NIMH: Director’s Blog
by Tom Insel
May 15, 2015
…It’s true that most of the neuroscience and genomics findings are not yet actionable for psychiatry. No one doubts that the brain is the organ of affect and behavior, but no one can point to a biomarker that is essential for clinical practice. In the short-run, we may do much more to bend the curve on suicide mortality by changing public policy [such as through restricting access to means] rather than finding a biomarker for suicidality. But in the long-run, and we need a long-run strategy, policy will hit a wall and we will need better diagnostics and therapeutics. That is where this new initiative can make a difference. The research of 2015 suggests that the clinician of 2025 and certainly the clinician of 2035 will need to know about cortical dynamics, neural networks, and genomic variation. Those entering the field today will need to know how to think about the brain and how to critique brain science. By changing the training of the next generation, we not only prepare for the future, we create it.
|
Genomic Biomarker for Suicidality Found!
Independent Georgia researcher, 1boringoldman, working closely with the CDCwebsite discovers clear evidence of…
|
[No, I haven’t gotten any calls from the Karolinska Institutet in Stockholm yet, but hope springs eternal. Should I perhaps add waiting room screening for the Y Chromosome?].
As an Internist, I never heard the term "biomarkers" or any of its synonyms. And yet they were part of everyday medical life. They were the tests we ordered to make or confirm diagnoses; the parameters we followed to monitor treatment; and the screening tests we collected to detect risk or early signs of disease. In Psychiatry, there weren’t any. Biomarkers were the thing that we didn’t have. Actually, by the time I showed up, we did have a couple in Melancholic Depression, Dr. Carroll’s Dexamethasone Suppression Test and Dr. Kupfer’s REM Sleep Latency studies. As is often also the case in physical medicine, neither was pathognomonic. But they stacked up well against many of our medical colleagues’ biological tests – particularly in combination. Then, in the most peculiar of moves, we discarded the diagnosis they marked [Melancholic Depression], and that was that for our brief biomarker years [until, of course, my recent breakthrough above!]. So we rely on the more subjective rating scales like the HAM-D or the PANSS.
Psychiatry’s greatest internal critic, Thomas Szasz, actually fueled the modern psychiatric obsession with biomarkers. Szasz argued that Mental Illness was a false construct – a Myth.
"To be a true disease, the entity must first, somehow be capable of being approached, measured, or tested in scientific fashion. Second, to be confirmed as a disease, a condition must demonstrate pathology at the cellular or molecular level."
In spite of the fact that even the diseases in medicine long antedated the biomarkers that they became associated with, the body psychiatric seemed to accept Szasz’s argument as a gold standard. The neoKraepelinians made one of their tenets negating Szasz, but then in another set about the business of searching for biomarkers Szasz’s definition required:
5. There are discrete mental illnesses. They are not myths, and there are many of them.
6. The focus of psychiatric physicians should be on the biological aspects of illness.
And Robert Spitzer’s claim of the DSM-III being "Atheoretical" …
Undoubtedly, with time, some of the disorders of unknown etiology will be found to have specific biological etiologies, others to have specific psychological causes, and still others to result mainly from a particular interplay of psychological, social, and biological factors…
… was soon forgotten in the biomedical transformation of psychiatry that followed him. So we now have the paradox of certain modern biopsychiatrists who speak of Mental Illness broadly as Brain Disease [as in biological] insisting on validation by biomarkers [as Szasz – and now Insel et al] but there aren’t any. For them, the search for biomarkers becomes, first and foremost, a quest for legitimacy. To add to the parodox, they also assume something like the universality of biomarkers [as in Tom Insel’s fantasy of a biomarker for suicidality or Nemeroff’s even more fanciful biomarker for PTSD]. And then there’s Insel’s newest scheme [RDoC], as in bring the diagnoses to the biomarker, rather than the other way around.
Okay. You win. I don’t think my Suicide Biomarker is a breakthrough. And I don’t think Thomas Szasz was a very nice man [he thought a lot of the patients were Malingerers who didn’t want to take responsibility for their lives]. I don’t think Insel is a clinician. And I kind of think the NNCI is forced at best. I watched one of their videos about talking to the patient using neuroscience talk. An obviously competent resident was illustrating talking to a anxious young woman who had a Heroin Relapse when her boyfriend broke up with her. He started with a [downloadable] brain picture and explained about the brain parts involved in "a circuit" [the Ventral Tegmental Area, the Nucleus Accumbens, the Orbital Frontal Cortex, the Prefrontal Cortex, the Amygdala, and the Hippocampus which he referred to as V+N, O, P, A, and H respectively] explaining each one’s part in the relapse and where and how different modalities of treatment worked, making an analogy to a car [the typing part is mine]. She got to take the completed diagram home as a reminder [click image to enlarge].
