Over the course of my life, I have been given no fewer than five different diagnoses for mental illnesses, under the diagnostic system laid out in psychiatry’s “bible,” the DSM. But it was a sixth diagnosis— one that ironically will no longer appear in the edition being rolled out this week, DSM-5— that probably most accurately describes what is genuinely different about me. I’m sharing this because my experience is a case study for explaining why the latest revision to the manual is raising such ire.
My journey from diagnosis to diagnosis illustrates both the pitfalls and the promise of psychiatry and why we can expect to improve some ways in which we identify mental illness, and why there are other aspects of diagnosing these conditions that will remain unsatisfactory without further scientific advances.
Dr. Allen Frances, who chaired the publishing process for the previous revision, the DSM-IV, and is critical of the DSM-5 in his new book, Saving Normal, published an editorial in the Annals of Internal Medicine highlighting some of his issues with the latest revision. Citing the “crisis in confidence” in psychiatry over diagnosis, he calls on physicians to “use the DSM cautiously, if at all.” DSM 5, he argues, is overrun with “diagnostic inflation”— for example, labeling grief as depression and placing the 40% of college students who binge drink at risk of diagnosis equivalent to alcoholism.
Frances isn’t the only one who has concerns about DSM-5. Last week, the the director of the National Institute of Mental Health, Dr. Thomas Insel, posted a blog in which he announced that even for research purposes, the DSM had outlived its usefulness. “NIMH will be re-orienting its research away from DSM categories,” he wrote. The problem, he said, is that the DSM is based on subjective descriptions of collections of symptoms that tend to occur together — but not on the physiological or psychological mechanisms that cause them.
As Insel put it, “Unlike our definitions of heart disease, lymphoma or AIDS, the DSM diagnoses are based on a consensus about clusters of clinical symptoms, not any objective laboratory measures. In the rest of medicine, this would be equivalent to creating diagnostic systems based on the nature of chest pain.” In other words, such a system could mistake heartburn for a heart attack and classify both as the same type of problem. Unfortunately, as he also admits, there isn’t a better system: NIMH is seeking to develop one based on brain research, but this does nothing for patients who need help now.
I can attest to that. My multiple diagnoses are the rule, not the exception, and one criticism of the DSM structure is that if you qualify for one diagnosis, you typically also qualify for others. Which one should be treated? Or do they all require interventions? And what if the therapies conflict with each other? You see the problem.
My symptoms started in early childhood and were linked to a collection of both positive and negative attributes. I was both obsessively interested in ideas and completely overwhelmed by my senses. Loud sounds, itchy clothes, new tastes, being held and any type of novelty that wasn’t intellectual disturbed me. I started reading at age three and withdrew into a whirl of academic achievement and social awkwardness. I had obsessive, repetitive behavior such as ritually counting certain actions like swinging on the swings — although I mostly managed to hide it, except for in being unable to shut up about my weird interests.
Given those symptoms, today I would almost certainly be diagnosed with Asperger’s Syndrome, the one diagnosis DSM-5 seems to have tightened, merging it into the spectrum of conditions that constitute autism (Asperger’s doesn’t even appear in the index as a separate condition).
At the time, however, the diagnosis didn’t exist. So instead of being seen as having some type of disorder, I was labeled as “selfish” and “gifted.” I folded both of these characteristics into my identity, convincing myself they made me a “bad person,” who might only redeem the fact that she preferred ideas to people by achieving overwhelming intellectual success.
While some argue that medicalizing labels like those in the DSM only do harm, my case was probably one in which they might have helped. Had I known I had Asperger’s for example, I wouldn’t have felt so bad about my bossiness and apparent disregard for other people — I would have realized that they were part of a brain difference that came with both advantages and disadvantages, not a matter of moral deficits. I would have also been explicitly taught how to do better in ways I could understand.
By the time I got to junior high school, I desperately wanted friends but had no idea how to make them, and my frustration made me unhappy. My social cluelessness also made me a target for bullies and by high school, I was quite depressed. That, however, was not something people expected in teens at the time.
Enter drugs, of the nonpsychiatric variety. In high school, I discovered that getting high not only gave me comfort and a sense of belonging, but an obsession that wouldn’t bore others when I pursued it endlessly. Instead of isolating me, this obsession allowed me to connect. By my second semester of college, I was addicted to cocaine and by the end of sophomore year, I was injecting heroin and had to leave Columbia University.
At this point, I finally started getting diagnosed — properly— when prison was a real possibility. When I chose to enter rehab at 23, I was correctly diagnosed with cocaine and heroin dependence — not exactly a difficult categorization to make of an 80 pound woman covered with track marks who tested positive for both drugs. Those were my first two official diagnoses and they were accurate. Score one for DSM.
