Carlat 播客：DSM 缺失的一页——如何区分成人 ADHD 与其他认知问题成因

The modern DSM began as a short list of 12 well-validated diagnoses. But the list expanded as the authors changed the goals from validity to reliability. Lost in this history is a missing page that helps separate adult ADHD from its mimics.

Duration: 14 minutes, 39 seconds

Transcript:

CHRIS AIKEN: There’s a missing page in DSM, and today you’ll learn how it can help distinguish adult ADHD from other, better-validated causes of cognitive problems.

KELLIE NEWSOME: Welcome to the Carlat Psychiatry Podcast, keeping psychiatry honest since 2003.

CHRIS AIKEN:

 I'm Chris Aiken, the editor in chief of the Carlat Report. 

KELLIE NEWSOME:

And I'm Kellie Newsome, a psychiatric NP and a dedicated reader of every issue. Last week, we looked at the research on adult-onset ADHD, and we ended up agreeing with the DSM. ADHD is a neurodevelopmental disorder, so the symptoms should trace back to childhood. In rare cases, they don't show up until the teenage years, possibly because a high IQ or a supportive parent covered for the child. But onset in middle age is unheard of. Nearly every case of adult-onset ADHD in the literature turned out to be better explained by other causes. But try telling that to a 35-year-old man who walks with new cognitive problems that he’s convinced are due to ADHD. Today, we look at what "better explained by other causes" means in practice. But first, a preview of the quiz. You can earn CME credit through the link in the show notes.

1. Which population is less likely to recover fully from the cognitive effects of cannabis use?

A. ADHD

B. Anxiety Disorders

C. People who start using late in life

D. People who start using as teens

CHRIS AIKEN:

Every disorder in the DSM carries this line: don't diagnose it if the symptoms are better explained by something else. It’s the most boring part of the criteria, but it carries the most weight, and its origin story traces back to the 1960s and a brand new technology: the videocassette recorder.

KELLIE NEWSOME:

This was a time when psychoanalysts ruled psychiatry. To them, diagnosis was just the surface; they were more interested in the unconscious origins of the symptoms. And so, psychiatric diagnosis was pretty lax. If patients had mild symptoms, they got diagnosed with anxiety-neurosis, and if they were more severe, they usually got a schizophrenia diagnosis. Nobody paid much attention to this until an NIMH researcher, Morton Kramer,  noticed that diagnoses in the US were way out of sync with those in the UK – sometimes by a factor of 10. Kramer purchased video recorders and recorded structured interviews of hospitalized patients – both British and American. He then had British and American doctors watch the tapes and come up with a diagnosis. The difference was striking. The Brits, using strict Kraepelinian thinking, diagnosed manic depression, where the Americans, using a looser, Freudian framework, diagnosed schizophrenia.

CHRIS AIKEN: The study was a major embarrassment, but for a group of psychiatrists working in the Midwest, it created an opening. In the 1950’s, Washington University in St. Louis, Missouri, started recruiting psychiatrists who wanted to align psychiatric diagnoses with the medical model. They were renegades in their time, and they didn’t fit in with the psychoanalytic schools that dominated in the Northeast. In 1974, two years after the videotape study came out, they published a new psychiatric textbook that aimed to take a “just the facts” approach to psychiatry, avoiding all speculation: Psychiatric Diagnosis by Eli Robins Samuel Guze. Robins and Guze concluded that only twelve psychiatric syndromes qualified as valid disorders. Their standard for validity came from the medical model: reliable symptoms that separate one disorder from another, a predictable course of illness, a link to biological markers such as lab values or treatment response, and a known pathophysiology or genetic basis. Here’s the twelve diagnoses met that bar. If you memorize one list in psychiatry, this is it:

1. Schizophrenia

2. Bipolar disorder

3. Major depressive disorder

4. OCD

5. Panic disorder & phobias

6. Anorexia & bulimia

7. Conversion disorder and somatization

8. Antisocial personality disorder

9. Substance use disorders

10. Dementia

11. Delirium

12. Psychiatric disorders due to a medical cause

KELLIE NEWSOME:

The book is still updated today, and in fifty years the authors have added only two diagnoses to that core list, both with provisional status:

13. PTSD

14. Borderline personality disorder

I know, it sounds like we’re a long way off from adult ADHD, but this was the backstory to DSM-III, and knowing it will help your diagnostics. While this Washington School was refining the diagnostic criteria, Robert Spitzer was working at Columbia to get one diagnosis removed from the DSM entirely: homosexuality. These neo-Kraepelinians weren’t just bothered by loose, Freudian diagnostics. They worried that psychiatric diagnoses would be abused, used to label people who didn’t fit into society or actively opposed the government, unless they were grounded in science. In the Soviet Union, political dissidents were diagnosed with “sluggish schizophrenia,” and in the US, they diagnosed 1960’s youth with "Protest Psychosis.”

AUDIO: " What is your diagnosis of what's going on inside the Democratic Party?" "The problem with the Democratic Party is that they've got Trump Derangement Syndrome."

