Friday, May 31, 2013

A Child Psychiatrist's Thoughts on DSM-5: Disruptive Mood Dysregulation Disorder

This is part 6 of my series on the DSM-5. Previous posts covered general impressions, PTSD, ADHD, autism, and bipolar.

One of the most controversial additions to DSM-5 has to be disruptive mood dysregulation disorder (DMDD), née temper dysregulation disorder with dysphoria (TDDD). The name was supposedly changed because some objected to the negative connotations of saying a child has a temper disorder. This diagnosis tries to answer the question: "So what do we call children who display chronic irritability and have severe tantrums, but do not really have the classic manic or hypomanic symptoms of bipolar disorder?"

The DMDD diagnosis largely grew out of research by Ellen Leibenluft at the NIMH. Her studies examining "broad phenotype" bipolar disorder (the kind of BD advanced by Wozniak, Biederman, et al, which Leibenluft prefers to call "severe mood dysregulation", or SMD) showed that children with this condition were more likely to be diagnosed with depression as young adults, but did not appear at elevated risk of developing actual manic episodes. JAACAP just published a very good review article in May that summarizes the differences between SMD and bipolar.

The DSM-5 mood disorders work group took SMD and modified it to become DMDD. The two are largely similar, with the main difference being that DMDD does not require hyperarousal symptoms such as insomnia, racing thoughts, or intrusiveness. One can certainly argue that there's less evidence to support SMD/DMDD as a distinct entity than there is for other conditions (such as melancholia) which did not end up in the DSM-5 as a separate disorder. Obviously, the DSM is based on an expert consensus process, which reflects politics and cultural trends, not just science. This is neither good nor bad, but just the way the DSM has always been.

Most online reports about DMDD that I have seen have been wrong or misleading (example 1, example 2) because they tend to focus on the mistaken notion that DMDD is just about temper tantrums. The actual diagnostic criteria include: 1) Severe recurrent temper tantrums, inconsistent with developmental level, that are out of proportion to the situation, occurring 3+ times per week; 2) Mood between tantrums is angry or irritable for most of the day, almost every day; 3) Symptoms have lasted more than a year; 4) Must be present in at least 2 settings (school, home, peers).

The symptoms must have an onset by the age of 10, so that the irritable mood of adolescence does not become a confounding factor, and the diagnosis cannot be made for the first time before age 6 or after age 18. The diagnosis also supersedes oppositional defiant disorder (ODD), and can be viewed as a more severe form of ODD. These are pretty stringent criteria, designed specifically to capture severe impairment. In Dr. Leibenluft's SMD population, around 38% of those children have had at least one psychiatric hospitalization. In the past year, I have seen 2, maybe 3 children who would meet the criteria for DMDD.

At last year's American Academy of Child and Adolescent Psychiatry meeting, Dr. Leibenluft, when asked about the new DSM-5 diagnosis, said "I'm about 65% pro-DMDD." I think that's as good an illustration as any that there's very little certainty in psychiatry, and there are pros and cons to the creation of any new diagnosis.

In this case, I think it's good that there is an alternative to labeling irritable, tantrum-prone children as having bipolar. Currently, like ODD, there are no medications for DMDD, so the focus should be on psychotherapeutic and family interventions. On the other hand, I'm sure many folks will end up ignoring the diagnostic criteria altogether and call any tantrum-prone child "DMDD." Also, I'd be surprised if Abilify does not become FDA-approved for treating DMDD by 2016.

Update (4/15/15): I've written a new post with my current thoughts on DMDD, including a review of recent research. You can read it here.

Wednesday, May 29, 2013

A Child Psychiatrist's Thoughts on DSM-5: Bipolar Disorder

This is part 5 of my series on the DSM-5. Previous posts covered general impressions, PTSD, ADHD, and autism.

Though I did not want to focus on the whole pediatric bipolar controversy in this post, it would be hard to discuss bipolar disorder in the DSM-5 without some historical context. Briefly, in the 1990's, different groups of researchers started to apply the label of bipolar disorder to children. Janet Wozniak at Massachusetts General Hospital (see this NPR story for a good summary) characterized a population of children with ADHD (and often oppositional defiant disorder) whose severe irritable moods and tantrums seemed extreme to the point of appearing "manic-like."

Another major proponent of childhood bipolar disorder was Barbara Geller at Washington University in St. Louis. Her perspective was that children who had bipolar disorder were not just irritable or had severe tantrums, but displayed classic manic symptoms like grandiosity, euphoria, and increased goal-directed activity. However, her child patients had these symptoms in much briefer episodes than adults did, with ultra-rapid cycling of moods over the course of a day for many months or even years in a row. This eventually led to a study in which the children and teens who participated had "manic episodes" lasting almost 5 years in duration.

