The Royal Free Epidemic of 1955:
Was it really mass hysteria?
EM Goudsmit PhD C.Psychol.
ME Research Online
This article was written in 1987.
This review challenges the hypothesis, forwarded by McEvedy and Beard (1970a) that the outbreak of encephalitis at the Royal Free Hospital in 1955 was the result of mass hysteria. Detailed examination of the literature shows that many of the clinical features of the Royal Free Hospital epidemic were inconsistent with the generally accepted view of mass hysteria. Moreover, recent evidence strongly favours the view that myalgic encephalomyelitis has an organic basis and this precludes it from being a hysterical disorder. Nevertheless, it is certainly possible that a number of the affected nurses at the Royal Free Hospital may not have suffered from encephalomyelitis, but from an anxiety-related disorder.
Mass hysteria or epidemic hysteria has been described (Sirois 1982) as a spontaneous, unorganised, uncontrolled and contagious outbreak of aberrant behaviour amongst a group of individuals. One of the most frequently cited examples of mass hysteria is the outbreak of encephalomyelitis, which occurred at the Royal Free Hospital, London in 1955 (Roy 1982). The mass hysteria hypothesis of the Royal Free Hospital epidemic, which was widely accepted, is mainly based on the work of two psychiatrists, McEvedy and Beard, who reassessed the available information on the outbreak and concluded that it could be regarded as an instance of epidemic hysteria (McEvedy and Beard, 1970a, 1970b, 1973).
As a result, there has been relatively little research into the causes of this condition and many people currently suffering from myalgic encephalomyelitis (M.E.) as this illness is now called, have experienced great difficulty in obtaining support and treatment.
In this paper, I will examine and challenge McEvedy and Beard’s explanation of the outbreak at the Royal Free Hospital and their view of M.E. as a psychogenic condition.
Many of the outbreaks of mass hysteria, which have been detected and investigated, share a number of characteristic features (Sirois 1982). Perhaps the most significant of these is the susceptibility of women, particularly young women grouped together in institutions such as schools. The literature portrays these women as generally badly educated, suggestible, suspicious and neurotic.
The type of symptoms reported during outbreaks of mass hysteria include fainting, nausea, malaise, abdominal pain, headaches, convulsions, tremors and hyperventilation, all of which can be explained as manifestations of anxiety.
A further feature of such outbreaks is their short duration. The vast majority only last a few days although a few have lasted several months. Moreover, outbreaks are generally assumed to be sparked off by a specific incident which gives rise to anxiety in the community.
A typical example of such an epidemic occurred in a girls-only school (McEvedy et al., 1966). It began after a number of girls from one class complained of abdominal pain and vomiting and were taken to hospital. The next day, there was an “explosive epidemic” which started during assembly and involved most of the classes. The main symptoms were faintness and “feeling peculiar”. The epidemic lasted 9 days although there were few cases on days 4 and 5 (the weekend). There were no positive laboratory findings but those affected were later found to have higher Neuroticism scores than those who had remained well.
At the Royal Free Hospital, the majority of sufferers were also women and some of the symptoms could be explained in terms of anxiety. However, numerous features were inconsistent with the classical picture of mass hysteria.
Details of 200 cases for which complete records are available are to be found in the report by the Medical Staff of the Royal Free Hospital (1957) and in the monograph by Ramsay (1986). The outbreak began on July the 13th 1955 when a resident doctor and a ward sister became ill and were admitted as patients. It lasted until October the 5th although sporadic cases continued to be seen long after that date. In total, 292 members of staff, including 149 nurses were affected. Of the patients, 265 were women and 27 were men.
The symptoms included profound malaise, headache, low-grade fever, sore throat and nausea as well as unusually severe depression and emotional lability. Pains in the neck, back, limbs and chest, plus dizziness and vertigo were also common. Moreover, in almost all cases, the cervical lymph nodes were enlarged and tender and in at least 50 per cent, there was generalised lymphadenopathy. The liver was enlarged in about one tenth of the cases and 74 per cent of patients showed objective evidence of involvement of the central nervous system. In addition, one quarter of the patients suffered from bladder dysfunctions. Spasm, tingling, twitching and rippling of muscle (fasciculation) were commonly found and some patients experienced sensory loss, particularly in the lower limbs.
