Mental illness is complex too

Mental illness causes significant morbidity and mortality around the world but we know very little about its physiology or about how to treat it.  Yesterday was World Mental Health Day, so it seems appropriate to talk about this now.  A piece in Science last week addresses this issue* and the news isn't good.  The problems are multiple -- the brain is complex and difficult to study, it's hard to know which aspects of the functioning of the brain to focus on in trying to determine what goes wrong and thus how to make it right again, animal models may or may not be adequate stand-ins for human disease and drug response.  And, current medications are no more effective than drugs first used 50 years ago, but the problem is so complex that pharmaceutical companies find that it's just not cost-effective to invest in research into psychiatric drugs because progress is so uncertain.

"We just don't know enough,” says Thomas Insel, director of the U.S. National Institute of Mental Health (NIMH) in Bethesda, Maryland. “Research and development in this area has been almost entirely dependent on the serendipitous discoveries of medications. From the get-go, none of it was ever based on an understanding of the pathophysiology of any of the illnesses involved.”

This is no different from the story of most other complex diseases, of course.  GWAS and other genetic approaches have not provided any significant breakthroughs, even though, as with most complex diseases, the genetic hammer has been thrown at these disorders time and time again.  The basic biology is poorly understood (at best) so one hasn't much clue which regions of the genome to look at most intensely. Leads arise, excite,...., and then fade.  Knowledge of many psychiatric illnesses may be even more rudimentary than that of non-psychiatric illnesses: schizophrenia rates may or may not differ across time and place, autism rates may or may not be increasing, ADHD may or may not be a disorder.

As it happens, a $10 million study of the use of fish oil to reduce suicide risk in US soldiers was announced this week, as reported in multiple places including here.

The Medical University of South Carolina, the Veterans Administration and the National Institutes of Health announced the study of omega-3 fatty acids on Monday, which is being conducted for the U.S. Army.

In the controlled study, veterans already receiving mental health services will be given smoothies high in omega-3s for a six-month period. Others will be given a placebo.

"One of the questions this study hopes to address is do we see a clinical effect that is strong enough that the military would then consider providing supplements to all military personnel, not just those who are already experiencing depression," said Bernadette Marriott, a professor in the Institute of Psychiatry at the Medical University of South Carolina and the principal investigator in the study.

This might be laughable, really, given how much we know we don't know about mental illness.  Except that this kind of study presumably can be done empirically, without any real understanding of the pathophysiology of either depression or any possible effect of omega-3 on the disease.  If the suicide rate is importantly lower among those given fish oil than those given placebo, and that's really the only significant difference between the groups, it suggests fish oil might well be why.  Well, with all the usual caveats about confounders and so on.  And even if it's, say, the taste of fish oil rather than any effect on the brain, really, who cares as long as it works.

In the same vein, a September episode of the BBC Radio 4 program, The Life Scientific, with British psychiatrist David Nutt was a discussion of, among things, the potential for magic mushrooms, ecstasy, LSD, cannabis or mephedrone as treatments for PTSD or depression.  His group has published a recent paper in PNAS reporting a study of fMRI's of the brains of people on psilicybin, the active ingredient in magic mushrooms.  He was surprised at the results, he told Life Scientific presenter Jim Al-khalili-- he expected to see much higher brain activity but instead cortical activity was greatly slowed.  And, from the paper:

These results strongly imply that the subjective effects of psychedelic drugs are caused by decreased activity and connectivity in the brain's key connector hubs, enabling a state of unconstrained cognition.

Why?  If you read the "Discussion" section of the paper you'll see it is full of uncertainty about what these results are telling them or might mean; lots of mays, ifs and mights.  They do draw some conclusions; "Depression been characterized as an “overstable” state, in which cognition is rigidly pessimistic," pessimism has been linked to the areas of the brain in which this study found lowered activity in people on psilocybin, and brain "hyperactivity has been linked to pathological brooding."  So, the idea is that reducing brain activity might interrupt pathologically negative or self-destructive thoughts.

