The public health agency of Seattle and King County in the US state of Washington received a call alerting it to an outbreak of acute gastroenteritis among participants at a soccer tournament held in the county one weekend. About 2,000 children from 120 schools in Washington and Oregon participated in the tournament. The players, chaperones and organisers had eaten at many hotels and restaurants in the area. The potential fallout was huge. Officials quickly contacted colleagues at the health departments in Washington and Oregon to help ascertain the scope and reach of the problem. At the Oregon Public Health Division, Dr William Keene, senior epidemiologist with the Acute and Communicable Disease programme, took the call. The three health agencies swiftly co-ordinated the investigations. Washington health officials contacted the organisers of the tournament - a task made easier by the comprehensive list of participating teams at the tournament website. Seattle and King County contacted the food establishments in the area to check for reports of food poisoning. Keene and his assistant, Dr Kimberly Repp, followed up with the known victims of gastroenteritis who were from an Oregon soccer team. Within an hour or two, it became apparent that the outbreak was limited to members of that team, as organisers and eateries reported no complaints of illness. Although the outbreak was much more contained than Keene and his colleagues had initially feared, Keene now needed to find out how the illness had spread. Although the impact of the outbreak was limited and unremarkable, investigations led Keene to an exciting discovery that affirmed a hard-to-prove theory that scientists had held for years. Keene and Repp interviewed the people who had fallen ill - seven in all - to establish a time line and to reconstruct the events. Repp also visited the victims to obtain stool specimens. They found that all seven people started falling ill with vomiting and/or diarrhoea on the Monday night or early Tuesday morning after returning from the tournament on Sunday. Based on the symptoms, Keene suspected that likely culprit was the norovirus, which is the leading cause of gastroenteritis worldwide. The stool specimens confirmed Keene's suspicions. Norovirus victims usually start exhibiting symptoms 36 to 38 hours after exposure because it is highly contagious. Therefore, Sunday's lunch became the main focus of investigations to unearth the source of the contagion. Keene and Repp learned that the victims had eaten Sunday's lunch in the hotel room - the food was purchased in Oregon before the drive to Washington for the tournament. Lunch consisted of sandwich bread, sliced meat, potato crisps, commercially made cookies and fresh grapes. Their findings showed that no single item was the likely source of the virus. Although the cookies had a higher statistical association with the sickness, only four of the victims had eaten the cookies. However, when Keene treated the cookies, crisps and grapes, which had been stored in one bag, as a single variable, he found the statistical link to the illness. The investigators were baffled, however, as to how the virus came to be on a bag of food. It was highly unlikely that the items were tainted at the source of production or sale. There were no other significant reports of similar sickness in the area. Repp went back to interview the team members. She returned with a crucial piece of news: a 13-year-old girl had fallen ill before everyone else. The girl started feeling ill on Saturday night and had left her roommates to spend the night with a chaperone. While in the chaperone's hotel room, she started to vomit repeatedly and also suffered bouts of diarrhoea. The next day, the chaperone drove her back to Portland. The chaperone later also fell ill with acute gastroenteritis. However, the girl did not have any contact with her teammates once she started being symptomatic. She was also absent from the Sunday lunch and had not touched any of the food items. Keene says that an ill person becomes infectious only when they start vomiting and having diarrhoea as the virus is carried in the body fluids. To infect another person, the virus has to be ingested. If the girl did not even return to her room to retrieve her belongings, how was the virus transmitted? Also, there was a 12-hour gap between the onset of her symptoms and the exposure of the rest of the team. Keene and Repp continued to dig for answers, making round after round of calls to mine the team members for more information. The mystery was finally solved when it came to light that the chaperone looking after the girl had done a most unusual thing: store the reusable grocery bag containing the lunch items in the bathroom. When the girl vomited, the virus aerosolised in the bathroom and settled on various surfaces, including the grocery bag and its contents. Then people handled the contaminated bag and food items. By a stroke of luck, Keene learned that the grocery bag in question had eventually been left on the kitchen table in the home of the chaperone. The chaperone then left on a short business trip and fell ill during that time. In the meantime, her family members also handled the grocery bag and also became ill. Keene obtained the grocery bag two weeks after the tournament and had it tested for the norovirus. The results were positive. Scientists have known that a norovirus can be transmitted via fomites or inanimate objects. But the role of fomites in an outbreak have been difficult to assess decisively, as environmental surfaces are seldom tested for the virus in the event of an outbreak. In this case, though, Keene was able to reconstruct the sequence of events and prove that the grocery bag and its contents had been contaminated through airborne transmission and became a source of the disease outbreak. So, people can get sick by simply touching contaminated surfaces. The difficulty with controlling a norovirus outbreak is that only a minute amount of virus is needed to make you ill and it is hard to kill. Keene recommends that victims be confined to using a designated bathroom, if possible, for the duration of the illness. All environmental surfaces, including door knobs, taps, toilet flushers and other such items, should then be cleaned with a 10 per cent household bleach solution.