Over-Diagnosing Mental Illness
Is the tide starting to turn?
“Every time we have a mental health awareness week my spirits sink. We don’t need people to be more aware. We can’t deal with the ones who already are aware.”
-Simon Wessely
An issue that seems to be getting more and more attention lately is the over-diagnosis of mental illness. In the last few weeks, I’ve come across several articles in major publications dealing with the topic - one from the U.K. and one from the U.S. Here’s a brief excerpt from each, to give you the flavor of what looks like an increasingly important conversation.
The first is from an article published in The Economist. (Economist articles usually don’t have bylines, so I don’t know who the author is.) Among other things, the article touches on the perverse incentives we’ve created for people to interpret everyday stresses as psychological disorders, and the idea that broadening the definition of mental illness makes it harder to help individuals with the most severe psychological problems.
For all their good intentions, campaigns intended to raise awareness are leading some people to conflate normal responses to life’s difficulties with mental-health disorders. Special treatment creates incentives for people to seek diagnoses and to medicalise problems unnecessarily. The need to treat people with milder conditions competes with care for those who have the most severe ones.
Start with the idea that mental health has become a catch-all term. The sheer proportions of people who say they have a disorder is a red flag. Some 57% of university students claim to suffer from a mental-health issue; over three-quarters of parents with school-age children sought help or advice over their child’s mental health in 2021-22. In surveys Britons increasingly describe grief and stress as mental illnesses, redefining how sickness is understood...
People have incentives to label mild forms of distress as a disorder. In 2022 more than a quarter of 16- and 18-year-olds in British schools were given extra time in official exams because of a health condition. Evidence of a mental-health problem can unlock welfare payments... Firms may prefer to label stress a disorder rather than deal with the consequences of acknowledging that working conditions are poor…
Britain needs to avoid the mass medicalisation of mild forms of distress. Funnelling people into an overstretched health-care system is having predictable effects.
The second excerpt is from a recent New York Times article by clinical psychologist Darby Saxbe. The article touches on some of the same themes as the Economist piece, including the pathologizing of everyday stresses and strains. But it also hits some new notes, such as the iatrogenic effects of school-based and online mental-health awareness programs, the prevalence of the therapeutic mindset among teenagers, and the dangers of self-diagnosis.
Recent studies have found that several [mental-health awareness] programs not only failed to help young people, they also made their mental-health problems worse…
[B]y focusing teenagers’ attention on mental health issues, these interventions may have unwittingly exacerbated their problems. Lucy Foulkes, an Oxford psychologist, calls this phenomenon “prevalence inflation” - when greater awareness of mental illness leads people to talk of normal life struggles in terms of “symptoms” and “diagnoses.” These sorts of labels begin to dictate how people view themselves, in ways that can become self-fulfilling.
Teenagers, who are still developing their identities, are especially prone to take psychological labels to heart. Instead of “I am nervous about X,” a teenager might say, “I can’t do X because I have anxiety” - a reframing that research shows undermines resilience by encouraging people to view everyday challenges as insurmountable.
It’s generally a sign of progress when diagnoses that were once whispered in shameful secrecy enter our everyday vocabulary and shed their stigma. But especially online, where therapy “influencers” flood social media feeds with content about trauma, panic attacks and personality disorders, greater awareness of mental health problems risks encouraging self-diagnosis and the pathologizing of commonplace emotions.
One objection to this line of thought is that, if taken too far, it could lead us to deny the reality or importance of any but the most extreme psychological difficulties. If psychological services could help ease people’s everyday stresses and strains, why deprive them of that help just because they’re not quite miserable enough to meet traditional definitions of mental illness? These are reasonable concerns. But they need to be weighed against parallel concerns about pushing too far in the other direction. After all, over-diagnosis isn’t cost free. Not only could it harm the mental health of those caught in its net, but it could siphon resources away from the people who need them most.
