Koro Epidemics

Summary

In late October 1967, hundreds of men in Singapore became convinced that their genitals were retracting into their bodies and that this would kill them. They gripped, clamped and tied themselves, and many were injured doing it. Hospitals filled. Then the health authorities went on the record to say it was not happening, and within a month it stopped. The same sequence has run in India and in China, and the ending is always the same, which is the whole point of the entry.

What is documented

The condition. Koro, in Chinese suo yang, is the acute conviction that the genitals are retracting into the abdomen and that complete retraction will be fatal. It is classified as a culture-related syndrome. No retraction occurs. The terror is real, and so is the injury that follows from trying to prevent it.

Singapore, 1967. The epidemic began in late October and ran for about ten days. It followed rumours that pork from pigs inoculated against swine fever would cause koro. There was pre-existing public anxiety about meat: people were already worried about chickens being injected with oestrogen.

At its peak, Singapore General Hospital was seeing 70 to 80 cases a week. In total, 469 cases were formally recorded. The true figure was certainly higher, because the survey covered only Western-style hospitals and did not capture patients who went to traditional Chinese physicians.

Patients arrived holding onto themselves, and arrived with clamps, string, rubber bands, chopsticks and clothes pegs applied to prevent the retraction. The injuries came from the remedy. Most patients were under twenty; the oldest recorded was forty.

The finding that matters. Every patient interviewed had heard about koro before experiencing it. There are no index cases who invented the symptom independently. The belief precedes the symptom, without exception, in the recorded series.

How it ended. The Singapore Medical Association and the Ministry of Health made a public statement: koro was an effect of fear, and the retraction was not occurring. Cases fell sharply. Within a month there were none. The findings were published by the Koro Study Team in the Singapore Medical Journal in 1969, which concluded that koro was a panic syndrome linked to cultural indoctrination. The Chinese Physician Association, working from a different framework, attributed the outbreak to fear and rumour-mongering.

Lower Assam, India, 1982. An epidemic beginning in September 1982. It subsided, without a single fatality, after what the treating physicians described as massive public reassurance.

Guangdong and Hainan, China, 1984-1985. Local records show genital-retraction epidemics in the region from as early as the late nineteenth century, with outbreaks in 1948, 1955, 1966 and 1974, recurring in periods of social tension. The 1984-1985 epidemic was the largest recorded: it lasted more than a year and affected over 3,000 people across 16 cities and counties. The local explanatory framework was a folk belief in evil spirits that steal male essence. A smaller outbreak followed in 1987. A public mental health campaign was conducted, and there has been no epidemic in China since. The episode was documented by Tseng and colleagues in the American Journal of Psychiatry in 1988, who noted that the region's geographic isolation and the localism of its folk beliefs kept the epidemics confined to it.

Leading explanations

Mass psychogenic illness (established). Koro epidemics display every diagnostic feature of mass psychogenic illness, and they display them cleanly:

• A pre-existing cultural belief that the event is possible and lethal. Without it there is no epidemic; with it, an epidemic is available to be triggered.
• A trigger, usually a rumour attached to an existing anxiety. In Singapore it was contaminated pork, landing on a population already uneasy about what was being injected into its food.
• Transmission by information, not by contact. Every patient had heard of it first.
• A susceptible demographic: overwhelmingly young men.
• Real physical harm, produced by the response rather than the condition.
• Termination on authoritative public reassurance, which is the signature that distinguishes it from everything else.

Why the ending is the evidence. A toxin does not stop when the Ministry of Health holds a press conference. A pathogen does not care what the Singapore Medical Association says on the radio. An epidemic that ends within days of being publicly contradicted by a credible authority, and that ends without a single death, is an epidemic of belief. Singapore 1967, Assam 1982 and Guangdong 1984-1985 all ended this way, in three different countries, across two decades. That is not an anecdote. It is a replication.

Current status

Later explained. The mechanism is understood, the epidemic curve is documented, the trigger is identified in each case, and the intervention that ends it is known and has worked three times in three countries. The case is kept in the register because it is one of the best-documented instances of a real, injurious, population-scale medical emergency in which the disease did not exist, and because the thing that cured it was a true sentence said out loud by someone people believed.

Last reviewed: July 2026. Records are provisional. Where the evidence changes, the entry changes. Found an error? Tell us.