Clinicians who are familiar with functional neurological disorders (FND) recognise that antecedent stressors do not always take the form of traumatic or severely threatening events – as illustrated in two recent popular books by the UK neurologist Suzanne O'Sullivan (Reference O'Sullivan2015, Reference O'Sullivan2021) and in a review of the published research evidence (Ludwig et al., Reference Ludwig, Pasman, Nicholson, Aybek, David, Tuck and Stone2018).

In their review, Morsy et al. (Morsy et al. Reference Morsy, Aybek, Carson, Nicholson, Stone, Kamal and Kanaan2021) therefore seek to explore other characteristics of life adversity that might be relevant. In truth, the literature they have reviewed is hardly able to answer the questions we want to put to it. Many studies did not use a method of eliciting life events that would allow for their nuanced characterisation and nor did all of them, surprisingly, specify the timing of reported life events in relation to the onset of neurological symptoms. The result is that the review is reliant upon relatively uninformative categories (uninformative because they give no clue as to psychological or social meaning) of events that could have occurred at some unspecified time before the onset of FND.

There are further limitations. One is the imprecision that comes from bundling different clinical presentations into a single diagnostic category – including the loss of potential explanatory power that comes from separating disorders with acute onset from those with insidious onset. As important is a failure in many cases to distinguish between episodic (event-like) adversity and the more extended adversity sometimes called difficulties. When life difficulties have been explored in some of the key primary research studies, such as the included studies of functional dysphonia, they have been found to be at least as common as events and it is a limitation that they are not considered in the present review: functional disorders frequently arise in the setting of interpersonal predicaments to which final (precipitating) events are often linked and from which they take their meaning.

The most useful data come from those studies that have used a detailed interview capable of characterising events and difficulties in more meaningful ways, included a control or comparison group, and timed the experience of adversity in relation to the onset of specific symptoms. Unfortunately for reviewers, such studies are few and have used differing definitions of event type that have for the most part not been tested in replication studies.

Morsy et al. (Reference Morsy, Aybek, Carson, Nicholson, Stone, Kamal and Kanaan2021) have identified four such studies that they regard as reporting on events with ‘escape’ features, and they present pooled data from these studies in a forest plot as part of their results. Their redefinition of the nature of these life events and its presentation in a data-pooling analysis is not justified.

There is a non-trivial difference between the results of the study described by Raskin (Raskin, Talbott, & Meyerson, Reference Raskin, Talbott and Meyerson1966) and colleagues where the quoted prevalence refers to illness that was judged as definitely having been ‘used to solve conflict brought about by precipitating distress’ and those described by Nicholson and colleagues where the quoted prevalence refers to events described according to the degree to which they ‘could be ameliorated by subsequently developing an illness’.

Our own life events dimension of Conflict over Speaking Out (Baker, Ben-Tovim, Butcher, Esterman, & McLaughlin, Reference Baker, Ben-Tovim, Butcher, Esterman and McLaughlin2013; House & Andrews, Reference House and Andrews1988) could not be described as reflecting Rankin's concept nor that of escape. For many of the social situations under consideration in the dysphonia studies, there was no realistic prospect that the onset of symptoms could resolve what was by definition an intractable conflict. The more likely formulation is that functional disorder emerged as a symptomatic breakdown with a meaningful connection to the stressor – much as depression can be seen as a meaningful response to loss. To quote Baker et al. (Reference Baker, Ben-Tovim, Butcher, Esterman and McLaughlin2013): ‘…many women with functional voice disorders may have become both vocally and emotionally burnt out, having lost not only their physiological voice, but “their voice” in the larger systems to which they belonged’.

These two different ideas (escape, conflict) are reflected in other literature on psychosomatic disorders without functional loss – for example in Craig and colleagues' study of events with secondary gain potential (Craig, Drake, Mills, & Boardman, Reference Craig, Drake, Mills and Boardman1994) and our study applying the more generalised concept of conflict not about speech (dilemmas) to the study of chronic fatigue syndrome (Hatcher & House, Reference Hatcher and House2003). Given the limitations of methods of eliciting life events and difficulties – they typically involve a single contact with a single informant - it is striking that a relevant history is found in the majority of patients interviewed. Clinical experience suggests that even these findings are underestimates, with more evidence likely to emerge in extended (therapeutic) contact, in contact with others in the intimate social network, and in observation of the response to rehabilitation efforts.

Future research should avoid conflating the two ideas since adversities with these characteristics may have very different associations, for example with social (interpersonal) vulnerabilities or the specific features of the clinical presentation.

Pending further research into the central role of life adversity in FND, two important conclusions drawn by the review's authors cannot go unchallenged. Emphasis on the importance of proximate events interacting with general vulnerability comes as a result of too little attention to life difficulties. And the claim that ‘events where symptoms may provide a solution’ have been demonstrated to have an association with an onset that is ‘strongest of all’ represents a serious misinterpretation of the evidence.