By David Tuller, DrPH
Several colleagues and I recently wrote to the journal NeuroImage: Clinical to request a correction in a 2021 article about functional neurological disorder (FND). The article included the false claim that a seminal 2010 study found that FND was the second-most-common diagnosis at outpatient neurology clinics. In fact, FND—called conversion disorder at the time—was the eighth-most-common diagnosis, with a prevalence of 5.5%. The #2 claim was based on the notion that the prevalence was almost triple that, at 16%.
The lead and senior authors of the paper, David Perez and Selma Aybek, are experts in the field. The many co-authors included a virtual who’s who of FND investigators. The journal responded within days to let us know that the authors had agreed to a correction; the editorial team asked for our input on the proposed language. We took this quick response as a welcome acknowledgement that the argument for a correction was air-tight and indisputable. We were, however, dissatisfied with the authors’ proposal, and we offered alternate language.
Last week, we learned that the authors had not taken our advice but had nonetheless elaborated on their correction. We still believed it fell short of what was appropriate, given the substantive nature of the initial error, but accepted the new version as a done deal. I described the correction but did not publish the text because it was not ours to publish.
Unfortunately, the article in question is only one of dozens of papers by leading FND experts that have also cited Stone et al to argue that FND was the second-most-common diagnosis and/or had a prevalence at outpatient neurology clinics of 16%. To start the process of seeking corrections in these additional papers as well, I have sent a letter to Perez and Aybek. The letter notes that we found at least nine additional papers for which one of them served as lead or senior author that contained similar mis-citations of Stone et al.
Here’s the letter:
Dear Dr Perez and Professor Aybek–
On behalf of our group, I want to thank you both for agreeing to correct an erroneous statement in a 2021 article published by the journal NeuroImage: Clinical. The paper, “Neuroimaging in functional neurological disorder: state of the field and research agenda,” cited Stone et al, a 2010 study, to assert categorically that functional neurological disorder (FND) was the second-most-common diagnosis at outpatient neurology clinics.
Stone et al was one of several papers emerging from a research project called the Scottish Neurological Symptoms Study (SNSS). The assertion that FND was the #2 diagnosis was based on the parallel claim that it was identified in 16% of study participants.
As my colleagues and I noted in our letter to the journal, these twin prevalence claims regarding the SNSS are unsustainable, given the actual findings. We presume the self-evident discrepancy between the statement in the paper and the evidence from the SNSS is why the journal’s editorial team responded within days to our request for a correction. (I have cc’d the co-signatories of the letter to the journal.)
As should be clear to anyone reading Stone et al and other papers from the SNSS, the research reported a prevalence rate of less than 6% for conversion disorder, the standard name for FND at the time–not 16%. At the lower rate, it was the eighth-most-common presentation at neurology outpatient clinics, not the second-most.
It goes without saying—or at least it should–that it is inappropriate to reinterpret these data post-hoc and effectively triple the reported prevalence by lumping in many other patients with “functional” or “medically unexplained” disorders but not FND. The former are diagnoses of exclusion; in contrast, an FND diagnosis requires the presence of positive rule-in clinical signs. Combining these two sets of study participants into one larger group and then categorically asserting that they all had FND is unwarranted and unjustifiable.
Subsequent prevalence studies are irrelevant when assessing the accuracy of statements citing Stone et al or other SNSS papers. So are investigators’ personal notions about prevalence rates, no matter how sincerely held. We remain perplexed that the entire FND field seems to have been disseminating misinformation about the SNSS results for so many years.
Now that the correction for Perez et al is in process, it is critical that the dozens of other papers containing the same questionable claims also undergo scrutiny. Articles citing the exact same set of data should not be providing completely different prevalence rates; some top FND investigators have endorsed a prevalence of less than 6% in some papers and a rate of 16% in others. These conflicting accounts render this body of work epidemiologically incoherent and suggest an eagerness to hype prevalence in ways that favor investigators’ beliefs and interests.
We have identified similar mis-citations of Stone et al in at least nine other papers for which one of you was either the lead or the senior author. I have listed them below, along with the sentences requiring corrections. (It is possible we have missed some publications. In addition, the list does not include papers requiring corrections for which one or both of you were co-authors but not the lead or senior author; our group will address those separately.)