The course of the illness is interesting, for the condition tended to worsen during the second and third week. Those who showed no evidence of invasion of the nervous system tended to be symptom-free within a month but in others, the illness was protracted. The symptoms often fluctuated in severity and in some cases, relapses occurred after patients had been fit enough to return home. A significant number of those affected are still ill.
In the same year, there were three similar outbreaks. The first outbreak occurred in Addington Hospital, Durban, at the same time as a poliomyelitis epidemic (Hill et al., 1959). Ninety-eight nurses were affected, of whom 11 were still unfit for duty three years later and 10 had to seek other occupations because of the residual effects of the illness. Like the outbreak at the Royal Free Hospital, this epidemic followed a number of cases in the general population living in the vicinity, and the reason for the increased virulence of the infecting organism was probably the semi-isolated nature of the hospital community and the close physical contact between the members of staff.
The second commenced in Dalston, Cumbria in February 1955 and lasted until July, affecting 233 members of the general population. The ratio of female to male sufferers in this outbreak was 1:1 (Wallis 1955).
The third outbreak occurred in a very large area of North West London, extending from East Ham in the North to Shepherds Bush in the South. It is not known exactly how many numbers were affected but one hospital alone admitted 53 cases between May 1955 and March 1958 (Ramsay 1957, 1986). It preceded the epidemic at the Royal Free Hospital, which served part of this area.
In all these outbreaks, the marked and persistent muscle fatigability was the dominant clinical feature. The two hospital outbreaks, but not the epidemic in Dalston, were regarded by McEvedy and Beard as examples of mass hysteria despite the fact that the symptoms in all four outbreaks were remarkably similar (McEvedy and Beard 1970b).
McEvedy and Beard (1970a, 1973) based their hypothesis on the following arguments:
1. The vast majority of those affected were young women.
2. These women were socially segregated.
3. No organic cause could be found and the results of the laboratory tests were not significant.
4. Some of the symptoms could be explained in terms of anxiety and hyperventilation.
5. The illness failed to “propagate beyond the institutional population.”
6. The mean Neuroticism score of a small number of affected nurses was higher than that of a control group of unaffected nurses.
7. Those nurses who became ill had suffered from more illnesses requiring hospital admissions and had borne fewer children than the unaffected nurses.
8. Two of the affected nurses had ‘cooked’ their thermometers.
The majority of sufferers at the Royal Free Hospital were young females and some were socially segregated to a certain degree. However, the literature portrays those who succumb to mass hysteria as poorly educated and suggestible. Yet most of those affected at the Royal Free Hospital were well-qualified members of staff and arguably, not the most suggestib1e group of people within the general population. One group of people who were probably quite anxious and therefore more ‘suggestible’ were the patients, but only 12 of them contracted ‘Royal Free Disease’.
The apparent susceptibility of nurses to this illness may reflect, as McEvedy and Beard suggested, the propensity of young women to react hysterically under certain circumstances. However, it has been observed that this condition tends to affect the more physically active members of a community. As the vast majority of nurses are women, any illness that attacks the most active people in a hospital will therefore claim a disproportionately large number of women sufferers.
That the susceptibility of women in hospital outbreaks of M.E. may be linked to factors other than the psychological make-up of women is supported by data from outbreaks amongst the general population. Many of these outbreaks, such as the ones in Dalston (Wallis 1955) and Adelaide (Ramsay 1986, Pellew 1951) affected an equal number of men and women and several, including ones in Switzerland (Gsell 1949) and Berlin (Sumner 1956) only involved men. Interestingly enough, McEvedy and Beard (1970b) did not regard the latter, a relatively mild outbreak which affected only 7 soldiers, as mass hysteria.
