Trial By Error: Finally, Our Letter on Inflated Claims of FND Prevalence Is Published

By David Tuller, DrPH

For more than a year, I have been criticizing experts in the field of functional neurological disorder for misrepresenting the findings of a seminal study, in effect tripling the reported prevalence rate of the condition. These untrue claims about the Scottish Neurological Symptoms Study (SNSS)—specifically, that the prevalence of FND among outpatient neurology clinics was 16% and/or that it the second-most-common diagnosis–have been repeated in more than 50 papers in the last dozen year. (I did not discover this discrepancy by myself; a shrewd observer brought it to my attention.)

FND is the updated name for the Freudian entity called conversion disorder, although the two differ in diagnostic criteria. Conversion disorder required the presence of some identified trauma or psychological distress that purportedly triggered the symptoms; FND requires the presence of rule-in clinical signs said to be inconsistent with known neurological disease. In fact, long before these changes were included in 2013 in the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders, neurologists used such rule-in signs to identify cases of psychogenic or “functional” gait and motor disorders, non-epileptic seizures, and sensory deficits—the three main categories of conversion disorder.

In the SNSS, the rate of patients in the study identified as having conversion disorder symptoms was 5.5%, not 16%–209 out of 3781 participants. And at 5.5%, it was way down the list of presentations at outpatient neurology clinics.

However, the FND field since then has expanded its purview beyond these three main categories of conversion disorder to encompass a range of cognitive and related symptoms–part of an overall argument that FND is a common presentation rather than relatively rare. Of the 16% in the SNSS identified as having FND, two-thirds were identified as having “psychological” diagnoses and symptoms that could not fully be explained by neurological disease. However, as we noted in our letter, there is no evidence that any of these additional 10% met or could have met the rule-in criteria for FND. The fact that their symptoms could not fully be accounted does not justify a post-hoc, categorical reinterpretation of their diagnoses as FND.

Epidemiology graduate students at Berkeley would be strongly reprimanded for making such logical leaps absent any convincing evidence to support the assumption. It is perplexing to witness such problematic behavior from so many prominent investigators—especially because in other papers some of them have endorsed the much lower rate of FND prevalence, with no explanation of the discrepancy.

In seeking to counter this untrue account of the SNSS findings, I posted a series of blogs about the matter and appealed to multiple journals to pursue corrections. That ultimately led to a letter that was finally published this week by the journal NeuroImage: Clinical after a somewhat tortuous journey. I have posted it in full below. (It’s in the public domain, since the revised version was posted as a pre-print.) The journal has also posted a response from the authors; I’ll respond to that in a later post.


To the editor:

An article in NeuroImage: Clinical“Neuroimaging in functional neurological disorder: state of the field and research agenda” (Perez et al., 2021), cited a prominent paper (Stone et al., 2010) as evidence for the assertion that functional neurological disorder (FND) is the “2nd most common outpatient neurologic diagnosis.” Although studies have yielded varying FND prevalence rates, the claim that it is the second-most common diagnosis at outpatient neurology clinics represents an erroneous interpretation of the findings of the referenced 2010 paper.

FND is the current name for what was formerly called conversion disorder, the diagnosis previously given to patients believed to have psychogenic motor and gait dysfunctions, sensory deficits, and non-epileptic seizures. According to the 2013 edition of the Diagnostic and Statistical Manual of Mental Disorders and as noted in Perez et al, FND is not a diagnosis of exclusion but requires the presence of specific “rule-in” clinical signs believed to be incompatible with known neurological disease. Some of these clinical signs have long been used by neurologists and other clinicians to help them identify cases of conversion disorder.

Stone et al.’s (2010) paper was one of several arising from the Scottish Neurological Symptoms Study (SNSS). The study reviewed records from multiple outpatient neurology clinics and reported that 209 of 3781 attendees, or less than 6 %, received diagnoses compatible with conversion disorder–in other words, what would now be called FND. In terms of ranking, this group of patients—labeled in the SNSS as having “functional” symptoms or diagnoses–was far down the list. The study found higher rates of many other conditions, including headache (19 %), epilepsy (14 %), peripheral nerve disorders (11 %), miscellaneous neurological disorders (10 %), multiple sclerosis/demyelination (7 %), spinal disorders (6 %) and Parkinson’s disease/movement disorders (6 %).

Earlier this year, a paper in the European Journal of Neurology (Mason et al., 2023) cited a different SNSS paper (Stone et al., 2009) to support the assertion that FND prevalence at outpatient neurology clinics was 5.4 %—far lower than the percentage needed to be the “2nd most common” diagnosis. Moreover, the authors of another paper (Foley et al., 2022) have recently issued a correction for the same misstatement of FND prevalence from the SNSS findings as the one identified in Perez et al.

The assertion that the SNSS found FND to be the “2nd most common” diagnosis at outpatient neurology clinics is based on a parallel and commonly repeated claim that the study found the prevalence in these settings to be 16 % (e.g. Ludwig et al., 2018). That rate is almost three times the 5.4 % prevalence recently highlighted in Mason et al. The extra patients included in this greatly expanded FND category were another 10 % collectively identified in the SNSS as having “psychological” symptoms or diagnoses. These “psychological” patients fell into a range of clinical sub-categories, among them hyperventilation, anxiety and depression, atypical facial/temporomandibular joint pain, post-head injury symptoms, fibromyalgia, repetitive strain injury, and alcohol excess. The SNSS paper cited in Perez et al reported that a combined grouping of the patients with “functional and psychological” symptoms or diagnoses had a prevalence of 16 % but did not provide any evidence that the 10 % included under the “psychological” label met, or could have met, the explicit FND requirement for rule-in clinical signs.

