some awakening thoughts…

Posted on Thursday 18 August 2011

"The STAR*D trial is now complete. More than 100 articles have been published from this decade-long effort… As is often the case, however, such large research efforts raise many new questions while answering others. This commentary highlights those STAR*D findings that I believe have important implications for either present clinical practice or for future investigations"…
from Star-D: Lessons Learned and Future Implications, by Dr. A. John Rush
in Depression and Anxiety 28:521-524 [2011]

Only the first page of this article is available on-line. The whole article summarizes Rush’s conclusions and thoughts about future directions for further clinical research. I’ve already been perhaps too vocal about what I think about STAR*D, so I’ll forgo any detailed comments on this particular version. I thought instead I’d take a shot at summarizing the STAR*D view of human depression and clinical research:

Major Depressive Disorder  is a Unitary Symptom Complex. Its presence is defined by the DSM-IV Criteria and its intensity is measured by rating scales [HAM-D, ICDS, QIDS, BDI, etc.]
    This model is acausal in that it doesn’t matter what causes the symptoms. Likewise, the rationale used for grouping the symptoms as a unified entity isn’t considered.
The treatments for Major Depressive Disorder are empirically derived interventions such as medications, psychotherapies [CBT], electrical [VNS, TMS, etc.]
    While the mechanism of action of the various treatments is of interest as an aside, it’s the effect on the rating scales of the treatment that matters:
Treatment success in Major Depressive Disorder is determined by a statistically significant decrease in the rating scale values compared to placebo treated subjects, or when scale values return to the range seen in normal subjects

Clinical Research involves testing these treatments in Clinical Trials using various strategies [sequencing, augmentation, combining, etc.] to obtain the maximal response from the tested treatments 

The leading edge of Clinical Research involves predicting individual patient’s response to various known treatments – called "personalized medicine"
    "We have a history of biomarker studies but I think that .?. is very appropriately talking about trying to find some biosignatures for treatment response. And there I think what we have to start focusing on, not so much and I’ve been indirectly saying that throughout in my questions, is that we should not only focus on whether we can predict somebody’s bad outcome or good outcome, because that clinically doesn’t help. If you tell the patient, "bingo you have a bad disease," I don’t think they say, "wonderful doc I’m so glad you told me." So clinically that is a big problem. The second issue is that clinicians have to decide on treatments and therefore the best goal, interim goal at least, we may want to understand the pathophysiology better and I’m not against that, but I think we have to help clinicians decide on one versus the other treatment." Dr. Madhukar Trivedi at the Mayflower Conference

My heading, A Midsummer Night’s Dream, doesn’t refer to my opinion of the article quoted above by Dr. Rush, though maybe I do think a lot of it is fantasy based. It refers to an actual dream I must’ve had last night but don’t remember. I ran across the quoted article late last night, up trying to finally conquer lingering jet-lag. In my opinion, the article, true to form, way overvalues and over-interprets the results from the STAR*D study and reaches conclusions, many of which would usually be grist for the mill. When I awoke this morning, however, I found myself thinking about the article in a different way – not my usual grinding internal argument with the analysis or conclusions. I was thinking that these STAR*D reports are based on a way of thinking that is foreign to me. Thus my hypothesized dream. I stand by my endless complaints that the methodology and analysis is flawed, and my over-riding concern about the pharmaceutical industry’s intrusive agenda, but a lot of my angst about STAR*D and many other similar studies is a complaint about the whole system of thinking they represent [also also what’s missing]. Thus my attempt to define that system of thought.

The medical model of disease goes like this:
    What we as physicians attend to are signs and symptoms – things we see and hear about. We see these things as markers of an underlying cause, a disease. We do things [history, physical, tests] to intuit the underlying cause [diagnosis]. Treatment is aimed, if possible, at the underlying cause. In cases where the underlying cause is unknown, physicians use whatever data possible to group the signs and symptoms as if they had a unitary cause [a syndrome] and represented a disease. Research is directed towards further refining the descriptions and searching for a cause or causes. Symptomatic treatment is fine so long as it doesn’t mask the underlying cause or harm the patient. Many treatments are empirical – things found to be effective without knowing why. Other treatments are based on known medical science. There are corollaries such as preventive medicine etc.
This is an inelegant description of the medical model, but it gets the job done. We were taught this model in medical school [and teach it when we’re older] – it’s the science of medicine and it becomes an automatic reflexive way of thinking.

