Overdiagnosis in Psychiatry
The Carlat Psychiatry Report, Volume 19, Number 9, September 2021
Joel Paris, MD
Professor emeritus at McGill University, and senior psychiatrist and research associate at Sir Mortimer B. Davis-Jewish General Hospital. Dr. Paris is the former editor of the Canadian Journal of Psychiatry and the author of 25 books including Overdiagnosis in Psychiatry, 2nd ed. (Oxford University Press; 2020).
Dr. Paris has disclosed no relevant financial or other interests in any commercial companies pertaining to this educational activity.
TCPR: Which disorders are the most overdiagnosed in psychiatry?
Dr. Paris: At the top of the list are major depression, PTSD, bipolar II, adult ADHD, and—this one is a recent addition because it used to be underdiagnosed—borderline personality disorder. Each of these conditions may be underdiagnosed in some settings as well, but they have also been pushed toward diagnostic expansion.
TCPR: That sounds like a lot of my patients. What are we doing wrong?
Dr. Paris: Part of the problem is not using the DSM criteria. Many clinicians are rushed, and the criteria are difficult to memorize. Nearly half of patients whose primary care provider diagnosed them with depression don’t actually meet the DSM criteria for it (Mojtabai R, Psychother Psychosom 2013;82(3):161–169). But even if the criteria are used, the DSM still has a pretty low bar for major depression. In reality, depression is a universal phenomenon that is on a continuum, and you make a slice somewhere. But where you make that slice is not really evidence based.
TCPR: When you say “a low bar for depression,” what do you mean?
Dr. Paris: The two-week cutoff was a big mistake. Lots of people have depression that self-resolves after a few weeks. If the DSM had said six or seven weeks, I think we’d be in less trouble. Another problem is that the DSM lumps both mild and severe cases under the diagnostic umbrella of “major depressive disorder.” That was not the case before 1980 with DSM-III, and there are reasons to separate mild and severe cases. We know, for example, that antidepressants don’t separate very well from placebo in mild cases—their efficacy is only undisputed in severe depression (Fournier JC et al, JAMA 2010;303(1):47–53). Treating everyone who meets criteria for major depression with an antidepressant is simply not evidence based. The UK has stepped away from this with the NICE guidelines, which do not recommend antidepressants for mild depression. But the APA and Canadian guidelines still allow it.
TCPR: Do clinicians in the UK use psychotherapy for mild depression?
Dr. Paris: Yes, and patients there have better access to it because the British hired psychologists to bring cognitive behavioral therapy (CBT) for anxiety and depression into the National Health Service.
PTSD, Bipolar Disorder, and ADHD
TCPR: How is PTSD overdiagnosed?
Dr. Paris: PTSD has a big problem with criterion A, which requires the presence of a trauma. Originally the patient had to experience or witness the trauma, but now just hearing about a traumatic event is enough to meet the criterion. And PTSD is a diagnosis beloved not only by clinicians, but by patients too. They actually want to have PTSD because it means somebody else is at fault, not them, and they’re a victim.
TCPR: How is bipolar disorder overdiagnosed?
Dr. Paris: The problem is not with standard bipolar I. That kind of mania is unmistakable. But bipolar II only came in with DSM-IV, and the criteria are kind of fuzzy. I think clinicians reach for their prescription pad when they see mood swings of any kind because they think they’re seeing a case of bipolar II. Mark Zimmerman showed that patients who screen positive for bipolar disorder on the Mood Disorder Questionnaire are just as likely to have borderline personality disorder as they are to have bipolar on closer examination (Zimmerman M et al, J Pers Disord 2019;33(4):533–543). Mood swings are a key feature of borderline personality disorder, and that can confuse the picture for many clinicians.
TCPR: And ADHD?
Dr. Paris: That diagnosis has gone up dramatically in the last 20 years in adults and children. Ten times as many people are taking stimulant drugs compared to a couple of decades ago. Nearly one in five adolescent boys are diagnosed with ADHD (Visser SN et al, J Am Acad Child Adolesc Psychiatry 2014;53(1):34–46). Adult ADHD used to be rare, but in the National Comorbidity Survey the rate was one in 23, and that was while using the childhood-onset requirement, which is often ignored in practice (Kessler RC et al, Am J Psychiatry 2006;163(4):716–723). Previously, symptoms had to start before age 7 to make a diagnosis, and then DSM-5 extended that to age 12.
TCPR: We are seeing adults present with symptoms of ADHD but no evidence of the disorder in childhood. What could be causing that?
Dr. Paris: Patients with “adult ADHD” have a lot of psychiatric comorbidities that could explain their cognitive problems. Problems with attention and executive functioning are well documented in mood, anxiety, personality, sleep, and substance use disorders (Paris J. Overdiagnosis in Psychiatry. New York, NY: Oxford University Press; 2020). Terrie Moffitt’s group followed a New Zealand birth cohort for 40 years and found that most people who had symptoms of ADHD in adulthood did not have those symptoms as children (Agnew-Blais JC et al, JAMA Psychiatry 2016;73(7):713–720). This suggests that the symptoms do not have the same etiology as true ADHD.
TCPR: The problem with age of onset in that study does cast doubt on the concept of adult ADHD. But the study ended up having a different effect. It generated a lot of headlines announcing the discovery of a new diagnosis: adult-onset ADHD.
