INTRODUCTION — Conversion disorder (functional neurologic symptom disorder) is characterized by neurologic symptoms (eg, weakness, abnormal movements, or nonepileptic seizures) that are inconsistent with a neurologic disease, but cause distress and/or impairment [1]. The disorder is common in clinical settings and often has a poor prognosis [2-5].
This topic reviews the epidemiology, pathogenesis, and prognosis of conversion disorder. The terminology, diagnosis, differential diagnosis, clinical features, assessment, and treatment are discussed separately, as are specific subtypes of conversion disorder (psychogenic nonepileptic seizures and functional movement disorders):
●(See "Conversion disorder in adults: Terminology, diagnosis, and differential diagnosis".)
●(See "Conversion disorder in adults: Clinical features, assessment, and comorbidity".)
●(See "Conversion disorder in adults: Treatment".)
●(See "Psychogenic nonepileptic seizures: Etiology, clinical features, and diagnosis".)
●(See "Functional movement disorders".)
EPIDEMIOLOGY — The prevalence of conversion disorder (functional neurologic symptom disorder) has been more widely studied in clinical settings than in the general population.
General population — A review found that the estimated incidence of conversion disorder across disparate geographical settings was 4 to 12 per 100,000 per year (0.004 to 0.012 percent), and the community prevalence of conversion disorder based upon case registries was 50 per 100,000 per year (0.05 percent) [6].
Clinical settings — The point prevalence of conversion symptoms in clinical settings ranges from 2 to 6 percent:
●A prospective study of 157 internal medicine inpatients found that conversion disorder was present in 2 percent [7]
●A prospective study of 3781 neurology outpatients found that conversion disorder was present in 6 percent [2]
●A retrospective study of 7836 outpatients evaluated by one neurologist found that conversion disorder was present in 4 percent [8]
Sociodemographic correlates — It is not clear if there are any established sociodemographic correlates of conversion disorder, per se. However, a prospective study compared 1144 neurology outpatients with symptoms somewhat or not at all explained by a recognizable neurologic illness (including 209 patients with conversion disorder) with 2637 patients whose symptoms were explained by a neurologic illness [3]. Symptoms not explained by disease were associated with:
●Younger age
●Female sex
●Greater disability, including physical, role, and social functioning
Conversion disorder has been reported in patients of all ages, from early childhood to old age, but is rare before age 10 years [9,10]. A prospective study of 50 patients with conversion disorder found that the mean age of onset was later for psychogenic weakness and movement disorders, compared with psychogenic nonepileptic seizures (39 versus 27 years) [11]. Nevertheless, there have been reports of nonepileptic seizures with onset in an older age group (mean age 61 years, n = 26), amongst whom comorbid disease was prominent [12].
Many studies have found that conversion disorder is more likely to occur in females than males [3,13,14]. However, the preponderance of females may vary by the subtype of the disorder. A pooled analysis of 52 studies that included 2270 patients with conversion disorder found that the proportion of females was less for psychogenic limb weakness than psychogenic nonepileptic seizures (48 versus 74 percent) [11].
ETIOLOGY AND PATHOGENESIS — The etiology and pathogenesis of conversion disorder (functional neurologic symptom disorder) are not clear [15]. However, there are many biological, psychological, and social factors that are more common in patients with functional symptoms than patients with comparable symptoms due to recognizable disease [16]. These factors may predispose patients to conversion disorder or precipitate and/or perpetuate symptoms (table 1).
Premorbid clinical factors — Psychological factors or life events such as trauma, interpersonal conflicts, and recent or old stressors may be associated with onset of conversion disorder [1]. However, psychological factors are not always reported, nor are they specific to conversion disorder [17,18]:
●A meta-analysis of 34 retrospective studies examined the history of stressful life events and maltreatment in patients (cases) with conversion disorder (n >1400) and in healthy or patient controls (n >2200) [19]. Stressful life events were more common in patients with conversion disorder than controls (odds ratio 2.8, 95% CI 1.4-6.0). In addition, maltreatment had occurred in more patients than controls:
•Emotional neglect (49 versus 20 percent)
•Sexual abuse (24 versus 10 percent)
•Physical abuse (30 versus 12 percent)
Among the 34 studies, 13 found that some patients with conversion disorder, ranging from 14 to 70 percent, did not have a history of either stressful life events or any type of maltreatment.
