In the aftermath of the accident, investigations focused on the amount of radiation released by the accident. According to the American Nuclear Society, using the official radiation emission figures, "The average radiation dose to people living within ten miles of the plant was eight millirem, and no more than 100 millirem to any single individual. Eight millirem is about equal to a chest X-ray, and 100 millirem is about a third of the average background level of radiation received by US residents in a year."

Based on these low emission figures, early scientific publications on the health effects of the fallout estimated one or two additional cancer deaths in the 10 mi (16 km) area around TMI. Disease rates in areas further than 10 miles from the plant were never examined. Local activism in the 1980s, based on anecdotal reports of negative health effects, led to scientific studies being commissioned. A variety of studies have been unable to conclude that the accident had substantial health effects.

The Radiation and Public Health Project cited calculations by Joseph Mangano, who has authored 19 medical journal articles and a book on Low Level Radiation and Immune Disease, that reported a spike in infant mortality in the downwind communities two years after the accident. Anecdotal evidence also records effects on the region's wildlife. For example, according to one anti-nuclear activist, Harvey Wasserman, the fallout caused "a plague of death and disease among the area's wild animals and farm livestock", including a sharp fall in the reproductive rate of the region's horses and cows, reflected in statistics from Pennsylvania's Department of Agriculture, though the Department denies a link with TMI.

The health effects of the 1979 Three Mile Island nuclear accident are widely, but not universally, agreed to be very low level. According to the official radiation release figures, average local radiation exposure was equivalent to a chest X-ray, and maximum local exposure equivalent to less than a year's background radiation. Local activism based on anecdotal reports of negative health effects led to scientific studies being commissioned. A variety of studies have been unable to conclude that the accident had substantial health effects, but a debate remains about some key data (such as the amount of radiation released, and where it went) and gaps in the literature.

Initial investigations

In the aftermath of the accident, investigations focused on the amount of radiation released by the accident. The official figures have been disputed by a number of insiders, who have suggested figures hundreds or thousands of times higher. According to the American Nuclear Society, using the relatively low official radiation emission figures, "The average radiation dose to people living within ten miles of the plant was eight millirem, and no more than 100 millirem to any single individual. Eight millirem is about equal to a chest X-ray, and 100 millirem is about a third of the average background level of radiation received by US residents in a year." To put this dose into context, while the average background radiation in the US is about 360 millirem per year, the Nuclear Regulatory Commission regulates all workers' of any US nuclear power plant exposure to radiation to a total of 5000 millirem per year. Based on these low emission figures, early scientific publications on the health effects of the fallout estimated one or two additional cancer deaths in the 10-mile area around TMI. Disease rates in areas further than 10 miles from the plant were never examined.

Local resident reports

The official figures are too low to account for the acute health effects reported by some local residents and documented in two books; such health effects require exposure to at least 100,000 millirems (100 rems) to the whole body - 1000 times more than the official estimates. The reported health effects are consistent with high doses of radiation, and comparable to the experiences of cancer patients undergoing radio-therapy,. but have many other potential causes. The effects included "metallic taste, erythema, nausea, vomiting, diarrhea, hair loss, deaths of pets and farm and wild animals, and damage to plants." Some local statistics showed dramatic one-year changes among the most vulnerable: "In Dauphin County, where the Three Mile Island plant is located, the 1979 death rate among infants under one year represented a 28 percent increase over that of 1978, and among infants under one month, the death rate increased by 54 percent." Physicist Ernest Sternglass, a specialist in low-level radiation, noted these statistics in the 1981 edition of his book Secret Fallout: low-level radiation from Hiroshima to Three-Mile Island. In their final 1981 report, however, the Pennsylvania Department of Health, examining death rates within the 10-mile area around TMI for the 6 months after the accident, said that the TMI-2 accident did not cause local deaths of infants or fetuses.

Scientific work continued in the 1980s, but focused heavily on the mental health effects due to stress, as the Kemeny Commission had concluded that this was the sole public health effect. A 1984 survey by a local psychologist of 450 local residents, documenting acute radiation health effects (as well as 19 cancers 1980-84 amongst the residents against an expected 2.6), ultimately led the TMI Public Health Fund reviewing the data and supporting a comprehensive epidemiological study by a team at Columbia University.