The STAR*D model represents a certain kind of medical thinking in situations where empiricism is the order of the day – paraphrased these days as "evidence-based medicine." When we don’t know a cause, we evaluate proposed treatments based on the measured results. We actually do the same thing when causality is known to test the effect in patients [we test an antibiotic that kills a bacteria in vitro to see if it works in vivo in infected patients]. In many medical specialties, a lot of the conditions have unknown causes [oncology, neurology, rheumatology, etc.]. In others like infectious disease or orthopedics, the specialty itself is defined by causes. Psychiatry is obviously in the first group – a specialty where most causality is unknown or speculative. I reject Dr. Insel’s redefining psychiatry as "clinical neuroscience." That’s his opinion [or wish], not a fact. I lived through a time when my former specialty, rheumatology, was seen as "clinical immunology." That was a fad, now long passed.

What my Midsummer Night’s dream must’ve been about is that the STAR*D way of thinking is just not my cup of tea. As a practicing physician, it’s incumbent on me to know as much about empirical medicine as possible. It’s my job to suggest and advise patients about the best known path for them to follow and warn them about the dangers along the way. A practicing doctor is the interface between medical science and the patient. But as a scientist, what interests me is causality and pathophysiology. I’m not conflicted about that difference. "Interest" just is what it is. A major force in my pique about modern psychiatry is that it seems to me to be only empirical, but my unknown dream told me that my "interest" was carrying too much weight in my mind. My complaint about modern psychiatry is that the pressure to understand causes, refine diagnoses, find the mechanisms of disease or treatments, seems relegated to the background. But empirical medicine is a perfectly legitimate part of medical science – applying scientific principles to validate outcomes.

That doesn’t change what I think about STAR*D. I think they’ve done a huge study on a heterogeneous diagnostic group pretending it’s a single disease. I can’t fathom how they came up with their sequencing scheme. They changed outcome parameters in mid-stream for questionable reasons and with-held the advertised parameters [which they have]. They pretended that the massive dropout rate didn’t matter, even though the subjects fled like rats on a sinking ship. It did matter. And they wrote a jillion papers in deceptive and jury-rigged ways. I honestly don’t think the study is interpretable, even if they were forthcoming. It’s lousy and distorted "empiricism." And I don’t think the cutting edge is predicting drug response with biomarkers. I see that as another piece of work driven by industry.

If that were all, I’d just say "bunk!" and move on. But that’s not what I’ve done. At my current rate, I’ll have as many blog posts about STAR*D as they have articles [that’s an exaggeration]. I think I’m using STAR*D as a way of expressing other things. My real complaint about modern psychiatry, besides the corrupting influence of industries of various flavors, is that it lacks the tension and curiosity about things unknown that drives medicine forward. For all the talk of neuroscience, what I mostly see in our literature is "product testing" and the application of any new technology of science that comes along to conditions without creative ideas and hypotheses driving the endeavor. I don’t know why people have Schizophrenia, and I want to know about that. There are thousands of similar questions, and not all of them have to do with neuroscience. I’m sure that people get depressed, have bad anxiety, or behave destructively because of things in their minds, or biographies, or lives – not in their brains. That seems self-evident to me. And I’m tired of having those of us who are interested in that causality discounted as kooks based on complaints about the early theories of Freud or the arrogance of our predecessors.

