Trial By Error: An FND Patient’s View–and More on Those Inflated Prevalence Rates

By David Tuller, DrPH

In a post last week, I noted that experts in FND have a tendency to assert prevalence rates that ignore their own diagnostic criteria. Before offering further thoughts on that score, I want to make one point very explicit: I am in no way questioning whether people with the diagnosis have serious disorders and very debilitating symptoms. Of course they do! And of course they deserve the best medical care available, like everyone does.

I have not suggested and do not believe that FND patients are hypochondriacs, hysterics, fakers or whiners. Perhaps that hasn’t been clear from previous posts. If anyone has interpreted my work that way or has criticized FND patients based on my posts, I really apologize for the misunderstanding. That has never been my intention. It has, of course, been my intention to raise concerns about how experts in the field have themselves generated confusion with contradictory statements about FND diagnoses and prevalence.

In talks and presentations, FND experts come across as sensitive and caring clinicians who clearly want the best for their patients. I have zero reason to doubt that. But whether doctors demonstrate compassion is a separate issue from whether it is acceptable to disseminate prevalence claims that go way beyond the established clinical criteria. Perhaps someone raised this problematic issue at the recent gathering, in Boston, of the Functional Neurological Disorder Society—but I doubt it.

It is true, as some have noted, that I haven’t included the voices of FND patients in my half dozen or so posts on this issue. While my focus is on questionable claims in the medical literature, I understand why this absence has bothered some patients. I’d like to rectify the omission by recommending an excellent, recently posted essay called “Cadenza for Fractured Consciousness: A Personal History of the World’s Most Misunderstood Illness.”

The author, a patient who goes by @FNDPortal, provides an overview of the history and development of the construct as well as a harrowing account of his own condition. Here’s an excerpt from the opening section:

“Late in 2013, I inexplicably lost my ability to walk. It happened over the course of just a few weeks; fast enough to be terrifying, slow enough to feel everything in crystalline detail…First, I could no longer manage my usual walk across town. Then I couldn’t reach the end of the parking lot. Then the other end of the house.”

Yikes! Terrifying indeed!  

If anyone believes FND patients are not struggling with something really awful, this essay will quickly alleviate them of their misconceptions. They will also learn a great deal about the FND patient perspective—or at least this particular FND patient’s perspective. He describes his search for answers, his appreciation of the FND diagnosis and related treatment approaches, and the improvements he has experienced over time. However, as enlightening as the essay is about many aspects of the issue, it does not resolve some of the questions I have raised since I first posted on FND in December, 2019.


A “rule-in” diagnosis, or not?

My post last week examined how FND experts have presented inconsistent information about prevalence rates of the condition. Based on the requirement that FND is now a “rule-in” diagnosis relying on positive clinical findings and not a diagnosis of exclusion, these cases account for close to 6% of outpatient visits to neurology clinics—per an exhaustive study into attendance at Scottish neurology clinics. Here, as I noted in my previous post, is a 2016 statement on the numbers from FND experts Alan Carson and Alexander Lehn, from UK and Australia, respectively:

“The recent changes…to a definition based on positive identification of physical symptoms which are incongruent and inconsistent with neurologic disease and the lack of need for any psychopathology represent a significant step forward in clarifying the disorder. On this basis, FND account for approximately 6% of neurology outpatient contacts.”

Yet FND experts themselves routinely declare that prevalence in outpatient neurology clinics is 16%, or even a third. These numbers represent either a tripling or heptupling (is that a word?) of the rule-in rate. 

Here’s just one example, from Richard Grunewald, formerly the clinical director of neurosciences at Central Sheffield University Hospitals NHS Trust, in a recent blog for an organization called Inneg. (Inneg describes itself as “a leading agency specialising in providing Clinical Negligence, Serious Injury and Complex Case reports.” In other words, if you’re involved in litigation related to FND, better call these folks for help.) Grunewald’s post is called “The Challenge of Distinguishing Unusual Neurological Symptoms from. Malingering.”

Here’s what he writes:

“One common problem in neurological medicolegal practice is diagnosing Functional Neurological Disorder (FND), sometimes known as somatization, conversion disorder, hysteria or medically unexplained neurological symptoms…It is common, and has been estimated that it comprises about one third of the workload of most neurologists, yet is poorly understood.”