Of course, they also point out that "Further work is required to test this hypothesis and the putative utility of psilocybin in depression."  Again, this can be done empirically, and understanding the pathophysiology is not required.  But, fMRI's are not required either.  It looks as though Nutt had a hypothesis about these compounds having potential therapeutic utility, did the study and then constructed a hypothesis to fit the results.  But ok, no one knew how aspirin worked for a long time either.  And if these compounds work, if fish oil works, and actual progress can be made in treating mental illness, this could change millions of lives for the better, no matter how or why.

But, given what we know about complexity, we're betting there's no such simple answer to the problem of brain function and its variation and how to fix what can go wrong. 
-----------------------------

*Science hosted a live chat on treating mental illness yesterday afternoon and it's archived here.

Redefining phenotypes: human nature, or Nature itself?

Marcia Angell has the first of a disturbing two-part article on the doping of America in the current issue of the New York Review of Books.  It is a reflection of the nature of science in our society, of the challenge of complex biological traits and their genetic basis, and on the way we try to work out challenges to research when they are inevitably embedded in a large, integrated society.

She poses a number of questions, drawn from the three books on the current state of mental health and mental health care that she reviews.  She points out that prevalence of mental illness has risen sharply in the past few decades.

What is going on here? Is the prevalence of mental illness really that high and still climbing? Particularly if these disorders are biologically determined and not a result of environmental influences, is it plausible to suppose that such an increase is real? Or are we learning to recognize and diagnose mental disorders that were always there? On the other hand, are we simply expanding the criteria for mental illness so that nearly everyone has one? And what about the drugs that are now the mainstay of treatment? Do they work? If they do, shouldn’t we expect the prevalence of mental illness to be declining, not rising?

One issue that she raises is that the phenotype keeps changing.  We become more aware of a problem, we examine people more closely to get more sensitively detected aspects of their traits (in this case, behavioral or psychological). As a result we move asymptotically to a point at which everyone will have some form or other of diseases that needs particular treatment--often continued sessions with counselors, physicians, trainers, tutors, or years on maintenance meds.  We can see that some disorders are rising rapidly in incidence, but have a difficult time determining the extent to which this is 'true' or is a matter of our redefining the nature and detectability of the phenotypes.  (This is true of traits like autism as well--is autism really more prevalent now, or simply more frequently noticed and diagnosed?)

In a kind of naturally accelerating evolution of vested interests (special education funds, pharmaceutical profits, research topics for professors and news topics for journalists, and funding portfolios for bureaucrats), pressure not to eliminate these diseases might develop.  This may not be an inevitable aspect of society, but it does seem to be a kind of built-in rachet in our type of society, with little sense that we can ratchet down without some major crisis to distract us from the luxury of personalized disease and 'care'.

This is artificial and something to be noted and, we think, resisted.  But there is an interesting relevance to topics we discuss here on MT and evolution.  Environments, whatever they be, are always changing.  Sometimes this is glacially slow, other times rapid.  For each individual, what counts as 'environment' is local, specific, and need not be similar to what counts as such for a different individual.  From Darwin to the present, the theory of evolution is that natural selection is based on relative fitness.  The relative number of copies of a genetic variant next generation depends on its relative reproductive abilities in this generation.

This fact--or this notion, at least--by its own nature implies that the environment basically must continually change.  That is because natural selection changes genotype frequencies and that, by definition, changes the 'environment' of each individual!  With perhaps some technical exceptions (e.g., 'balancing selection' for readers familiar with that concept) this environmental flux is ineluctable, at least because the relative proportion of each type of competitor, not to mention how they modify what in the physical environment is available to an individual, by definition changes when selection is operating.

So what? you might ask!  So what is that even in Nature the definition of the phenotype, and its success in the life of each individual, is changing all the time.  Just as culture defines the phenotype when it comes to social position, treatment, privilege or disadvantage, and so on, so Nature defines the phenotype in terms of the adaptive and evolutionary future of its underlying genotype.

In either case, and perhaps unavoidably, we face an ever-changing landscape of what genes do and how it matters.

Cryola, in all colors

Existential angst

We've already posted some on this (here), but we've been thinking some more about last week's reports of revisions to the formal categorization of psychiatric diseases, the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM). Given the reordering and recategorizing done in this revision, many people will have woken up last week no longer having the disorder that they had just the day before. That's difficult enough, but the deeper question is what it means that diseases can come and go at the drop of a 943 page tome. 