But about halfway through my 28-day inpatient stay, I had what cannot have been more than a five minute visit with the program’s psychiatrist. Although I doubt he could have picked me out of a line up later that day, within two minutes of questioning me, I was diagnosed as bipolar and prescribed lithium. It is certainly true that I have a habit of talking quickly and while detoxing from cocaine and heroin, I absolutely had severe mood swings. However, I have never had anything close to a real manic episode and didn’t suggest otherwise in the conversation.
Bipolar is a classic example of the DSM’s diagnostic inflation. It is now possible to be diagnosed with types of the disorder that do not include what was once its defining characteristic — becoming so elated or agitated that you lose touch with reality. It is quite likely that I met the criteria for a type called bipolar II — a disorder where you have periods of depression that alternate with periods of upbeat mood that do not cross the line into mania. But the diagnosis didn’t accurately characterize what was actually happening to me at the time, which was basically that my already odd brain was recovering from several years of severe addiction.
Later, I would pick up several other diagnoses: in early recovery, a therapist noted my obsessive tendencies and added the obsessive-compulsive disorder (OCD) label and when I suffered a new bout of depression, that, too, got added to the list.
So, what did the DSM do for me? I collected diagnoses, but none of them— aside from the one I never officially received — fully described my real problems. The addictions were real— but they didn’t simply arise because I took drugs. I took drugs because I didn’t know how to deal with the depression and social isolation of what I now suspect is Asperger’s. The addiction treatment system failed to correctly identify my underlying issues and gave me a label with little consideration.
Indeed, if I’d held on to the bipolar diagnosis, I could have been severely harmed by inappropriate medications. The OCD diagnosis at least accurately characterized my obsessive nature — but virtually everyone with Asperger’s could also be diagnosed OCD; it doesn’t provide the whole story. Like the addictions, the depression was certainly real but it, too, was probably secondary to the social isolation caused by the Asperger’s. However, my treatment for depression with antidepressants was probably the most useful therapy I received: it actually reduced the sensory and emotional overload I’d tried to address by self-medicating with the illegal drugs.
The problem of multiple diagnoses like mine is one reason NIMH wants to abandon DSM and replace it with a system that looks at the brain systems that are going awry rather than focusing solely on symptoms. That, some experts like Insel believe, can lead to better understanding of how best to treat specific issues in these circuits. So maybe, if someone had recognized and treated my sensory and emotional overload early on, rather than labeling them as signs of selfishness or simply being “gifted” or “different,” it might have prevented both the depression and the addiction.
The NIMH’s “research domain criteria (RDoC)” classification system however, isn’t ideal either. For one, it assumes a pretty robust understanding of the circuitry in the brain that dictates normal function, as well as how these networks go awry in mental illness— and that’s a level of knowledge we still haven’t achieved. The search for genetic and chemical markers of specific problems has also been hampered by the failure of these problems to line up with DSM diagnoses. There might, for example, be a gene that predisposes someone to “get stuck” on negative thoughts — but it might not show up in studies of people with depression because there are many roads to depression that may not involve that gene. Focusing simply on single symptoms and their genetics in this way may help clarify the situation — as NIMH plans to do — but we are far from having these answers.
And if researchers do eventually trace specific symptoms to their chemical and neural-circuit roots, how will they sort out the variety of unique symptoms and diagnose them properly? The more symptoms that psychiatrists may uncover, the more the combinations of symptoms will multiply and the more difficult it will be for both clinicians and patients to properly interpret the information.
And such genetic and biochemical markers will only take us so far. My problems developed not just because of my genetic predispositions, of which I’m sure there are many, but because of my environment and my psychological reactions to them. Those responses are sometimes amenable to medication and may indeed require it — but typically, cognitive and social changes are needed as well. A diagnostic system that tries to reduce the brain to biology would be like a computer technician who knows only how to fix hardware but not the software: not very effective if the software is what’s acting up.
Frances argues — and I agree — that such biology-based criteria for defining mental illness are still far off, and that for now, the DSM criteria are the best way that doctors can help patients today, in the clinic. But he argues that we need to tighten up the DSM criteria and recognize that college binge drinking, for example, shouldn’t lead to a lifelong diagnosis of having had a “substance use disorder” that doesn’t categorically distinguish between alcoholism and milder drinking problems.
Such measures could also reduce the over-treatment of conditions that don’t require psychiatric care, such as ordinary grief. Yes, the current DSM-based system is no doubt perpetuated by a drug industry that markets both disorders and their treatments, and, as my own case showed, it can lead to incorrect labeling and risky medication treatments. The symptom-based definition of mental illness also encourages labeling of conditions and disorders in a way that implies far more knowledge of mental disorders than is actually the case; government programs, school systems and insurers all rely on being able to tag symptoms with a name to determine eligibility for services. We do need diagnoses: but both psychiatrists and patients alike should recognize that these labels are neither immutable nor perfect.