Around the same time, there was certainly a strong trend of adults with chronic irritability (e.g. those with borderline personality disorder, substance dependence, PTSD, etc.) being diagnosed with bipolar disorder because of "mood swings." This is despite the fact that manic or hypomanic episodes have to be episodic (instead of chronic) and present for at least 4 days (for bipolar type II) or 7 days (for bipolar type I) in order to meet the DSM-IV criteria. Clearly, many clinicians just ignore those criteria and go with their gut or with what's diagnostically in vogue. And even though I typically like NOS diagnoses, "bipolar disorder, not otherwise specified" is especially insidious since it allows someone to be diagnosed as "bipolar" even though they've never really had a manic or hypomanic episode.

Using a nonspecific bipolar diagnosis seemed to happen more often with children, since so few have "classic" (or "narrow phenotype") bipolar disorder. It certainly didn't help that many of the manic symptoms seen in adults with bipolar disorder (increased energy, racing thoughts, hypertalkativeness, and distractibility) are also found in kids with ADHD. With the publication of the popular book The Bipolar Child, rates of diagnosis of bipolar disorder in children reached epidemic proportions by the mid-2000's, and many kids were put on antipsychotic medications (see this detailed post at 1 Boring Old Man for a good summary).

With respect to bipolar disorder in the DSM-5, it would appear that the proponents of narrow spectrum bipolar disorder have won the day. Thus, there is no change in the DSM-5 definition of mania or hypomania, and no modifications to loosen the criteria in children. The DSM-5 also places greater emphasis on the fact that bipolar is an episodic disorder. The mood disorders work group tried to clean up bipolar NOS as well, adding a category called "other specified bipolar and related disorder," which includes those who have hypomanic episodes lasting 2-3 days, hypomania without a depressive episode, depressive episodes with some hypomanic symptoms, and short duration cyclothymia. Additionally, "rapid cycling" still refers to 4 or more mood episodes per year, rather than what Dr. Geller and her colleagues described.

So what about those kids who seem like they're constantly "manic?" Stay tuned for my post on disruptive mood dysregulation disorder (DMDD).

Friday, May 24, 2013

Preventing Transition to Schizophrenia: What Doesn't Work, What Might

One of the reasons I decided to become a child psychiatrist was seeing the devastating effects of schizophrenia in adult patients during medical school, and knowing even then how inadequate the treatments were. I thought that by working with children and adolescents, that someday I might be a part of preventing someone from developing schizophrenia in the first place.

However, during residency, I became more jaded about early intervention in treating "prodromal" symptoms, in large part because the institution where I was training emphasized biological treatments above all else. One professor, who had close ties to multiple pharmaceutical companies, exhorted us to use antipsychotic medications not only for teenagers with odd thinking or behaviors, but also for those with paranoid or schizotypal personalities, despite there not being a damn bit of evidence this would help anything.

When the evidence did start coming in for the use of antipsychotics to prevent transition to schizophrenia, it was not encouraging. From the most recent meta-analysis I could find:
One study compared CBT and risperidone with CBT and placebo. Very low quality evidence within the first six months of treatment suggested no difference in transition to psychosis (risk ratio 1.02 (95% confidence interval 0.15 to 6.94)), which remained at 12 months (1.02 (0.39 to 2.67)). Differences in symptoms of psychosis (total, positive, or negative), depression, and quality of life were not significant. Dropout was similar between groups (1.09 (0.62 to 1.92)), although the evidence was also rated as very low quality.

