Illness representations of people with later-onset functional seizures

Purpose: Although functional seizures can start at any age, little is known about the individuals for whom onset occurs after the age of 40. It has been proposed that health-related traumatic events are more relevant causal factors for people with ‘later-onset functional seizures ’ than for those whose functional seizures begin earlier in life, however, the illness representations of people with later-onset functional seizures have not yet been investigated. This study aimed to understand the experiences and illness representations of people with later-onset functional seizures. Methods: This was a mixed-methods study. People with later-onset functional seizures were recruited via a neurologist ’ s caseload and online membership-led organisations. Semi-structured interview transcripts were analysed using Template Analysis according to the Common-Sense Model (CSM). Self-report measures of demographic and clinical details were collected to characterise the sample and verify themes. Results: Eight people with later-onset functional seizures participated in the study. Illness representations relating to all domains of the CSM as well as an additional theme of ‘Triggers ’ were identified. Functional seizures were characterised as a mysterious brain disorder analogous to a computer malfunction and involving involuntary movements associated with alterations in consciousness. Perceptions of duration were indefinite, and triggers were unknown or at the extremes of autonomic arousal. Half of the sample identified health-related events/ trauma as causal. Opinions were divided on ‘cumulative life stress ’ as a causal factor. Most perceived themselves to have limited or no control but having ‘control ’ over seizures was conceptualised as different to reducing their likelihood, frequency, or impact. Later-onset functional seizures were viewed as being more detrimental for caring and financial responsibilities but to have advantages for acceptance. Conclusions: This is the first study to assess the illness representations of people with later-onset functional seizures. Many themes were similar to those identified in samples including people with earlier-onset functional seizures. Health-related trauma or events were the most strongly endorsed perceived causal factor, but with the exception of ‘consequences ’ , all representations were characterised by uncertainty. Clinicians should hold in mind the interaction between life stage and the consequences of later-onset functional seizures.


Introduction
Functional seizures are a subtype of Functional Neurological Disorder (FND), which outwardly resemble epileptic seizures or syncope, but are not explained by epileptiform activity or cerebral hypoperfusion, respectively [1].The Integrative Cognitive Model proposes that functional seizures arise from the automatic and non-deliberate activation of a mental representation of seizures (i.e. the "seizure scaffold"), which is comprised of material related to seizure-like experiences and/or behavioural responses (e.g., altered consciousness or convulsive movements).Hypothetically, this can occur in response to elevated autonomic arousal and/or psychological distress, generalising over time to more emotionally neutral settings [2].FND is common and disabling [3,4].
Functional seizures have historically been thought of as a disorder that mainly affects 'young people' [5,6].Indeed, the median and modal ages of onset in one of the largest studies people with functional seizures to date were 28 years and 19 years old, respectively [7].Nevertheless, approximately 27% of the 698 participants in this study reported onset after the age of 40 years old, with the latest age of onset recorded at 76 years of age [7].Other studies have found that between 9.6% and 44% of patients over 60 years of age referred to epilepsy specialist centres for diagnostic videotelemetry received diagnoses of functional seizures [8][9][10].Therefore, later-onset functional seizures represent a significant, if overlooked, problem.
Although there are no large-scale, prospective studies of later-onset functional seizures, some evidence suggests that later-onset functional seizures may constitute a clinically distinct entity to earlier-onset functional seizures.In a sample of patients recruited to the screening stage of a Cognitive Behavioural Therapy Randomised Controlled Trial, female predominance was more marked prior to the age of 40 (although there continued to be a greater incidence in females beyond the age of 40) [7].Findings from a retrospective chart review comparing 26 patients with 'late-onset' functional seizures (>55 years) against 241 patients with earlier onset functional seizures attending a specialist clinic, provided preliminary evidence that health-related traumatic events might be more relevant precipitators for the former group [5].Similarly, a retrospective study of 130 consecutive adults with video-EEGconfirmed diagnoses found an increased likelihood of reporting "health problems pre-[functional seizure] onset" in patients whose first functional seizure occurred after 50 years of age (n = 55) [11].A retrospective chart review and analysis of 365 adults (18 to 88 years of age) admitted to a videotelemetry unit showed that there was a greater delay to diagnosis and a higher rate of co-morbidities in patients with functional seizures who presented later in life [12].However, selection biases necessary for Randomised Controlled Trials and those set within specialist units may limit the generalisability of these findings to the broader functional seizure population.Additionally, limitations inherent in retrospective chart review designs (e.g., selection and recall biases, incomplete data, the presence of confounding variables etc.) tempers the strength of conclusion that can be drawn about potential differences between earlier-and later-onset samples.Nevertheless, people with later-onset functional seizures and older people with functional seizures are at a different biopsychosocial developmental stage than their younger counterparts [e.g., 13]; Therefore, they likely have different life-and health-related experiences before and after the onset of their functional seizures.
The Common-Sense Model of Self-Regulation is a widely used theoretical framework outlining how individuals think about and adapt to chronic illness [14,15].The model asserts that before an individual can cope with an illness, they must make sense of it through the development of 'illness representations'.Illness representations are theorised to consist of six cognitive dimensions: identity, consequences, timeline, perceived control, illness coherence, and emotional representation.Illness representations in turn help to determine coping behaviours, which are associated with outcomes [16].Therefore, illness representations are also important factors in psychological therapy for functional seizures and indeed other conditions in neurological practice such as epilepsy.Although there are methodological limitations in the evidence base [17], most psychological treatment modalities (e.g., Cognitive Behavioural Therapy, Mindfulness-Based Therapies, psychodynamic psychotherapies) require an understanding of individuals' perceptions about the causes, triggers, and maintaining factors for their condition [18].Indeed, existing models of functional seizures propose that cognitions and illness models can be precipitating and perpetuating factors in and of themselves [e.g., the Integrative Cognitive Model; 2].
Illness representations have been studied using quantitative and qualitative methodologies in samples with functional seizures.Quantitative studies, which employ the Illness Perception Questionnaire [IPQ; 19] or variations thereof, have found that illness representations are associated with treatment-and seizure-related characteristics, as well as distress, quality of life, and perceived stigma [20].Qualitative investigations have added nuance to the picture; a systematic review found that the illness representations of people with functional seizures were characterised by uncertainty, with a reluctance to consider psychological factors as causal, but associated with significant consequences for financial and psychosocial wellbeing [21].However, there are no qualitative studies focusing on illness representations in later-onset patients.
The aim of this study was therefore to conduct a qualitative investigation of the illness representations of people with later-onset functional seizures according to the Common-Sense Model.It was hoped that findings would inform current theoretical frameworks of functional seizures as well as psychological intervention for this patient group.

