Triggers and Clinical Presentations of Functional Neurological Disorders: Lessons from World War 1

Functional symptoms that defy standard medical explanations pose a considerable challenge to public health. They are highly prevalent in the general population [1] and affect over 40% of primary care patients [2] and about a third of patients who attend general neurology clinics [3, 4].

Functional disorders have challenged physicians for several centuries and came into focus in the late 19th century when Charcot’s theatrical demonstrations and hypnosis of mainly young female patients with functional neurological disorders (FNDs) attracted great audiences [5]. Another important boost of knowledge came with the shell shock epidemic of World War 1 (WW1), during which tens of thousands of traumatized soldiers from all combatant countries developed functional syndromes. However, post-traumatic reactions of civilian populations during the major conflicts of the 20th century (and through to today) are still a neglected topic. Another hitherto unexplored question is if specific triggers (war related or unrelated to the war) brought on a specific set of functional symptoms. Historical records allow us to study phenomenology and natural history of functional disorders in a setting that did not have modern treatments at its disposal. The study analyses 937 case records of patients with functional disorders treated at the National Hospital (NH) for the Paralysed and Epileptic in London during WW1. The records provide the unique opportunity to compare military and civilian patients who were treated in one hospital by the same physicians in a defined period of time. In addition to the rare opportunity to study presentations of functional disorders in a well-defined historical context, these case records offer a deep insight into patients’ individual histories. They allow us to explore the exact nature of triggering events (sudden vs. chronic and “physical” vs. “psychological”), the cluster of physical and psychological symptoms, and also sex differences in clinical presentations.

The study analyses 937 case records of patients with functional disorders treated at the National Hospital for the Paralysed and Epileptic (NH) (today the National Hospital for Neurology and Neurosurgery) in London during WW1 (all admissions from August 4, 1914 – the day of the British entry into the war – to November 11, 1918 – the Armistice). The author read all approximately 5,000 case records from that time (including case records from early 1914 and 1919 that could also contain material covering the war years) and excluded cases with a clear organic pathology (e.g., malignancies of the brain and spine, nerve injuries, general paralysis of the insane [syphilis], strokes, and muscular dystrophies) from the analysis. Patients with functional disorders were then divided into 3 main groups: soldiers (n = 436), civilian males (n = 153), and (civilian) females (n = 321). Twenty-seven children with functional symptoms were also admitted to the NH during that time (age < 16 years). The sample of military patients has partly been described in a previous article [6] but has been reanalysed for the purposes of the present study. Because this study aimed at documenting all functional symptoms in the civilian and military populations and also recording more subtle functional presentations, the total number of functional symptoms increased. In the historical study on military admissions, a maximum of symptoms from 4 different symptom categories could be rated, whereas the current study documented symptoms from up to 6 different symptom categories (6 categories were enough to document all symptoms that were present in each patient). The same strict criteria were used for civilian and military admissions (see online suppl. methods; see www.karger.com/doi/10.1159/000507698 for all online suppl. material).

The case records provide a detailed account of psychological and physical symptoms as well as an in-depth neurological and general physical examination. Already a century ago, the NH neurologists diagnosed a functional disorder not only through exclusion of organic pathology but also based on positive diagnostic signs (e.g., Hoover’s sign). The diagnosis of a functional disorder in this study (instead of or in some cases in addition to a defined neurological illness entity) was based on the concordance of the assessment of the physicians of the time and the author’s own reading of the records. In cases of doubt, the author consulted with experts in movement disorders.

For this study, functional symptoms were allocated to 16 major categories (Table 2) following a set of inclusion/exclusion criteria that are described in detail in the online suppl. material. Contemporary diagnoses, treatment approaches, and treatment outcomes were also analysed for the different patient groups. Statistical analysis for group differences was conducted in SPSS version 25 (IBM SPSS Statistics for Windows, version 25.0. IBM Corp., Armonk, NY, USA) using the appropriate tests for categorical (χ²) and continuous (one-way ANOVA) variables.