I liked the resident, and thought he did a reasonably credible job in his presentation. But I hardly think it prepared residents for the coming decades, and it came too close to the dumbed-down-ness of those "chemical imbalance" explanations of yore. This was part of the accompanying case presentation:
Today, she presents to your outpatient clinic feeling anxious and demanding that you prescribe alprazolam to her because she has heard that it helps with feeling nervous. A urine drug screen is performed in the clinic and returns positive for heroin. The resident confronts her with the results, and she breaks down crying about how she recently broke up with her boyfriend, happened to drive by her former drug dealer’s home and relapsed. She expresses to the resident how guilty she feels about using again after working so hard to stay clean for 5 years.
You can watch the video here. He made the point that explaining things this way helped him to not shame the patient [a good point]. But I sure wish he’d mentioned the part in red and commented on it. If she actually makes it to that NA meeting, they’ll do that for sure with laughter and rolling eyes. I know that because I played this video for an addict in long term recovery whose reaction was pretty negative ["
No addict is going to listen to all of that!"]. But I’m not here to judge.
"If you’re not here to judge, 1boringoldman, why are you here?" said the voice from the corner of the room. It’s because I’m an old training director, and while I agree that we learn from watching others, we also learn by watching ourselves. I wanted to ask Chris [the resident in the video] what he thought. Do we know enough to really say those things about the brain? Did the brain "prop" help? Will this young lady feel talked down to? Will she feel heard? Does this much bio take us away from the psycho·social? An alternative…
"One thing any addict whose gotten clean knows is that the Heroin solution to pain is seared deep into the brain for all times. It’s like a reflex, whether it has been 5 days or 5 years. The only tools you’ve got in the face of that reflex is what you’ve learned in your mind and what others can do to help you. So when you heard yourself saying that you "happened to drive by [your] former drug dealer’s home," you were already in deep do-do. Did you really believe that? You needed to be driving by your nearest NA meeting, or a Sponsor’s house, or here. Did that occur to you? I have a feeling that with 5 years clean, you know all of that. Here’s what we can do about the relapse, «blah, blah, blah». But maybe right now, the thing we really ought to talk about is the breakup with this boyfriend. You’ve been in a whole lot of hurt, and it obviously threw you for quite a loop. So you just said the hell with it and did something really pretty self-destructive. Tell me about all of that…"
I’m well aware that Chris is making a training video to show how one might use neuroscience metaphors in explaining things to a patient, and that his natural behavior in real life might have been quite different. But my point remains. The NNCI proposes to "help residents incorporate relevant neuroscience into their own emerging clinical ‘voice’." That’s okay with me, when appropriate. This young woman isn’t in need of knowing how her brain works, or how treatment for her Heroin relapse works. She’s needs someone who addresses her obviously painful life crisis and who finds out if her relapse was a suicidal equivalent. And I expect that I might have that same objection to many such clinical vignettes, that the brain talk might well get in the way of the talk that matters.
Your discovery just goes to show how the Y-chromosome messes life up. Your graph belongs with some of these classics: http://tinyurl.com/os78px2
I enjoyed the Suicide Chart immensely, though actually, it was homicide that was on my mind this morning– and not for the usual reasons. (My commute is pretty easy these days, and my road rage has improved a great deal.)
Funny you mentioned Karolinska, Dr. N– that’s where Jari Tiihonen is from, the author of a study of the “link” between psychotropic drugs and homicide that got plastered all over the Internet this morning– Drudge Report, Google News, Science Daily, CBS Atlanta– man those folks in Finland sure have one hell of an advance team, huh?
And– shocking news! The antidepressants have been exonerated– again! In fact, the nasty drugs that are correlated with suicide all seem to be older drugs that are available in generic form.
Well, not that shocking. Let’s look at Dr. Tihonen’s COI from some of his other work:
http://www.ahrp.org/cms/content/view/615/110/
“The principle investigator, Dr. Tiihonen, has served as a consultant to Lundbeck, Organon, Janssen-Cilag, EH Lilly, and Bristol-Myers Squibb, and has received fees for giving expert opinions to Bristol-Myers Squibb and GlaxoSrnithKlme, and lecture fees from Janssen-Cilag, Bristol Myers-Squibb, Eli Lilly, Pfizer, Lundbeck, GlaxoSmithKline, and Astra Zeneca.”