Study design
A mixed-methods design using semi-structured interviews was considered most appropriate for an in-depth exploration of how people with later-onset functional seizures experience and make sense of their condition [22].Survey data were collected to help characterise the sample and verify themes.The study was designed and reported using the consolidated criteria for reporting qualitative studies [23].

Participants
Participants were eligible for inclusion if they were currently experiencing functional seizures that had started for the first time after the age of 40, and they were diagnosed by a consultant neurologist.There is no universally agreed definition for what constitutes 'later-onset' functional seizures, with other studies somewhat arbitrarily selecting 50 [11], 55 [5] and 60 years old [9] as the cut-off for 'late onset' functional seizures.Forty years old was selected as the cut-off for 'later-onset' (as opposed to earlier-or late-onset) functional seizures because this was the age at which the female preponderance for functional seizure onset began to balance out in a sample of 698 people with functional seizures screened for a UK Randomised Controlled Trial [7].Participants with 1) comorbidities that could partially or wholly explain their seizures (e.g., epilepsy), 2) a diagnosed intellectual disability, 3) who were unable to complete the survey, or 4) interview in English were excluded.

Sampling and recruitment
Participants were recruited via two routes: social media and the NHS.Potential participants recruited from the internet expressed their interest in the study via an online form.Potential participants from the NHS were identified by their neurologist.We then used purposive, stratified sampling [24] to select equal numbers of eligible participants across age deciles (forties, fifties, and sixties) for age of onset and across a range of illness durations; this was with the aim of balancing the depth against breadth of illness experiences and representations captured.Diagnoses by neurologists were confirmed in writing by participants' General Practitioners or their own neurologist (JS).The sample size was deemed sufficient when saturation had been reached; that is, the point at which further interview data produced little or no new information to answer the research question [25,26].See Fig. 1 for a breakdown of participant flow through the study.

Procedure
Data were collected between January 2022 and April 2022.Eligible participants were invited to complete an online survey capturing written informed consent as well as relevant demographic and clinical characteristics.Participants then completed a remote semi-structured interview comprised of non-leading questions, for example, "What (if anything) do you think triggers your functional seizures on a day-to-day basis?".Interviews lasted between 59 and 79 min (Mean = 64 min).Interview recordings were divided by two members of the research team (IW and KF), transcribed verbatim and anonymised in parallel.To aid later analysis and help to establish quality and validity, the researcher kept a reflective diary following each interview and during the coding process [27].