The third argument for the mass hysteria hypothesis is that no organic cause could be identified. However, this does not prove that there was no organic cause, or that the cause was psychological. In fact, there were a considerable number of symptoms indicating the presence of an infective agent. For instance, in addition to the number of people with low-grade fever (89 per cent) and lymphadenopathy (79 per cent), many showed unmistakable evidence of central nervous system involvement. Forty per cent had ocular paresis, 19 per cent had facial palsies and 11 per cent had bulbar paresis. Not only are these symptoms atypical of both mass hysteria and conversion disorder; they cannot easily be faked.
Furthermore, while symptoms such as malaise, pain and dizziness may be attributed to anxiety states, many of the other symptoms noted during the Royal Free Hospital outbreak, including the enlarged tender glands, are not linked to the autonomic nervous system and are not characteristic of anxiety states and hysteria.
The protracted course of ‘Royal Free Disease’ is a further argument against the McEvedy and Beard hypothesis. A typical outbreak of mass hysteria tends to be of short duration and is both benign and self-limiting. However, the epidemic at the Royal Free Hospital lasted several months and many of the sufferers remained ill for a year or more (Ramsay 1986). The chronicity of the illness was also observed in Los Angeles, Iceland and Durban and in sporadic cases not associated with an epidemic (Ramsay 1986, Wookey 1986). Another prominent feature of ‘Royal Free Disease’ is the susceptibility to and frequency of relapses, particularly after physical and/or mental exertion. This too, is not a characteristic of hysterical disorders.
McEvedy and Beard’s claim that the illness did not “propagate beyond the institutional population” is inaccurate since sporadic cases continued to be seen in North West London long after the hospital epidemic had ended (Ramsay 1957).
During 1968 and 1969, McEvedy and Beard sent questionnaires including the Eysenck Personality Inventory (EPI) to 98 nurses who had been ill during the 1955 outbreak and to 91 unaffected nurses. The 71 affected nurses who returned the completed questionnaires were then paired with 71 unaffected nurses who acted as a control group.
The results were published in 1973 and showed that the mean Neuroticism score of the affected nurses was 12.2 whilst that of the control group was 10.3. Although the score of the affected nurses was significantly higher than that of the controls, it was well below the normative scores (Eysenck and Eysenck 1964) for hysterics (15.2) and anxious people (15.8). Nevertheless, McEvedy and Beard took their findings as evidence that the sufferers of ‘Royal Free Disease’ were either “pathological hysterics” or “normal women who behave hysterically under stress”.
They did not consider the possibility that the higher Neuroticism scores of the affected group may have been the result of the ambiguity of some of the questions which make up the EPI and of ‘confounding’. At least five of the items on the EPI enquire about symptoms which are not only experienced by emotionally oversensitive people but also by many sufferers of ‘Royal Free Disease’. Consequently, if a small number of respondents were indeed still suffering from the sequelae of this condition, their Neuroticism scores would probably be higher than those of the nurses who had not been ill and this factor alone may have resulted in a higher group mean. This alternative explanation for the data is supported by the preliminary results of an ongoing study into the personality of people suffering from M.E. (Goudsmit).
Two other findings which McEvedy and Beard took as support for their mass hysteria hypothesis was that the affected nurses had spent longer in hospital as patients than the unaffected nurses and that they had borne fewer children. Since hysterics are considered to be less healthy than ‘non-hysterics’ and more likely to have social and sexual problems, these findings suggest that at least some of the sufferers may have been hysterical personalities. However, one should not deduce from this that all the sufferers were hysterics or that the outbreak at the Royal Free Hospital was mass hysteria. Moreover, it is not known whether the nurses who returned their questionnaires were representative of all those who became ill.
The revelation that two of the nurses had ‘cooked’ their thermometers does not support McEvedy and Beard’s hypothesis either, because hysterics do not consciously simulate disease (DSM III, 1980, p. 246-247). While it is certainly plausible that a number of nurses may have been anxious about the nature of the illness (although according to the report by the Medical Staff of the Royal Free Hospital, it was originally thought to be infectious mononucleosis), it is unlikely that well over a hundred nurses would have become ill simply as a result of anxiety about an illness which at the time was not associated with serious long-term complications or a poor prognosis. However, any existing anxiety could have lowered the resistance to disease of some of the nurses and this could have made them more susceptible to the illness (Weiner 1986).