1) Aybek S , Nicholson TR, Zelaya F, O’Daly OG, Craig TJ, David AS, Kanaan RA. Neural correlates of recall of life events in conversion disorder. JAMA Psychiatry. 2014 Jan;71(1):52-60.
***Sentence requiring correction: “Conversion disorder (CD), also known as hysteria, is by no means a historical entity, accounting for 16% of neurology outpatients.”
2) Apazoglou K, Mazzola V, Wegrzyk J, Frasca Polara G, Aybek S. Biological and perceived stress in motor functional neurological disorders. Psychoneuroendocrinology. 2017 Nov;85:142-150.
***Sentence requiring correction: “FND represents the second commonest cause for a neurological consultation after headache.”
3) Frasca Polara G, Fleury V, Stone J, Barbey A, Burkhard PR, Vingerhoets F, Aybek S. Prevalence of functional (psychogenic) parkinsonism in two Swiss movement disorders clinics and review of the literature. J Neurol Sci. 2018 Apr 15;387:37-45.
***Sentence requiring correction: “Prevalence estimates of functional neurological disorders varies from 6 to 15% of patients presenting to neurological services and represent the second commonest cause for a neurology consultation after headache.”
4) Jalilianhasanpour R, Ospina JP, Williams B, Mello J, MacLean J, Ranford J, Fricchione GL, LaFrance WC Jr, Perez DL. Secure attachment and depression predict 6-month outcome in motor functional neurological disorders: a prospective pilot study. Psychosomatics. 2019 Jul-Aug;60(4):365-375.
***Sentence requiring correction: “Functional neurological disorder (FND)/conversion disorder is a common neuropsychiatric condition, estimated to be the second most common outpatient referral to neurology clinics.”
5) Ospina JP, Jalilianhasanpour R, Perez DL. The role of the anterior and midcingulate cortex in the neurobiology of functional neurologic disorder. Handb Clin Neurol. 2019;166:267-279.
***Sentence requiring correction: “Importantly, FND is the second-most common referral to outpatient neurology.”
6) Cretton A, Brown R, LaFrance WC Jr, Aybek S. What does neuroscience tell us about the conversion model of functional neurological disorders? J Neuropsychiatry Clin Neurosci. 2020 Winter;32(1):24-32.
***Sentence requiring correction: “FND is the second most common cause of a neurological outpatient visit after headache.”
7) Perez DL, Hunt A, Sharma N, Flaherty A, Caplan D, Schmahmann JD. Cautionary notes on diagnosing functional neurologic disorder as a neurologist-in-training. Neurol Clin Pract. 2020 Dec;10(6):484-487.
***Sentence requiring correction: “FND is among the most common conditions seen by neurologists across outpatient (2nd only to headache) inpatient, and emergency department settings.”
8) Diez I, Williams B, Kubicki MR, Makris N, Perez DL. Reduced limbic microstructural integrity in functional neurological disorder. Psychol Med. 2021 Feb; 51(3): 485-493.
***Sentence requiring correction: “For much of the 20th century, functional neurological (conversion) disorder (FND) was marginalized across neurology and psychiatry despite being the second most common reason for neurological referral and incurring significant healthcare costs.”
9) Paredes-Echeverri S, Maggio J, Begue I, Pick S, Nicholson TR, Perez DL. Autonomic, endocrine, and inflammation profiles in functional neurological disorder: a systematic review and meta-analysis. J Neuropsychiatry Clin Neurosci. 2022 Winter;34(1):30-43.
***Sentence requiring correction: “FND is the second most common condition seen in outpatient neurology clinics, a finding compounded by observations of high health care costs and poor prognoses among many patients.”
As a courtesy, I wanted to alert both of you that our group intends to continue our efforts by contacting these journals, noting the acknowledgement of error in the NeuroImage: Clinical paper, and requesting corrections. It would be helpful if you could let us know whether you yourselves have already informed these journals of the need for corrections, or have any plans to do so. Such steps on your part would obviously facilitate and speed up the process of ensuring that the medical literature on FND adheres to the facts.
Thanks again for acknowledging the error in the NeuroImage: Clinical paper.
Best–David
David Tuller, DrPH
Senior Fellow in Public Health and Journalism
Center for Global Public Health
School of Public Health
University of California, Berkeley