Measures
Demographic questionnaire.A questionnaire was developed to capture demographics, seizure characteristics, and health history.
The Work and Social Adjustment Scale (WSAS) [28] is a five-item self-report scale, collecting ratings of how much their 'problem' interferes with work, home management, private and social leisure, and relationships on a 0-8 Likert Scale.The maximum possible score is 40, indicating the most severe impact on functioning.For the purposes of this study, 'problem' was changed to 'functional seizures'.The WSAS has previously been used in studies of people with functional seizures [e.g., 29].
The Patient Health Questionnaire (PHQ-9) is a nine item self-report questionnaire measuring symptoms of Major Depressive Disorder over the last two weeks.The PHQ-9 has high internal consistency (α = 0.87) and a cut-off set at ≥ 13 is considered to give the best balance between sensitivity and specificity for a diagnosis of Major Depressive Disorder in people with functional seizures [30].Respondents indicate the frequency of each symptom on a 0-3 Likert scale, with a maximum possible score of 27 indicating the most severe symptoms.
The Generalized Anxiety Disorder (GAD-7) scale is a seven item self-report questionnaire, assessing symptoms of Generalized Anxiety Disorder experienced over the last two weeks.Participants indicate the frequency of each symptom on a 0-3 Likert scale with a maximum possible score of 21 indicating the most severe symptoms.The GAD-7 has high internal consistency (α = 0.92) and a cut-off of ≥ 12 is considered to give the best balance between specificity and sensitivity for a diagnosis of Generalized Anxiety Disorder in people with functional seizures [31].
The Brief Illness Perception Questionnaire (BIPQ) [32] is a nineitem scale designed to rapidly assess illness representations.
Participants indicate how much they agree with statements on a 0-10 Likert scale with a maximum possible score of 80 indicating the most threatening illness perceptions.It is a widely used measure in people with functional seizures [20].In accordance with the scoring manual, items 3, 4, and 7 were reverse scored so that higher scores represented more threatening views of illness.The questionnaire was adapted to ask specifically about 'functional seizures'.

Analysis
To protect participant anonymity, all participants were assigned pseudonyms and the amount or specificity of information 'linked' to each individual was limited where possible.Free-text responses to questions about causal representations in the BIPQ were grouped posthoc into themes.Because the aim of the study was to identify participant illness representations using an a priori theory to guide theme development (the Common-Sense Model), Template Analysis was employed [33].Template Analysis is consistent with the epistemological position of 'subtle realism'; that knowledge is never independent of the of the researcher's perspective, whilst also acknowledging the existence of reality that unknowable to the researcher [34].
An initial template was constructed using the domains of the Common-Sense Model [35] (identity, timeline, cause, cure/control, consequences) as top level themes.Transcripts were then coded according to the initial template (while also remaining open to any other important themes), working sequentially through one transcript at a time to identify sub-themes and top-level themes, and modifying the template as necessary until all relevant data were satisfactorily coded.The final template was refined by discarding subthemes that were not strongly represented in the data and considered complete when the final template applied to the whole data set.To enhance the reliability/ dependability of themes, two members of the research team met to discuss extracts and reach an initial consensus (IW and DG).Themes and supporting quotes were discussed again with the Edinburgh FND  Research Group (including JS) to further refine and enhance the quality and reliability of interpretation.

Researchers' role and reflexivity
To ensure quality through reflexive analysis, the researcher's relationship to the participant group and the research theme should be made explicit [36].To that end, the lead researcher had clinical experience of working with people with functional seizures as a trainee Clinical Psychologist in an acute neuropsychology setting, as well as prior experience of conducting doctoral research in FND.Other members of the research team also had significant research and clinical experience in FND.The analysis was therefore approached, unavoidably, with certain pre-conceptions about participant experience.Where possible, the researcher aimed to mitigate the influence of these biases through an iterative analytical process which began with 'staying close' to the intended meaning of participant's narratives in the initial coding, discussion of themes with other members of the research team from different clinical different backgrounds, and through verification of themes with self-report measures.

Ethical review
This study received full ethical review and approval from the University of Edinburgh East of Scotland Research and Ethics Committee (REC reference 21/ES/0094), and NHS Lothian Research and Development (R&D) in December 2021.

Participants
Eight people (five females, three males) with later-onset functional seizures from across the United Kingdom participated in the study (Table 1).Five participants were recruited from the internet and three via the NHS.Participants' age at the time of interview ranged between 42 and 66 years old.Six participants were not in work due to their functional seizures; however, their educational history was comparable to the general population.
Functional seizure characteristics are reported in Table 1.Duration ranged between three and seventeen years.Reported seizure semiologies included "Whole/parts of my body shake," (n = 8), 'I go very still (n = 2), 'I collapse' (n = 3) and 'Absences' (n = 6) with participants often having more than one type.Participants reported retaining awareness but being unresponsive (n = 6) or being unaware/loss of consciousness (n = 8).

Health history
The reported frequency of lifetime mental health diagnoses and significant physical health events other than FND (e.g., serious accidents/injuries, major operations or 'serious' illnesses) are detailed in Table 1.Four participants reported at least one 'health event' within two years of the onset of their functional seizures.

Psychopathology and illness representations
At the time of survey completion, no participants scored above the diagnostic cut-off for Generalised Anxiety Disorder (≥12) and five participants scored within the range for Major Depressive Disorder (≥13) in people with functional seizures [30,31].Scores on the WSAS were in the range usually considered to represent a 'clinically significant functional impairment' in seven of the eight participants (Table 2).A notable exception to this pattern was Liz, who endorsed relatively low symptoms of anxiety and depression and a minimal impact of her functionals seizures on her functioning.