Right now, a large cohort of academic psychiatrists are leading us down a garden path. They know their science, but they’re not scientists in the sense that they’re not thinking deeply about what we don’t know. They’re tinkering with the same things over and over, more driven by method and a "race to market" [translational science] than scientific curiosity or creative thought. I expect that same criticism could and should be leveled at all flavors of psychiatry, including my own [and, in fact, modern medicine in general]. But based on my interpretation of my unremembered dream, I’m going to try to separate my frustration with what’s missing in modern psychiatry from complaints about how what’s present is being conducted. They’re just two different things…
  1.  
    Tom
    August 18, 2011 | 7:21 PM
     

    I’m sure that people get depressed, have bad anxiety, or behave destructively because of things in their minds, or biographies, or lives – not in their brains. That seems self-evident to me. And I’m tired of having those of us who are interested in that causality discounted as kooks based on complaints about the early theories of Freud or the arrogance of our predecessors.

    RIGHT ON! DAMN STRAIGHT! SPOT ON!

  2.  
    Patient 0
    August 18, 2011 | 11:36 PM
     

    Amen to this:

    “I’m sure that people get depressed, have bad anxiety, or behave destructively because of things in their minds, or biographies, or lives – not in their brains. That seems self-evident to me. And I’m tired of having those of us who are interested in that causality discounted as kooks based on complaints about the early theories of Freud or the arrogance of our predecessors.”

    We are creatures of our environment… And if I had an endless supply of cocaine, like Freud did, I’m sure I could come up with all sorts of theories and make everyone happy- until the cocaine runs out. When I was studying Psych2 I noticed they refuse to even tell the truth about Freud and what a damned failure he actually was OR the damage he did. All you ever hear about are a couple of BS sexual theories (fixated by the cocaine). I can’t even respect his theories, as most are drug driven bunk.

  3.  
    Autistic Lurker
    August 19, 2011 | 10:42 AM
     

    short time lurker and first time poster but I’d be interested in your opinion about the various genetic studies for MDD, Bipolar and Schizophrenia.

    A.L.

  4.  
    Peggi
    August 19, 2011 | 11:13 AM
     

    I do have questions about how some people can have similar life experiences – similar challenges, tragedies, whatever – and some have the resilience (or whatever you want to call it) to stand up to those challenges while other folks crumble. I wish I knew more about what it is that lets some people pick themselves up and dust themselves off while others can’t.

  5.  
    August 19, 2011 | 5:47 PM
     

    Peggi

    I’ll admit that I have trouble with the concept of “resiliency.” It’s a tautology. We say it’s something people have, yet it’s defined after the fact. Since the person came through a trauma seemingly unscathed, we hypothesize that they had resiliency, yet the resiliency was defined by their coming through unscathed. It just goes around in circles. In treating traumatized people, over time, the concept seems to fade. Some very damaged people appear resilient because they split off or dissociate from their inner injury. Some who appear to have little resiliency, actually were more traumatized than initially thought, often repeatedly. And there as many variations along that theme as ripples in a stream. It’s hard to talk about without long examples, but I haven’t seen cases where there’s some mysterious protective force operating that transcends explanation. So I wonder as you do, but experience has made me skeptical that the concept of resiliency has much value…

  6.  
    August 19, 2011 | 5:53 PM
     

    Autistic Lurker,

    It’s not an area of great expertise, but I believe that there are genetic factors in both Schizophrenia and Manic-Depressive Illness [bipolar]. But I have trouble with such ideas about “MDD.” I don’t think MDD is precisely defined and is likely a heterogeneous collection of a variety of forms of depression, some with genetic loading, others without. It’s my main beef with the DSM-III,IIIR, IV, and IVTR. MDD as it is currently formulated seems just too broad to study effectively without partitioning.

  7.  
    Autistic Lurker
    August 20, 2011 | 9:46 PM
     

    Thanks you very much for your answer.

    A.L.

  8.  
    August 24, 2011 | 9:45 AM
     

    Awakening thoughts: “My real complaint about modern psychiatry…”…

    I have grown fond of Psychiatrist-blogger 1 Boring Old Man. I generally agree with him on things, and I respect his efforts to be more up to date on the latest things than I am. From one of his Awakening Thoughts: My real complaint about modern ps…

Sorry, the comment form is closed at this time.