There was very low quality evidence for the benefits and harms associated with olanzapine, from one study comparing olanzapine with placebo. We saw no difference in transition to psychosis after 12 months (risk ratio 0.44 (95% confidence interval 0.17 to 1.08)). Dropout was similar between groups at 12 months (1.59 (0.88 to 2.88)). For participants taking olanzapine, there was a large effect on weight during the first eight weeks (standardised mean difference 0.81 (0.28 to 1.34)), which remained large at 12 months (1.18 (0.62 to 1.73)). Effects on symptoms of psychosis (total, positive, or negative), depression, and mania were not significant. Data at 24 months were not analysed because fewer than 50% of participants remained.
Thus, I was somewhat skeptical when I saw this on the Twitter yesterday:
Looking at the article (link is to Google's cached version in case you don't have a Medscape login), it described a comprehensive treatment program, implemented community-wide in Portland, Maine, that has "significantly reduced hospitalization rates for initial psychosis by one third." More encouraging was the fact that this program has successfully been replicated at 5 other locations across the country, and according to lead investigator Dr. William McFarlane, rates of conversion to psychosis were "almost identical" between the prodromal group and a control group. What the program actually entails is interesting, and a very different approach to just using an antipsychotic or doing CBT:
As part of the program, at-risk youth, identified with the Structured Interview for the Prodromal Syndromes (SIPS), are offered a comprehensive package of treatment consisting of family education, assertive community treatment, supported education/employment, and low-dose psychotropic medication.
Looking elsewhere, I found more details about the 8 components of the family-aided community treatment (FACT) program in this dissertation. I've truncated each of the bullet points to save space:
  • Community education and outreach: Early Assessment and Support Team/Alliance representatives go into the community to increase awareness about psychosis and to encourage early referrals. These education efforts are offered to a wide range of audiences...
  • Targeted outreach to those in need: Psychosis is often frightening, and even the thought of being diagnosed with such a serious mental health condition may cause a young person to refuse to seek help. Team members meet the youth and family at their level of readiness to form a relationship built on trust. Services are strengths focused and oriented toward issues young people find relevant...
  • Consistent services in the transition from adolescence to adulthood: Services are provided to teens and young adults by the same team...there is no discontinuity of care or caregiver teams just because a person ages out of childhood services.
  • Supported employment/education specialist: This specialist works closely with each program participant...the majority of young people involved with [this program] do not pursue federal disability funding.
  • Psychopharmacology treatment options: When it comes to medications, [the program] emphasizes education and choice. Medications are used cautiously, and close attention is paid to the side effects experienced by the individual.
  • Occupational therapists: These specialists are available to help assess and provide treatment for underlying sensory, cognitive, and functional issues.
  • Family inclusion: Families are viewed as essential partners in the decision making process. Most families participate in evidence-based multifamily psychoeducation treatment focused on increasing knowledge, reducing conflict, and problem solving.
  • Commitment from systems leaders: State and regional leaders work together to develop and realign funding streams, regulations, and workplace policies to best serve individuals in a flexible way, without barriers such as insurance restrictions and gaps between child and adult systems.
Since I was not at the APA, I have no idea how many of the participants in this program were on antipsychotic medications, or whether they could estimate how much the medications actually contributed to the overall outcome. However, I would be willing to bet that the program would work quite well even without the medication component.

Notably, this research is supported not by the NIMH, but by the Robert Wood Johnson Foundation, "the nation's largest philanthropy devoted solely to the public's health." This public health approach is what we need more of in mental health: It takes a village, not just a pill.

Sunday, May 19, 2013

A Most Influential Professor

On the first day of Abnormal Psychology class, The Professor sat on stage, just him with his cane, no notes or Powerpoint. His jowls hung low, giving him a bulldog look. Instead of appearing mean, however, he seemed almost bored, in a wizened sort of way. The Professor began by asking the class to come up with all of the different names that exist in our culture for someone who is "crazy." Students got into it, enthusiastically shouting out dozens of nouns, ranging from the clinical (psychotic, delusional) to the pejorative (nutso, wacko). A graduate assistant wrote all of the words down on a chalkboard while The Professor continued to sit, expressionless. There were almost 50 words on the board by the time people started running out of ideas. "What is the purpose of all these words," The Professor asked us, before answering himself that they are labels, used by those who were "well" or "normal," to define those who were "not normal."

He proceeded to launch into an explanation of his background. "I am an insight-oriented object-relations psychodynamic psychotherapist," he began, and while I had no idea what that actually meant, I was impressed by his certainty. Mental illnesses, he explained, are nothing like physical illnesses. Psychiatric labels are cultural inventions, a "word game" that cannot be separated from the time and the place in which those words originated. "I do not believe in biological reductionism or determinism," he continued, speaking in composed paragraphs to students who were used to hearing bullet points. He lamented how biological treatments have taken over much of mental health, and he told us that we would spend little class time covering conditions like depression, anxiety, or schizophrenia. Instead, he focused on conditions for which there were no medications (at least at the time): Conversion and other somatoform disorders, dissociation, addictions, eating disorders, and of course, personality disorders.

Over the course of the semester, The Professor repeatedly emphasized to us that human traits are on a continuum, even though the DSM tried to fit everything into discrete categories. He used his own dimensional scale to rate each patient on various traits such as impulsivity, neuroticism, and even adaptive regression in the service of the ego. He staged live demonstrations in front of the class in which he interviewed actors trained to portray patients with various psychopathologies; the auditorium was so crowded on those days that I'm convinced he could have sold tickets.