Illness representations
Inspection of total scores on the BIPQ indicated that, except for Liz, most participants held adverse or 'threatening' illness representations (Table 3).Participants reported a negative impact of functional seizures on their lives (Consequences) and that they expected a long duration Note.Health history = number of reported lifetime physical health events (including major accidents, operations, illnesses, and diagnoses) and mental health diagnoses.Note. a = Above cut-off for Major Depressive Disorder in people with functional seizures.b = Above cut-off for clinically significant functional impairment.
(Timeline).Responses also indicated that most participants thought that there was little they could do personally to influence their functional seizures (Personal control) but perceptions regarding the potential influence of treatment (Treatment control) were more divided.More participants than not reported a high number of symptoms associated with their functional seizures (Identity) but again, opinions were more divided on how much their functional seizures concerned them (Concern).Scores on the 'Coherence' subscale indicated that participants' understanding of their functional seizures ranged from "no understanding at all" (Diane), to uncertainty (Jane, James, Susan, and Roger), to a relatively good understanding (Margaret, Liz, and Brian).Finally, scores on the 'Emotional representation' subscale indicated that more participants than not were distressed by their functional seizures.In total, 'physical health' (e.g., "Brain injury", "surgery", "migraines") and 'stress/psychological trauma' (e.g., "stress", "trauma", "emotional health") were the two most frequently endorsed causal factors for functional seizures (both reported seven times) (Table 4), however, 'physical health' was given the highest ranking for importance.The next most common response was 'Don't know' (e.g., "Can't pinpoint anything for certain", "no idea") followed by 'Over-work/ work-related factors' (e.g., "Working too hard", "work/life balance").

Qualitative findings
Six top level themes were identified in the qualitative analysis.Five of these related directly to the Common-Sense Model; identity, timeline, cause, control/cures, and consequences.As functional seizures are paroxysmal, the super-ordinate theme of 'triggers' (i.e.day-to-day triggers) was considered conceptually distinct from 'cause' (i.e. the initial cause) and therefore emerged as an additional theme in the template (Table 5).

Identity
'Identity', pertains to symptoms participants associate with their illness and how they label them.Each participant had different names for their functional seizures, perhaps reflecting the lack of consensus around the condition in the medical and wider community.Nevertheless, functional seizures were almost unanimously conceptualised as a brain disorder (Table 6).Participants frequently drew on computerbased analogies such as that of a 'software' issue or 'signalling' problem resulting in involuntary movements.At the same time, participants also spoke about their functional seizures as somewhat mysterious, both to themselves and to medical professionals.There was considerable variability in the frequency and severity of seizures, which made them unpredictable and added to a sense of frustration and hopelessness.
Alterations of consciousness were referred to explicitly as "dissociation" or described as such, for example, participants spoke about Note.Cause 1 = most important of top three causes, Cause 3 = least important of top three causes.'zoning out' or being 'not there', consistent with depersonalisation/ derealisation [37].Diane used the metaphor of "going into another world like a dream" to describe her experience.Indeed, others have reported that people with functional seizures are more likely than those with epilepsy to use space/place metaphors to conceptualise these events, reflecting an important aspect of their subjective experience [38].

Timeline
Participants who referred to the expected duration of their functional seizures seemed to be unsure or conveyed a sense of hopelessness that they would ever be cured.
'I hope it will go away but I don't think so…' (Margaret).

Causes
Three main sub-themes for the 'root causes' of functional seizures were identified (Table 7).Half of the sample attributed their onset to a 'health-related trauma or event'; seemingly because these were salient events that occurred either immediately prior to their first functional seizure or in the months preceding; reflecting an underlying belief that for an event to be causal it must occur immediately prior to onset.

' …it's obvious to me that it was tied into this one event…'. (Jane).
Jane then goes on to describe a traumatic event during a routine medical procedure that resulted in her admission to intensive care.When asked what it was about this experience that caused her functional seizure, she explained: 'I think the shock of that because I was aware something went wrong as I say to you… I remember thinking, "I'm gonna die," I remember having that feeling as everything went black that I was gonna die'.(Jane).
Other participants entertained the possibility that medical events or conditions such as spinal surgery, Covid-19, menopause, sleep apnoea, stroke, or repeated (accidental) knocks to the head over their lifetime were causal.For some, the psychological impact of these medical events was perceived to be the mechanism of onset; for others, it was the events' physical impacts.
The presence or absence of 'cumulative life stress' was also identified as a subtheme.Every participant reported several outwardly challenging events over their lifetime (prior to the onset of their functional seizures) including physically, mentally, or emotionally strenuous occupation or caring responsibilities, experiences of mental and/or physical illness, or experiences of domestic abuse.Some recognised the emotional and physical impact of these events and linked them directly to the onset of their functional seizures.As Diane put it, "I've broken my nervous system effectively."For some others, cumulative life stress made less sense as a cause because they were feeling relatively happy and healthy around the time of onset.Indeed, for most, while they had tried hard to identify causal explanations, they were held lightly and characterised by uncertainty.