Is Grunewald aware that FND is now a rule-in diagnosis and not simply a placeholder term for anything unexplained by current neurological understanding? He seems not to know that the established prevalence via positive identification is “approximately 6%”–as Carson and Lehn reported. Whether Grunewald is familiar with these data or not, he and other experts in the field continue to fudge the diagnostic boundaries by citing prevalence rates of 16% or a third.

In a 2021 article, the authors described confusion among neurologists about how and when to code patients as having FND and noted that a major factor was “the outdated belief that FND is a diagnosis of exclusion.” (In other words, some or many neurologists do not understand that FND is now considered a diagnosis of inclusion, based on the rule-in signs.) The authors also stated that FND “accounts for 16% of consultations”—even thought that number combines both the “approximately 6%” identified by rule-in signs, per Carson and Lehn, plus another 10% who were assessed as having “psychological symptoms,” according to the data from Scottish neurology clinics.

Yet having “psychological symptoms” is not considered a clinical rule-in sign for FND, as is required for the diagnosis. So what is the authors’ rationale for combining these two smaller groups into one larger FND category? They provide no explanation. Perhaps they don’t recognize that they are contradicting themselves. The FND literature is full of examples of this sort of internal inconsistency and epidemiological incoherence.

People, this is not rocket science! (Or perhaps that should be: “People, this is not neuroscience!”) It is basic statistics, logic, and public health. After all, investigating disease prevalence rates is a core public health function. I have never pretended to have a medical education or an advanced degree in biology–much less neuroscience. But such training is not needed in order to notice that experts are citing multiple prevalence rates for the condition.

If FND is now a rule-in condition, I would argue that anyone without rule-in signs who has been given the label has been misdiagnosed. Much clinical guidance in recent years has focused on how to execute these rule-in strategies and why positive identification through such incongruities in clinical presentation is necessary to render the diagnosis accurately. Don’t take my word for it. Here is Jon Stone, a colleague of Alan Carson’s from the University of Edinburgh, in a 2021 presentation to the Encephalitis Society:

“Some people think that FND is a condition you diagnose when someone has neurological symptoms but you can’t find a brain disease to go along with it. And that’s absolutely not the case. Some people [i.e. clinicians] do that, but if they’re doing it like that they’re doing it wrong.”

From my interpretation of this authoritative statement, anyone claiming that the prevalence of FND is greater than the “approximately 6%” cited by Carson and Lehn is “doing it wrong.” Yet Stone and Carson themselves wrote in a 2020 paper that FND is the “second commonest reason for new neurology consultations”—a claim based on the 16% prevalence rate. In contrast, based on the “approximately 6%” prevalence, FND is much farther down the list of common neurological presentations.

This is all kind of weird. Why aren’t neurologists following their own FND criteria in their prevalence pronouncements?


Rebranding and reframing FND

Here’s how this confusion might have developed over time. By all or most accounts, it seems that around a third of outpatient neurology attendees present with symptoms that do not conform to known patterns of neurological disease, as currently understood. The Scottish neurology clinic study, for example, reported that 30% of outpatient consultations–not quite a third–involved cases that could not not be explained by “organic disease.” Rather than identifying some or all such unexplained symptoms and conditions by old standbys like conversion disorder, hysteria or other terms often perceived as insulting, neurologists began decades ago promoting the need to call them “functional” instead.

Ok, rebranding a disorder with a less stigmatizing term is a step forward. And many patients report that they feel more respected, believed and cared for than when the condition was called something else. (Other patients report that, in their experience, doctors have changed their language but not their dismissive attitudes toward the illness.)

Along with the new name, neurologists pushed for a revision of the diagnostic criteria. They recognized that not every patient in this cohort had identifiable psychological distress or trauma that could, in their view, account for the symptoms—a requirement under the longstanding definition of conversion disorder. That requirement was therefore removed in the description of the illness enshrined in the fifth iteration of the so-called psychiatric bible known as the Diagnostic and Statistical Manual (DSM).

