Coca-Cola withdrew 30 million cans and bottles from sale in Belgium last month when nearly 100 people reported suffering stomach cramps, nausea, headaches and palpitations after drinking Coca-Cola products. The safety scare followed an unrelated alert over potentially cancer-causing dioxins in Belgian meat and poultry which helped bring down the government.
However, a preliminary inquiry has suggested that the Coca-Cola alert may have been due to nothing more toxic than an outbreak of mass hysteria. In a letter to the Lancet this week, four members of Belgium's Health Council say that preliminary inquiries suggest last month's epidemic was a case of "mass sociogenic illness" (the more politically-correct term for mass hysteria).
The story started on June 8, when 26 children from one school began feeling sick, tired and complaining of headaches, palpitations and stomach-ache after drinking bottled Coca-Cola. The next day the evening news announced that Coca-Cola was recalling the incriminated product, and within the next 48 hours several other children from the same school complained of the same symptoms, followed by children from other schools.
The national health authorities stepped in and banned the sale of Coca-Cola, while hundreds more people phoned the National Poisons Centre to say they, too, had been poisoned. People in Northern France also reported symptoms of poisoning but no one was found to be seriously ill. "Most complaints consisted of vague constitutional symptoms and were transient," says the letter to the Lancet. "Neither physical findings nor laboratory results revealed any significant abnormality."
Despite the negative findings, on June 15 the Coca-Cola company claimed it had identified two causes for this apparent epidemic of cola-induced symptoms - a fungicide that had contaminated the outside of the cans and "bad carbon dioxide". But when toxicology experts examined analyses of the cans, they could not find anything present in any concentration that could explain the outbreak.
The authors of the letter, one of whom studied the toxicology reports, say the outbreak suggests mass sociogenic illness. "MSI is described as a constellation of symptoms of an organic illness, but without identifiable cause, which occurs among two or more persons who share beliefs related to those symptoms," they write.
The symptoms of a mass sociogenic illness are unpleasant and fearsome, regardless of being psychological in origin. Outbreaks of MSI typically occur in workplaces or schools. They share symptoms, often including overbreathing (hyperventilation that causes cramp in the hands and feelings of dizziness and panic), faintness, abdominal pain, sickness, headache, weakness and itching. The symptoms are many, the abnormal findings, on examination, non-existent.
Women (or girls) succumb more than men, and symptoms are triggered by seeing someone else affected. Describing an epidemic in an elementary school in Georgia, Dr Richard Levine and his research team wrote in the Lancet in 1974: "Some remarks by affected individuals suggest a psychosomatic mechanism. 'It was like seeing through a fog.' 'My eyes felt like they were popping out of my head.'"
As Levine points out, "The diagnosis of hysteria is one of exclusion. Physical examination and laboratory data when obtained, revealed few or no abnormalities." In this case, parents were concerned about airborne toxins. Nothing was found, parents and children were reassured and the cases stopped.
No one dies from the symptoms and there is usually another group in the same environment that is well, with those who report being ill having typically been under unusual physical or mental stress. The epidemic usually starts and ends quickly. News of the symptoms is spread by the media and social networks, particularly parents. As the authors of the Lancet letter point out: "It is probably significant that a company with such high visibility and symbolic image was involved in this episode. Besides the important role of the media, the scale of the outbreak may have been amplified by the radical measures taken by the health authorities, as well as deficient communication by the Coca-Cola company."
Working out what has caused a group of people to get acutely ill is a tricky business. In the UK, public health doctors collect details of symptoms, sort out a definition of cases and identify risk factors that might be implicated. Tracking down the agent responsible is laborious and requires painstaking work - the public health team actively look for cases, writing letters to GPs asking if they've come across patients with the same set of symptoms, and looking through records of accident and emergency departments.
In Belgium a previous scare about dioxins contaminating animal feed had helped oust the government in the national elections in June. There are suggestions that public health officials might have suspected mass hysteria in the Coca-Cola case within 48 hours of their investigation but were not confident enough to go public with their suspicions.
In the UK, there have been some dramatic examples of mass sociogenic illness. In 1965, in a Blackburn school, a few girls said they felt dizzy and peculiar and by midday 85 of them were in hospital, lying with their teeth chattering and overbreathing. Although most were rapidly sent home, the school had two more days of the epidemic.
As is usual with such epidemics, pollution of the air and surrounding land and food were investigated, a viral cause was looked for, but a psychiatric cause was suspected by the second day. There were probably two triggers - a polio outbreak in the town, and an incident in which some of the same schoolgirls had genuinely fainted after standing for three hours waiting for a church service.
In their description of the case in the British Medical Journal in 1966 Dr Peter Moss and Dr Colin McEvedy write: "What became epidemic was a piece of behaviour consequent on an emotional state; excitement or, in the latter stages, frank fear, led to overbreathing. "Once learned, this reinforcing piece of behaviour started spontaneously whenever school was assembled."
After the extensive media coverage, two other schools, in Portsmouth and Wrexham, had similar epidemics. Other epidemics with various symptoms, have occurred in marching bands (girls falling like ninepins), at football games and among nurses at a leprosy hospital in Japan.
But not all mysterious illnesses in schoolchildren can be put down to mass hysteria. In 1994, 50 children from one London school suffering with sickness and abdominal pain arrived at an accident and emergency department. Within six hours they had recovered. Initially, investigators felt that many had been prompted to feel sick by the sight of their friends being ill and the chance of a ride in an ambulance. Detailed investigations, however, suggested that cucumber, contaminated by a toxic chemical, was to blame.
In the report of the incident in the Journal of Epidemiology and Community Health, the researchers write, "Investigators should beware of too readily attributing a psychogenic cause to unusual outbreaks of acute illness in school children."