Dr. Paris: Well, Dr. Moffitt did open the door for this in her article, but I would say, “Why call it ADHD?” In the Moffitt study, most of the patients with “adult-onset ADHD” had a history of mood, conduct, or substance use disorders that might have explained their adult ADHD symptoms. Another problem with ADHD is the inclusion of the inattentive subtype, which took place with DSM-IV in 1994.
TCPR: What is the problem with inattentive ADHD?
Dr. Paris: It opened the door to overdiagnosis. Now almost anybody who is troubled can come up with a story about their inattention, and they do. Hyperactivity is more observable, so it’s harder to overdiagnose, and our research is most robust with the hyperactive types. We know what these patients are at risk for later on—things like substance use and antisocial behavior.
Underdiagnosis in Schizophrenia
TCPR: If some disorders are overdiagnosed, are others underdiagnosed?
Dr. Paris: Schizophrenia is probably underdiagnosed. I think the problem there is not with the definition (although the overlap with bipolarity is still pretty controversial), but with its difficult prognosis. I think clinicians would rather call it schizoaffective disorder, which has a better prognosis because it lacks the negative symptoms, or just nonspecific psychosis. However, there has been a movement to recognize early signs of schizophrenia, and that is changing things.
TCPR: How so?
Dr. Paris: The idea is that if you recognize prodromal signs and intervene early, you may be able to obviate that bad prognosis. Early intervention proponents haven’t proven their case completely at all, but they have helped shift the perspective. Interestingly, the evidence is stronger for CBT than it is for antipsychotics when it comes to preventing schizophrenia in youth at risk (Nelson B and McGorry P, Child Adolesc Psychiatr Clin N Am 2020;29(1):57–69).
Borderline Personality Disorder
TCPR: What is lost by missing a personality disorder?
Dr. Paris: Outcomes are not as good. Whether it’s pharmacotherapy, psychotherapy, or even electroconvulsive therapy, a lot of research shows that people with personality disorders do not do as well when given standard treatments for other disorders. These patients require more specialized methods.
TCPR: Which personality disorders are most important to recognize in practice?
Dr. Paris: Borderline and antisocial. Those are the only ones that have serious research behind them. We have a good etiological theory for borderline personality disorder. Emotional dysregulation is a heritable trait that is then amplified by an invalidating environment. There’s more to the theory, but that explanation captures its essence, and it was one of Marsha Linehan’s greatest contributions.
TCPR: What is an invalidating environment?
Dr. Paris: It refers to growing up in a situation where your family isn’t interested in your emotions. “Emotional dysregulation” means a patient is unusually sensitive—they have “thin skin” and high neuroticism, where everything bothers them and they can’t calm down. And that is a highly heritable trait. If they grow up in an environment where somebody around them says, “I understand how you feel. You have every right to feel that way. Let’s figure out how you can get past this. You know how to calm yourself and think of a way to solve the problem,” then it doesn’t turn out so bad. But what patients with borderline personality disorder describe to me is, “Nobody was interested in my feelings. I was told to buck up, get over it, and not make a big deal out of it.”
TCPR: So it’s not necessarily trauma that causes it.
Dr. Paris: Correct. One of the great myths about borderline personality disorder is that it’s due to trauma. While trauma is common in borderline personality disorder, fewer than half the cases actually have a significant history of trauma. Often, when you take a closer look at a patient’s history, you’ll find they had evidence of a personality disorder before the trauma, and borderline personality disorder affects judgment, impulsivity, and relationships in ways that can put people at risk for trauma (Paris J. Treatment of Borderline Personality Disorder. 2nd ed. New York, NY: Guilford Press; 2020).
TCPR: Some have suggested that cyclothymia is the underlying temperament behind borderline personality disorder.
Dr. Paris: I think emotional dysregulation is a more accurate term. Cyclothymic disorder involves frequent mood cycles—hypomania, depression, mixed states. But moods last longer than emotional dysregulation. Patients will describe it like this: “I get upset and I blow sky high. It takes me hours to calm down.” And this is nothing like hypomania, which is a consistent state of high energy and irritable or elevated mood for four days or more. With emotional dysregulation, the symptoms usually come on as a direct reaction to an interpersonal stress. This persists for a few hours and then the patient sleeps it off, and the next day something else happens.
TCPR: You mentioned that borderline personality disorder used to be underdiagnosed, but now that’s changing.
Dr. Paris: Yes, and this gets back to prognosis. People used to think borderline personality disorder was incurable, but now that we have effective treatments for it like DBT and other psychotherapies, I’m having to tell patients, “No, you actually don’t have borderline personality disorder” and they’re disappointed.
TCPR: How would you explain to a patient that they don’t have a psychiatric disorder or that their problems are not severe enough to warrant medication?
Dr. Paris: Some people might be relieved to hear that, while others might be outraged. I think it depends on how they see themselves. One of my colleagues at McLean Hospital described something she called “emotional hypochondriasis” where people feel that their psychology is sick and they hold onto that as a kind of identity. So this is a complicated question. I tell residents, “Some people are going to walk out of here mad, and you’ll have to get used to it.” But I don’t have a simple answer.
TCPR: Thank you for your time, Dr. Paris.