The experts further posited that certain clinical tests provide affirmative evidence of the “functional” rather than “organic” nature of the symptoms—Hoover’s sign being the most prominent. (The essay from @FNDPortal provides a helpful discussion of these strategies.) So the updated description in DSM-5 also noted that evidence from such rule-in signs is necessary for rendering an accurate diagnosis.

It would be a significant improvement if patients with symptoms not consistent with standard neurological understanding but without the mandated rule-in signs were said to have idiopathic neurological disorders or neurological disorders of unknown etiology. But perhaps because the field has become so used to regarding all unexplained symptoms as “functional” rather than “organic,” experts still seem to use the FND label when writing or talking about this larger group of patients without rule-in signs.

So who cares? Don’t all these patients need medical care and treatment, whether their illness is called FND or neurological disorder of unknown etiology?

Of course they do. But clinicians conduct medical investigations so that they can render a diagnosis. Once they have rendered the diagnosis, they move on to providing care and treatment based on that assessment. The literature on “medically unexplained symptoms,” a larger category of which FND is a major subset, is replete with warnings that once the diagnosis has been established, continuing to investigate possible medical causes of somatic complaints serves to feed patients’ purportedly unjustified fears that they have underlying but undiscovered pathophysiological dysfunctions.

I don’t know exactly how this dynamic is playing out in clinical care. But it is indisputable that those referencing prevalence rates of 16% and a third, whether in the medical literature or in public presentations, are not adhering to the rule-in rule they have themselves promoted. In medicine, as in public health, consistency in communication and messaging on core issues like disease prevalence is essential. The FND field so far has fared poorly on this front.

FND is either a diagnosis of inclusion, or it isn’t. Citing prevalence rates far higher than the “approximately 6%” identified by rule-in signs risks turning FND into another wastebasket diagnosis of exclusion—a dumping ground for any symptoms as yet unexplained.

15 thoughts on “Trial By Error: An FND Patient’s View–and More on Those Inflated Prevalence Rates”

  1. Lady Shambles

    Thank you for your continued examination of this peculiar state of affairs. That old idiom about the left hand not knowing what the right hand is doing seems apt. And it doesn’t inspire any confidence in me that their 6% is a meaningful diagnosis either.

  2. Some people with ME are finding FND written into their medical notes, either as a revised diagnosis instead of ME, or as a new diagnosis after they asked for, and had, medical investigations for new symptoms. Frankly the possibility is putting some pwme off trying to get medical investigations into new symptoms. Some people with ME are stating that they had no idea they had been given a diagnosis of FND until they requested to see their medical records.

  3. I guess this isn’t just a problem for the UK but I think the UK’s Chief Medical Officer should be informed about it. Surely something’s going badly wrong in medicine for things like this to happen and them not be swiftly challenged and corrected? Is it that nobody notices or that nobody cares?

  4. I wonder if the discrepancy in the stats depends on whether or not all the p/w the subtypes are being included (IE those for which the Hoover’s ruling in sign wouldn’t apply but where the new ‘hallmark’ of FN(C)D – internal inconsistencies – might). These would include people with a PPPD dx (PPPD is being over diagnosed and, in some cases, other neuro/vestibular conditions are being missed) and people thought to have Functional Cognitive Disorder (cognitive/memory issues in the absence of a structural neurodegenerative condition). This could also explain why the rule in signs (eg Hoover’s etc) are not being referred to so much in the literature now and the shift is towards ‘internal inconsistencies’.

    And the problem remains that the term ‘functional’ is described differently by different people eg Stone says ‘symptoms where there is no disease’ (does migraine count as a disease?) but others say ‘a change in function, not structure’ (so that’s most people with migraine, seasickness etc etc). What we do know is that the term has remained ‘vibrant’ due to it’s ambiguity and can be used to convey one message to patients and another to doctors, rendering shared-care and the Patients as Partners deal impossible.

    I tried to get all the links to some of the literature about how the term ‘functional’ is used in different settings on one page but I don’t think it’s worked, however googling the titles of the papers should bring them up:

    The function of ‘functional’: a mixed methods investigation
    Richard A Kanaan  1 , David Armstrong, Simon C Wessely

    PMID: 22250186 PMCID: PMC3277687 DOI: 10.1136/jnnp-2011-300992

    Published online 2020 Jun 11. doi: 10.12688/wellcomeopenres.16022.1
    PMCID: PMC7338913
    PMID: 32685699
    What is the functional/organic distinction actually doing in psychiatry and neurology?

