**4. Discussion**

Based on the recollections of all the 27 ex-Royal Free hospital staff and medical students who provided data for this study, hysteria as the underlying cause of the Royal Free outbreak seems inconceivable. Our study group's accounts are based on their firsthand personal experiences. McEvedy and Beard based their epidemic hysteria hypothesis on an analysis of some selected patient case notes [7]. They did not provide any evidence from the follow-ups of patients who had had the disease or from hospital staff. Epidemic hysteria is a diagnosis of exclusion, but McEvedy and Beard provided no data to exclude an infectious disease as a cause of the outbreak.

#### *4.1. Evidence for Infectious Illness*

Although no causal pathogen was found in the Royal Free outbreak, epidemiological and clinical features were consistent with an outbreak of an infectious illness. The disease was present in the wider community of north London as well as in all five hospitals of the Royal Free group. It affected male and female, young and older staff. Case to case infection clearly showed an incubation period of several days and no immediate visual transmission. Initially, prodromal constitutional symptoms and upper respiratory signs of low-grade fever, pharyngitis, and cervical lymphadenopathy were present. After a few days, diverse muscular and neurological symptoms and signs appeared in many patients. Lymphocytes typical of viral infection were seen in some patients. The duration of the illness ranged from a few days to many months. Its severity ranged from patients with the possible abortive disease to patients with severe disease. The authors of this paper consider that the closure of a large teaching hospital in London for three months might be necessary to control an outbreak of a persisting, highly infectious disease, that affected a large number of hospital staff and that might be transmitted to hospital patients. On the other hand, an outbreak of epidemic hysteria would not be a sufficient cause to close a hospital.

#### *4.2. Arguments for Epidemic Hysteria*

McEvedy and Beard asserted that there was little evidence of organic disease affecting the central nervous system [7]. Our study group contradicted this assertion and reported diverse neurological manifestations in many patients and permanent paralysis in a few. The original hospital medical staff report describes 148 patients with involvement of cranial nerves and/or motor or sensory defects in the limbs and trunk [3]. The undoubted neurological manifestations in this outbreak are not found in epidemic hysteria.

Our study group confirmed that the majority of those affected were female nurses. In McEvedy and Beard's cases selected for study, they found an attack rate of 0.8% in males and 11% in females and said this supported their epidemic hysteria hypothesis [7]. However, the original Royal Free hospital staff reports showed that females comprised 70% of the population at risk, and the attack rate was 10.4% for females and 2.8% for males [4]. These attack rates are comparable to those found in other outbreaks of ME, which ranged from 1.6% to 4% in males and 6.4% to 8.4% in females [23]. A high attack rate in females compared with males does not distinguish epidemic hysteria from ME.

McEvedy and Beard stated that the intensity of the malaise compared with the slight pyrexia supported their epidemic hysteria hypothesis [7]. Our study group confirmed malaise and mild pyrexia. Severe malaise that worsens with exertion is a cardinal feature of ME, but malaise and pyrexia are not features of mass hysteria [27,28]. Pyrexia is characteristic of infectious disease.

McEvedy and Beard noted the presence of subjective features similar to those seen in a previous epidemic (described by McEvedy [29]), of hysterical over-breathing in schoolgirls as evidence of epidemic hysteria. In this previous epidemic, many reported symptoms resembled the constitutional prodromal symptoms exhibited in the Royal Free outbreak. However, notably, hyperventilation was reported in 40% of the schoolgirls and tetany occurred in one-third of them [29]. Hyperventilation has been reported in 19%–32% of cases in outbreaks of mass hysteria [27,28]. McEvedy and Beard noted a raised respiratory rate only in four severely ill, Royal Free patients and speculated "this was a frightened and hysterical population whose over-breathing was intermittent and covert" [7]. Hyperventilation cannot be covert. Overt hyperventilation was not reported by our study group, nor reported in the original medical staff reports [3,4]. The notable absence of hyperventilation does not support the argumen<sup>t</sup> that the Royal Free outbreak resembled this outbreak of hysterical over-breathing in schoolgirls.