CHRIS AIKEN: That was Secretary of Health and Human Services Robert F Kennedy, Jr, carrying forth a time-honored political tradition. Back to 1974. Robert Spitzer successfully expunged homosexuality from the DSM, using the medical logic of the Washington School. The APA was so impressed with his feat that they handed over the entire DSM to Spitzer to update. Spitzer front-loaded the DSM committee with psychiatrists from the Washington School, and they drafted a brief list of validated psychiatric disorders called the Research Diagnostic Criteria. But the APA committee wouldn't accept such a short list. The DSM is a billing manual as well as a diagnostic one, and psychiatrists have to bill for whatever walks through the door, valid disorder or not. Spitzer achieved this compromise by moving the goal posts from validity to reliability. As long as everyone could agree on whether a patient met the criteria, the diagnosis was reliable and suitable for the new DSM. Validity, he hoped, would come later, as we investigated these new syndromes with brain imaging, genetics, and treatment studies.

KELLIE NEWSOME:

Psychiatry isn't the only field guilty of this diagnostic fluff. Here’s a few from the ICD:

V97.33XD: Sucked into jet engine, subsequent encounter.

R46.1: Bizarre personal appearance. There’s also “Other contact with cow,” “Walked into lamppost,” “Problems in relationship with in-laws…”

CHRIS AIKEN:

Patients present with all of those problems, but that doesn't mean they are valid disorders. When Spitzer released DSM-III, the original valid dozen was expanded to 265 disorders. Today, the manual lists more than 300. That’s all fine, and in fairness, many of these 300 are subtypes of the valid dozen, but what is missing in the manual is a clear statement about which are real and which are not. I suspect that if they did include such a list, insurers would hold it up and say, “We’re only going to pay for these 12.”

KELLIE NEWSOME:

We call that the missing page in DSM. Instead of a clear list, it’s buried in that boring part of the criteria that says you should not diagnose it if the symptoms are better explained by a more valid diagnosis. ADHD was not on the original list of valid diagnoses, but a case could be made for including childhood ADHD. In genetic studies, it is highly heritable, 75%; it has a characteristic treatment response and neuroimaging findings, and reliable follow up data. But others argue that most patients with ADHD have comorbidities, and it’s not clear that ADHD separates cleanly from other disorders, particularly among adults. Next week, we’ll look at how to separate it. One cause of ADHD symptoms is cannabis use, and today’s study looks at how long it takes the brain to recover from that. It's a systematic review by Valerio Ricci and colleagues in the American Journal of Addictions that pulls together 26 studies involving over 2,800 participants.

CHRIS AIKEN:

 Before we get into the study, let’s note a controversy. It’s not clear that cannabis causes ADHD, and it may be the other way around – people with ADHD are more likely to use cannabis. But we can say this – they aren’t using it to treat their symptoms. Cannabis was tested as a treatment for ADHD, and it didn’t work. In one trial, it actually made the cognitive symptoms worse, while the subjects thought it was making their cognition better. Multiple randomized, placebo-controlled trials find that acute THC or cannabis impairs working memory, attention, executive function, and verbal learning in a dose-dependent manner. So we do know that cannabis worsens cognition in the short-term; what’s missing are randomized trials of the long-term effects. Today’s update looked at the other side – what happens when cannabis is removed. Most of the studies here were cross-sectional or longitudinal, but three were randomized trials. They tracked people with neuroimaging, cognitive testing, and biomarkers, comparing cannabis users to controls or former users across abstinence periods, from 72 hours out to 8 years.

KELLIE NEWSOME:

 Verbal memory recovers fast. After quitting, teens and young adults showed gains within a week, and those held up through four weeks. Working memory took a little longer, two to three weeks, to improve. Attention and executive function lagged behind; one study found them still impaired at three weeks. Chronic heavy use down regulates CB1 receptors, and these returned to normal about a month after quitting.

CHRIS AIKEN:

 But the benefits of quitting were not the same across the lifespan. Adults who quit showed no lasting impairments, but teens who started using cannabis young didn't fare as well after quitting. One of the studies that confirmed this lasting problem with teen cannabis use was the famous New Zealand cohort – ironically, the same one we mentioned that found adult-onset ADHD. Hmmm… maybe some of these were the same subjects?

KELLIE NEWSOME:

 Another factor that predicted poor recovery was heavy use, more than five joints a week.  The bottom line. Cannabis can mimic ADHD, and those problems usually improve within a few weeks after stopping. But the patient’s subjective experience tells a different story – they may think it is helping. So try this in practice. Have them test their cognition objectively, like with the free Thinc-it app (that’s T H I N C – I T) or another computerized cognitive test. Then test again 4 and 8 weeks after quitting. Do you work with patients who don’t know they have an illness? It’s called anosagnosia, and in the August issue of the Carlat Report, we interview Xavier Amador, who developed a therapeutic way to engage them in treatment. Also, in the issue, pharmacotherapy and autism. Start your subscription with the promo code PODCAST for $30 off, and thank you for helping us stay free of industry support.