The Professor told us many stories, colorful and memorable. He mesmerized us with tales of 18th century mass hysteria. He lectured authoritatively on the superego lacunae present in those with narcissistic and antisocial personalities, and I could not help but visualize Swiss cheese. He recounted many of his own experiences with patients, especially those with histrionic and borderline personalities, whom he described as very "kiss kiss bang bang." Though I missed the James Bond reference (there was no Google back then), that description still strikes me as particularly apt.

Of course, The Professor was far from infallible. Humility and self-doubt were not part of his repertoire. He was prone to broad generalizations, delivered matter-of-factly: Bulemics were histrionic and attention-seeking, while anorexics had more severe super-egos and conflict with their mothers. Women who were sexually abused as children became obese as an unconscious defense against further advances. People who suffered severe enough abuse could develop multiple personalities as a way of coping. The Professor, after the first day of class, never stooped to acknowledge any other perspectives besides those which he knew to be true.

A friend of mine, who took Abnormal Psychology with a different professor, hated the class because his professor treated the DSM diagnoses as if they were naturally-occurring phenomena like planets or animal species. Looking back on my journey through psychiatry, this was often the perspective of the teachers I had from medical school onward. Therefore, I'm especially glad I had The Professor so early on. He was not a big name at the university, as a clinical psychologist in a department filled with researchers and "cognitive neuroscientists." Yet he has influenced me more than anyone else with regard to how I think about psychopathology and psychiatric diagnosis. In particular, the dimensional system just made sense. The DSM-5 even incorporates some dimensional scales to rate symptom severity, though I was disappointed when the APA Trustees voted down efforts to add a dimensional element to personality disorders in DSM-5.

I also learned from The Professor the importance of recognizing the limits of our knowledge and perhaps why psychoanalysis had fallen out of favor, even though those were not lessons he was explicitly trying to teach.

Thursday, May 16, 2013

Challenges of Psychiatry Blogging: Why Start? Why a Pseudonym?

This is part 4 of my series on the challenges of psychiatry blogging. Previous posts covered content, audience, and design.

Why Start?

I'm sure everyone who blogs does so for intensely personal reasons; thus, I will not speculate on anyone's motivations but my own here. I've always enjoyed writing but have not done much of it since college. I enjoy reading other psychiatry and neuroscience-related blogs, and I've long wished that there were more blogs focused on the many important issues related to child psychiatry. Eventually, I decided to start a blog myself for that niche. I hope to be critical in the sense of "involving skillful judgment as to truth, merit, etc.; judicial," rather than being "inclined to find fault or to judge with severity, often too readily." Though I've focused a lot on the DSM thus far, it is no accident that my first post was a movie review, and I hope to write about a broad range of topics, including vignettes from my own clinical experiences.

Why Now?

Why did I choose to start the blog last month, instead of at some other time? This question is harder to answer, and so let me examine it as I would any complex human behavior. Could it be that some life change increased the free time I have? Could I have been inspired by reading other blogs and being caught up in the zeitgeist with the DSM-5 about to be released? Is it that I finished writing a self-reflective piece offline, and encouraged by how rewarding it felt, managed to overcome previous inertia? Does it have anything to do with the fact that, after having knee pain for the last few years, I finally found a way to deal with the pain in April, thus allowing me to run again (and I'm getting good ideas while running)? Could it be the increased daylight and seasonal change helped me be more alert and focused? Is it that I've been drinking stronger coffee? I can't give you a simple explanation for why now, except to say it's likely a combination of several of those factors.

The Name Thing

Why am I using a pseudonym instead of my real name? Well, first let me say it's not so that I can vent my frustrations or write mean things. I'm more likely to use profanity in real life than on this blog. I'm fully aware that there is no true anonymity on the Internet, so I write everything as if it were under my actual name. I treat a blog post like I would an essay, refining each one until I am somewhat satisfied with the result.

Neuroskeptic has actually written a paper on the topic of anonymity in science, and how it can "facilitate the free expression of interpretations and ideas, and can help to ensure that suggestions and criticisms are evaluated dispassionately, regardless of their source." I'd like to think that's one of the main reasons why I use a pseudonym. Another reason is simply convenience. It's much easier for someone to Google "psycritic" to look up my blog (as well as my comments on other blogs), than to search for a name that belongs to many people. It helps me compartmentalize; though my patients are unlikely to bring up this obscure blog, I would still prefer to not have to chat about the blog during our sessions, and instead focus on what is going on in their lives.

Lastly, we come to the name itself. Why Psycritic? Well, as I said on Twitter:
What can I say? It's certainly not original, but the domain name was available. Now, we just need a psychskeptic who blogs, and we'll have a complete tetrarchy!

Saturday, May 11, 2013

What If the NIMH Succeeds? What Then?

