Nature kicks off the 2010s with an editorial pep-talk for psychiatry: A decade for psychiatric disorders.New techniques — genome-wide association studies, imaging and the optical manipulation of neural circuits — are ushering in an era in which the neural circuitry underlying cognitive dysfunctions, for example, will be delineated... Whether for schizophrenia, depression, autism or any other psychiatric disorders, it is clear... that understanding of these conditions is entering a scientific phase more penetratingly insightful than has hitherto been possible.But I don't feel too peppy.
The 2010s is not the decade for psychiatric disorders. Clinically, that decade was the 1950s. The 50s was when the first generation of psychiatric drugs were discovered - neuroleptics for psychosis (1952), MAOis (1952) and tricyclics (1957) for depression, and lithium for mania (1949, although it took a while to catch on).
Since then, there have been plenty of new drugs invented, but not a single one has proven more effective than those available in 1959. New antidepressants like Prozac are safer in overdose, and have milder side effects, than older ones. New "atypical" antipsychotics have different side effects to older ones. But they work no better. Compared to lithium, newer "mood stabilizers" probably aren't even as good. (The only exception is clozapine, a powerful antipsychotic, but dangerous side-effects limit its use.)
Scientifically, the 1960s were the decade of psychiatry. We learned that antipsychotics block dopamine receptors in the brain, and that antidepressants inhibit the reuptake or breakdown of monoamines: noradrenaline and serotonin. So it was natural, if unimaginative, to hypothesise that psychosis is caused by "too much dopamine", and that depression is a case of "not enough monoamines". (As for lithium, we still don't know how it works. Two out of three ain't bad.)
These are still the core dogmas of biological psychiatry. Since the 60s, the amount of money and people involved in the field has exploded, but today's research is still essentially making footnotes to the work done 30 or 40 years ago. It would be somewhat unfair to say that we haven't made any solid advances since then, but only somewhat.
The double helix structure of DNA was discovered in 1953, just after antipsychotics and antidepressants. Imagine if biologists had learned about the double helix, but instead of using it to understand genetics, or catch criminals, or sequence genomes, they spent 50 years arguing about whether all DNA was shaped like that, or only some of it.The standard response to the charge that psychiatry has lagged behind the rest of medicine is that "It's hard". And it is, because it's about human life, which is complex. But so is the subject matter of every science: the whole point is to seek simplicity in the complexity. Genetics was hard, until we worked out how to do it.
What's remarkable is that so many things in psychiatry are simple. For example: any drug which blocks the dopamine transporter (DAT) in the brain has stimulant effects: increased energy, focus, and motivation, and at high doses, euphoria, grandiosity, and potentially addiction. Cocaine, amphetamine, Ritalin etc all work this way. There are no cocaine-like drugs that don't block DAT and no DAT inhibitors that aren't cocaine-like. Simple. The stimulant high looks strikingly like the mania seen in bipolar disorder, and is pretty much the exact opposite of what happens in clinical depression. Couldn't be easier.
There are plenty of cases just like this. What's also striking is that neuroscience has advanced in leaps and bounds since the 1960s. A 60s, or even a 90s, textbook about neuroscience looks incredibly dated - a 60s psychiatry textbook is essentially still up-to-date except for the drug names. Contemporary neuroscience is far from being a mature science like genetics, it has its problems (see: all my previous posts) but compared to psychiatry, "basic" neuroscience is rock-solid. Although I trained as basic neuroscientist, so I would say that.
Why? That's an excellent question. But if you ask me, and judging by the academic literature I'm not alone, the answer is: diagnosis. The weak link in psychiatry research is the diagnoses we are forced to use: "major depressive disorder", "schizophrenia", etc.
Basic neuroscientists don't use these. If a neuroscientist wants to study the effect of, say, pepperoni pizza on the human caudate nucleus, they can order a Dominos, recruit their friends as research subjects, pop them in an MRI scanner and get to work doing rigorous (and delicious) science. They've got the pepperoni pizza, they've got the human caudate nucleus - away they go.
Whereas in order to do research in psychiatry, you need patients, and to decide who's a patient and who isn't you basically have to use DSM-IV criteria, which are all but meaningless in most cases. It doesn't matter what amazing new scientific tools you have - genome-wide association studies, proteomics, brain imaging, whatever. If you're using them to study differences between "depressed people" and "normal people", and your "depressed people" are a mix of people who aren't ill and just need a holiday or a divorce, undiagnosed thyroid cases, local bums lying about being depressed to get paid for being in the study, and (if you're lucky) a few "really" clinically depressed people, you'll not get very far.
Edit 10.1.2009 - Changed the date of the discovery of the structure of DNA from 1952 to the correct 1953, oops.