Science magazine has a piece (subscription required) on this.  They write about co-morbid conditions, in which mental disorders don't exist in their 'pure form', but instead categories overlap, and, for example, substance abuse and anxiety tend to ride along with other mental disorders, particularly depression. 

DSM-IV, currently in use, gives a menu of nine symptoms for "major depression," a diagnosis that afflicts about 17% of the population at some point in life, according to the U.S. National Comorbidity Survey. (Bipolar illness—depression alternating with mania—affects another 1%.) Missing from the list is anxiety; yet, says Fawcett [psychiatrist at the University of New Mexico School of Medicine in Albuquerque, and chair of the DSM-V work group on mood disorders], anxious depressives are at greater risk for suicide, and there's a "staggering" fivefold difference in response to antidepressants, with nonanxious depressives doing much better. Fawcett's group is therefore recommending that "mixed anxiety depression," a condition that has been residing in the Appendix of DSM-IV, be promoted to a freestanding diagnosis.

But giving anxiety a higher status within depression raises other categorical questions. The "anxiety disorders" are currently a separate category that includes generalized anxiety disorder, phobias, and panic, as well as obsessive-compulsive disorders and post-traumatic stress disorder. The symptoms defining the latter two are far more varied than those associated with depression and anxiety.

The piece further reports:

Indeed, DSM-V authors are debating whether the relationship between anxiety and depression is so close that they should be subsumed into a supercategory of human hopelessness, fear, and existential angst.

Don't we all feel this at some point, to some degree, whether or not we meet the criteria for diagnosis and treatment for a disorder? This is not to say that some people don't suffer terribly from depression or anxiety, by any means. These are very real conditions, and can be utterly debilitating.

But much of this is about medicalizing what used to be considered normal, driven at least in part by pharmaceutical companies out to make huge profits by selling medications that people will take for life. Watch commercials by these companies and you'll never feel normal again: shyness has been turned into 'social anxiety disorder', and of course there's a medication for it. And similarly, baldness or erectile dysfunction or toenail fungus can all be treated now, and while this might be welcome news to sufferers, it's much more welcome to the keepers of the pharmaceutical bottom line.  Indeed, the BBC reports that 'experts' say that 'pain should be viewed as a disease in its own right' and treated appropriately.    

We've written before about the need for science to be based on trust. Is the struggle to categorize anxiety and depression going to bring us any closer to a 'truth' about mental illness, and to easing the pain of the sufferers, or is all this recategorizing primarily about further enriching big pharma?  What should we believe?

And of course this category creep makes studies of causation even more difficult because the more heterogeneous the cases, the less likely they'll share a cause, be it genes or environmental factors or a combination of both. 

Crayola Deficit Syndrome

In that same light, we were interested to read the story in the NYTimes last week about families in which occupational therapists are now being hired to help children improve their handwriting.

These days, many little fingers are being drilled. Twenty-five years ago, pediatric occupational therapists primarily served children with severe disabilities like spina bifida, autism or cerebral palsy. Nowadays, these therapists are just as focused on helping children without obvious disabilities to hold a pencil.

So even handwriting difficulties are being medicalized. Parents desperate to have their children hold their Crayons right! If not, they need therapy, lest they become Signature Challenged adults! To qualify for that, a diagnosis! Crayola Deficit Syndrome!  

This a push-pull situation that is either good for everybody or bad, depending on your point of view.  The parents pull for diagnoses that can make them feel better (a label seems to make the worrisome scrawls understood and qualified for treatment, and even curable), and the various professional interests push (to expand their client base, drug customers, special-ed teachers). 

The reality is more continuous, a spectrum of variable traits. At the extremes, no one would argue that diagnosis and treatment aren't meaningful. But we're pushing rapidly from the extremes toward the middle. Maybe some day being 'average' will be considered a diagnostic trait, since if your kid is 'average' s/he is behind other kids and you'll demand a diagnosis and a therapist to fix this.

This may seem rather silly, but not for the reason you think. We would not want our progeny to be Signature Challenged, but after all, when our kids grow up, there won't be any signatures or handwriting. It will all be by bar-scans, keyboards or voice recognition. The really savvy parents will be checking their kids voiceprints, not their scrawls. So, relax! There's no need to Cryola over misspelt 'milk'.