Ever since National Institute of Mental Health (NIMH) Director Thomas Insel wrote his Transforming Diagnosis article on how the NIMH is moving away from the DSM to a new system called the Research Domain Criteria (RDoC) for future research studies, there have been countless articles and blog posts written about what this may mean for the future of mental health.

One of the most insightful perspectives comes from 1 Boring Old Man, who points out that the NIMH is trying to do the same thing as the DSM-5 all over again by focusing on "biology, genetics and neuroscience so that scientists can define disorders by their causes, rather than their symptoms" [quote is from the NYTimes article]. Dr. Allen Frances, the Chair of DSM-IV, thinks the new NIMH approach has merit, but he strongly criticized the NIMH for over-promising advances that won't arrive for a very long time, while ignoring the present plight of the chronically mentally ill.

Neuroskeptic likened the controversy to the Protestant Reformation, with the NIMH's RDoC (Protestantism) rising to rival the DSM approach (Catholicism), but in the end they worship the same God (biological psychiatry). This focus on the biological basis of mental illness troubles me, since I think it is terribly limiting. So much of a person's well-being is dependent on relational aspects and influenced by culture and society, as the Child in Mind blog pointed out. According to NIMH's mission statement:
The mission of NIMH is to transform the understanding and treatment of mental illnesses through basic and clinical research, paving the way for prevention, recovery, and cure. For the Institute to continue fulfilling this vital public health mission, it must foster innovative thinking and ensure that a full array of novel scientific perspectives are used to further discovery in the evolving science of brain, behavior, and experience. In this way, breakthroughs in science can become breakthroughs for all people with mental illnesses.
Though this statement does not explicitly constrain the NIMH to only fund studies looking at the biological aspects of mental illness, the language of "curing" someone obviously reflects a biological perspective. Dr. Insel is a quite a True Believer in the premises behind biological psychiatry, as shown in his TEDxCaltech talk:
“If we waited for the ‘heart attack,’ we would be sacrificing 1.1 million lives every year in this country,” he said. “That is precisely what we do today when we decide that everyone with one of these brain disorders, brain circuit disorders, has a behavior disorder. We wait until the behavior emerges. That’s not early detection, that’s not early prevention.”
As ridiculous as the above position sounds to me, let me play devil's advocate and future-think. Suppose the NIMH succeeds beyond anyone's—even Dr. Insel's—wildest dreams. What if their biological paradigm is able to elucidate at the brain circuit level (including all the circuits for positive/negative emotional valence, cognition, social processes, and attention/arousal) exactly what is happening when a person is depressed, or anxious, or hallucinating, and technology advances enough so that treatments can directly target those dysfunctional brain circuits, what then?

Well first, to make the diagnosis, there will have to be some kind of brain imaging examining the circuitry, likely coupled with obtaining a person's genetic profile. Given the complicated wiring in the brain, this will have to be done by a computer instead of a human. Treatments clearly won't be like today's medications that just target a receptor or set of receptors. To target a circuit, I can envision several methods: 1) The circuit would either have to be ablated using precise neurosurgery or interventional neuroradiology; 2) Some kind of medication would be used in conjunction with a device outside the brain that allows the medication to become active only in certain targeted areas of the brain; 3) Some sort of nanotechnology with tiny smart robots inside the brain reprogramming circuits. Because the brain is so plastic and easily influenced by the environment, a person will likely need repeated procedures or continuous treatment to prevent the circuitry from reverting to its previous state. And we haven't even talked about prevention, which seems to be Dr. Insel's goal. To do that, everyone would have to get brain-scanned on a regular basis and genotyped.

Certainly, new technologies will come along that I can't even imagine today. However, none of this will be cheap. Even certain cancer drugs today (which aren't that high-tech in the grand scheme of things) can cost hundreds of thousands of dollars per year. So once those super-expensive new brain treatments come out, who will get them? As we've seen with cancer treatment, rich people like Steve Jobs can get genotype-specific treatments and out-of-state liver transplants that ordinary folks cannot afford. Thus, it's hard for me to envision these advances in understanding brain circuitry doing much, if anything, for "public health."

Even trickier are the ethical issues that these new advances would pose for society. If you can correct the circuits causing a person's cognitive dysfunction and hallucinations, then you can certainly damage them as well. Who would we trust with such technology? Pharma? The government? China already locks up dissidents in mental hospitals; imagine if the Chinese authorities could rewire the circuits contributing to a person's desire to protest injustice. And what would happen if we no longer need any human contact, sunshine, exercise, or purpose in life to ward off depression or anxiety? Would we be content to live like the oblivious human batteries in The Matrix?

