Scientists agree radiation poses long-term risk, but they are unsure how great

Besides the workers who have tried to cool the melting fuel at Fukushima, most residents within about 50 miles of the stricken nuclear plant have so far been exposed to radiation levels lower than those from a typical CT scan.

With reports of contaminated food and water as far away as Tokyo, millions of people in Japan will be exposed to some radiation, and they will be closely watched. Experts expect that plant workers with the highest exposure will face an increased lifetime risk of cancer.

Scientists also generally agree that any radioactive particles blowing all the way across the Pacific would be far too diluted to cause any harm.

Where the experts do not agree is on how much cancer risk people are likely to face in nearby parts of Japan. Those exposures are significantly above the natural background but below levels where science has clear-cut data on cancer risk.

Some experts state emphatically that radiation is dangerous even in small amounts. Others say that at very low doses it is harmless or even beneficial.

More relevant to Americans is the profusion of CT scans, which can deliver many times the radiation dose of conventional X-rays. Several scientists have recently begun raising alarms that excess use of scans could be causing thousands of additional cancer cases each year, though that, too, is controversial.

As of late this week, scientists had measured elevated levels of radioactive iodine and cesium in air, and, more recently, radioactive iodine in tap water as far away as Tokyo, 150 miles away. They have also detected radioactive fallout on local spinach, cabbage, broccoli, and other vegetables.

How this will affect the population will be long studied and debated.

Scientists are not entirely to blame for their lack of certainty. They cannot just collect people, blast them with radiation, and record the results. Most of what we know about ionizing radiation comes from studies of survivors of the atomic-bomb attacks on Hiroshima and Nagasaki, with some additional data gathered from medical exposures.

Monitoring of the atomic-bomb survivors began five years after the bombs were dropped and included more than 100,000 individuals, said Kathryn Higley, head of the department of nuclear engineering and radiation health physics at Oregon State University.

Scientists doing the study could not know exactly how much exposure those people had, she said, but they made good estimates by finding people's locations during the atomic blasts and measuring radioactive by-products in materials from nearby buildings.

What that revealed was a modest increase in cancer rates among the most exposed. The increased risk ranged from 1 percent to about 10 percent over a lifetime.

That means those who get exposed to high doses are not doomed to get cancer. The higher the exposure, the greater the risk, said John Keklak, director of radiation safety at Thomas Jefferson University Hospital.

"Every unit of radiation dose is like getting another ticket to a lottery, though not a lottery you'd want to win," he said.

Although scientists use a baffling array of units to measure radiation, the most common seems to be the millisievert, or mSv.

We get exposed on average to about 2.8 mSv a year from nature. Dental X-rays expose people to about .1 mSv. A typical CT scan exposes people to about 10 mSv, though studies have shown they can range from 3 to 90 mSv.

The data on the bomb survivors show a steady relationship between cancer risk and dose, but only down to about 100 mSv, said Higley, below which there is no measurable increase in cancer.

"The statistical noise is so great that there's too much uncertainty in that range," she said. Therefore, scientists agreed to calculate risk of doses lower than this by assuming that they can extrapolate down to zero, the risk growing ever smaller.

Under that assumption, there is no threshold below which radiation is safe.

Scientists acknowledge this is an assumption, and some vehemently disagree with it, but it is still considered the most reasonable one to make, according to the International Atomic Energy Agency and the National Academy of Sciences. It forms the basis of deciding our legal occupational exposures.

"There's no safe level of radiation," said Ira Helfand, a radiation expert and former head of Physicians for Social Responsibility, an antinuclear activist group.

At 100 millisieverts, the added cancer risk is about 1 percent, which may sound small, he said, but it means that of 100 people exposed, one will get cancer from that exposure. If millions of people are exposed, thousands will get cancer.

By the same logic, one group of researchers estimated that Americans will suffer 29,000 additional cancer cases from medical imaging.

That is based on the idea that even a small risk leads to cancers when millions of people are involved. About 70 million CT scans are done each year to identify problems from arterial plaque to tumors. A 2009 study reported that four million Americans each year undergo enough scans to expose them to more than the recommended safety limit for workers in nuclear power plants.

Scientists who study radiation say that is certainly an acceptable trade-off if such a procedure saves a patient's life, but not if its purpose is to protect doctors from lawsuits. A 2010 study found that 27 percent of a sample of 284 scans were "inappropriate."

Low-level radiation has led to long-standing disagreement over the effects of releases from Chernobyl and Three Mile Island. Most scientists agree the releases at Three Mile Island were too small to cause harm.

Chernobyl was a much worse accident. Thirty workers, most of them firefighters, died of radiation poisoning.

The science is less clear about what happened beyond the plant's perimeter. There was a spike in thyroid cancer among children who consumed milk contaminated with radioactive iodine, but beyond that, the epidemiologists could not find any measurable increase in cancer, the International Atomic Energy Agency reported.

Helfand argues that the Chernobyl disaster killed thousands of people by giving them cancer they would not have gotten otherwise. In his estimate, the radioactive contamination from the plant would have led to an additional 250,000 cancer cases.

But that number is based on some extrapolations and assumptions, and others doubt it, just as there will be doubts and disagreement about Fukushima for years to come.