●In a subsequent retrospective study of 322 patients with motor conversion disorder and 644 psychiatric control patients, the proportion of patients with a history of childhood sexual abuse was similar in both groups (20 and 22 percent), as was the proportion with a history of physical abuse (23 and 22 percent) [14].
Pre-existing disorders and symptoms may predispose patients to developing conversion disorder; in some cases, this process is considered symptom amplification [20]. Compared with controls with recognizable diseases, patients with conversion disorder are more likely to have a history of pre-existing psychiatric disorders (eg, depressive, anxiety, and personality disorders) and other somatic symptoms (eg, pain, fatigue, and cognitive impairment) [11,14,21], or other types of preexisting functional somatic disorders, such as irritable bowel syndrome [22]. In addition, conversion disorder is often preceded by neurologic illnesses (eg, migraine, peripheral nerve pathology, or stroke) that appear to trigger, but do not explain the conversion symptoms [23]. Physical injury may also precede conversion symptoms [24,25].
An excess of symptom modeling (exposure to similar symptoms in others) has been reported in some studies of conversion disorder [26-28] but not others [29-31]. In a negative study, which was relatively large and included 132 patients with functional movement disorders and 148 healthy controls, the proportion of health care workers in each group was comparable (25 and 20 percent) [32].
Perpetuating factors — Conversion symptoms may be perpetuated by beliefs that there is an irreversible neurologic disease, which may create a powerful obstacle to rehabilitation. Clinicians may also make things worse by failing to give a clear explanation and positive diagnosis of the symptoms [33], excessively investigating symptoms [34], prescribing inappropriate drugs and appliances (eg, crutches and wheelchairs when these exceed the patient’s needs) [35], performing unnecessary operations, and misattributing symptoms to recognizable disease or to irrelevant radiologic or laboratory findings (eg, age-related degenerative changes in the vertebrae) [36]. In addition, conversion symptoms may be perpetuated by physical deconditioning, comorbid psychiatric disorders, disability-related financial benefits, and litigation [16,37]. Another potential perpetuating factor is life stress (eg, work or family) that was relieved by onset of the disorder and would probably recur if the patient was to recover from conversion disorder [38].
Hypotheses — Several hypothetical models attempt to explain how conversion symptoms develop; these models are not mutually exclusive.
Cognitive-behavioral models — Multiple overlapping cognitive-behavioral models have been proposed to explain the etiology and pathogenesis of conversion disorder.
One cognitive-behavioral model is based upon findings that processing of perception and behavior mostly occurs outside of awareness, and proposes that conversion symptoms may result from psychological influences at these lower levels of processing [39]. It is suggested that patients with conversion disorder initially encounter (in themselves or others) a stimulus for a particular symptom, such as mild weakness from migraine. This generates a mental representation or memory of paralysis. Excessive anxiety about becoming paralyzed, and/or hypervigilance in looking for evidence (eg, weakness) of paralysis in oneself, may activate the mental representation to the point that it overrides sensory input and distorts awareness and behavior. Selective attentional bias leads patients to persistently focus upon evidence that they are paralyzed and ignore evidence that they are not [40]. Some studies of patients with conversion disorder suggest that their explicit awareness of emotional symptoms, such as anxiety, is low [41,42].
Another cognitive-behavioral model proposes that conversion disorder involves dissociative experiences [6]. Dissociation is subjectively perceived as disconnection from oneself (depersonalization) or the environment (derealization). During dissociation, awareness and integration of thoughts, feelings, memories, and identity is altered, as is integration of somatic experiences and functions, and patients lose functioning of motor control or sensory awareness [43]. Dissociation may occur as a consequence of fatigue, panic attacks [44], physical injuries [24,25,45], recognizable diseases [23,46], iatrogenic triggers such as general anesthesia [44,47], or drug side effects [25]. In this model, the symptom of paralysis or abnormal movement arises during the dissociated state when the patient is “depersonalized” and generally loses feelings of ownership over body movements. Attention paid asymmetrically to this experience, combined with fear of what the symptom might represent (eg, stroke), may intensify and cause more “localized” depersonalization, thus prolonging the symptom. In nonepileptic seizures, prodromal symptoms of panic attacks may become so unbearable that patients respond by losing awareness in an apparent blackout [48-50], with subsequent nonepileptic seizures occurring as a conditioned response to unpleasant thoughts, feelings, and situations, even in the absence of a prodrome.