Triggers
The most common experience amongst participants was of not being able to identify any specific or reliable triggers for functional seizures in their day-to-day lives.All but one participant spoke about their seizures as 'coming out of the blue'.Liz's account (Table 8) conveys the effort that she has made to understand her triggers and the frustration that comes with not being able to reliably identify any.
While some individuals held strong beliefs that stress was an occasional or consistent trigger, others could not identify it as such because they could not find a pattern.These opposing views are reflected in the accounts of Jane and Margaret (Table 8).For some participants, situations that might reasonably be expected to lead to feelings of stress, such as being overwhelmed, fatigue from lack of sleep or 'overdoing it', or returning to work, were identified as potential triggerseven if participants did not specifically refer to feeling 'stressed' at the time.
Relaxation was also identified as a trigger by several participants (some of whom also identified stress as a trigger).These participants reported functional seizures when falling asleep at night or when being encouraged to try guided relaxation as an intervention.This made some participants wary of ever allowing themselves to fully relax, presumably causing them to make concerted attempts to remain in a persistent state of physiological arousal.

Controls and cures
Most participants perceived themselves to have no or very limited control over their seizures (Table 9).In several cases, participants described a sense of 'shaking it out/letting it happen' -an almost wilful submission to the seizures because they need to be 'released' from the body.These accounts suggest that functional seizures serve, in some individuals at least, as a way to provide relief from uncomfortable prodromal experiences, which has been reported before [39].Some participants spoke about trying to think their way out of seizures if they retained awarenessalmost as if seizures could be stopped through sheer will (as opposed to using a particular cognitive strategy).However, this did not seem to be an effective approach, and added to a sense of helplessness.It is understandable that people with functional seizures would increase their focus on their seizures in-the-moment, but this may inadvertently serve as a maintaining factor (as with other functional symptoms) [40].Despite a consensus that participants had limited control over their seizures, all were able to identify strategies that either helped to stop functional seizures in-the-moment, reduce their likelihood, or cope with the sequalae.Notably, none of the participants who initially claimed an absence of control over their seizures later went on to revise these initial statements, perhaps reflecting a lack of coherence in their illness representations, or a different understanding (compared to that of healthcare professionals) of what 'controlling' their seizures meant.
Regarding stopping functional seizures in-the-moment, the five participants who were able to identify 'warning signs' of an imminent seizure in sufficient time could use these to take steps to help avert them.Warning signs were therefore viewed positively, and participants who did not get warnings thought that they might be helpful for this reason.Strategies that participants could then employ in-the-moment included 'distraction, grounding, and mindfulness'.Participants like Diane and Brian had learned these through attending psychological therapy, through their professional roles, or through practicing mindfulness over their lifetimes.
Avoidance strategies were also employed to control the likelihood of a functional seizure.These included situational avoidance, such as escaping doctors' waiting rooms or public spaces in case the overwhelming feelings brought about by these settings triggered a seizure.Others spoke about using experiential avoidance (i.e.avoidance of emotions) to decrease the likelihood of functional seizures.Diane talked about learning to 'walk away' from stressful experiences for this reason.Making use of 'social support' was discussed as a strategy that could help participants to either employ in-the-moment strategies to ward off a functional seizure, or to cope with their condition.One participant in his early forties spoke about how his young children helped to distract him.With regards to coping more generally with the impact functional seizures, participants also described the value of peer-support groups which could provide a sense of understanding, as well as support from partners or friends who help to promote safety or comfort during a functional seizure.Liz explained the importance of people staying calm around her because of retained awareness during seizures.
Another identified strategy for coping with functional seizures was acceptance.For some, this took the form of a passive resignation to the fact that they had functional seizures so they just needed to 'let them happen', while participants like Diane took a more open, hopeful, and engaged approach.Being at a later-stage in life when functional seizures started made it easier to accept as there was a sense in which individuals had not been prevented from setting up their lives.For others, living with functional seizures for an extended period gave them time to come to terms with their situation.