    Provider Perspectives
    Limits to truth-telling: Neurologists’ communication in conversion disorder

  5. “It has, of course, been my intention to raise concerns about how experts in the field have themselves generated confusion with contradictory statements about FND diagnoses and prevalence.” (Dr. Tuller)

    Very good.

    It is fundamental that scientific research include testing of current practices, because these things still happen and have been for some time:

    “At least six small alternative-frame open reading frames (ORFs) overlapping well-characterized SARS-CoV-2 genes have been hypothesized to encode accessory proteins. Researchers have used different names for the same ORF or the same name for different ORFs, resulting in erroneous homological and functional inferences. ”

    There are still unsound practices that need to be pointed out … its the only way such things get fixed.

  6. Dr Tuller,

    Suppressing criticism is what you are doing, just like those you are criticizing, my research.


  7. I only half agree with Dr. Tuller here.

    Where I disagree is I think substantial uncertainty over prevalence rates is to be expected when you don’t have a good test (or even when you do have a good test, but the test too expensive to apply to the population of interest … cf. U.K. government scaling down rapid antigen tests for covid19).

    I agree the way the data is typically reported is pretty poor.

    For sake of argument, assume the rag bag diagnosis of exclusion is 30% of patients, and the those with rule in signs are about 6%. That 6% isn’t even the majority of the 30%. The assumption that the 24% of patients without rule in signs and without some alternative dx must have the same condition as the 6%, even though they don’t exhibit the same signs, seems logically unjustified.

    FND isn’t my area, but I do sometimes work in some autism research. “Spectrum condition” – severity of symptoms varies widely. Prevalence rates therefore highly susceptible to what the threshold for dx is. Questions such as “is autism actually becoming more prevalent” or “what is the male/female sex ratio for the condition” require considerable attention to the possibility of being confounded by differences in diagnostic criteria.

  8. In reply to SusanC’s comment

    I disagree, I think Tuller is spot-on here.

    If FND has to be rule-in, then the stated prevalence surely has to be the prevalence according to rule-in diagnosis, so around 6%? Surely it can’t be anything else? These things appear to be carved in stone. I really can’t see how they can be disputed.

  9. In regards to Susan C’s comments, my point is not that FND is actually 6%. I don’t know what it is. But I’m just pointing out that that is what the experts themselves say based on their own criteria and data. They strongly advocate that FND is only a rule-in diagnosis and that others are “doing it wrong” if they go beyond that, as Stone says. But they themselves constantly go beyond that and assume all with unexplained symptoms have FND, at least when they cite prevalence rates. That’s not right.

  10. Just for clarification, in my comment above I didn’t mean that the 6% FND prevalence can’t be disputed, of course it can. What I meant is that I think it’s hard to dispute the inconsistency here because FND experts seem to have carved these things in stone for themselves by being very clear that FND must be diagnosed with rule-in signs and by giving a figure for the rate (around 6%) when it is so defined. Providing or suggesting FND rates in neurology of 16% or 30% or ‘a third’, could encourage the misdiagnosis of people who don’t have FND according to the experts’ own definition/criteria. I think it’s hard to argue with that.

  11. In fact, the problem of health is much deeper than we think. We are taught to limit ourselves to junk food, but they open fast food cafes on every corner. We are told to eat more vegetables, but the ads show chips. In fact, the problem is in the portion, not in the usefulness. We shouldn’t demonize food with every health issue.

  12. Steve Hawkins

    The over all problem is still the same as with MUS etc.: the vague abbreviation is taken to be a distinct disease entity, with the result that a heterogenous group of patients are lumped together so that the only way to research and treat them is individually, and that is the very thing that they were lumped together to discourage, under the ‘too much testing’ mantra that is rife in an NHS that is being starved of funding in preparation for selling it off like everything else.

  13. Pingback: Trial By Error: A Letter About the Inflated Prevalence Rate of Functional Neurological Disorder

Comments are closed.

Scroll to Top