Some cognitive behavioral models hypothesize that conversion disorder is perpetuated by maladaptive thoughts. Evidence for this hypothesis includes findings that recovery can be impeded by thoughts that conversion symptoms are due to a disease process with a known pathologic basis and beliefs that psychological factors are not involved [9,21,51]. In addition, beliefs that symptoms are caused by recognizable disease and are therefore not reversible appear to be a more powerful predictor of poor prognosis than the number of conversion symptoms or severity of disability [52].
Neurobiologic model — Conversion disorder may involve abnormalities in neural networks of grey matter brain regions rather than a disturbance in one specific structure [6]. These networks are thought to include frontal (orbitofrontal and anterior cingulate cortex) and subcortical (limbic) structures that may be activated by emotional stress, and in turn provide input to inhibitory basal ganglia-thalamocortical circuits that reduce conscious motor or sensory processing [53,54]. In addition, there may be aberrant activity in brain networks that include the inferior parietal lobe/temporoparietal junction and are involved in self-agency, which is the experience that one causes one’s own actions [53,55,56]. One hypothesis is that overly sensitive amygdala responses to fear (ie, abnormal responses to stimuli, including those that are objectively neutral) lead to changes in networks mediating sensory and motor function. In the setting of abnormal self-directed attention (eg, depersonalization), these changes produce sensations or movements that are associated with an abnormal sense of self control and are interpreted as involuntary symptoms of a disease [6,40].
Neuroimaging of patients with conversion disorder has suggested predominantly functional central nervous system changes, but also possibly structural changes. However, it is not known if these differences indicate causal factors, confounding comorbidities, or merely sequelae of the disorder. Functional magnetic resonance imaging comparing patients with conversion disorder (weakness subtype) with healthy controls during recall of traumatic events have found differences in regional brain activity (eg, prefrontal cortex and hippocampus) [57], which correlated with symptom severity [58]. In addition, structural magnetic resonance imaging studies that compared patients with functional movement disorder to healthy controls found evidence of altered brain structure in patients (eg, increased thalamic volume and decreased sensorimotor cortical thickness) [59,60].
One study of patients with conversion disorder found increased levels of basal cortisol [61], whereas another did not [62].
Psychodynamic models — The classic psychodynamic hypothesis, which gives conversion disorder its name, is that patients experience an unconscious conflict that is converted into a somatic symptom [16,63,64]. The symptom serves as a defense against anxiety and distress, which are reduced by the conflict remaining unconscious [43]. In this model, the conversion symptom symbolizes the conflict and may help the patient avoid an overwhelming situation [65]. The hypothesis is difficult to test scientifically, has been hard to apply clinically, and thus has little empirical support [43].
Subsequent psychodynamic hypotheses emphasize abnormal interpersonal relationships that develop in the context of problematic early relationships (eg, poor parenting) or traumatic events [39]. In these models, a new conflict or traumatic event leads to recurrence of previous patterns of abnormal behavior and the development of physical symptoms. The physical symptoms are regarded as a coping response, secondary to emotional dysregulation. As an example, somatic symptoms that lead to contact with clinicians may enable patients to meet dependency needs established earlier in life. Such hypotheses are also hard to test.
Evidence supporting the psychodynamic model includes a retrospective study that assessed patients with conversion disorder (n = 43) and patients with depression (n = 28) for stressors prior to symptom onset, and asked healthy controls (n = 28) about stressors in the prior two years [38]. The stressors/life events were rated for the extent to which symptoms enabled one to escape from an aversive situation or ameliorate the stressor (eg, providing escape from going to work and facing an abusive boss). The incidence of “escape” life events was greater in the month before symptom onset for patients with conversion disorder, than for patients with depression and healthy controls (53 versus 14 and 0 percent).
PROGNOSIS — Prognosis for conversion disorder (functional neurologic symptom disorder) is generally poor. Reviews of observational studies suggest that symptoms persist or worsen in approximately 40 to 66 percent of patients [66]. As an example, a 14 year prospective study of 107 patients with functional limb weakness found that symptoms remitted in 20 percent, improved in 31 percent, persisted in 23 percent, and worsened in 26 percent [67]. Most follow-up studies have also found persistent physical symptoms (eg, fatigue and pain) and impaired quality of life [66].