----------------------------

**Louis Menand expands on many of the same points we make here in a fine review of a book on the history of depression and another on the history of the drug industry in the March 1 New Yorker.  He says, among other things, that the redefinition of traits that were once considered normal, such as shyness, has 'juggernaut effects'.  Scientists then get grants to find 'the gene for shyness', and build research institutes to study it, 'even though there was never any evidence that the condition has an organic basis'.  

The moveable feast of madness

There was an interesting story in the New York Times on Jan 8 called "The Americanization of Mental Illness," that we are just now getting around to blogging about.  It turns out that mental illness varies by time and place. One culture's madness is completely unknown in another.

In some Southeast Asian cultures, men have been known to experience what is called amok, an episode of murderous rage followed by amnesia; men in the region also suffer from koro, which is characterized by the debilitating certainty that their genitals are retracting into their bodies.

Schizophrenia takes a different course in different cultures, being more severe in the US than in other cultures.  The symptoms of what we in the US know as anorexia nervosa -- an obsession with weight and dieting and a view of one's body as fatter than it actually is -- were, until recently, very different in Hong Kong, where anorexic patients primarily complained of bloated stomachs. In the 1990s, however, the symptoms began to mimic the more profound American disease.  And so on.

Thus,

[f]or more than a generation now, we in the West have aggressively spread our modern knowledge of mental illness around the world. We have done this in the name of science, believing that our approaches reveal the biological basis of psychic suffering and dispel prescientific myths and harmful stigma. There is now good evidence to suggest that in the process of teaching the rest of the world to think like us, we’ve been exporting our Western “symptom repertoire” as well. That is, we’ve been changing not only the treatments but also the expression of mental illness in other cultures. Indeed, a handful of mental-health disorders — depression, post-traumatic stress disorder and anorexia among them — now appear to be spreading across cultures with the speed of contagious diseases. These symptom clusters are becoming the lingua franca of human suffering, replacing indigenous forms of mental illness.

And yet, we in the West have been looking for decades for the genes 'for' mental illnesses such as schizophrenia and depression. The searches, as we've written here numerous times, have been essentially fruitless. In part this has been because it has been so difficult to know how to to define the traits so that a study population convincingly has the same disease. But now it seems that much of the difficulty may be that these diseases are products of a particular time in a particular culture, not 'simple' genetic traits like cystic fibrosis or Tay Sachs disease, in which the biological consequences of the causative mutation are clear.

Unless someone wants to argue the unrelenting geneticistic view that Asians have different mental illnesses -- of the extent and type described above -- because they are genetically different. That's possible in principle, though almost impossible to disentangle from culture, geography, and history. However, there are three strong counter arguments, and some supported by actual data. First, there are secular (time) trends in these kinds of traits, that become more common or more rare within lifetimes within a culture.

Second, many classical migrant studies have been done that show that, to a great extent, migrants adopt the characteristics of their new home population. Clearly-genetic traits like facial appearance may not change, and unless largely environmental, cannot change so fast. But culturally-based traits can, and they do. The formal studies largely have to do with disease, but anyone with eyes can see the extent to which second and third generation immigrants adopt all aspects of their local culture that their conservative parents can't prevent.

The third argument is the manifest way in which culture, including in this case ideas of behavioral traits including disease, spreads around the world these days, and so quickly that we know they're not due to genetic change.

There are many profound, if uncomfortable implications. We know what the problem is, but in the absence of a Lone Ranger with a magic silver bullet better idea, we cling to old beliefs or wishful thinking.

For example, does its moveable feast-ness imply that mental illness isn't a biological trait? Not at all. It 'simply' means that there is a much more complex interaction between culture and biology than geneticists are allowing for. And that genetic characteristics are not pre-wiring for specific behaviors, but may be more probabilistically likely to be manifest in various types of environment. This is why most deterministic stories about the genetics of behavior are basically Just-So stories. Once you throw in human culture, all bets about the rules of the game are off. Overall, humans are a constant, relative to variation in culture. But we vary a lot biologically, too.

Biological variation is not a good predictor, in most cases, of specific traits. This is just what the most successful, and thoughtful new GWAS kinds of analysis are also finding about disease -- even without considering the globally varying and changing cultural landscape.