During the time of war, about 3,770 inpatients were treated at the NH (among them about 1,000 soldiers). About one-quarter of these patients suffered from a functional disorder. The following analysis is based on the adult patients (n = 910); the group of patients younger than 16 years (n = 27) is described separately. Sociodemographic characteristics of the adult sample are shown in Table 1.

Functional symptom categories are shown in Table 2. Nearly one-third of patients from each group suffered from hyperkinetic movements (e.g., shaking, tremor, and choreic movements). Paralysis/weakness/gait disorder, somatosensory disturbances, pain, and somatic symptoms were also common, occurring in more than 20% of cases in each group.

χ² tests of independence were performed to examine if certain symptom categories were more common in any of the 3 groups. Paralysis/weakness/gait disorder and speech disturbances were significantly more common in soldiers than in civilian male and female patients. Pain and somatic symptoms were more common in female than in both categories of male patients. Somatosensory disturbances were less common in civilian men than in the other groups. The nature of somatic symptoms was generally similar across groups, but gastrointestinal complaints and general loss of energy were less common in soldiers (see Table 3).

According to the current diagnostic criteria (DSM-5; DSM = Diagnostic and Statistical Manual of Mental Disorders), about three-quarters of the adult patients with functional disorders (668/910; 73.4%) would have been classified as suffering from conversion disorder (functional neurological symptom disorder, DSM-5), including all patients with weakness/paralysis/gait disorder, hyperkinetic movements, swallowing problems, speech symptoms, seizures, anaesthesia or sensory loss, and/or special sensory symptoms, including visual and hearing disturbance (see Table 2). The majority of the remaining patients with functional disorders would have fulfilled the criteria for somatic symptom disorder. It is noteworthy that 81.7% of soldiers presented with functional neurological symptoms, most of them with a combination of functional neurological symptoms (see Table 2).

More than a third of the civilian functional disorder patients received a contemporaneous diagnosis of neurasthenia (online suppl. Table 1). Overall, 21.2% of female patients were diagnosed with “hysteria,” making it the second most common diagnosis in females. In contrast, only 5.2% of male civilians were diagnosed with hysteria. “Functional disorder” was the most common diagnosis among military patients (32.6%); this label was chosen twice as often in military as opposed to civilian patients (15.7% in male and 16.2% in female civilian patients).

Twenty-seven children, 16 girls and 11 boys, with functional disorders (average age 11.9 ± 2.9 years) were treated at the NH during WW1 (online suppl. Table 3). In all, 44.4% of children obtained a diagnosis of hysteria (27.3% of boys and 56.3% of girls), a rate which is much higher than that in the adult patients. In children with functional symptoms, motor symptoms (mainly paralyses) were the most common presentation, in over one third of cases (37.0%), followed by hyperkinetic movements (33.3%) and seizures (29.6%). Somatic symptoms occurred at a lower rate in children than in the adult population (29.6%). Twenty-four children would have fulfilled the criteria for an FND diagnosis.

The major categories of treatment are listed in Table 1. Most civilian patients with functional disorders (about 70%) were prescribed physical treatments, such as massages, baths, and heat treatments, in combination with rest and a high-caloric diet. Soldiers also received physical treatments (42%). However, the most common treatment applied to military patients with functional disorders, in particular functional motor disorders, was electrotherapy. Women received electrotherapy as frequently as soldiers, despite the lower rates of functional paralysis/weakness; conversely, the rates of isolation therapy in women were higher than those in the male patient groups.

Treatment outcomes were rated by the physician and documented on the patient record as “cured,” “improved,” “in status quo,” or “worse” (see Table 1). Military patients showed better treatment outcomes than male (χ² [3, N = 589] = 35.1, p < 0.001) and female civilians (χ² [3, N = 757] = 32.1, p < 0.001). Nearly half of the soldiers with functional disorders who were treated with electrotherapy (48.7%) were cured; in 43.4%, symptoms improved. The cure rate for civilian male patients who received electrotherapy was only 20.5% and that for female patients was 28.3%. Treatment outcomes were dependent on the chronicity of symptoms. About one-third (28.6%) of patients with a chronic course showed no improvement at all during their hospital treatment (compared to 12.2% in the non-chronic group).