In recent years, there has been a considerable increase in research into the pathogenesis of myalgic encephalomyelitis, both in its epidemic and in its more common endemic form. The main findings of the research have been reviewed elsewhere (Ramsay 1986) but can be summarised as follows:-
In a study of 12 patients whose illness had all the features found in the North West London and Royal Free Hospital epidemics, Behan and Behan (1980) found abnormalities in the lymphocytes of the peripheral blood, increased IgM and decreased IgA levels, reduced complement factor 4 concentrations and serum anti-complimentary activity. This corresponded with the findings of Dillon et al., (1974) who found that lymphocytes from a number of patients grew and multiplied in tissue culture, a phenomenon strongly suggestive of a persistent virus.
In a further study of 50 patients, all of whom had gross fatigue made worse by exercise (Behan et al., 1985), lymphocyte function was found to be abnormal in the group as a whole. In 35 it was highly abnormal and 6 of these showed the same changes for up to 2 years. High titres of serum autoantibodies were found to smooth muscle in 18 patients, to thyroglobulin in 13, to nuclear constituents in 6 and to gastric parietal cells in 4. Muscle biopsies were abnormal in 20 patients and showed widely scattered necrotic muscle fibres in 15. Behan and his co-workers suggested that the syndrome was “due to the interaction of viral infection and immunological processes which produce damage to intracellular enzymes and result in abnormal muscle metabolism, especially on exercise”.
This hypothesis is supported by the work of Professor G.K. Radda and his team at Oxford University. In 1984 they reported on a young General Practitioner who suffered from M.E. and whose muscles showed an abnormally early intracellular acidosis when tested by nuclear magnetic resonance. They considered that the fatigue in his case to have been the result of excessive glycosis (Arnold et al., 1984). There is also a great deal of evidence that the muscle fatigue can result from an enzyme deficiency (Ramsay 1986).
Clear electrophysiological evidence of an abnormality in the peripheral part of the motor unit in patients suffering from M.E. was provided by Jamal and Hansen (1985). Single fibre electromyography recordings of 40 patients revealed that 75 per cent had definite abnormalities of jitter. In addition, all the muscle biopsies examined were found to be abnormal, the most important findings being type II fibre predominance, subtle and scattered muscle fibre necrosis, bizarre tubular structures and mitochondrial abnormalities. The very low jitter values found during the electromyography were consistent with the presence of scattered muscle fibre necrosis seen in the muscle biopsies.
In 1986, Southern and Oldstone reviewed the medical consequences of persistent viral infection and postulated that once inside a cell, a virus may block or divert enzymes necessary for cellular metabolism; they may also disrupt the cell’s plasma membrane and alter its permeability sufficiently to lead to its destruction. They also discussed both how persistent infections could be established and the new concept that viruses can cause disease not by destroying cells, but by altering the specialised function of the infected cell.
At present, the mechanism by which viruses are maintained and evade or suppress the immune responses remain unclear and routine laboratory tests still cannot always detect the presence of a persistent viral infection. Nevertheless, the various lines of research, some of which are described above, are gradually establishing incontrovertible evidence that M.E. has an organic basis and that it is not, as McEvedy and Beard suggested, a manifestation of a hysterical disorder.
In conclusion, some of McEvedy and Beard’s arguments were based on supposition and conjecture and they ignored many of the objective signs of the illness. The clinical features of the Royal Free Hospital outbreak including the generalised lymphadenopathy, the heavy involvement of the cranial nerves, its protracted course and the tendency to relapse following exertion are all compatible with organic disease processes. Conversely, they are not characteristic of mass hysteria.
Given the available information on the outbreak plus the results of recent research into the aetiology of M.E., the mass hysteria hypothesis of the Royal Free Hospital epidemic is no longer tenable.
I would like to thank Dr. A.M. Ramsay MA MD, Honorary Consultant Physician in Infectious Diseases, Royal Free Hospital for all his help and support during the preparation of this manuscript.