Consequences
Participants reported widespread consequences of their functional seizures (Table 10).More immediately, most participants reported physical pain and discomfort.The pain came from headaches after their functional seizures, falls or injuries (dislocated joints, bitten tongues), or intense, widespread, muscle spasms and cramps during their seizures.Distress was another commonly reported immediate consequence, functional seizures were described as frightening because of their unpredictability or upsetting because of the loss of control.Some participants spoke about how fear can turn into frustration over time, perhaps as they begin to habituate to repeated experiences.
'Participation restrictions' are defined by the World Health Organisation [41] as restrictions from an impairment or disability that impact on a person's involvement in life situations including activities of daily living, engaging in employment or social life.Participants spoke about how the more general sense of insecurity resulting from their functional seizures affected their social and leisure time, withdrawing from the outside world, resulting in isolation and loss of friendships.
The impact of developing later-onset functional seizures on employment had serious financial implications for people who had mortgages or children to provide for.Loss of the ability to drive was also a commonly reported restriction, which was tied in with a loss of freedom and independence.Loss or changes to plans were reported by most participants, who mourned the future they had envisaged for themselves.For one participant in mid-adulthood, this meant retiring decades earlier than anticipated: for those in later adulthood, the loss of holidays with their spouses, opportunities for volunteering in the community, or feeling settled and comfortable.For all, this seemed to have a negative impact on their sense of identity.
Roles in the family system changed.Participants experienced concern for the emotional impact of others witnessing their seizures.They also conveyed frustration around loss of independence and the role reversal when, at this stage of their lives, they would otherwise have been caring for their spouses or children.
Perhaps the most consistent sub-theme within 'consequences' was participants' experiences of 'guilt, shame, and stigma'.All participants reported at least one experience of perceiving stigmatising attitudes from other healthcare professionals, or from people in their community.Fear of being stigmatised caused them to avoid social events like watching their children's sporting events or visiting supermarkets in their local areas.There was also evidence that at points in their lives, participants had begun to internalise this stigma, wondering to themselves if they were making up their conditions or describing their functional seizures in somewhat pejorative terms, such as 'going mad' or 'crazy'.However, at the time of interview, it appeared that all participants had accepted the legitimacy of their condition, even if wider society had not.

Discussion
The purpose of this study was to understand the illness representations, as defined by the Common-Sense Model [14], of people with lateronset functional seizures.Participants characterised their functional seizures as a mysterious brain disorder analogous to a computer malfunction.Symptoms were experienced as episodic, involuntary movements accompanied by dissociation with variable intensity and frequency.The perceived 'Timeline' of functional seizures was indefinite.Causal explanations, if present, were held lightly but linked most commonly to health-related events and to a lesser extent, cumulative life stress.Several participants reported that they could not identify any reliable triggers, however, did go on to identify some: including stress and relaxation.Similarly for cures/controls, several participants perceived there was nothing they could do to control their functional seizures, and yet were able to describe strategies for warding off functional seizures in-the-moment, reducing their likelihood, and for coping with the impact of functional seizures on their lives.There was a level of uncertainty and contradiction throughout participants' illness representations, reflecting the confusing nature of their experiences.Nevertheless, participants were clear about the impact of functional seizures.Consequences included physical pain and discomfort, distress, guilt, shame, stigma, and restricted participation.
For the most part, the content of illness representations identified during interviews was verified by the quantitative measures used to characterise the sample; depression scores achieved 'caseness' for five of the eight participants (i.e.consistent with Consequences -Distress) and symptoms had a significant impact on functioning (i.e.Consequences -Participation Restrictions) for seven of the eight participants.BIPQ scores suggested that most participants held adverse or 'threatening' illness representations, with a long expected duration, low degree of personal control, and with significant consequences on their lives and emotional wellbeing.Physical health and chronic stress were cited as the two most common causal factors on the BIPQ.Two participants indicated on the coherence subscale that they had a clear understanding of their functional seizures, which was somewhat at odds with elements of their respective narratives; this sort of contradiction is difficult to reconcile but has been noted in other qualitative studies of adults with functional seizures when discussing their understanding of causes [42,43], and so perhaps also reflects how difficult it can be for people to make sense of their functional seizures.
The idea that two opposing states -stress and relaxation -might both act as triggers, can be accommodated by the 'Window of Tolerance' model [44,45].This model proposes that the central nervous system functions optimally when the autonomic nervous system is not in a state of extreme hyper-or hypo-arousal.Outwith their 'window of tolerance', individuals are more likely to experience so-called 'freeze' or dissociative responses.Stress (hyper-arousal) and relaxation (hypo-arousal) may therefore both serve to trigger the hypothetical 'seizure scaffold' [2] and bring about a functional seizure.Indeed, other models of functional seizure pathophysiology have incorporated intolerance of arousal into biopsychosocial accounts of the disorder [e.g., 46,47].
When questioned, many participants' initial responses indicated that they could not identify triggers, warning signs, or controls.This could be explained by the presence of multiple or inconsistent triggers.An alternative explanation might be the presence of alexithymic traits and/ or abnormalities in interoceptive processing which have been identified in some studies of people with FND [e.g., 48,[49][50][51], and would make it harder for this patient group to have a clear sense of their internal milieu.What is more, interoceptive sensitivity has been shown to decrease with age [52], which may further exacerbate patients' difficulty identifying warning signs, triggers, and controls.Interestingly, later in their interviews, most were able to identify strategies that helped to reduce the likelihood of seizures occurring or mitigate the impact of them on their lives, although they did not go on to revise their earlier claims about having 'no control'.Conflating being unable to fully prevent a health event with having 'no control' over it, perhaps reflects a degree of 'black-and-white thinking' which would be a sensible target for treatment in psychological therapy.
While there have been no studies exclusively examining illness representations of people whose functional seizures start before 40 years old, many of the representations identified in this sample were similar to those endorsed by adult onset samples [21].For example, Green et al., [42] also applied the Common-Sense Model using an Interpretative Phenomenological Analysis approach to understand the illness representations of nine adults with functional seizures (age of onset ranging between 11 years and 57 years) attending a neuropsychiatry clinic.Representations around 'identity' were perhaps less clear than in the present study, but arguably public understanding has advanced in the 19 years since this study was published.Nevertheless, participants endorsed similar variability in symptoms, an indefinite 'timeline', and an uncertainty about cause with a tentative link to stress [42].Uncertainty but a willingness to entertain both physical and psychological causes [42,43,[53][54][55] as well as a belief that people with functional seizures have limited or no control [42,43,[55][56][57] have also been identified in qualitative investigations of adult samples.Similarly, people with adult-onset functional seizures endorsed widespread impact on participation including isolation, difficulties with daily life, loss of valued activities and the experience of stigma [43,[53][54][55]58,59].
Other illness representations seemed perhaps more specific to people with later-onset functional seizures.In mid-to late-adulthood, participants had greater caring, professional, or financial responsibilities, or they were planning for retirement.Developing functional seizures at this stage of life was therefore viewed as more disruptive in some ways (Consequences).However, increased age or living longer with functional seizures was viewed by some as an advantage for acceptance of the condition, which it seems helped participants to adjust to and therefore cope with their functional seizures (Cures/Controls).Indeed, increased age has previously been positively correlated with acceptance [60] although given the minimal amount of longitudinal data it is difficult to establish whether the positive association between age and acceptance stems from cohort effects (e.g., generational differences in economic and social contexts) or biopsychosocial effects of ageing.It has been proposed that health-related trauma might be a more relevant causal factor for people with later-onset functional seizures [5,61].Indeed, several (but not all) participants did identify healthrelated events or trauma as a potential cause.Given that physical ill health is more likely with increasing age, it may also be that healthrelated traumatic events are more common [62].Notably, four participants indicated in the survey that they had experienced a significant health-related event within two years of the onset of their functional seizures.However, the other four participants did not indicate any such events near the onset of their seizures.A cross-sectional study of 365 adults across the lifespan with functional seizures found that the number of comorbidities increased with age of presentation from early adulthood, however, this association followed a similar trend to what would be expected in the general population who do not go on to develop functional seizures [12].As such, it cannot be argued that people with later-onset functional seizures are different to their age-matched peers in terms of the frequency of health-related events, but perhaps an underlying vulnerability to the development of functional seizures may be exposed by health-related events occurring in later life [63,64].

