Public Release: 

No apparent increase in cancer deaths among Three Mile Island residents, report University of Pittsburgh researchers

University of Pittsburgh Medical Center

Radioactivity released during the Three Mile Island (TMI) accident in 1979 does not appear to have caused an increase in cancer mortality among people living within a five-mile radius of the nuclear accident, according to researchers at the University of Pittsburgh's Graduate School of Public Health (GSPH). The findings were published Friday, April 28, on the website of Environmental Health Perspectives, a journal of the National Institutes of Health's National Institute of Environmental Health Sciences. The paper will also appear in the June issue of the journal.

While several previous follow-up studies on the TMI population have been conducted in the past, this one is the most extensive due to its longer, 13-year time-frame and the use of information about residents' lifestyles (such as smoking habits and education levels) and everyday background radiation exposure beyond what was caused by the TMI incident.

"This study helps put to rest the lingering question of whether the residents of Three Mile Island are experiencing an increase in cancer deaths as a result of the nuclear accident," said Evelyn Talbott, Dr. P.H., associate professor, department of epidemiology and principal investigator on the study.

The TMI incident occurred at a nuclear power plant near Harrisburg, Pa., on March 29, 1979, when a reactor leaked small amounts of radioactive gases. It is often referred to as the worst nuclear accident in American history. Scientists have calculated that the average person present in the area during the 10 days after the incident was exposed to considerably less radiation than the annual dose an individual receives from the everyday environment in the United States.

The University of Pittsburgh study covered the years 1979-1992. For demographic and lifestyle data on the individuals living in the TMI area, researchers used information collected by the Pennsylvania Department of Health in interviews conducted with TMI residents within two months of the accident. Information collected on these 32,135 individuals included education, occupation, smoking status, residential history, medical history, previous radiation exposure and daily travel in and out of the area during the 10 days following the accident.

Investigators from the University of Pittsburgh determined these individuals' maximum and likely gamma exposure during the 10 days after the accident and combined the results with the interview data as well as mortality data from the Pennsylvania Department of Health. In analyzing the information they used standard calculations such as statistical modeling and standard mortality ratios -- direct comparisons of the number of observed deaths with the expected number of deaths in a general population.

Mortality indicators examined in the study included all heart disease, all malignancies as well as specific cancers that are known to be sensitive to radioactivity: bronchus, trachea and lung; breast (women only); lymphatic and hematopoietic tissue (blood-forming organs), excluding chronic lymphocytic leukemia and Hodgkin's disease; and the central nervous system. Thyroid cancer was considered, but no deaths were reported during the study period.

Initial results indicated a significantly higher mortality from all causes among the TMI population as compared with residents of the surrounding three-county area. The largest contributor to that mortality was heart disease. However, after adjusting for smoking and education, the increases were no longer apparent.

"The effects of smoking and education levels on the incidence of heart disease are well known," said Dr. Talbott. "When we controlled for these risk factors, along with background radiation factors, we found that the elevations in mortality were not note-worthy."

Likewise, elevations in mortality from cancers of the bronchus, trachea and lung were observed in women, as well as an increase in lymphatic and hematopoietic tissue cancers in men, even after controlling for background radiation exposure, education and smoking. However, neither showed a significant dose-response trend.

Investigators noted an increasing pattern of relative risk for breast cancer in relation to increasing levels of likely exposure to gamma radiation, suggesting a possible link between dose of radiation and increased risk. However, they concluded that overall there was no significant relationship between likely exposure to gamma radiation and breast cancer mortality.

"A relationship between gamma radiation and breast cancer has been noted in other investigations, but emissions from the TMI incident were significantly lower than in other documented studies," said Dr. Talbott. "Therefore, it is unlikely that this observed dose trend is related to radiation exposure on the day of the accident." To fully explore the relationship between low-level radiation and breast cancer, an in-depth case-control study would be necessary, she said.

While the authors found no consistent evidence suggesting that the low-dose radiation released during the TMI accident had a measurable impact on the mortality of those living in the area for 13 years after the event, they acknowledge that further study is warranted.

"Because the latency period for many cancers is 20 years or more, continued follow-up on the TMI residents will provide a more comprehensive look at their mortality, as well as morbidity, from various cancers," said Dr. Talbott. The University of Pittsburgh research team is currently analyzing data collected through 1999 on the same population.

Established in 1948, the GSPH at the University of Pittsburgh is world-renowned for contributions that have influenced public health practices and medical care for millions of people. It is the only fully accredited school of public health in the Commonwealth of Pennsylvania and is one of the top-ranked schools of public health in the United States.


For more information about the GSPH at the University of Pittsburgh, access the school's website at

This research was supported by a grant from the Three Mile Island Public Health Fund.

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.