Sep 14 2009

ADHD Revisited

Last week I wrote about some new evidence regarding the biological basis for ADHD. As commonly occurs with such pieces, the comments attracted some discussion of what I call mental illness denial. For those who are interested in this topic I recommend reading my five-part blog series on this topic beginning here. But I would also like to address some of the specific points raised.

Commeter Pious Fraud relates feedback from a friend who is a follower of Thomas Szasz. Szasz is definitely the father of mental illness denial. Forty years ago he had a point, and I admire his defense of autonomy, informed consent, and universal human rights. At the time a psychiatric diagnosis was akin to a loss of rights and autonomy. But there are two major malfunctions of the Szasz position today.

The first is that whatever legitimate points he had forty years ago were fought and won. He should just accept his victory – psychiatry has been reformed, partly from a series of legal precedents that reaffirm the rights of the patient. Second, Szasz committed a fallacy that is all-too common – making the scientific facts fit his ideology. Even when the ideology is a good one, like human dignity and autonomy, nature does not have to comply to our moral wishes.

Specifically, Szasz went too far is saying that because psychiatry abused the notion of mental illness, that mental illness in fact does not exist, and all of psychiatry is nothing but political oppression. He zoomed past his legitimate points headlong into pseudoscience and denial.

I find this very similar to other areas in which in order to support a noble moral value, like equality of the sexes, science is abused. Some feminists insist, against the evidence, that there are no significant differences between the sexes. I think Steven Pinker gets it exactly right in The Blank Slate when he points out the fallacy of that position – if you make your moral position dependent upon certain scientific facts, then you make your moral position vulnerable to falsification or you may be motivated to alter the facts to suit your moral needs. Rather, the facts are what they are and we can accept them while simultaneously taking a moral position. Men and women can be equal without being equivalent.

By the same token we can decry the abuses of psychiatry without denying the science of psychiatry.

Now onto the comment:

So he says, “My reading of the literature is that the evidence strongly supports the conclusion that ADHD is a brain disorder. Of course it is modified by the environment – but at its core, it’s a problem of brain function.” Then he says,“To clarify, no one says that ADHD is a disease, or that there is some pathological process going on. It may be just one end of the Bell curve of typical human variation.”

It may just be my reading of it, but those two statements sound a bit conflicting.

They do not conflict, but the commenter is conveying Szasz’s exact error in failing to see the difference between a disease and a disorder. My point is that ADHD can be a disorder without being a pathological disease.

The reason for this has to do with the unique complexity of the brain. Brain function depends upon aspects of brain physiology and anatomy such as the activity of certain neurotransmitters and the particular pattern of connections among neurons. These do not fit any classic definitions of pathology – but they do affect brain function, and therefore can produce a disorder. There is no analogy to the liver because the liver does not contain this layer of complexity.

In other words – brain cells may be normal (lacking pathology) but the connections among them may be dysfunctional.

The commenter continues:

But besides that, first, the definition of disease and disorder need to be worked out. Lots of things are on the far end of the bell curve, like having your heart in the right side of your chest (but it retains function), being left-handed, and murdering your family, but those are not considered diseases.

There is more confusion about disease and disorder here, despite the fact that I defined disorder quite specifically – a disorder is an alteration or deficiency of a typical feature or function that results in demonstrable harm. This may contain features that are at one end of the Bell curve of variation.

The commenter gives examples of things which do not cause demonstrable harm and are therefore not disorders, but ignores other examples which do. High blood pressure, for example is a disorder. Even if we ruled out secondary hypertension – when high blood pressure is due to an underlying disease, and only consider primary hypertension, when high blood pressure is present in the absence of an underlying disease. Blood pressure is simply at the high end of the Bell curve of human variation. Where do we arbitrarily draw the line to say that beyond this point there is high blood pressure? Simple – we use data to determine at what level high blood pressure causes demonstrable harm, when does it increase the risk of strokes and heart attacks.

We can see the straw man approach that Szasz takes to deny mental illness in these comments. First the confusion of disease and disorder, then the focus on variation while leaving out the “demonstrable harm” bit. But then the commenter migrates over to another straw man.

But I think he is arguing the “harmful dysfunction” idea, which holds 2 criteria for “mental disorder”: 1) there is a dysfunction, generally based on an evolutionary model of natural function, and 2) the dysfunction is deemed harmful by the standards of the person’s culture. So, for example, without understanding how a hallucination works, you can infer it is a disorder with the explanation that it represents something gone wrong with how our perceptual mechanisms are designed to function. One problem with this idea is in determining what is “natural function”. For example, drapetomania – the disease of runaway slaves – was based on the idea that black people were naturally designed to be submissive and to serve,…of course a false idea of what is “natural”. But it illustrates that the whole harmful dysfunction framework is limited by our current cultural understandings of what is “natural”, as well as what is “harmful”. Using the bell curve as the basis for determining what is natural (i.e., most people don’t hallucinate – or don’t claim to hallucinate – so hallucinating is not natural or normal) has similar and other flaws.

This is a field of straw men, and the internal logic is not consistent. I laid out the two criteria above and in my previous posts – lack or alteration in a typical trait, and causing demonstrable harm. The commenter mangles both criteria, the first as being based on an evolutionary model of “natural” (and they also confusingly incorporate the notion of “dysfunction” which really is part of the second criterion). But this first criterion is not at all dependent on what is “natural” – it simply reflects a relative deficiency of a common trait (like lack of executive function) or deviation from the statistical norm.

By itself this is not sufficient for the definition of disorder. It is  a common denialist tactic to separate two linked criteria and argue that neither stands alone. Variation and natural selection alone cannot generate evolutionary change – you need both working together. By the same token, you need deviation from the typical and demonstrable harm to be a disorder.