Clinical and research implications
While many of the themes identified in this study are similar to those identified in samples including individuals with earlier ages of onset [21], the illness representations of people with later-onset functional seizures have clinical implications for working psychologically with this patient group.Healthcare professionals may support people with lateronset functional seizures by exploring their illness representations in more detail and building awareness.Given a possible age-associated decline in interoceptive sensitivity, people with later-onset functional seizures may require more training than their younger counterparts to identify signs of physiological hyper-or hypo-arousal that herald the onset of a functional seizure.Once 'tuned in' to warning signs, distraction and grounding techniques could be employed to reduce their frequency.If functional seizures cannot be prevented in the moment, individuals could be encouraged to take steps to manage their impact (e. g., through accessing social support).This may help to reduce distress through enhancing a sense of personal control.Many participants cited experiential/situational avoidance as a strategythis was perceived to help in the short-term but can become unhelpful in the long term if used inflexibly [65].Therapeutic protocols that challenge the inflexible use of avoidance or explain how it can serve to maintain difficulties may therefore also be indicated.
It might be argued that the main point of difference from earlier onset samples is the interaction between life stage and the theme of 'consequences'.It is therefore important for the clinician to hold in mind that later-onset functional seizures will likely carry significant implications for finances, caring responsibilities, and retirement.Indeed, a recent systematic review has demonstrated the significant economic costs to patients associated with a diagnosis of FND, comparable to or greater than other chronic neurological conditions [66].Psychological adjustment to the 'consequences' of later-onset functional seizure and participants' beliefs around the value of acceptance point towards the potential utility of interventions such as Acceptance and Commitment Therapy [ACT ;67].The theoretical underpinnings of ACT as a transdiagnostic intervention which makes no assertions regarding aetiology but aims to reduce distress in part through building willing acceptance to 'live well in a difficult context' (as opposed to a more passive resignation) may therefore be a 'good fit' for people with later-onset functional seizures.Indeed, there is preliminary evidence that ACT is effective for reducing symptom interference and improving mood in a case series of people with FND [68].This may be enhanced by wider multi-disciplinary team support involving occupational therapy to increase engagement with valued activities [69].