Before you accuse me of being a nutty conspiratorialist, consider this: If I were to time travel to the 1960's, and I told people that in 50 years time, everyone would have pocket-sized devices that would combine the functionality of TV, radio, telephone, telegraph, camera, newspapers, magazines, books, and myriad other games and diversions; that no one has to remember anything anymore because they can just ask an entity called "Google"; and that people would stare at this device for hours a day, even during social situations like group dinners, I think they would have put me in a psychiatric hospital.

Obviously, not much of what I am saying is new or original. Many science fiction authors have imagined such a dystopia. You can argue that it's the not the NIMH's job to consider all the potential consequences decades or centuries away, and you may be right. But I will say this: The risks of biological psychiatry are great, with uncertain payoffs. Directing those billions of dollars to address issues like transgenerational poverty, child abuse/neglect, interpersonal violence, and the housing of mentally ill in jails and prisons, while boring, will almost certainly reduce the burden of mental illness and help make our society a better place.

Friday, May 10, 2013

A Child Psychiatrist's Thoughts on DSM-5: Autism

This is part 4 of my series on the DSM-5. Previous posts covered general impressions, PTSD, and ADHD.

Quick, according to the DSM-IV section on pervasive developmental disorders (PDDs), what are the differences between autism (offically "autistic disorder"), high-functioning autism, Asperger syndrome, and PDD not otherwise specified (PDD-NOS)? If you can tell me without looking it up that autism requires 6 total symptoms from 3 categories, Asperger syndrome requires 3 symptoms from 2 categories, PDD-NOS requires only 1 symptom, and there's no such thing as "high-functioning autism" in DSM-IV, then you're much more knowledgeable than I am.

Hopefully, it is easy to see why the workgroup tasked with revising the autism diagnosis in DSM-5 tried to simplify this complex chimera into something more comprehensible. Most experts post-DSM-IV have concluded that all of these conditions overlap quite a bit and are best conceptualized as different presentations along a continuum (clinically-speaking, of course; research into etiology is a whole different ball of wax). Hence the term autism spectrum disorder (ASD). Academia has favored using ASD instead of PDD for a while now, and this seems to have filtered into the popular culture over the past decade, as you can see in Google Trends. So what does the DSM-5 change when it comes to the diagnostic criteria for autism?

The 3 categories of autism symptoms in DSM-IV encompass impairments in social interaction, communication, and repetitive or stereotyped patterns of behavior. DSM-5 eliminates the communication category, partly by combining social communication deficits (such as difficulty initiating or sustaining a conversation) with the other social impairments (such as inability to make eye contact or share enjoyment) into one category. Thus, the 2 categories in DSM-5 are "social communication & interaction" and "restricted, repetitive behavior," and impairment in both have to be present to diagnose someone with ASD. The restricted behaviors can now include hypo- or hypersensitivity to sensory stimuli, which is seen in many who are on the spectrum. Language delay, which formerly distinguished autism from Asperger syndrome, is no longer part of the criteria for diagnosing ASD, since that is now felt to mainly reflect differences in IQ.

The impairments can also be "by history," which means that a child who previously had inability to make eye contact but improved after behavioral therapy can still have that count towards the diagnosis. The age criterion has been changed from "onset prior to 3 years" to onset during the "early developmental period." As with many other parts of DSM-5, there is now a dimensional aspect, with 3 different levels of ASD severity based on the extent of impairment in each individual. Also, different specifiers can be added to the diagnosis, e.g. ASD with intellectual disability, or speech delay, or regression in functioning, or specific genetic conditions.

No change is without controversy, and there are some big ones here. Not surprisingly, the elimination of Asperger syndrome as a separate entity sparked an outcry amongst some in the Aspie community. Another controversial aspect is that ASD, unlike other entities such as ADHD, now actually has tighter diagnostic criteria than some of the previous PDDs such as Asperger's and PDD-NOS, resulting in fear that some who already have those labels will lose their diagnosis after the DSM-5 comes out. In this month's Journal of the American Academy of Child and Adolescent Psychiatry, members of the DSM-5 Neurodevelopmental Disorders Workgroup sought to reassure the public (subscription required):
With respect to the PDDs, the DSM-5 has essentially moved from letter grades to a “pass-or-fail” system. Everyone with an existing autistic disorder, Asperger disorder, or PDD-NOS diagnosis (e.g., “A,” “B,” “C”) should simply be reassigned (e.g., “pass”), and not formally rediagnosed unless there is some clinical reason to do so.
It will be interesting to see whether this guidance is followed out in the real world. While the DSM-5 taketh away, it also giveth, in the form of a new category called social communication disorder (SCD). This diagnosis is for someone who has difficulties with "using verbal and nonverbal communication for social purposes" but no restricted, repetitive behaviors. Some of the people who previously would have been diagnosed with PDD-NOS would likely fall under this category. It will also be all too easy to conclude that the DSM-5 is pathologizing social awkwardness and turning it into a disorder, just as many felt that DSM-IV turned shyness into a disorder called social phobia. The truth, unfortunately, will take time to emerge and will likely be complicated, as in the case of shyness vs. social phobia.

