PENNSYLVANIA — Ever since the March 28, 1979, accident at Three Mile Island, Pa., people who lived near the nuclear power plant have been left wondering whether enough radiation was spewed into the air to trigger any cases of cancer -- cancers that might not appear until years later.
Twenty years ought to be enough time to know. But linking cancer to any subtle environmental effect is no easy business, so scientists are still wrangling over what they've observed.
Most who have studied the accident have concluded that even those living closest to the plant couldn't have been harmed by the tiny amounts of radiation released during those fearful few days.
Any cancers or other ailments that have emerged, they say, would have happened anyway. But a few scientists question that thinking, suggesting that radioactive releases were much higher than anyone realized.
The final answer was left to the epidemiologists, who have combed hospital records and interviewed thousands of people to seek connections between cancer and exposure to plumes of radioactive gases.
A study published in 1990 reassured people that little, if any, harm was done.
A re-analysis published two years ago suggested a connection to leukemia and lung cancer.
The largest study, which tracked 35,000 people living within five miles of the plant for 18 years, is due out this year.
Evelyn Talbot and colleagues from the University of Pittsburgh won't reveal details of their findings, which include lifestyle information, such as smoking, gleaned from interviews. But she said that the overall cancer rate was no higher than would have been expected if the area had no nuclear power plant, and that for individual cancers, her paper would not drop any bombshells.
"If that doesn't finally close the case, maybe it will never be closed," said Jonathan Berger, a Philadelphia environmental scientist who manages a fund for research into the accident.
The studies all run up against the same problem, he said.
There weren't sufficient measuring devices around the site to accurately assess the releases.
That oversight may seem surprising, but according to Harold Denton, a nuclear engineer appointed by President Jimmy Carter to give advice during the accident, the idea of a meltdown looked about as likely as the sinking of the Titanic.
That leaves local residents wondering whether cancer deaths of friends and relatives were caused by the accident, and whether they themselves carry invisible, radiation-induced cancer time bombs.
Paula Kinney, 51, used to live so close to the plant that she could see the imposing concrete cooling towers from her kitchen window.
"It was sheer terror when the news first got out. It was like, we were all going to die," she said.
Now she wonders whether the radioactive releases have anything to do with the ovarian cysts her older daughter suffered at age 13, a problem she said affected many teen-age girls in the neighborhood.
She wonders about the menstrual irregularity and other odd symptoms that still plague her younger one, who was 4 at the time.
She wonders whether the radiation caused the death of an infant born to a friend, and whether she and her family are saddled with a high risk of cancer.
Sharon Arndt, 40, who lived in Middletown, Pa., with her family at the time, connects the accident to her mother's death from colon cancer at 65.
They made it sound like everything was fine, she said of officials who made news announcements during the accident.
Arndt, now a legal secretary in Harrisburg, Pa., said she wishes she had evacuated -- something many people did.
Experts disagree over both the quantity of radiation released and which of several releases were most serious.
Some radiation escaped soon after a stuck valve started the whole thing at 4 in the morning on March 28.
Then, radioactive gases evaporated from contaminated water that had spilled over the floor of the reactor and an auxiliary building.
During the first day of the accident, people on the farms and in the towns of Middletown and Royalton, Pa., reported classic symptoms of radiation sickness -- nausea, vomiting and hair loss.
Those symptoms might also be attributed to the stress of thinking they were about to witness a nuclear apocalypse.
But it's extremely unlikely that hazardous amounts of radiation were released during the first day or two, said David Lochbaum, a nuclear engineer and former consultant at the Susquehanna, Pa., plant who now works for the anti-nuclear group Union of Concerned Scientists.
The worst releases began on the fourth day, he said, after the 100-ton uranium core had begun to melt.
(The reactor shut off within minutes of the accident, but there was no way to shut off the heat-generating radioactivity produced by nuclear fission in the core.) That day, plant operators tried to bleed off a bubble of hydrogen gas that had collected in the reactor.
Some engineers feared this bubble would blow up the steel reactor vessel and perhaps destroy the concrete containment building that housed it, sending highly radioactive materials all over Pennsylvania.