Among the subtypes of conversion disorder:
●Sensory symptoms [68,69] may have a better prognosis than weakness/paralysis [69], dystonia [70], and tremor [1,4].
●The prognosis of psychogenic nonepileptic seizures varies, but is often poor. In a systematic review of 25 observational studies, seizure remission occurred in 40 percent or fewer patients [66].
However, prognosis may be better if the disorder is diagnosed early. A prospective study of 54 patients with new onset psychogenic nonepileptic seizures found that within three months of diagnosis, remission had occurred in 50 percent [71].
Baseline predictors of outcome in patients with conversion symptoms vary between studies, and much of what is known comes from studies that included not only patients with conversion disorder, but other somatic symptom and related disorders as well. Across different studies with follow-up ranging from 8 months to 14 years, prognostic factors included the following (table 2) [4,6,23,52,66,67,69,72-76]:
●Factors associated with a positive outcome included:
•Onset in childhood or adolescence
•Early diagnosis
•Good response to initial treatment
•Comorbid anxiety or depression
•Subsequent change in marital status
•Good therapeutic alliance with the clinician
●Factors associated with a poor outcome included:
•Multiple physical symptoms
•Longer duration of symptoms
•Poor physical functioning
•Comorbid personality disorder
•Beliefs that symptoms are irreversible and caused by a disease with a known pathologic basis
•Illness-related financial benefits
However, these data do not enable clinicians to reliably predict outcome in an individual patient.
SUMMARY
●The estimated prevalence of conversion disorder (functional neurologic symptom disorder) in community studies ranges from 0.004 to 0.2 percent, and in clinical settings, from 2 to 6 percent. Sociodemographic correlates of conversion disorder may include younger age, female sex, and greater disability. Onset of conversion disorder rarely occurs prior to age 10 years. (See 'Epidemiology' above.)
●The etiology and pathogenesis of conversion disorder are not known. However, many factors have been identified that may predispose patients to conversion disorder or precipitate and/or perpetuate symptoms (table 1). (See 'Etiology and pathogenesis' above.)
●Psychological factors or life events such as trauma, interpersonal conflicts, and recent or old stressors may be associated with onset of conversion disorder. However, psychological factors are not always reported, nor are they specific to conversion disorder, and many patients with conversion symptoms may not have identifiable psychological factors. (See 'Premorbid clinical factors' above.)
●Pre-existing psychiatric disorders and functional somatic symptoms may predispose patients to developing conversion disorder. In addition, conversion disorder may be triggered by physical injury and neurologic illnesses. (See 'Premorbid clinical factors' above.)
●Conversion symptoms may be inadvertently perpetuated by clinicians who fail to give a clear explanation and positive diagnosis of the symptoms, excessively investigate symptoms, prescribe inappropriate drugs and appliances, perform unnecessary operations, and misattribute symptoms to recognizable disease or to irrelevant radiologic or laboratory findings. (See 'Perpetuating factors' above.)
●Several hypothetical models attempt to explain how conversion symptoms develop and are not mutually exclusive:
•Cognitive-behavioral models emphasize factors such as processing of perception and behavior outside of awareness, selective attentional bias that amplifies minor physiological stimuli and asymmetries, dissociation, and maladaptive thoughts that perpetuate symptoms. (See 'Cognitive-behavioral models' above.)
•The neurobiologic model hypothesizes that abnormalities in neural networks of grey matter brain regions involving frontal and subcortical structures are activated by emotional stress, and in turn provide input to inhibitory basal ganglia-thalamocortical circuits that reduce conscious motor or sensory processing. Networks involved in the experience that one causes one’s own actions also appear important. (See 'Neurobiologic model' above.)
•The psychodynamic hypothesis is that patients experience an unconscious conflict that is converted into a somatic symptom. The symptom is a defense against anxiety and distress, which are reduced by the conflict remaining unconscious. Subsequent psychodynamic hypotheses emphasize abnormal interpersonal relationships that develop in the context of problematic early relationships or traumatic events. (See 'Psychodynamic models' above.)
●Prognosis for conversion disorder is generally poor. Factors associated with positive and poor outcomes are listed in the table (table 2). (See 'Prognosis' above.)
