Children are particularly at risk of thyroid cancer from radioactivity in contaminated food and drink. Credit: Reuters/Kyodo

Even as the damaged reactors at the Fukushima Daiichi nuclear power station continue to leak radiation, researchers have begun laying the groundwork for studies that will look for any long-term effects on public health.

Academic scientists face major obstacles as they try to collate baseline data on radiation doses in the face of the enormous disruption caused by the earthquake and tsunami that hit the country last month. But the experience of the 1986 Chernobyl nuclear accident shows that such baseline data are vital. Without them, drawing firm conclusions about any adverse health effects will be much more difficult.

Researchers emphasize, however, that environmental levels of radiation outside the 20-kilo­metre evacuation zone around the power plant are currently far below levels that warrant concerns about human health. The greatest threat to human health from the disaster is consuming contaminated food and drink, they say.

Assessing the impact of any exposure to radiation in the environment may require cohort studies to look for a raised incidence of cancers years from now among people living in regions with raised levels of contamination. Just how far-reaching those studies need to be, or whether they are needed at all, will depend on the extent of the ongoing contamination from the damaged reactors. Although the prevailing winds are blowing the bulk of radio­isotopes from the plant out over the Pacific Ocean, periodic changes in weather patterns are dumping fallout inland, increasing the doses that residents receive. The Japanese authorities acknowledged last week that it may be months before the reactors are brought under control. For now, "it is difficult to predict what the health effects might be", says Dillwyn Williams, a cancer researcher at the Strangeways Research Laboratory in Cambridge, UK.

Plant workers are being exposed to much higher levels of radiation than the general population, and will be monitored for long-term health effects. The Tokyo-based Radiation Effects Association already has an ongoing study of the health of Japanese nuclear-power workers, and new dosimetric data for Fukushima workers will be merged into that study.

The problem is that it is very difficult to get a real picture of the exposure of the population. ,

But the Radiation Effects Research Foundation (RERF), based in Hiroshima and Nagasaki, which is responsible for radiation epidemiology studies on survivors of atomic-bomb explosions, is already initiating discussions on broader Fukushima studies. In a joint statement to Nature (see go.nature.com/cckfoe), the RERF's vice-chairman and chief of research Roy Shore, and Kotaro Ozasa, its head of epidemiology, say that it is vital to gather baseline data — such as the exact locations of people exposed to fallout — as soon as possible. Several agencies, including Japan's science ministry, local authorities and the Tokyo Electric Power Company, the plant's operator, are already publishing measurements. But compiling and evaluating the information will be a challenge, say Shore and Ozasa, as these data are currently "scattered and uncoordinated".

"The problem is that it is very difficult to get a real picture of the exposure of the population," says Elisabeth Cardis, a radiation epidemiologist at the Centre for Research in Environmental Epidemiology in Barcelona, Spain. A critical review of all the available data is desperately needed, she says (see go.nature.com/ejlpny).

Questionnaires should also be sent to people in higher-risk areas to identify details such as the time spent outdoors on various dates, and what food and water were consumed, say Shore and Ozasa. "Obviously, it is important to obtain those data sooner rather than later, but at this point, coping with the huge effects of the earthquake and tsunami has to take precedence," their statement says.

Japan's prompt evacuation of the 20-kilometre zone around Fukushima, and bans on suspect produce, should have helped to curb exposure to isotopes of concern. Iodine-131, which has a half-life of just 8 days but accumulates quickly in the thyroid gland, is still the major component of the emissions from the nuclear plant and remains the greatest acute radiation health threat to the public, says Richard Wake­ford, an epidemiologist at the Dalton Nuclear Institute, University of Manchester, UK. Some of the radioactivity levels detected in food since the accident have been "pretty hefty", he adds.

One of the largest health impacts from Chernobyl has been the 6,000 or more cases of thyroid cancer, mostly affecting people who were children at the time of the accident. In most of these cases, people received high radiation doses through drinking milk from cows that had grazed on iodine-contaminated pasture. Children are particularly at risk because their thyroids are still developing and are more prone to radiation damage than adults' mature thyroids.

The Japanese authorities are distributing potassium iodide tablets in affected areas, and Williams says that this is a crucial precaution. The tablets swamp the thyroid with non-radioactive iodine, preventing uptake of the radioactive form. Japanese children may also have a cultural advantage that lowers their risk from radioiodine. Whereas the children of Chernobyl tended to be iodine-deficient, the Japanese diet, rich in fish and seaweed, is "one of the most iodine-rich diets in the world", says Williams. Milk is also far less important in the Japanese diet than it was for the rural populations around Chernobyl.

Radioiodine doses in the thyroids of children in the most contaminated areas are already being monitored by the Japanese authorities. Nature has learned that the first results of that survey show minimal thyroid doses in 946 children living in areas northwest of the plant, a region where some of the highest fallout over land has been reported. Measurements during 28–30 March found maximum doses of 0.07 microsieverts per hour. This would suggest that the children had received total doses of less than 100 microsieverts, many thousands of times lower than was received by people living in contaminated areas around Chernobyl. The results "seem reassuring that not much iodine-131 has got into children", says Wake­ford, adding that if the food bans are being effective, "Japan will have got a grip on what is the major concern in this sort of situation".

Vadim Chumak, a health physicist at the Research Center for Radiation Medicine at the Academy of Medical Sciences of Ukraine in Kiev, who has coordinated Chernobyl health studies, says that Japanese radiation researchers should heed a key lesson from that disaster. Dose data are fleeting, he warns, and if they are not collected now, any eventual research would be much more prone to uncertainty. Dosimetric monitoring after Chernobyl was sub-standard, he says, "so in our research we had to invest enormous time and effort in the retrospective estimation of doses".