So, plant operators let out more water to carry away some of the hydrogen dissolved in it.
They didn't realize that the core had melted, said Lochbaum, so that water flowing through the reactor was picking up radioactive contaminants.
That water went to storage tanks where dissolved gases were vented outside.
It was only three years later, when a camera was lowered into the damaged reactor, that the extent of the meltdown was known.
Scientists on a government panel analyzing the accident concluded that the releases were too small to pose a danger to neighbors of the plant.
According to Ralph DeSantis, a spokesman for plant operator GPU Nuclear, a person in the worst possible position, standing at the boundary of the plant through the whole week after the accident, would have received a dose of about 100 millirems -- about what people ordinarily receive from natural sources of radiation in the environment during about three months.
It's important to note that no one actually received that much, he said.
The most anyone actually received was 8 millirems, hardly more than the natural radiation of Earth and space that regularly bombards people, he said.
But, according to Lochbaum, those estimates included data from a crucial meter that should have revealed what was going up the stacks but was not designed to read such high radiation levels.
Epidemiology -- which tries to find causes of disease in large populations -- is not an exact science, and the field is rife with conflicting studies.
In recent months new studies have refuted earlier ones that showed eating fiber protects against colon cancer and that low-fat diets helped thwart breast cancer.
Cancer is relatively common, so some people in any large enough region are bound to develop it, said David Savitz, an epidemiologist from the University of North Carolina.
It's also particularly difficult to link any disease with a specific event.
"We can't rerun this experiment," Savitz said.
In studying TMI, epidemiologists have combed hospital records trying to determine whether any additional cancers resulted from the radiation releases.
The first such study, published in 1990, found no evidence that TMI led to an increase in cancers.
"We went through a pretty arduous effort of going to all hospitals in a 30- or 40-mile radius where people might have gone for cancer care," said one of the authors, Maureen Hatch of Columbia University.
An increase in cancers in 1982, she said, might have resulted from more people seeking cancer screening, fearing they had been harmed by the accident.
Hatch said that most people near TMI were not exposed to radioactive releases, because the prevailing winds sent the radiation out in a narrow plume.
Therefore, she said, their best bet for determining whether the releases led to cancers would be to compare people who lived within that plume with residents outside it. In the end, they did see a slightly greater number of lung cancers in the plume, but Hatch said that probably had more to do with stress, which could trigger smoking, as well as a conventional power plant adding pollution to the air.
Hatch estimated about a 30 percent increase in risk -- tiny compared with smoking, which causes an increased risk of 3,000 percent.
Hatch did not find more instances of leukemia or lymphoma, which are thought to be the cancers most sensitive to radiation -- a determination based on Hiroshima survivors.
Hatch said her results don't prove that the accident was harmless -- only that any cancer effects are too subtle to show up in this kind of study.
Steven Wing, an epidemiologist from the University of North Carolina, disagrees.
He re-analyzed the data for lawyers representing 2,000 people suing TMI, and found a stronger incidence of lung cancer -- 30 cases compared with the 13 expected for the area had there been no radiation.
That corresponds to an 85 percent increase in risk.
He said he also saw an increase in leukemia but that it was hard to determine its statistical significance because they found only a handful of cases inside and outside the plume.
Savitz suspects that the scientists' differing conclusions have been influenced by their opinions about how much radiation was released.
"If you believe that minuscule amounts of radiation were released, you're less likely to see a connection than if you thought people were hit with a higher dose," he said.
Wing notes that, beyond 20 years, any effects of the accident are diluted by the migrations of people to and from the area.
So the forthcoming study will either close the case or leave people still fearful.
* The reactor's fuel core became uncovered and more than one-third of the fuel melted.
* Inadequate instrumentation and training programs at the time hampered operators' ability to respond to the accident.
* The accident was accompanied by communications problems that led to conflicting information being released to the public, contributing to the public's fears.
* Radiation was released from the plant. The releases were not serious and were not health hazards. This was confirmed by thousands of environmental and other samples and measurements taken during the accident.
* The containment building worked as designed. And despite melting of more than one-third of the fuel core, the reactor vessel maintained its integrity and contained the damaged fuel.
Source: GPU Inc. Web site