All right, now, y’all have warned me that simply taking money from drug companies doesn’t disqualify someone from doing good research… I don’t want to be a pharmascold… but that’s an awfully long list.
I can’t get to the damn article on EBSCO, Science Daily has a synopsis here:
http://www.sciencedaily.com/releases/2015/06/150601082529.htm
“In order to properly study the link between drug use and the risk of committing a crime, the following criteria must be fulfilled: the sample needs to be representative, the reason for using the drug needs to be taken into consideration, and the effect needs to be controlled for. Furthermore, the effects of any other drugs and intox-icants used simultaneously also need to be considered. No other studies like the present one have been pub-lished thus far.”
Notice that the article does not explicitly claim that THIS study fulfilled those criteria, only that none like the present one have been published before. We are also assured that:
“After confounding factors were controlled for, the results show that the use of anti-psychotics was not associated with a significantly increased risk of committing a homicide, whereas the use of anti-depressants was associated with a slightly elevated risk…”
I wonder how those confounding factors were controlled for– whose bookie or bag man I will have to pay to get access to that particular piece of information. We are told, however, that a 31% increased risk from antidepressants is “slightly elevated” while the 45% increase associated with benzodiazepenes is considered “significantly” elevated– all this in the context of finding an increase of 206% for clients taking… antinflammatory painkillers.
I presume they are referring to NSAIDs (Though who knows? None of the articles I looked at seemed to mention exactly which drugs were being studied.) With those kinds of numbers, you’d think the headline of the story would be, “Ibuprofen doubles risk of homicide!”
Again: I barely passed research and writing. Maybe these are really solid numbers, and these guys have really got it going on. But if I did a study where the noise correction couldn’t filter out a 100% increase in the homicide rate due to antinflammatories, I wouldn’t put that on my demo tape, much less have it blasted to every corner of cyberspace. I’d figure something was probably very, very wrong with my study, because my results did not make very much sense, and no one I’ve ever heard of ever reported observing such a thing in clinical practice.
Thank you for this follow-up post, particularly the additional information about NCCI. The video clip was illuminating. The psychiatry resident Chris struck me as knowledgeable, likable, and passionate. I agree with your observation that he emphasized the bio over the psychosocial, which I suppose was the point of this video – showcasing an attempt to discuss psychological problems with an emphasis on brain circuits. What struck me was how much the video also showcased the assumption that psychological processes are *caused* by the brain. In the video, the person doesn’t scan the environment, the ventral tegmental area does. Behaviors like picking up a piece of chocolate off the counter are enacted because the orbital frontal cortex drives and motivates us to enact them. The prefrontal cortex makes us moderate our chocolate intake. This video depicts the clinical application of a peculiar brand of reductionism – the idea that psychology is caused by the brain (i.e., eliminative reductionism), as opposed to the uncontroversial idea that psychological events are ultimately rooted in the brain and are correlated with brain activity (i.e., constitutive reductionism; http://ruby.fgcu.edu/courses/twimberley/10199/psy/Cognitive.pdf). Because of my research interests, my second general reaction to this video was to wonder how blaming the brain for client problems affects critical variables like self-efficacy and prognostic optimism. An emerging body of research consistently shows that blaming psychological problems on biological causes, without highlighting the malleability of biological factors, causes reduced self-efficacy and prognostic pessimism (http://onlinelibrary.wiley.com/doi/10.1111/cpsp.12056/abstract). Something tells me this research is either not known or seen as inconvenient/irrelevant to NCCI and Insel.
Brett,
That last point is similar to what can happen with blaming behavior on “unconscious” forces. It can lead to a loss of agency. I see it up here in the rural world in people who have been told [inappropriately] that they have “bipolar disorder.” They say it directly about emotionally driven behavior. “I guess that’s my bipolar.” It can be a way to evade responsibility. But it can also be, as you say, a “reduced self-efficacy and prognostic pessimism..” Thanks for the link.
Catalyzt, three studies about NSAIDs’s side-effects.
About the Lundbeck, eh, Tiihonen study.