The most commonly documented triggers are listed in Table 4. In about one-third of both military and civilian cases, no trigger could be identified. Soldiers typically developed symptoms following combat trauma, such as shell explosion, injury during battle, burial, or witnessing a comrade’s violent injury/death. However, soldiers could also develop symptoms after a minor injury/shock during home leave [6]. For civilians, the most important identified categories of triggers were medical illness and physical accidents, both for the general cohort (Table 4) and for the approximately two-thirds of the cohort who had an FND (online suppl. Table 2).

When only considering those patients whose symptoms developed after accidents (n = 80; also including 14 soldiers who had accidents while not serving at the front), a different symptom pattern emerged, with over half of these patients (53.8%) suffering from functional paralysis/weakness/gait disorder (loss of motor function) – commonly combined with somatosensory deficits – and also somatic symptoms (47.5%) and pain (46.3%) (Fig. 1). About a quarter of these patients (22.9%) were diagnosed with “hysteria,” another quarter with “functional disorder” (25.3%), and another sizeable group (18.1%) with “traumatic hysteria/neurasthenia/neurosis.”

Conversely, in patients whose symptoms were triggered by (chronic) work-related stress, loss of motor function was relatively rare (23.3%), but somatic symptoms occurred in more than two-thirds of cases. Medical conditions (commonly infectious diseases such as influenza or typhoid) triggered a symptom pattern that was intermediate between the accident and work-related stress groups, with 35.6% of patients with loss of motor function, 61.1% with somatic symptoms, and 40.0% with pain.

This study analysed 937 case histories of patients with functional disorders treated at the NH in London during the time of WW1. These records allow us to explore the impact of culture and triggering events on clinical presentations in a well-powered sample.

While somatic symptoms such as headache and fatigue are universal, culturally independent ways of expressing distress [8], functional paralysis/weakness is a rarer phenomenon in the general population, with a 6-month period prevalence of 1% (1). Recent studies of neurology inpatients have suggested frequencies of functional paralysis between 1 and 18% [9]. At the NH, about 9% of WW1 inpatients suffered from functional paralysis.

Although the soldiers presented with significantly higher rates of functional paralysis than the civilians, similarly high rates were seen for civilians whose symptoms were triggered by an accident. Conversely, more chronic, less intense stressors (medical conditions and work-related stress) produced much lower rates of functional paralysis (see Fig. 1) but resulted in increasing rates of somatic symptoms. Thus, there may have been other factors, such as the acuteness of the stressor and/or the presence of acute pain/an unexpected injury rather than the military setting that triggered the development of functional paralysis. One symptom that stood out in the soldier group was loss of speech, which was significantly more common among soldiers than civilian patients. The link between functional speech disorders and war trauma has been noticed before [10].

The technological changes that accompanied WW1, particularly the use of zeppelins and airplanes to attack civilian populations, added a new unexpected dimension to warfare and challenged the idea that the home front and the war front were naturally separated. The experience of air raids or just the threat of an attack from the skies caused constant uneasiness and strain. Tellingly, many of those patients whose breakdown was triggered by air raids (among them 4 children) developed anxiety and hyperkinetic movements (online suppl. Fig. 1). Many patients would return to the hospital years after the war, still suffering from hyperkinetic movements brought on by zeppelin or other air raids (author’s own ongoing research).

Comparative analyses of male and female patients with functional disorders are scarce, partially due to low sampling in men [11]. In line with current-day figures showing that women are 2–3 times more likely than men to be diagnosed with FNDs [12], two-thirds of civilian patients with functional disorders treated at the NH were female. The finding that female patients had significantly higher rates of pain and somatic symptoms than men (soldiers and civilians) corresponds with more recent studies [13].