For me as a practicing clinician, the new ASD diagnosis is one of the best changes in DSM-5. Soon, I may not even have to reach for a reference book to see if someone meets criteria for an autism spectrum disorder.

Monday, May 6, 2013

NOS Is Dead; Long Live NEC!

In a recent Shrink Rap post on the DSM-5, Dr. Steven Reidbord wrote the following comment:
I don't have much to add regarding DSM 5. I use DSM codes on super-bills I give to patients, so they can receive partial reimbursement by their health plans after paying me directly. Most of the time I use 300.00 (nonspecific anxiety) or 311 (nonspecific depression), and occasionally "parity" codes for major depression etc. I've read all the controversy over the new edition, and it saddens me that our field is so distorted by politics and money. However, I don't see it affecting me much directly.
I have to say that I, too, often use ICD (not DSM) codes 300.00 and 311, which in the DSM-IV are called anxiety disorder not otherwise specified (NOS) and depressive disorder NOS. Working with kids, two other favorites of mine are mood disorder NOS (296.90) and disruptive behavior disorder NOS (312.9). Why such a penchant for such nonspecific diagnoses? Well, for one thing, I believe it truly reflects the patients that I see. I often hear patients or family members talk about "rages" or "anger outbursts" or "mood swings" lasting minutes to hours, none of which are well-captured by a current DSM diagnosis.

I also think that the NOS diagnoses allow clinicians to be humble, by not claiming to know more than we do about the etiology of a patient's condition. Since there are no medications approved to treat NOS diagnoses, to me they feel less "biological." For example, when I say a child has "mood disorder NOS, disruptive behavior disorder NOS, and a parent-child relational problem," I am describing a child that is moody, often has tantrums, and does not get along with his parents. I explicitly say to the parents that I'm not saying there's anything biologically wrong with their child, that these are behavioral problems that can have a multitude of causes, including family dynamics, and that what's going to ultimately make things better is not a pill.

One criticism I've often heard about using a NOS diagnosis is that it is lazy, something done by a clinician who hadn't bothered asking enough questions to reach a firmer conclusion. This is certainly possible. However, in my practice I've generally seen the exact opposite (especially in a Medicaid clinic where I worked): Many patients being diagnosed with schizophrenia or bipolar disorder type 1 or major depressive disorder, when it is clear to me that they never met the criteria for those conditions in the first place. This "mania" for reaching a firm diagnosis, I believe, is largely driven by a need to medicate. If all you have is a hammer, everything looks like a nail.

Thus, I was somewhat peeved to read last year that one of the explicit goals given to the DSM-5 workgroups was:
Eliminate "not otherwise specified" (NOS) diagnoses within categories
The rationale for them wanting to do this is obvious. As blogged about elsewhere, the DSM-5 leadership had hoped to move diagnosis onto a firmer biologically-based footing, and what can be more "fuzzy," biologically-speaking, than diagnoses for which there are only loosely defined criteria? Not surprisingly, the biological psychiatrists had an "epic fail" when it came to biologicalizing DSM-5. Thus, even though DSM-5 has technically eliminated NOS diagnoses, nonspecific conditions will live on in the form of Not Elsewhere Classified (NEC). Here's the difference in their definitions:
NOS: Not Otherwise Specified - This designation is equivalent to the word "unspecified" and indicates that the documentation does not provide enough information to assign a more specific code.

NEC: Not Elsewhere Classified - This designation is used to indicate there is no separate specific code available to represent the condition documented. In this case, the diagnostic statement is specific, but the coding system is not specific enough.
What can I say, except that this actually looks like progress? Clinicians like Dr. Reidbord and myself can not only continue using our fuzzy nonspecific diagnoses, but also instead of having NOS imply that we're lazy clinicians and didn't document enough, we'll be using NEC and implying that it's the coding system (DSM-5) that's inadequate.

Friday, May 3, 2013

A Child Psychiatrist's Thoughts on DSM-5: ADHD

This is part 3 of my series on the DSM-5. Previous posts covered general impressions and PTSD.

If you've been reading the nation's leading mental health blog, you will have been inundated with news and opinion about the rising rates of Attention-Deficit/Hyperactivity Disorder (yes, the hyphen and the slash are both there in the DSM-IV) and abuse of prescription stimulants by college kids. And you may even have come across DSM-5 Taskforce Chair David Kupfer's letter to the editor, in which he defended the DSM-5 by saying "revisions were aimed at helping clinicians more precisely recognize the symptoms of individuals with A.D.H.D. to facilitate the right care for the right person."