Nature (2010). A decade for psychiatric disorders Nature, 463 (7277), 9-9 DOI: 10.1038/463009a
23 comments:
So this begs the quiestion: how would you change the DSM?
How to you change the DSM? Start to get rid of it. This from someone who was diagnosed with major depression with psychotic tendencies.
Just a follow up. They are attempting to come out with the DSM V. It is delayed because the entire community is almost in civil war.
You raise many good points about the need for better diagnosis in order to recruit the most appropriate and uniform population as research subjects, but this piece is just too skeptical, even for you.
It's too easy to sit back and say that we aren't going to go anywhere in the next decade of work on psychiatric diseases, as naive, perhaps ignorant, and likely wrong as that statement is.
To suggest that the only avenues for future work include diagnosis and understanding brain chemistry is definitely what it means to be stuck in the 50s or 60s. We are moving into a post-genomic scientific world where recent studies are just now being able to test circuit-based and neurodevelopmental hypotheses underlying psychiatric disorders. Future research will likely not be looking to categorize people based on genetic similarity but rather perhaps on similar defects in circuitry or plasticity. And if genetics are used, the rise of personal genomics will allow researchers to get a fuller picture of the defects influencing a mental phenotype as opposed to relying on just a few islands of genetic markers. With that in mind, the DSM-(whatever) may eventually sink to being only a crude initial assessment of a potential problem (if it has not become that already), followed by more insightful and data-rich tests to determine defect patterns.
The next 10 years of psychiatric research will be nothing like the 50s or the 60s. As it should be. And any suggestion otherwise demonstrates a resistance against progress at worst or a "head-in-the-sand" stance at best.
Feel free to call me out on all of my misguided nonsense. But please wait until Jan 9, 2020 since there is no right answer at the moment. Don't worry, I won't forget to check back. I put it in my iCal...
"There are no cocaine-like drugs that don't block DAT and no DAT inhibitors that aren't cocaine-like."
This depends on your definition of "cocaine-like". Bupropion blocks DAT and isn't much like cocaine (though it does have stimulant properties), Modafinil is a stimulant that doesn't (to my knowledge) block DAT. This is not nearly as simple as you are making it sound.
Also, I think a large reason the psychological sciences lag behind basic neuroscience is the lack of applicable animal models. A lot of what we know of genetics and neuroscience comes from animals, particularly rodents. Psychological experiments have many ethical constraints that biological experiments do not have.
Non scientist here with a question to neuroskeptic (or anyone else knowledgeable),
Can you tell me if you believe that major psychiatric illnesses such as what are term "clinical depression," "bi-polar disorder" , "schizophrenia" will be found to have a pathogen trigger in most cases?
(I realize that diagnosing what is and what is not "depression" is a point of contention, but for simplicity's sake, I am talking about cases that are obvious--in the case of depression, people who've lost all affect with no desire to get up in the morning; in the case of bi-polar, people who spend their life savings in the course of a week taking two virtual strangers on a European vacation and then weeks later fall into a major depressive episode; in the case of schiz, people who hear voices...in other words I am referring to obvious cases of mental illness.)
Do you think Paul Ewald and Gregory Cochran and other evolutionary biologists are essentially right about their "germ theory"--that bugs will be found to be the trigger of such illnesses?
So the layman who assumes that much of psychiatry is tosh is not altogether wrong? Or is he confusing psychiatry with the theologies of Freud, Jung and so on?
Thank you for this post, Neuroskeptic. Your concerns about the state of psychiatric diagnosis go back to the appearance of DSM-III. Here is my take on it at the Dahlem conference in 1982.
In the clinic... a conceptual drift of alarming proportions has occurred in the name of diagnostic reliability. This drift threatens to retard the field by increasing the variance present in clinical populations and by obscuring historical insights that helped us construct the amine paradigm 20 years ago…. How is a theory of depression to be developed, how is it to be tested, and how is an antidepressant drug effect to be recognized if the clinical entity (or independent variable) no longer corresponds to the original?
More recently, in agreement with Neuroskeptic, I concluded that “the corruption of clinical research standards has brought us to an epistemologic quagmire…”
See Carroll, B.J. Neurobiologic dimensions of depression and mania. In The Origins of Depression: Current Concepts and Approaches, ed. J. Angst. Dahlem Conference. Life Sciences Research Report 26. Springer Verlag, Berlin, 1983, pg 163-186.
Also http://hcrenewal.blogspot.com/2009/04/in-defense-of-psychiatric-diagnoses-and.html
Cochran & Ewald is very interesting. They allow, as they must, that even if all disease comes in some ultimate sense from parasitism, the parasite need not touch the host, the obvious examples being sickle cell, thalassemia, and the rest of the course, awkward, or ad hoc adaptations to falciparum malaria.