The drapetomania example is, frankly, an absurd non sequitur as it has nothing to do with the modern concept of disorder, which is not dependent upon any subjective definition of what is “natural.” You cannot equate “natural” and “typical”, which I think is one of the mental errors being committed here.

Next they try to dismiss the concept of “harmful” as if it is as culturally subjective as the slave example. But by “demonstrable harm” we really mean demonstrable – as in there is a measurable negative outcome. For example, decreased life expectancy – that’s as concrete as you can get. If you dismiss life expectancy as a culturally determined value, then you are dismissing pretty much all of medicine. That is a major theme of my critique of mental illness denial – if you apply their arguments consistently, then you end up denying all of medicine.

For ADHD other criteria of harm are also used – like increased divorce rate, higher rate of incarceration, lower lifetime income, etc. These are the kinds of outcomes we use for a mental illness.

But also, to the extent that there is some legitimacy to the point that such criteria are culturally subjective, that does not render them useless. Because in effect the approach to ADHD and similar disorders is to say to individual people – if you think that being divorced, in prison, and broke is a bad thing that you would like to avoid, there are treatments available that have been shown to be safe and effective.

In other words – it is OK to allow individuals to define for themselves what matters in terms of their quality of life. Ironically, offering patients proven treatments for ADHD (as long as proper informed consent is given) reaffirms individual autonomy – exactly what Szasz was fighting for in the first place.

Next we get to what is actually the most legitmate point in the denial arsenal:

Second, his first statement I quoted underestimates the bi-directionality between biology and experience. There are many examples demonstrating that behavior, experience, and function can alter and determine brain structure – just as the other way around.

I completely agree that there is bi-directionality. I am not underestimating it. The brain has memory, it learns, and learning means changing brain function to reflect experience. Older studies of brain function in mental illness did not adequately control for this, but as the science evolved researchers got better and more careful at controlling for this. For example, including a group of subjects who have never been medicated rules out effects of medication on brain function.

But in my experience mental illness deniers continue to use the bi-directionality argument as blanket dismissal of all studies that show alterations in brain chemistry or function correlating with clinical entities, like ADHD – even when it makes no sense.

In the study I discussed last week, researchers found that dopamine regulation was different in subjects with ADHD – different in ways that reflect underlying genetics, not any known mechanism of brain memory or plasticity.

Finally we get to the core of the Szasz position:

He also makes the comparison between diagnosing ADHD and diagnosing a migraine, as we wouldn’t question the existence or reality of a migraine even though we have no objective markers (in other words, the diagnosis is based on subjective symptoms). But, in the words of Thomas Szasz, behaviors are not and can never be diseases. So, yes, I believe labeling and treating a person’s behavior is very different from labeling and treating a person’s physical complaint of pain. There is a moral component involved in the former that is not involved in the latter.

The analogy between migraine and ADHD is, in my opinion, perfectly valid. What we have here is most of the Szaszian fallacies brought together – such as the confusion of disease and disorder. But that aside, the commenter makes a fallacy of circular reasoning – that mental illness does not exist because behaviors cannot be diseases. These are just two formulations of the same premise. No independent reason is given for a behavior not being a legitimate criterion of a disorder.

For example, children with Lesch-Nyhan syndrome exhibit self-mutilation behavior – they will chew off their own lips if left to their own devices. This is a behavior caused by a disorder of brain function – an extreme example to make a point. No one seriously doubts that this behavior counts as a dysfunction. We don’t need to wring our hands about whether or not we are being culturally parochial is saying that it is not a good thing for kids to chew their own faces off. Yet it is just a behavior.

And again, in the end the denialists get caught up in semantics and labels. If you don’t want your child to chew their own face off, perhaps we can intervene to mitigate this behavior. If your child is failing in school and is exhibiting behaviors that correlate with poor life outcomes, there are treatments that have been proven safe and effective in improving those outcomes. Don’t worry about the label.

Also, there is a false dichotomy between symptoms, like pain, that are supposedly objective while other symptoms, like inability to focus attention, require a moral judgment. In fact all symptoms require a value judgment to some degree. It is just that some are more obvious or universal than others. Most people (not all) desire to avoid pain and premature death. Many people also would like to function better in their lives, and avoid what they consider to be undesirable outcomes such as incarceration and divorce. This is not functionally different from treating patients with dementia so that they can live at home longer and delay entry into a nursing home.

Medicine is not free of value judgments. What is important is that patients have autonomy and informed consent – again, principles for which Thomas Szasz orginally campaigned.

Finally, I just want to point out the denialist strategy of switching criteria mid-argument. The commenter constrantly raises different objections in order to make unrelated points. In other words, I made the point that migraine is a diagnosis based upon clinical criteria rather than objective biological markers – specifically countering the argument that it is not legitimate to criticize ADHD as a diagnosis on the basis that it is based on clinical criteria. The commenter raised the common counter that the specific clinical criteria are not behavior but rather are pain. This is a  non sequitur, and simply raises a separate point. What I never get is an admission that clinical criteria are OK – that point is just side-stepped.

In short, I can never seem to get to common ground with mental illness deniers. They just shift around until them come full circle. But when we try to nail them down as to their specific objections to mental diagnoses it amounts to either semantic games, or denying practices that are not only perfectly legitimate, they are ubiquitous in medicine – so the denial of mental illness in reality is a denial of all medicine.

In the end they want to carve out arbitrary exceptions for clinical diagnoses they don’t like for ideological reasons. And this leads them to circular reasoning – mental illnesses do not exist because behaviors are not illness.

40 responses so far