American Psychiatric Association Committee on Nomenclature and Statistics. (1980) Diagnostic and Statistical Manual of Mental Disorders (3rd ed.) Washington D.C.: APi.
Arnold, D.I., Bore, P.J., Radda, G.K., Styles, P. and Taylor, D.J. (1984) Excessive intracellular acidosis of skeletal muscle on exercise in a patient with post-viral exhaustion/fatigue syndrome. Lancet, 1, 1367-1369.
Behan, P.O., Behan, W.M.H. and. Bell, E.J. (1985) The post viral fatigue syndrome - an analysis of the findings in 50 cases. Journal of Infection, 10, 211-222.
Behan, P.O. and Behan, W.M.H. (1980) Epidemic Myalgic Encephalomyelitis. In Clinical Neuroepidemiology. F. Clifford Rose (Ed) pp 374-383. London: Pitman Medical.
Dillon, M.J., Marshall, W.,C., Dudgeon, J.A. and. Steigman, A.J. (1974) Epidemic Neuromyasthenia: outbreak among nurses at a children’s hospital. British Medical Journal, 1, 301-305.
Eysenck, H.J. and Eysenck, S.B.G. (1964) Manual of the Eysenck Personality Inventory. Sevenoaks: Hodder and Stoughton.
Gsell, 0. von. (1949) Abortive poliomyelitis. Helvetica Medica Acta, 16, 169-183.
Hill, R.C.J., Cheetham, R.W.S. and Wallace, H.L. (1959) Epidemic myalgic encephalomyelopathy. The Durban outbreak. Lancet, 1, 689-698.
Jamal, G.A. and Hansen, S. (1985) Electrophysiological studies in the post-viral fatigue syndrome. Journal of Neurology, Neurosurgery and Psychiatry, 48, 691-694.
McEvedy, C.P. and Beard, A.W. (1973) A controlled. Follow-up of cases involved in an epidemic of ‘benign myalgic encephalomyelitis’. British Journal of Psychiatry, 122, 141-150.
McEvedy, C.P. and Beard, A.W. (1970a) Royal Free epidemic of 1955: a reconsideration. British Medical Journal, 1, 7-11.
McEvedy, C.P. and Beard, A.W. (1970b) Concept of benign myalgic encephalomyelitis. British Medical Journal, 1, 11-15.
McEvedy, C.P., Griffith, A. and Hall, T. (1966) Two school epidemics. British Medical Journal, 2, 1300-1302.
Medical Staff of the Royal Free Hospital. (1957) An outbreak of encephalomyelitis in the Royal Free Hospital Group, London in 1955. British Medical Journal, 2, 895-904.
Pellew, R.A.A. (1951) A clinical description of a disease resembling poliomyelitis, seen in Adelaide, 1949-1951. Medical Journal of Australia, 1, 944—946.
Ramsay, A.M. (1986) Postviral Fatigue Syndrome. The Saga of Royal Free Disease. M.E. Association. P.O. Box 8, Stanford-le-Hope, Essex SS17 8EX.
Ramsay, A.M. (1957) Encephalomyelitis in north west London. An endemic infection simulating poliomyelitis and hysteria. Lancet, 2, 1196-1 200.
Roy, A. (Ed) (1982) Hysteria. Chichester: Wiley.
Sirois, F. (1982) Epidemic Hysteria. In Hysteria. A. Roy (Ed) pp 101-116. Chichester: Wiley.
Southern, P. and Oldstone, M.B.A. (1986) Medical consequences of persistent viral infection. New England Journal of Medicine, 314, 359-367.
Sumner, D.W. (1956) Further outbreak of a disease resembling poliomyelitis. Lancet, 1, 764-766.
Wallis, A.L. (1955) An unusual epidemic. Lancet, 2, 290-291.
Weiner, M.A. (1986) Maximum Immunity. Chapter 4. Bath: Gateway Books.
Wookey, C. (1986) Myalgic Encephalomyelitis. Postviral Fatigue Syndrome and How To Cope With It. London: Croom Helm.