Strengths and limitations
There is no formally agreed age cut-off for 'later-onset functional seizures' and the boundary for 'late onset functional seizures' has ranged between 50 and 60 years old in other studies [5,9,11].Although we used age-related demographic data from a comparatively large sample [7] to justify 40 years old as our cut-off, we acknowledge the somewhat arbitrary nature of this decision.While efforts were made to minimise selection bias by recruiting across different settings (NHS and internet), the self-selecting nature of recruitment means that the included sample may be more 'psychologically minded' or have different illness experiences and representations compared to the rest of the population with functional seizures.For example, the average seizure frequency reported by participants in this study was notably higher (approx.26 per month) than of participants included in the largest psychotherapy trial to date [15 per month; 29].Inclusion criteria were designed to facilitate recruitment of participants with reliable diagnoses whilst also capturing experiences across a broad range of ages (forties, fifties, and sixties).While this was consistent with the research question, it also necessarily resulted in some variability in experiences which may have contributed to the uncertainty identified in the analysis.Regrettably, no participants in their seventies or eighties were identified and participants experiencing severe fatigue opted out; so, these experiences were missing from the study.This study will also have missed participation from individuals who had not accepted the legitimacy of their diagnosis.Finally, we did not have access to information regarding the certainty of diagnosis according to the ILAE categories for all patients [1].
While the use of the Common-Sense Model as a theoretical framework was made explicit and helped to clearly operationalise the research question, it also meant that the researchers approached analysis with pre-conceived ideas about how to make sense of participants' narratives.Unfortunately, it was not possible for us to invite participant feedback on the coding scheme.However, verification of themes with participants' responses to self-report measures helped to increase confidence in the identified themes.Furthermore, discussion of themes and associated quotes within the immediate research team, and later, the wider research team further enhanced the trustworthiness of the analysis.

Conclusions
This is the first study to assess the illness representations of people with later-onset functional seizures.Later-onset functional seizures were characterised as a mysterious brain disorder analogous to a computer malfunction and involving involuntary movements associated with alterations in consciousness.Perceptions of duration were indefinite, and triggers were unknown or at the extremes of autonomic arousal.Healthrelated events were the most strongly endorsed causal factor, followed by cumulative life stress, but explanations were held lightly.Developing functional seizures after 40 was experienced as disadvantageous for the consequences of the disorder, but advantageous for building acceptance.While many themes were similar to those identified in earlier-onset samples, clinicians should consider the likelihood of interactions between life stage and participation restrictions, for example disadvantageous impacts on financial and caring commitments.

Table 1
Participant demographics, functional seizure characteristics, and health history.

Table 2
Participant psychopathology and functional impairment scores.

Table 3
Brief Illness Perception Questionnaire subscale and total scores.Higher scores indicate more 'threatening' illness perceptions.Max subscale score = 10, Max total score = 80.

Table 5
Final template of themes.

Table 6
Supporting quotes for 'Identity' theme.

Table 7
Supporting quotes for 'Causes' theme.
'I think it was the amount of stress I was under at the time…The seizures were simply overloading myself; I've broken my nervous system effectively.'(Diane) 'It was probably the most peaceful I'd been for a long time.I wasn't seeing anybody in a relationship, there was no stress, I loved my job, very, very odd.You know I wasn't stressed or anything.'(Susan) 3.3 Explanations held lightly 'Yeah, I would say that I have no strong beliefs that any of these things caused it…' (Brian)

Table 8
Supporting quotes for 'Triggers' theme.
'…when I have a bad turn [meaning a functional seizure] it's not that I'm stressed.I don't [need to] feel stressed to have a seizure.' (Margaret) '…or there's too much going on my brain can't really go nowhere and it just shuts down, you know?' (Roger) 4.3 Relaxation '…it took me a long time to switch my mind off in meditation, you know its long process getting to that space.And now I'm terrified to switch my mind off so I can't, I don't even try to because I'm worried, I'll have a seizure.' (Brian) I.A. Williams et al.

Table 9
Supporting quotes for the 'Cures/control' theme.

Table 10
Supporting quotes for 'Consequences' theme.
having seizures in the kitchen and your risk falling on the gas cooker and that… So… I think the impact when you're over 40 years is vast really.' (Jane)' 'I'm the one that has it…he's[husband]the one who has to deal with it' (Margaret) 6.4 Guilt, shame, and stigma 'People think "Oh you know, she's just more of a nutter than what we already thought she was," putting it bluntly, and it's just more mental health stigma piled on me.' (Jane)