I found three studies that link anti-inflammatory painkillers
http://www.ncbi.nlm.nih.gov/pubmed/8707453
1996 These findings suggest that NSAIDs can induce or exacerbate reproducible symptoms (depression, paranoia) in patients with either affective disorder or schizophrenia. These adverse effects may be more severe and frequent than thought previously. NSAID-treated patients should be studied for NSAID-induced psychiatric side effects.
http://www.ncbi.nlm.nih.gov/pubmed/10205728
1999 We wanted to show the NSAIDs’s side-effects. These are non specific, they can be either affective disorders, depersonalization states, hallucinations, or paranoid psychosis (which seems to be the most frequent). The symptomatology is easily suppressed, as soon as NSAIDs are stopped and/or psychotherapeutic symptomatic treatment is started.
http://link.springer.com/article/10.1007%2Fs100670050114
1998 The [5] cases presented suggest that NSAIDs can induce or exacerbate idiosyncratic reproducible adverse psychiatric symptoms in certain vulnerable patients, including those with a variety of psychotic or neurotic disorders, and also in elderly persons[ …]
Ironically, Rett’s Disorder was booted when a genetic marker was identified (I guess because it didn’t fit into the atheoretical paradigm), so I’m not sure they even understand what direction they are going. If a genetic marker were hypothetically found for major depression you have to wonder if they’d be dumb enough to throw that out too.
Now I’m confused. Once we find biomarkers (I’m less confident than they are), is the disorder now outside our specialty? Organized psychiatry lately has shown a weird streak of self-immolation (see Collabo-care).
Catalyzt, nice share. There is another question to consider besides whether taking antipsychotics increases the risk of a person committing homicide (causal link), namely whether taking antipsychotics reduces the risk of or prevents a person from committing homicide. The temporal link (ie. that the person was taking antipsychotics at the time of the homicide) is sufficient to debunk the myth forwarded by Pharma and Rx advocates that SGA’s can somehow prevent homicide or violence. Forest meet trees. Confounding factors and causal links are red herrings to distract our attention from the obvious fact standing right in front of us all along – There is no murder prevention pill, and their own temporal data proves that. So next time one of those folks tweets out “another preventable tragedy” after the next mass shooting, as they often do, we can collectively roll our eyes wondering which psychotropic drug it will turn out the killer was on. The new pivot for the Rx proponents is to say: “Forty percent of people taking psychotropic medications are taking the wrong medication or dose.” So how do you find out if a person is on the wrong medication or dose, after they kill someone? Best argument I’ve heard yet against Rx-assisted AOT and for not releasing the criminally dangerous from jail because they are ostensibly mentally ill. As for the broader question, no doubt that neurobabble is replacing psychobable. Genomics, MRI’s, etc. are the new tea leaves to interpret, replacing rating scales. As for Y Chromosome Disorder (still don’t get why that was not added to the DSM 5), it’s nothing new. Isn’t that why boys are diagnosed with AD/HD six times more than girls?
Y chromosome plus birthday from October-December (youngest in their class at least in Canada) predicts who is diagnosed with ADHD.
http://www.livescience.com/18841-adhd-diagnosed-youngest-kids.html
It’s interesting the DSM and organized psychiatry has frowned on diagnosing immaturity in adults (with emphasis away from Axis 2) but doesn’t mind it so much in kids where pharma has a bigger role.
I’ll go meta on their neuroscience:
The obsession on biological psychiatry to the exclusion of psychodynamics and environment may be due to miscommunication between three brain areas: the cortex, striatum, and thalamus (i.e. cortico-striato-thalamic pathways).
Radua et al. compiled the results of 34 studies that used methods like diffusion tensor imaging and discovered single-minded obsessiveness and inability to integrate complexity required to understand the BPS model may be due to white matter dysfunction involving the corpus callosum and cingulum.
Biological endophenotypes for executive dysfunction such as COMT gene variants may also play a role in the type of executive dysfunction that causes seeming bright people not to understand that light is a particle and wave and that psychopathology is both biological and environmental. The n-back test may “light up” the dorsolateral prefrontal cortex. In individuals who favor a bio-bio-bio outlook on the human condition, a dysfunction of the dorsal anterior cingulate gyrus may be responsible for selective attention toward one cause at the exclusion of another.
Or alternatively, institutional psychiatry is headed by careerists always looking to jump on the next fad to chase money after the last one fails.
“Neuroscience metaphors” to convince patients they have “brain diseases” requiring psychiatric intervention (usually drugs) are still unethical.”
Yes, and that would obviously include describing SSRI discontinuation as a form of epilepsy.