Columbia epidemiological study

In 1990-1 a Columbia University team, led by Maureen Hatch, carried out the first epidemiological study on local death rates before and after the accident, for the period 1975-1985, for the 10-mile area around TMI. Assigning fallout impact based on winds on the morning of March 28, 1979, the study found no link between fallout and cancer risk. The study found that cancer rates near the Three Mile Island plant peaked in 1982-3, but their mathematical model did not account for the observed increase in cancer rates, since they argued that latency periods for cancer are much longer than three years. The study concludes that stress may have been a factor (though no specific biological mechanism was identified), and speculated that changes in cancer screening were more important.

Wing review

Subsequently lawyers for 2000 residents asked epidemiologist Stephen Wing of the University of North Carolina at Chapel Hill, a specialist in nuclear radiation exposure, to re-examine the Columbia study. Wing was reluctant to get involved, later writing that "allegations of high radiation doses at TMI were considered by mainstream radiation scientists to be a product of radiation phobia or efforts to extort money from a blameless industry." Wing later noted that in order to obtain the relevant data, the Columbia study had to submit to what Wing called "a manipulation of research" in the form of a court order which prohibited "upper limit or worst case estimates of releases of radioactivity or population doses... [unless] such estimates would lead to a mathematical projection of less than 0.01 health effects." Wing found cancer rates raised within a 10-mile radius two years after the accident by 0.034% +/- 0.013%, 0.103% +/- 0.035%, and 0.139% +/- 0.073% for all cancer, lung cancer, and leukemia, respectively. An exchange of published responses between Wing and the Columbia team followed. Wing later noted a range of studies showing latency periods for cancer from radiation exposure between 1 and 5 years due to immune system suppression. Latencies between 1 and 9 years have been studied in a variety of contexts ranging from the Hiroshima survivors and the fallout from Chernobyl to therapeutic radiation; a 5-10 year latency is most common.

Further studies

On the recommendation of the Columbia team, the TMI Public Health Fund followed up its work with a longitudinal study. The 2000-3 University of Pittsburgh study compared post-TMI death rates in different parts of the local area, again using the wind direction on the morning of 28 March to assign fallout impact, even though, according to Joseph Mangano in the Bulletin of the Atomic Scientists, the areas of lowest fallout by this criterion had the highest mortality rates. In contrast to the Columbia study, which estimated exposure in 69 areas, the Pittsburgh study drew on the TMI Population Registry, compiled by the Pennsylvania Department of Health. This was based on radiation exposure information on 93% of the population living within five miles of the nuclear plant - nearly 36,000 people, gathered in door-to-door surveys shortly after the accident.^[19] The study found slight increases in cancer and mortality rates but "no consistent evidence" of causation by TMI. Wing et al. criticized the Pittsburgh study for making the same assumption as Columbia: that the official statistics on low doses of radiation were correct - leading to a study "in which the null hypothesis cannot be rejected due to a prior assumptions." Hatch et al. noted that their assumption had been backed up by dosimeter data, though Wing et al. noted the incompleteness of this data, particularly for releases early on.

In 2005 R. William Field, an epidemiologist at the University of Iowa, who first described radioactive contamination of the wild food chain from the accident suggested that some of the increased cancer rates noted around TMI were related to the area's very high levels of natural radon, noting that according to a 1994 EPA study, the Pennsylvania counties around TMI have the highest regional screening radon concentrations in the 38 states surveyed. The factor had also been considered by the Pittsburgh study and by the Columbia team, which had noted that "rates of childhood leukemia in the Three Mile Island area are low compared with national and regional rates." A 2006 study on the standard mortality rate in children in 34 counties downwind of TMI found an increase in the rate (for cancers other than leukemia) from 0.83 (1979–83) to 1.17 (1984–88), meaning a rise from below the national average to above it.

A 2008 study on thyroid cancer in the region found rates as expected in the county in which the reactor is located, and significantly higher than expected rates in two neighbouring counties beginning in 1990 and 1995 respectively. The research notes that "These findings, however, do not provide a causal link to the TMI accident." Mangano (2004) notes three large gaps in the literature: no study has focused on infant mortality data, or on data from outside the 10-mile zone, or on radioisotopes other than iodine, krypton, and xenon.