Antonio Damasio made a very compelling point in Looking for Spinoza: Emotions are best viewed as another way of restoring homeostasis. Negative emotions are spurs to change something to improve your external and internal environment; positive emotions are cues to keep doing whatever you’re doing. Interfering with the operation of that system is not likely to lead to better outcomes over the long-term.
The point of medicine should be to improve the outcomes for the patient, not for the people around them. With the exception of de facto crime, the interests or issues of bystanders should not constitute sufficient reason for medical intervention. Unfortunately in mental health that principle is frequently ignored despite a lot of talk about using a more medical model.
Start with a panel of experts coming up with a subjective definition for a subjective judgment of mental disorder (i.e. you aren’t normal enough, without ever quantifying what that is); another bunch of experts comes up with subjective measurements to confirm someone meets the subjective criteria for a diagnosis of the subjectively judged “disorder”. Then evaluate whether the subjective disorder is over or under diagnosed in a population. Since there is no way to objectively determine what the base rate is (since there are no objective criteria for the disorder, or that it really is a disorder), what on earth is a type 1 error or type 2 error in that context? I have no clue, but I know it’s not science. It’s not falsifiable (and if you ever nail things down enough to get close to doing so the definitions will be changed in the next edition of the DSM). It’s enough to make you side with Thomas Szasz.
It’s easy to blame the DSM, pharmaceutical companies and others for the drugging of America as it’s been called. Certainly there is some measure of fraud, self-interest, political lobbying, and the universal desire of authority for docile subjects, but I think the legislative changes requiring parity for mental health were the final piece of the puzzle leading to an explosion of diagnosis farming and rent-seeking from big pharma to practitioners to school administrators. It’s becoming a brave new world where normal keeps getting pseudo-scientifically redefined and restricted to open up new markets and methods of control, and ultimately where only a few in charge will be deemed normal and the rest are in need of help.
Environmentalists and others warn of the dangers of crop monocultures, but what about human monoculture? Genetic variation is an evolutionary advantage. That’s why we have sex. It’s possible the entire bell curve of humanity is not just normal but necessary, and not just the lump in the middle.
For an example of how the process of putting a scientific veneer on hazy subjective judgments leads to trouble, see this post on SciAm: ADHD: A Backlash to the Backlash. It’s by a parent who obviously cares and has been through a lot with her child. However, it presents no evidence that shows the typical treatments benefit anyone but third parties. The study that was cited to show benefits of treatment only shows improvements in subjective judgments of others and not in the one objective measure of outcomes used.
“These graphs reveal convergence of treatment groups from 36 months to 8 years and maintenance of improved functioning overall relative to baseline. An exception appears for WIAT math achievement, for which no randomized treatment group-related gains were detected at any assessment point “
The only other potentially objective measure was arrests and contacts with police but that was muddied by using parent reported numbers. So the subjective measures of improvement converged with time and the lone objective measure showed none of the treatments had any benefit. Is that a good reason to drug someone? I don’t think so, especially when the drugs used are completely non-specific shotgun approaches to treatment. I strongly suspect heroin-induced placidity would achieve statistically significant results with this assessment methodology. The questionable use of SSRI’s in depression is another example of the synergies between shaky science and hazy diagnosis.
This comparison to withholding insulin from diabetics inadvertently highlights the differences between the medical and mental health models.
As child and adolescent psychiatrist Harold S. Koplewicz noted in the Huffington Post in a response to Sroufe, “you can’t put a child on a placebo for his entire adolescence for the purpose of a study.” Koplewicz also points out that many therapies, including insulin therapy for diabetes, haven’t gone through long-term randomized studies for the same reason.
I’m going to give Harold the benefit of the doubt and assume he’s not being intentionally deceptive but if you withhold insulin from a Type I diabetic they will die in a fairly short period of time. Insulin targets a very specific physiologic deficiency and the treatment outcomes can be quantified using completely objective [physical] measures via blood glucose or HgA1c. None of those are the case with drugs in mental health.
The author also makes a point of discussing the cruelty and abuse directed toward her child. Though not intended, she makes an extremely convincing case for firing a teacher or two but not for drugging the victimized child. The underlying reasoning is if someone picks on you for being different, you should be drugged for your own good. Fortunately for the black kids schools were desegregated before this thinking took hold or they might have all been drugged too.
Arguing that overdiagnosis is not a problem with ADHD she says:
Furthermore, according to a 2012 publication from the U.S. Agency for Healthcare Research and Quality, the real prevalence of ADHD hasn’t changed since the 1980s, and changes in diagnostic rates are “consistent with changes in clinical guidelines.”
Here’s what the study says:
As noted by the authors of the original CER1, some evidence suggests that rates of identification and treatment for people with ADHD have increased in recent years;2-4 however, prescription patterns and variations indicate that increases in identification may be linked with changes in practice rather than an increase in the underlying prevalence of the disorder.5,6 In fact, the underlying prevalence of the disorder in children appears to have been relatively stable since the 1980s, to the extent that it has been measured using identical methods.7 Increases in identification and treatment have occurred primarily among girls and older children consistent with changes in clinical guidelines.3,8 Increases in off-label prescription of psychotropic medications for very young children have also been observed, presumably for preschoolers identified with ADHD or disruptive behavior.9
So there were increases in diagnosis rates from changes in practice while the base rate remained unchanged. If that’s not overdiagnosis what is? If the earlier rates of identification were too low the criteria were wrong, but then what reason is there to believe the current ones are right?
Finally, according to the author “research indicates that ADHD is “highly heritable” (about 60 to 80% inheritable [PDF])” Accepting the numbers from the article as a given ADHD affects 3-7% of the population and is about 60-80% inheritable. From an evolutionary perspective it seems more reasonable to conclude that ADHD is adaptively neutral than detrimental. Again not a reason to drug someone. Humans like to think they’re here for more than reproduction so that line of reasoning may not hold much water, but I tend to think we’re here for more than pleasing authority figures too.