FND is not synonymous with the broader “functional and psychological” category in the SNSS and should not be presented as if it were. The post-hoc reinterpretation of previously reported data in a way that conflates FND with other complex conditions—almost tripling its apparent prevalence in the process–is an example of the phenomenon known as diagnostic creep. In any event, the SNSS results are a matter of record. Whatever future studies might determine about FND rates, the published findings cited by Perez et al and addressed in this letter do not support either the claim that it is the “2nd most common” diagnosis in outpatient neurology clinics or the related claim that its prevalence at these venues is 16 %.


David Tuller (corresponding author)
Center for Global Public Health
School of Public Health
University of California, Berkeley
Berkeley, CA, USA

David Davies-Payne
Department of Radiology
Starship Children’s Hospital
Auckland, New Zealand

Jonathan Edwards
Department of Medicine
University College London
London, England, UK

Keith Geraghty
Centre for Primary Care and Health Services Research
Faculty of Biology, Medicine and Health
University of Manchester
Manchester, England, UK

Calliope Hollingue
Center for Autism and Related Disorders/Kennedy Krieger Institute
Department of Mental Health/Johns Hopkins Bloomberg School of Public Health
Johns Hopkins University
Baltimore, MD, USA

Mady Hornig
Department of Epidemiology
Columbia University Mailman School of Public Health
New York, NY, USA

Brian Hughes
School of Psychology
University of Galway
Galway, Ireland

Asad Khan
North West Lung Centre
Manchester University Hospitals
Manchester, England, UK

David Putrino
Department of Rehabilitation Medicine
Icahn School of Medicine at Mt Sinai
New York, NY, USA

John Swartzberg
Division of Infectious Diseases and Vaccinology
School of Public Health
University of California, Berkeley
Berkeley, CA, USA

3 thoughts on “Trial By Error: Finally, Our Letter on Inflated Claims of FND Prevalence Is Published”

  1. HURRAH! This calls for quite a celebration *!*!*!*!*!*!*!!!! And many thanks to David and all who signed the letter.

    I imagine this must now be a real GAME-CHANGER for all those with ME/CFS and long-covid, as well as those given spurious FND and MUS diagnoses by their neurologists for their unexplained symptoms.

    Why there hasn’t been a correction, goodness only knows. I seem to remember that one was offered or agreed in the first place, wasn’t it? Is the editor really content with this outcome? The letter makes it clear – the prevalence claim by the authors of that paper is not backed up by the reference that was supplied for it. If there is any doubt, one only has to go to this chapter on the epidemiology of FND - and read no further than the abstract. The chapter was co-written by one of the lead investigators of the SNSS. Both FND prevalence claims – around 6% and second most common – can’t be correct, can they? So if the editor and the paper’s authors consider the journal paper’s claim to be correct, can we assume that that epidemiology chapter will now be corrected?

    But WHY IS THIS KEY FOR PATIENTS WITH ME/CFS? What does it have to do with them? This is all about FND, isn’t it? Well, apart from reports that ME patients are now being diagnosed with FND instead, I would strongly urge the ME community to take notice for the following reason.

    David wrote -“These untrue claims about the Scottish Neurological Symptoms Study (SNSS)—specifically, that the prevalence of FND among outpatient neurology clinics was 16% and/or that it the second-most-common diagnosis–have been repeated in more than 50 papers in the last dozen year.”

    IN MORE THAN 50 PAPERS! So 50 or more papers should presumably now be corrected, OR that epidemiology chapter should be. One or the other, I’d suggest. And who were these 50 papers written by? A considerable number of the papers were co-authored by people who are also co-authors of this JNNP paper - – you may recall it, it’s the one that spouted forth about supposed anomalies with the ME/CFS NICE review process. So, assuming that having an accurate scientific record matters enough to them, I’d have thought that these authors should now be either rushing to get their FND papers corrected or pressing really hard for that epidemiology chapter to be put right. One or the other.

    And it’s not just journal papers that have put out this questionable FND prevalence information, it looks to me like it’s also been employed in the context of legal work and by charities and in petitions to governments. This perhaps demonstrates how important it is to get epidemiological information because it spreads quite readily. David mentioned the “epidemiology graduate students at Berkeley”. I can’t help but think that this letter and this story could be of great educational value for students all over to teach them that epidemiological information really does matter, that it’s important not to get it wrong and that to treat it casually might not be a great idea.

  2. Oh, silly me, I nearly forgot. Why do I think this is key in relation to Long covid as well as ME? If researchers have failed to get simple epidemiological information correct in their own field, why on earth should we trust them on anything else?

    (A reminder of some of what’s been going on long covid-wise - and -,associated%20with%20an%20underlying%20condition. )

  3. See the first paragraph here - (reference 3) . There’s quite a line-up of co-authors for that paper. Will it be corrected?

Comments are closed.

Scroll to Top