McEvedy and Beard noted "the glove-and-stocking distribution of the anesthesia" as evidence of epidemic hysteria and commented "It seems fair to say that the characteristic pattern of sensory loss is a classically hysterical one" [7]. They found a glove-and-stocking type of anesthesia recorded in the charts of 13 patients, 11 of whom were also severely ill [7]. Our study group did not report details of sensory losses, but the original medical staff reports stated that "objective sensory loss was usually maximal peripherally, and frequently coincided with motor weakness" [3]. Glove-and-stocking anesthesia may occur in patients with a hysterical conversion disorder, but it can also be due to peripheral neuropathy in many serious organic diseases. This type of conversion disorder has not been reported in any published outbreaks of mass hysteria [27,28] and its presence in patients who are also seriously ill is questionable. A glove-and-stocking type of anesthesia in a few seriously ill patients does not support a diagnosis of epidemic hysteria.

Our study group said that EMG recordings of muscles affected by Royal Free disease showed characteristic features. To support their epidemic hysteria hypothesis, McEvedy and Beard stated that "a deliberate attempt by one of the authors to produce an electromyographic record similar to that reported in Royal Free disease was successful" [7], with the implication that abnormal EMG tracings of patients with Royal Free disease might have been fabricated. They published an EMG tracing of the extensor digitorum of the arm from a healthy person while encouraging the outstretched arm to tremble and suggested a similarity between this tracing and the EMG tracing of a weak tibialis anterior muscle affected by Royal Free disease during maximal sustained volition, that had been published

by the Royal Free medical staff [4,7]. They proposed that the EMGs in the Royal Free patients could have been produced by "maximum effort" [7]. Whether an EMG of a healthy arm muscle, while encouraging the arm to tremble should be equated with an EMG of a weak leg muscle under maximal sustained volition from a patient suffering from Royal Free disease is questionable, but this attempt to imply that the experienced Royal Free medical staff might have misinterpreted EMG data or that one Royal Free patient might have fabricated an abnormal EMG does not provide evidence that the Royal Free outbreak was epidemic hysteria.

Distinguishing epidemic hysteria from an organic illness can be difficult, but characteristic features can help with diagnosis. In epidemic hysteria outbreaks, person-to-person i.e., visual transmission usually occurs within minutes [27,28]. Contrary to this, our study group reported an incubation period of several days. In epidemic hysteria, symptoms usually quickly resolve in patients separated from other patients and from the environment where the outbreak began [27,28]. In the Royal Free outbreak, patients sent home did not recover quickly. The incubation period and the failure of symptoms to resolve in isolated patients is not consistent with epidemic hysteria.

#### *4.3. Reactive Psychogenic Symptoms*

Diagnostic difficulties occurred in a minority of patients who were thought to be neurotic or to have exaggerated their symptoms. We sugges<sup>t</sup> that at least some patients might have developed "reactive psychological disaster syndrome" [26] as a result of knowing that they had been exposed to a serious, debilitating, infectious disease of an unknown cause. A minority of patients with possible reactive psychogenic symptoms does not invalidate an organic cause for the outbreak.

#### *4.4. SARS CoV 2*

Recent reports show that some patients infected with SARS CoV 2 have developed post-viral symptoms characteristic of ME/CFS [30]. Given the growing recognition of similarities between ME/CFS and post-viral SARS CoV 2 [30], we hope that these patients are not regarded as having a psychosomatic illness. We also hope that future studies investigating features of both diseases may lead to new treatments that could potentially be of benefit for both groups of chronically ill patients.

#### *4.5. Strengths and Limitations*

The study group all experienced the Royal Free outbreak of ME as hospital staff, medical students, and some as patients. The outbreak was dramatic and the participants provided clear first-hand eye-witness accounts. The authors of this study were medical students at the Royal Free medical school at the time of the outbreak. Our recollections are consistent with the findings of this study. The study participants were self-selected members of two organizations for staff who worked or trained at the Royal Free hospital and may not be representative of the hospital staff at the time of the outbreak. This study took place 58 years after the outbreak and the participants' recollected accounts are subject to recall bias, are dimmed by the passage of time, and lack specific details of clinical findings.
