Trial By Error: Suzanne O’Sullivan’s “Psychosomatic” Mis-Diagnoses

By David Tuller, DrPH

*This is a crowdfunding month at University of California, Berkeley. If you appreciate my work and would like to make a donation (tax-deductible to US taxpayers) to the university in support of my position, here’s the link: https://crowdfund.berkeley.edu/project/46120

Neurologist Suzanne O’Sullivan has been conducting a media blitz for her new book, “The Age of Diagnosis: How Our Obsession with Medical Labels Is Making Us Sicker.” The Guardian ran a lengthy excerpt in early March. O’Sullivan appeared on the popular podcast Freakonomics. Multiple publications have published profiles of her, with credulous assessments of her work.

O’Sullivan has long been a major promotor of psycho-behavioral explanations for complex illnesses. Her previous books in this vein–including It’s All in Your Head: True Stories of Imaginary Illness, and The Sleeping Beauties: And Other Stories of Mystery Illness—were bestsellers. In the SARS-CoV-2 era, it should not be surprising that she has now scooped up Long Covid and incorporated it into her expansive worldview. “As no one saw doctors during the pandemic, it’s inescapable that the period was a melting pot for psychosomatic conditions,” she explained in one recent interview.

The new book’s release has coincided with a public discussion in the UK of labor shortages, which have been blamed in part on the numbers currently designated as unable to work because of illness—often because of Long Covid. This discussion is itself linked to the government’s push to reduce unemployment and/or disability benefits and its assertion that many of those currently not working should in fact be working. O’Sullivan’s framework seems likely to provide perceived confirmation of the arguments for cutting social welfare payments as well as political cover for those seeking to promote that strategy.

Similar discussions regarding which people should be considered deserving of publicly funded benefits are occurring in Australia, where I just spent four weeks, and the US. Understandably, this debate has not been received well among many people with disabilities, including those with ME, ME/CFS, Long Covid, and other debilitating conditions whose underlying mechanisms remain unclear.

While “The Age of Diagnoses” has received some positive reviews for O’Sullivan’s storytelling abilities, the praise has not been universal. In one telling example, BJGP Life, an online offshoot of the British Journal of General Practice (BJGP), has published a refreshingly critical perspective. Elke Hausmann, a GP in Derby who has herself experienced prolonged postural orthostatic tachycardia syndrome and other symptoms after COVID-19, pushed back vigorously against O’Sullivan’s “curious characterisation of Long Covid as a psychosomatic condition.”

Here’s a compelling selection from Hausmann’s review:

“O’Sullivan suggests that people don’t like to be told that their physical symptoms might have an underlying psychological explanation. Her idea is that we simply ignore this possibility because mental health conditions are still stigmatised in society, and we don’t want to be seen through the lens of that stigma. Proponents of the psychosomatic explanation of Long Covid (and the same goes for ME), always make out as if we were really the ones stigmatising patients with mental health disorders, by inferring that mental health conditions are somehow lesser, when we represent our illness as ‘physical’…In fact, rather than being resistant to psychological explanations and therapeutic interventions for Long Covid, we generally tend to be quite open to them. Until, that is, experience shows us that this route to a cure is a dead end.”

**********

Guardian excerpt highlights Paul Garner as “poster patient”  

I have read The Guardian’s excerpt from “The Age of Diagnosis.” (I haven’t read the book itself.) O’Sullivan’s thesis is that we that we are being swamped by a tsunami of “over-diagnosis”–as if all those once dubbed the “worried well” have now received specific medical labels for their assorted complaints, aches and pains. In this view, people become so fixated on these labels that they keep themselves chronically sick as a way of affirming and sustaining their illness identity.

The problem with this thesis isn’t that what O’Sullivan describes doesn’t ever happen. Every time a diagnosis relies on symptoms rather than accurate biomarkers, some people with something else or nothing at all will nonetheless be deemed to have the illness. That’s unavoidable. The problem, or one problem, is that she seems to apply her thesis indiscriminately to broad disease categories as well as to the individuals whose cases she examines in detail. Nonetheless, I take little issue with some of Dr O’Sullivan’s general assertions—like the following:

“It could be that borderline medical problems are becoming iron-clad diagnoses and normal differences are being pathologised. These statistics [indicating increased numbers of diagnoses] could indicate that ordinary life experiences, bodily imperfections, sadness and social anxiety are being subsumed into the category of medical disorder.”

O’Sullivan is right. It “could be” that some of what she suggests is, in fact, happening. It could be, as she writes, that the increase in diagnoses means that, at least in part, “we are attributing more to sickness.” But if we are indeed attributing more illness to sickness, the question then is: How much more? And is that necessarily a bad thing, as O’Sullivan believes?

O’Sullivan blames much of the “over-diagnosis” on the metastasizing Diagnostic and Statistical Manual (DSM) of Mental Disorders, the so-called psychiatric Bible. The DSM has grown dramatically from its first edition in 1952 to its most recent, DSM-5, published in 2013, with a proliferating number of disorders. As O’Sullivan accurately notes, the categorizations of various conditions have shifted and expanded significantly, with boundaries that are in many respects porous and arbitrary. No argument with her on that score.