Personally, I find it easy to see a big possibility for autoimmune diseases not caused by infection of the actual diseased individual. Just consider the vast pressure for an avid adaptive immune system. The presence of various deleterious mutations (probably ~300 per human on average) introduces "noise" which makes it even harder for the immune system of every individual to have the optimal avidity that balances him between autoimmunity and infection resistance.
On the other hand, supposing we have an individual with a nice strong autoimmune diathesis, what is likely to push him into disease? Well, a microbe could be an excellent candidate, after all it bears PAMPs which might help do the job. But the microbe may not cause the disease in the sense of causing the difference from other people who don't have it. For example, the microbe could be EBV or HSV-1 or Chlamydia pneuomnia in which case almost everyone would have it chronically or at least acutely.
BrianW - "Bupropion blocks DAT and isn't much like cocaine (though it does have stimulant properties)"
But it's only a weak DRI: according to Wikipedia anyway...
The occupancy of dopamine transporter (DAT) by bupropion and its metabolites in the human brain as measured by positron emission tomography was 6–22% in an independent study[115] and 12–35% according to GlaxoSmithKline researchers.[116] Based on analogy with serotonin reuptake inhibitors, higher than 50% inhibition of DAT would be needed for the dopamine reuptake mechanism to be a major mechanism of the drug's action. By contrast, approximately 65% occupancy or greater of DAT is required to achieve euphoria and reach abuse potential.
"Modafinil is a stimulant that doesn't (to my knowledge) block DAT."
But it doesn't have cocaine-like effects. It promotes wakefulness, and maybe improves attention, but it doesn't cause euphoria etc. It has little if any abuse potential. Similarly caffeine is a 'stimulant', but not a cocaine-like one.
Thanks for tackling the question of lack of progress in psychiatry. How many researchers in the field do you think share your views? Is the diagnosis issue discussed publicly or privately at conferences, etc?
What about lack of reliable outcome measures for many disorders? Isn't that a fatal flaw in itself?
Why do "you basically have to use DSM-IV criteria" to do psychiatric research? What is stopping people from thinking outside the box?
For example, there is a lot of work demonstrating that severe working memory deficits are fairly common (~10%) in young children, sample research. So, there is a well defined phenotype, reliable diagnostic and outcome measures, extensively researched animal models and some understanding of basic neuroscience (at least as compared to most other mental disorders). Seems like if you want to make progress in understanding the biology of a specific brain malfunction, this model would be at the top of the list. Yet, as far as I know very little biomarker (ie genetics, imaging) research is being done with this group. Why? It's not in the DSM? Wouldn't scientists be highly motivated to work in an area where progress is most likely?
Art, one of the problems in clinical psychiatric research is what I call methodological imperialism. Though lip service is paid to the approach of using multiple diagnostic criteria sets, it almost never happens. Granting agencies and journals increasingly demanded conformity to DSM-III-IV criteria, even though these were announced as provisional hypotheses. Insistence on related structured interviews just compounded the problem.
Use of multiple, competing criteria would allow their alignments with biomarkers to be compared. That, in turn, could drive choice and refinement of diagnostic criteria in an iterative fashion. Because that mostly doesn't happen,the conceptual drift is reinforced rather than questioned.
Bernard, do people actually doing the research have a voice in this or are they just slaves of the funding agencies? It seems to me that researchers that share Neuroskeptic's views have a responsibility to do something. Too much is at stake to simply coast on autopilot.
Art, it is possible we may go back to the future, if only because younger investigators have a clearer view of the futility of the current approach. In mood disorders, which I know best, heterogeneity of case material is a very serious problem, as Neuroskeptic described.
But never underestimate the power of orthodoxy and inertia in science. For many established investigators the name of the game is mainly to keep the game going, and that involves a steady flow of funds from the agencies. So the investigators are not eager to rock the boat, even though they recognize the problem.
Pat: Unfortunately, I'm not familiar with Ewald or Cochran. It's certainly possible for pathogens to cause psychiatric symptoms - syphilis infection of the brain can cause symptoms very much like psychosis and mania, for example. There is some evidence that pre-natal viral infections can raise the risk of schizophrenia, although I believe the effect is small to tiny.
But overall I don't see pathogens being the One True Cause of mental illness. It doesn't seem likely given that we know there's strong genetic influence, and the fact that most severe mental illness strikes at around the same time (peak in late adolescence, less commonly later, much less commonly earlier). That's not a pattern associated with infectious disease. I suppose it's possible that an infection at a critical period could trigger the disorder, but I don't see much evidence for that yet.