Triggers in the civilian patient groups were similar in men and women (most common identifiable triggers: medical conditions, followed by minor accidents and work-related stress). In the female patient group, sex-specific triggers (e.g., abortion, childbirth, and menopause) could be identified in only 6.9% of cases. The fact that work accidents and work-related stress triggered functional symptoms in women in similar dimensions as in men can partly be explained by changing roles of women and their integration into the country’s workforce during WW1.

The treatments were predominantly physical, and treatment outcomes were promising, in line with recent studies on physical treatments for FNDs [14, 15]. The very high cure rate of about 50% in soldiers with functional disorders who received electrotherapy might be explained by the sophisticated treatment approach for military patients, with a combination of electrotherapy, suggestion, persuasion, and physical training with positive reinforcement and model learning [16]. Some elements of this approach, in particular the demonstration of movement in an affected limb, are part of today’s treatment protocols with transcranial magnetic or electrical stimulation which, however, have shown inconsistent results in functional disorders [17, 18].

Treatment approaches towards female patients were different from those of male military and civilian patients, as reflected in the higher rates of isolation therapy in female patients and the relatively high rates of electrotherapy (despite lower rates of functional paralyses/weakness) compared to military patients. These differences may be explained by general attitudes towards female hysteria and treatment protocols which had a clear punitive component.

The comparison with present-day cohorts has to take the lack of standardized instruments in the historical assessments into consideration. The study, therefore, focussed on syndromes and symptom clusters (which can be identified from detailed patient and clinician narratives in the records) rather than categorical diagnoses. One main outcome was the striking similarity of triggers a century ago and today, with the main groups being infections, work-related stress, and accidents. Functional disorders after infections, for example, Lyme disease [19], and also following work-related stress are in the focus of attention again [20]. Accidents without major physical injury frequently lead to chronic functional conditions, in particular chronic pain syndromes [21]. The results of this historical cohort are in line with recent studies highlighting physical injury and physical illness as common triggers for FNDs [22]. The important role of medical illness and accidents in triggering FNDs in civilians highlights a very relevant point: although psychological trauma very likely plays an important role, the actual triggering event (at least in the patient’s perception) often has a physical nature. Whether this implies that a physical stressor tips the balance of an already vulnerable physiological state or whether the psychological processes following the physical illness/injury (including psycho-immunological interactions) are responsible (or both) cannot be answered from the historical records but would be research questions for prospective studies today. The documented treatments also primarily addressed the physical side of the complaints, with considerable success (at least as reported in the records), which is in line with modern efforts to treat FNDs with physiotherapy and other physical interventions [7, 15].

In addition to the similarity of triggers now and then, clinical presentations also show many commonalities. The historical dataset confirms the overlap of functional symptoms (online suppl. Table 1). Considerable co-occurrence of somatic syndromes and symptoms has also been described for present-day patient groups [23-25]. Although military patients showed higher rates of paralysis/weakness than civilians, when correcting for the nature of the trigger (acuteness and level of physical injury), those differences disappear. The military/civilian context seems to be less relevant for the symptom expression than the nature of the trigger. What this analysis shows then, certainly for the civilian group but to some extent also for the soldier patients, is that presentations of FNDs and other functional syndromes have remained remarkably consistent over the last century. This does not exclude a major role for cultural factors – after all, most present-day comparison samples come from the same European/Anglo-American culture as the WW1 sample, and a comparison across a wider geographical range would be needed – but suggests some biological invariance in the presentation of functional symptoms.

The author is grateful to the staff of the QS Library and Archives, University College London.

Ethical approval was not required for this study because it is based on historical case records.

The author has no conflicts of interest to declare.

This research received no specific grant from any funding agency, commercial or not-for-profit sectors.

The author has designed the study, collected all material in the archives, analysed the data, and written the manuscript.

All case records can be studied at the Queen Square Library and Archives at University College London. A list of the cases included in the study can be obtained upon request from the archives.