In its discussion of Long Covid, however, the excerpt includes a section presenting Paul Garner as the “poster patient” for demonstrating that the illness is largely or in many cases a result of fear, anxiety, and related emotional or psychological states–and therefore amenable to self-cure. To anyone who has followed the Garner saga since its inception in 2020, O’Sullivan’s decision to highlight his “recovery” narrative exposes deep flaws in her storytelling methods and destroys her credibility. It also exhibits her own case of confirmation bias. A review of Garner’s public statements about his condition documents the following: He experienced a nasty bout of post-viral illness that appears to have self-resolved within six months of his SARS-CoV-2 infection.

After Garner got better, he attributed his recovery to an intervention that sounded very much like the Lightning Process–and, in particular, to the role played by his powerful manly cognitions. He also went out of his way to trash members of the ME community for having offered him advice about pacing, post-exertional malaise, and the like, during his prior months of suffering. In reality, it doesn’t appear that he was suffering from ME or ME/CFS, and certainly not the kind of Long Covid that persists for years. The evidence is that he was already back to a regular exercise schedule by the time he received the advice that he later claimed led to his cure.

Garner has since established himself as an evangelist for so-called “mind-body” solutions to Long Covid and other major illnesses. In this mode, he has suggested that others who fail to recover with his proposed self-empowerment approach have themselves and their negative outlooks to blame. Medical colleagues, like O’Sullivan, have contributed to the hype by disseminating his story and endowing it with unwarranted professional authority.

O’Sullivan transmits Garner’s self-serving interpretation of events without an iota of skepticism or critical thinking. Here’s her account:

“Viruses cause fatigue in order to make people rest, which promotes recovery. But, in Garner’s case, his recovery had gone awry because he inadvertently conditioned his body to stay tired. Garner realised he had to retrain his brain to react differently to the fatigue if he was to get better.

“‘I suddenly believed I would recover completely,’ he writes. ‘I stopped my constant monitoring of symptoms. I avoided reading stories about illness and discussing symptoms, research or treatments by dropping off the Facebook groups with other patients. I spent time seeking joy, happiness … and overcame my fear of exercise.’ By the end of 2020 he had made a full recovery.”

Come on. How does O’Sullivan know that Garner failed to recover because he “inadvertently conditioned his body to stay tired”? That’s Garner’s argument–but what evidence is there for it? Or for the staterment that he recovered because he learned to “retrain his brain”? None. But O’Sullivan buys it without question.

O’Sullivan’s gullibility when confronted with this self-mythologizing tale indicates that she suffers from impaired judgment and limited psychological insight–at least when someone’s story aligns with her own prejudices and inclinations. A large part of O’Sullivan’s narrative strategy involves offering similarly detailed, and generally readable, accounts of people’s hard-to-diagnose illnesses. Given her pre-existing psychosomatic template, however, these individual portraits cannot be taken at face value. Nor do they add up to proof of the book’s larger and frequently misguided thesis regarding mass psychosomatic illness.

In short, O’Sullivan knows how to tell compelling stories–and, it seems, compelling fairy tales as well.

8 thoughts on “Trial By Error: Suzanne O’Sullivan’s “Psychosomatic” Mis-Diagnoses”

  1. What if….. WHAT IF…..the psychosomatic model that doctors, (O’Sullivan included), have promoted has caused, or even contributed in part, to the rise in economic inactivity due to ill health that O’Sullivan and others now appear to attribute to people’s false beliefs about their symptoms? There’s good reason to believe that it might have done – the rise in economic inactivity began years before the pandemic and has risen much more in women than in men (see here -https://www.bankofengland.co.uk/speech/2022/may/michael-saunders-speech-at-the-resolution-foundation-event figure 6) . How can this be explained? It seems that the Society of Occupational Medicine was struggling to explain it (see here- https://www.som.org.uk/sites/som.org.uk/files/SOM_Deep_Dive_Research-compressed.pdf page 19 of 64) and suggested that factors affecting women’s long term health needed investigating to see if anything was causing a problem in relation to female-specific illnesses.

    Now I’m NOT jumping to any conclusions, I trained as as scientist and like to have good evidence before considering adopting any particular stance. I’m simply saying that if all these economists and health bodies are scratching their heads to try to work out what has cause this unexplained rise in economic inactivity that especially affects women, they might want to look at a medical model – the NHS’s model on medically unexplained symptoms – that I’d say has repeatedly painted women as the main offenders of wasting healthcare resources through their somatization. (I think I’m entitled to use that term “offenders” since it has been used before – see here -https://recipeforworkforceplanning.hee.nhs.uk/Portals/0/HEWM_LinksAndResources/Psychiatrist%20in-depth%20review_Main%20report%20-%20Nov%202014.pdf?ver=KSIcjYLQDeP8YR4Z5lJY9w%3D%3D – see section 5.7). A July 2014 Department of Health ‘Positive Practice Guide’ for the IAPT programme indicated that women were 3 to 4 times more likely than men to present with functional symptoms. (There’s nothing quite like producing a self-fulfilling prophecy, is there? ).