Less clear from those articles is what exactly is changing in the DSM-5 when it comes to ADHD, and I hope this post would shed some light on that. But first, some background on the challenges of diagnosing ADHD. Something as broad as an "attention deficit," as dependent on many different parts of the brain working together, and as prone to abuse is inherently going to be a mess when it comes to "accurate" diagnosis. Who can really say at what point "laziness" or "prefers hands-on learning" becomes "ADHD, predominately inattentive?" Thus, misdiagnosis is rampant, and there is a problem of both over- and under-diagnosis of ADHD.

This is nicely illustrated in this study from 2005, which asked parents if their children have been diagnosed with ADHD, and also had the parents complete a standardized questionnaire (the SDQ) about past ADHD symptoms. From the abstract:
Prevalence of clinically significant SDQ ADHD symptoms is 4.19% (males) and 1.77% (females). Male prevalence by race is 3.06% for Hispanics, 4.33% for Whites, and 5.65% for Blacks. Significant differences in prevalence occur across gender (p < .01) and among males across race (p < .01), age (p < .01), and income (p < .02). In the full sample, 6.80% of males and 2.50% of females have a parent-reported lifetime ADHD diagnosis but are negative for SDQ ADHD. Likewise, 1.59% of males and 0.81% of females are positive for SDQ ADHD but negative for parent report of ADHD diagnosis.
Thus, the results support the rough rule of thumb that only about half of kids with ADHD have actually been diagnosed, but about half of those who have the ADHD diagnosis don't really have ADHD. Yes, it's a mess. ADHD is almost certainly over-diagnosed not just among well-off suburban strivers needing "study pills," but also among poor urban youth who may be presenting with behavioral problems under conditions of chronic stress in the family. However, many kids who do have the syndrome are not being diagnosed, for a complex multitude of reasons. Those kids often end up marginalized in school, fall in with the wrong crowd, turn to drugs and alcohol, and have poor outcomes reaching far into adulthood.

The DSM-5 version of ADHD does not change much, but is clearly geared towards decreasing under-diagnosis. The biggest change is that instead of requiring some impairing symptoms before age 7, that age limit is now raised to 12. This is because many children with attention problems, especially if they are intelligent and not especially disruptive in class, can breeze through elementary school without adults noticing that they are not really paying attention. However, once those children reach middle school, they often become overwhelmed by the more challenging material, the demands of going to a different teacher for each subject, and having to keep track of multiple assignments with different due dates.

The 18 diagnostic criteria remain essentially the same. Instead of calling the different forms of ADHD (like combined or predominately inattentive) "subtypes," the term "presentation" is used, since they are not really distinct entities (i.e. most hyperactive children will become less hyperactive over time while continuing to have issues with attention). Also, kids with both autism and ADHD symptoms can now be officially diagnosed with ADHD, since the exclusion for Pervasive Developmental Disorder has been removed. The inattention symptoms have a much needed clarification that they are not due to oppositional behavior or difficulty understanding.

For diagnosis of ADHD in adults, the DSM-5 criteria have been loosened so that patients only need to meet 5 of 9 criteria for inattention, instead of 6 of 9. The manual also provides more examples of symptoms that are relevant to adults.

Ultimately, the question for me is, will the DSM-5 actually solve any problems associated with the ADHD diagnosis in the real world? Sure, it might make the lives of researchers easier since more people would qualify for the full ADHD diagnosis rather than ADHD, NOS. However, I don't think the change in criteria will do anything to improve under-diagnosis, since parents who didn't want to take their hyperactive or inattentive kids to see a doctor before probably still would not do so. Obviously, the DSM-5 won't do anything to stem the tide of over-diagnosis, but I'm also not sure that it'll make the problem any worse. In adults, the problem with diagnosing ADHD is that it is harder to get the third party data like parent or teacher reports that clinicians rely on when making the diagnosis in kids. Anyone can look up the diagnostic criteria and tell their doctor they have those things.

Perhaps it would have helped if the DSM-5 could specify that the symptoms cannot be purely subjective and that objective reports are needed from other sources, but for better or worse, that's not what the book is about. It won't change how I approach those in their late teens or adulthood who come to me saying they have previously undiagnosed ADHD, since I ask those patients to undergo neuropsychological testing so I would have some objective data.

Of course, even with proper diagnosis of ADHD, treatment doesn't necessarily address all the issues or lead to normalized functioning. But that's a topic for another time.