Further Reading
Lucy Foulkes and Argyris Stringaris have an open-access paper in BJPsych Bulletin titled “Do No Harm: Can School Mental Health Interventions Cause Iatrogenic Harm?” Here’s an excerpt:
To date, there has been very little investigation into why harms such as symptom increase occur in school-based mental health interventions. Here, we speculate that one relevant mechanism might be that interventions inadvertently encourage adolescents to ruminate on their negative thoughts and emotions. Indeed, qualitative studies highlight that although some adolescents find school mental health interventions helpful, others say the focus on negative thoughts made them feel more stressed and unhappy. Relatedly, if an adolescent is encouraged to label their negative thoughts and emotions with psychological or psychiatric terminology in school interventions, this might lead to changes in self-concept (e.g. ‘I have anxiety’) and changes in behaviour (e.g. avoidance) that ultimately increase distress and other symptoms in some adolescents.
The unique developmental features of adolescence may also be relevant. Adolescents are especially susceptible to peer influence, and school-based mental health interventions commonly occur in groups. It is well established that adolescents can influence each other's negative moods and can learn problematic behaviour from each other (sometimes known as ‘deviancy training’). It is therefore a reasonable hypothesis that encouraging adolescents to discuss negative thoughts, feelings and behaviours in group settings, as is so common in school-based interventions, could lead to an increase in these experiences in others via peer influence and social learning.
Scott Alexander (Astral Codex Ten) has a piece where he argues that, at least for some mental health conditions, awareness campaigns may exacerbate or even create the problems they’re supposedly shining a light on. Would we be better off, he asks, with Mental Health Unawareness Campaigns?
[D]oes naming and pointing to a mental health problem make it worse? This was clearest in Hong Kong, where a seemingly very low base rate of anorexia exploded as soon as people started launching mental health awareness campaigns saying that it was a common and important disease (as had apparently happened before in Victorian Europe and 70s/80s America). But it also showed up in the section on how increasing awareness of PTSD seems to be associated with more PTSD, and how debriefing trauma victims about how they might get PTSD makes them more likely to get it... A full treatment of this theory would go through the bizarre history of conversion disorder, multiple personality disorder, and various mass hysterias, tying it into some of the fad diagnoses of our own day…
Of course, the null hypothesis is that there are lots of people suffering in silence until people raise awareness of and destigmatize a mental illness, after which they break their silence, admit they have a problem, and seek treatment. I am slightly skeptical of this, because a lot of mental health problems are hard to suffer in silence - if nothing else, anorexia results in hospitalizations once a patient’s body weight becomes incompatible with healthy life. Still, this is an important counterargument, and one that I hope people do more research into…
I find myself imagining a culture that holds Mental Health Unawareness Campaigns. Every so often, they go around burning books about mental illness and cancelling anyone who talks about them... Whenever there is a major natural disaster, top experts and doctors go on television reassuring everyone that PTSD is fake and they will not get it. Whenever there’s a recession or something, psychiatrists tell the public that they definitely won’t get depressed, since “depression” only applies to cases much more severe than theirs, and if they feel really sad about losing all their money then that’s just a perfectly normal emotion under the circumstances…
I’m not sure if this culture would have more or less mental illness than our own. But we’re trying the opposite experiment now, so I guess we’ll get to see how that turns out.
I appreciated your cautionary question near the end because it seems to me that we can do better than just return to the old responses of “man up” and “you need to toughen up sweetie”. One approach might be to teach young folks (and ourselves) how to live a philosophy of resilience with modern Stoicism being one very promising model. Instead of a reactive approach to mental illness (which can activate a counterproductive process of auto-suggestion) we take a proactive approach of providing tools and habits of thinking that anyone can benefit from in an uncertain, hard and often cruel world.
I thought this quote also made an important point that can be applied to governments: “Firms may prefer to label stress a disorder rather than deal with the consequences of acknowledging that working conditions are poor…” Perhaps medicalization of stressors is one response to governments that refuse to take a more active role in creating a higher quality of life for those being left behind.
So the idea is that mental illnesses are over-diagnosed - a claim that many people love to make. This would mean that many people are diagnosed with mental illness than they should be, presumably because some people receive a diagnosis even though they do not have that mental illness.
But we can only say there is an over-diagnosis problem if we know the actual number of people who suffer from mental illness and discover that the number of people with a diagnosis exceeds that. So I ask: where's the data? Because if we do not have data to show that many people receive diagnoses when they shouldn't, we cannot talk about over-diagnosis.