    There is evidence, that David has flagged up (see here -https://virology.ws/2025/03/31/trial-by-error-new-study-documents-iatrogenic-harm-from-perceived-psychosomatic-and-psychiatric-misdiagnoses-of-rheumatic-diseases/), that women with lupus and other autoimmune diseases are being badly let down by being misdiagnosed with medically unexplained symptoms or other mental health diagnoses, so WHY, OH WHY, is nobody raising the psychosomatic MUS/functional model, (call it what you will), as a potential/likely cause of the rise in economic inactivity that is affecting the UK’s economy so badly now? I’m not saying that I’m definitely right, I’m saying this should be looked that, that it really needs to form part of the public debate.

    We know that there’s been a determined effort to confine Long Covid to the MUS/functional illness/psychosomatic category – see here -https://virology.ws/2022/12/14/trial-by-error-is-the-long-covid-phenomenon-an-expression-of-psychosocial-distress/ and I understand that Long Covid affects more women than men, (although it’s a man who I know who’s still suffering the effects of covid years after contracting it – no short-lived post-viral illness for him even though he wasn’t hospitalized and has shown no inclination towards adopting that Long Covid label ). To my mind, Long Covid sufferers should be pressing this point. I can’t quite understand why they’re not, unless it’s that their campaign group/s are being led by doctors who don’t want to believe that they could have been badly misled by their medical colleagues about MUS and somatization for fear that they may no longer be able to trust anything that they’re taught by those they’ve previously regarded as wholly trustworthy.

  2. It seems to me that many doctors err by mistaking theories for truth and by portraying them as such. (I suspect that may be due to how doctors were trained at medical school.) To me, theories remain theories until they are DISproved rather than proved. We can say that good quality evidence would appear to strongly support a particular theory, but we can’t say more than that. I’m asking that scientists and economists disprove – find strong evidence against – my theory that “the UK’s rise in economic activity due to ill health could be caused wholly or in part by the widespread promotion and adoption of the psychosomatic model within the NHS over the last decade or so, to the particular detriment of female patients.”

  3. Auntie Jellyfish

    Garner seemed (at least on social media) unusually panicky right from the off about not recovering immediately – I don’t think I’ve ever witnessed anyone quite so anxious, quite so driven in a comparable situation (admittedly early pandemic was a weird and not un-panicky time). And I wonder if his behaviour since is out of embarrassment at that.

  4. Marion Brown commented:
    There is something else too
    ….adverse effects of very commonly prescribed medicines
    …https://www.bmj.com/content/380/bmj.p221/rr-0

    Indeed, I’d say that this is another theory that deserves attention when we’re told that there has been a rise in mental health problems in association with the rise in economic inactivity due to ill health. That apparent rise in mental health problems could be due in part to people with MUS being labelled/coded with mental health diagnoses instead of physical health ones but it could also be due to more people seeking psychiatric help to mentally cope in our extremely stressful world. It’s possible that this theory regarding adverse effects of prescribed medicines feeds into the MUS theory, and vice versa. If people with MUS are told they have mental health problems instead of physical ones, that could result in them receiving inappropriate psych medicines that could cause adverse side-effects and withdrawal effects that cause them more medically unexplained symptoms, perpetuating or exacerbating the problem. This could represent a vicious cycle that both people (originally) presenting with either physical or mental health problems could be drawn into. Attributing physical health problems/symptoms to mental health issues is not risk-free; rather it could leave many patients exposed to significant iatrogenic harm, not least from underrecognized side effects and withdrawal effects of prescribed medication.

  5. Wrt the gender bias, SFN is a case in point. “..SFN typically presents in 2 main clinical patterns: length-dependent SFN (LD-SFN) or non-length-dependent SFN (NLD-SFN). LD-SFN starts in the distal extremities of the legs and hands symmetrically, before ascending upwards; NLD-SFN symptoms are asymmetric, patchy, or involve the face and trunk prior to the involvement of the extremities. The proportion of NLD-SFN to LD-SFN has been approximated at 1:4 [22] to 1:3 [25], but the exact proportion is unknown, as patients with NLD-SFN tend to be under-recognized and misdiagnosed as having a psychogenic disorder [25]. In particular, NLD-SFN is more common in women, presents at a younger age, and has a greater association with immune-mediated conditions than LD-SFN”.

    (Chan Amanda C. Y. et al. Biomarkers in small fiber neuropathy. Review. Exploration of Neuroprotective Therapy. December 2022. Peer Reviewed No Impact Factor yet but was accepted for inclusion in Scopus In October 2024.
    https://doi.org/10.37349/ent.2022.00033)

Leave a Comment

Your email address will not be published. Required fields are marked *