Recent outbreaks linked to fresh sprouts remind us that toxic microbes can sometimes travel in small and seemingly healthful packages.

The Japanese need no reminder. They learned the hard way, 15 years ago, when that health-conscious society was gripped by what may be history’s worst outbreak of E. coli O157:H7.  The epidemic sickened thousands, most of them children, and killed at least 12.

It also showed critical weaknesses in Japan’s public health system–problems that the government spent years trying to correct.

Up to that time, E. coli O157:H7 was virtually unknown in Japan. Just six small outbreaks had been reported at childcare centers and schools in the five years from 1991 to 1995.

That all changed in mid-July 1996, when health officials in Sakai City, a major port city with a population near a million, received reports of a large number of elementary schoolchildren who had come down with severe diarrhea. The next day, hundreds more sick children and teachers were reported, still more on the next.   

Within a week, thousands were ill, hundreds hospitalized, some with hemolytic uremic syndrome (HUS), requiring kidney dialysis. And some were dying. 

As doctors and hospitals dealt with the sick, health officials started searching for the source.    All the Sakai City patients were students or teachers, and Sakai City shares its water system with neighboring cities where there were no illnesses, so water did not appear to be the problem. These and other circumstances pointed to school lunches, which were prepared at a central facility, then trucked to neighborhood schools across the city.

Meanwhile, the epidemic seemed to be expanding. Workers at a factory in Kyoto, about 30 miles from Sakai City, were showing up at clinics with bloody diarrhea. Several developed HUS. Officials learned that all the patients had eaten in the factory cafeteria.

Something was horribly wrong, but what?

Dr. John Kobayashi, a University of Washington professor and former chief epidemiologist for the state of Washington, was monitoring the outbreak from Seattle and wanted to help. “But the public was in something of a panic, and the Japanese government was embarrassed,” he recalls. “Here they were, the world’s second-richest nation, and then they had this huge outbreak.”

It was terrible timing. A few years earlier, Japan’s booming economy collapsed. Then, in 1995, an earthquake flattened the city of Kobe, not far from Sakai City. A few months later, terrorists released deadly sarin gas in the Tokyo subway, killing 13 people. “They were feeling snakebit,” Kobayashi says.

To make matters worse, Japan’s well-respected health care system was ill-equipped to respond to an outbreak of foodborne illness, Kobayashi says. Their epidemiologists do fine long-term research, but were not trained to respond quickly to an E. coli outbreak.

“Their strategy was to begin a long-term cohort study of 50,000 schoolchildren, which would take forever. They had no experience with field epidemiology.”

American epidemiologists like Kobayashi are accustomed to rapid responses to outbreaks– intensive interviews of victims in an attempt to zero in on the exact source of the illness.

By the time he was contacted by Japanese officials, the epidemic had been going on for weeks, Kobayashi says. And they still didn’t know the source.

Many of the obstacles were cultural. The Sakai City outbreak was enormous because school meals are centrally prepared–an efficient way to feed people, or to distribute foodborne illness. And children are strongly encouraged to eat everything on their plates, which gave field workers few clues to a possible source.

“And the Japanese are reluctant to talk about their illnesses,” Kobayashi says. “The whole idea of officials calling people and asking about their diarrhea is very difficult in Japanese society. And then it is not polite to ask people for the names of other sick people. This is important information to epidemiology, but almost inconceivable in Japanese culture.”

Eventually, the outbreak sickened at least 9,441 people, most of them school children–the  worst toll of any recorded outbreak.

Investigators eventually blamed a seemingly unlikely source–radish sprouts, produced by a single Japanese farm and served with school lunches along with cooked chicken and noodles. The sprouts were fingered by a process of elimination. No E. coli was found on samples, but the sprouts were served raw, while most other ingredients were cooked.  

Sprouts are susceptible to contamination because they are cultivated with heat and moisture, conditions that also favor E. coli, Salmonella and other microbes, Kobayashi explains.  Japanese officials suspected that the contamination originated with the radish seeds, which had been grown and shipped from Oregon.

“But it’s not clear,” Kobayashi adds. “Who knows? The investigation took so long that, by the time samples were taken, everything had been wiped clean.”

Regardless of the source, the outbreak shed light on profound deficiencies with Japan’s public health system, and Kobayashi was hired by the Japanese government to help fix the problem. Over a span of five years, he made frequent trips to Japan, helping train epidemiologists and to redesign their system for responding to foodborne illnesses.

If another major outbreak comes along, he says, the Japanese will be far better prepared.   

  • Ayaka Kubo Lau

    I believe if the Japanese government humbled themselves and quickly reached out for the help Dr. Kobayashi offered, the outcome would have been better, I’m sure. What I’m about to write might be a little off topic, but it’s about humbling ourselves to improve food safety.
    The biggest difference between the food manufacturing environment in Japan and U.S. is the educational background of the workers. Almost all the workers in Japan have received at least high school level education in one language, Japanese. Any regulatory changes, procedural changes, possible outbreak, recall, etc. can be quickly communicated to them, there is no need to translate food safety training materials into another language. Their managers, customers, consumers all speak the language they speak. Their government speaks the language they speak.
    On the other hand, although I have not seen any statistics, based on my observation as a QMS consultant, hardly any food processing facility workers in U.S. have received high school level education in English or worse, in their native languages. They cannot read the standard operating procedures, they cannot write down the corrective actions they took, they cannot tell which numbers and letters on the packaging is the lot number they need to record, they often do not have access to high quality food safety training materials, and even if they did, they might not be able to read them.
    Yet, the reality is that they are our eyes and ears on the production floors. They are the ones who wash, peel, mix, grind, bake, cut, package the foods we eat every day sometimes in extremely cold, humid or hot environment. They are the ones who need to do more and more work for no extra pay because the regulations, customers and consumers demand it.
    If we are serious about improving the food safety system in this country, we need to humble ourselves, recognize their role in the food safety system and ask them to help us. I wouldn’t be surprised if half of those workers do not know anything about the new food safety law. We need to elect some government officials who speak their language and we need to provide them with the tools they need.
    I do not speak their language, but I often bring snacks, sit down with them and share some meals before I begin implementing new procedures. I want to show my respect and appreciation and let them know I cannot help deliver safe foods without their help. I was not doing this at the beginning of my career. Oh, what a difference a small change like this has made.

  • crs

    That is excellent and very thoughtful advice.

  • I understand why you won a Pulitzer.

  • I didn’t know that you have such an important language problem in the US however: you couldn’t be more to the point! Here in Switzerland it’s probably even worse. With a population of over 20% migrants from eastern Europe and beyond I have seen food industries with a few hundred workers representing up to seven languages. Even if they would be willing to learn German, French or Italian (depending on the region in Switzerland)the women among them are not allowed to do so thanks to their Muslim husbands.
    These are real challenges in the work of a consultant! And probably just like you: sitting down with them, sharing lunch in the canteen, “pointing and grunting” opens ways of communication. First steps like learning proper hand washing and losing the fear of automatic hand-disinfection locks are perceived as a success by the workers (and me!). The next steps are issueing leaflets with key hygienic behaviour points – in seven languages! Never give up, however; food safety is too important – and luckily, the management of the food industries see the point.