomenology alone. I think you are doing a great service by raising awareness of the need for careful differential diagnosis, especially when it comes to disorders conventional western medicine either fails to consider or too quickly dismisses.

    It would be interesting to see data comparing careful clinical diagnostic methods (looking at family history, course, and treatment response in addition to phenomenology) to neuroimaging results when it comes to diagnosis. I’ll have to look to see if anyone has done that study. Perhaps you know of such a study.

    Your points will certainly make me think even more carefully about differential diagnosis.


  9. Brian,
    While I especially appreciate the concept of “coalescence,” in pulling together challenging content, I am also hopeful that the process of dialog will arrive at sense of “equanimity.”

    This discussion here seems to find a reasonable ending on both content and process – and thanks for your kind remarks.


  10. A comment on the original post:

    We documented about 40% misdiagnosis of bipolar disorder as unipolar depression, AFTER correcting for such misdiagnosis occurring after the first manic episode, i.e., correcting for the course presentation in of depression before mania (which happens in about one-half of persons with bipolar disorder). (Ghaemi SN, Boiman EE, Goodwin FK. J Clin Psychiatry. 2000 Oct;61(10):804-8) So that critique in the original post has been scientifically disproven. Claims about science should be made after studying the science.

    As to Kuhn and philosophies of science, I fully agree on this matter and have discussed them at length in my book Concepts of Psychiatry.

    As to differences between practice and studies, errors can work in both directions. Don’t be so sure clinical experience is correct: For millenia clinicians strongly believed in bleeding, to be disproven by Pierre Louis’ “numerical method” research, which many clinicians in the 19th century decried as devaluing clinical experience. Some philosophers of science indeed believe that studies tend to be most informative when they refute, rather than confirm, inductive experience.

  11. Hi,

    I came across this site via a random search, and usually would keep my mouth shut about stuff like this, but felt very strongly motivated to leave a comment.

    I don’t expect what I say to change your mind. I was diagnosed with depression and when reading up on antidepressants, I came across similar hype about the dangers of medicating. And I thought it was mostly a bunch of misconceptions and ignorance. I thought that it made sense that people being treated for depression would be suicidal (duh) and didn’t connect that with the medications.

    …And then I took Effexor.

    And I don’t expect that you will believe me, but I wanted to write it here anyway.

    I was NOT suicidal before Effexor. But it sank me quickly and brutally into a more intense depression than I have ever experienced. To the point that there was concern that I would be a danger to myself. I knew that something was terribly wrong from the start, but I second-guessed myself, and didn’t think that it was really possible for antidepressants to make me more depressed. But having seen the dramatic change in me taking the medication vs. not taking it (the brutal level of depression resolved after coming off the medication). The change was not subtle. Not in the least. And it was perfectly timed with getting on and off of the antidepressant. I have also experienced extreme aggression that was COMPLETELY out of character for me. I think it is foolhardy to try to draw direct connections between the actions of those teens and the medications they were on without more information. But I can see how some people could, indeed, go that way as a result of the influence of their medications. Again, I don’t expect you to believe me, but this was not a mild impulse, or even close to the scale of anything I’ve experienced in my life. It was unprovoked, almost overwhelming violent rage from someone most people describe as “sweet.” Triggered by a woman in a grocery store who didn’t seem “polite enough”. It boggles my mind now that I could ever have a reaction like that, let alone with such a completely trivial provocation.

    I don’t advocate against antidepressants to treat depressive disorders. I think that they save lives, and are valuable tools. But I DO need to say that for a small subgroup of the population they can absolutely cause paradoxical reactions, and other unintended personality changes. I have lived it. And though I’m not sure that I would have believed me either, I couldn’t leave this here without sharing my story. I am a smart, educated, rational, scientifically-minded person. And I am not hysterical. Not that I begrudge you your opinion. I shared it myself before it happened to me.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.