Many infectious diseases have quite high heritability. Some over 0.50. Leprosy and TB I believe have some of the highest.
Late adolescent modal onset is consistent with transmission of an agent by amorous kissing, etc. Occasional early onset is consistent with inefficient non-amorous transmission, or inefficient vertical transmission.
That's some over 0.50, *I believe*.
Anyway some of them are a good deal higher than some of those of the unsolved "complex" diseases.
as bad as DSM-IV is (and I agree with your assessment of its defects), psychiatry without DSM would be worse. If you've ever treated elderly patients you may have had the opportunity to read the clinic notes from pre-DSM-era psychiatrists. They are uniformly incomprehensible, and useless.
sadly, DSM-V is shaping up to be an unworkable disaster of its own.
By the way, if anyone feels like reading Cochran & Ewald, it's very easy. Both their main paper and the one on schizophrenia are extremely well written and clearly developed. And there is no obscurantic use of barely-explained technica to make the paper look more sophisticated.
But to repeat, I'm not 100% convinced by it.
There's been quite a serious focus on immunological pathogenesis in psychiatric illnesses in the past (from my experience the early nineties had a fair bit) but the problem was that no one could find much evidence for the hypothesis.
As to whether our wonderful post-genomic world will lead to a cornucopia of understanding and new treatments I reckon that's unlikely. For a start there is a lot less work being targeted at treatments than basic neuroscience, and what therapeutic development there is has not been fruitful - I'm not aware of any promising medications in development for schizophenia for example.
I would say that part of the problem with the diagnostic lack of specifity in psychiatry is that (a) attempts to come up with more specific diagnoses have largely failed, (b) institutional inertia and conservatism, (c) the massive difficulty in doing clinical research.
On the latter point, and touching on the genomics question, you only have to look at research being done at NIH looking at novel dopaminergic treatments for schizophrenia while simultaneously examining dopamine metabolism genotype - they've been doing it since the late nineties and still haven't got enough subjects - this is not easy research - I am seriously doubtful that we have the resources or skills to translate all this (fairly non-specific) basic neuroscientific research into the clinical realm.
Personally I think that the genomic/pathological focus is much more of a scientific/funding fad than a necessarily productive avenue - if the money sunk into this research had gone into pure clinical research I think we'd be much further along right now - but I guess we'll see in 10-20 years because it will realistically take that long to see any benefit from the basic research insights of the last decade or so.
PJ,
The questions "why" and "how" (etiology and pathogenesis) can't be answered separately from the question "what" (classification). While the latter is being answered incorrectly (DSM's), we can't know if the answers to the former questions are meaningful or relevant.
The concept of "spectrum" disorders is a step in the right direction. In Bipolar disorder, Akiskal's approach and his diagnoses (most of which aren't in DSM IV)are much more on the money, than DSM IV. It's instructive that when Akiskal wanted DSM V to include a diagnosis which recognized Bipolar uncovered by anti-depressants i.e. AD induced hypomania or mania, it wasn't accepted, because Big Pharma thought their profits might suffer. A reason for lack of progress in psychopharmacology since the 1960s is Big Pharma. Since profits are their only consideration, they approach "developing" new psychotropic drugs with the following philosophy: Finding a truly innovative or a clearly more effective drug is relatively riskier if profits are the only concern. It's more certain to element a troublesome side effect of commonly used medication. Big Pharma knows they can "spin" their new medication as more effective whether it is or not. Antipsychotics are very instructive. Instead of really attempting to find an antipsychotic which is truly more effective such as Clozaril, Eli Lilly gave us Zyprexa which initially was touted as Clozaril w/o the severe adverse effects. Of course, not only is it no more effective than a drug such Perphenazine, it introduced Diabetes Mellitus (along with Risperidone) as a side effect in treating Chronic Schizophrenia. We are seeing chronic renal insufficiency and will not doubt see further devastating effects of
DM. Clozaril was initially developed in the late 1950s or early 1960s. So it can be said the field of psychopharmacology has actually regressed in some areas. The most effective medications: MAOIs for depression, Lithium for Bipolar, Clozaril for Chronic Schizophrenia, are all more effective than anything which has come out afterwards. For the category of anxiety disorders in terms of effectiveness there has been little improvement in terms of absolute effectiveness: Immipramine is as effective for Panic Disorder as SSRIs,for treatment of combat related PTSD although the US military recommends SSRIs, they are totally ineffective, while MAOIs, and TCAs are effective. The lack of progress is a result , at least in part, to ceding the development of new drugs to Big Pharma . In this one area, the federal government is more effective than private industry.
Has Paul Ewald published any orginal experimental research in the last decade?
Post a Comment