The National Academy of Sciences (NAS) has started the initial planning step of the Nuclear Regulatory Commission-sponsored pilot study of cancer risks in populations around six U.S. nuclear power plants and a nuclear fuel-cycle facility. The NRC is asking the Academy to carry out this pilot to help the agency determine whether to extend the study to additional U.S. reactors and fuel-cycle facilities.

(Pictured: Millstone Power Station from above)

The pilot study will examine cancer risks around each of the seven nuclear sites using two types of epidemiological studies. The first will examine multiple cancer types in populations of all ages living near the nuclear sites; the second will be a record-linkage-based case-control study of cancers in children born near the sites.
The six nuclear power plants are:

  • Dresden Nuclear Power Station, Morris, ll.
  • Millstone Power Station, Waterford, Conn.
  • Oyster Creek Nuclear Generating Station, Forked River, N.J.
  • Haddam Neck (decommissioned), Haddam Neck, Conn.
  • Big Rock Point Nuclear Power Plant (decommissioned), Charlevoix, Mich.
  • San Onofre Nuclear Generating Station (permanently shut down), San Clemente, Calif.

The Dresden and Millstone sites include both operating reactors and a decommissioned reactor. The pilot effort will also study Nuclear Fuel Services in Erwin, Tenn. NAS recommended these sites because they provide a good sampling of facilities with different operating histories, population sizes around them, and expected levels of complexity in data retrieval from the relevant state cancer registries.

The NAS study aims to update and improve information on potential cancer risks around nuclear sites from the 1990 U.S. National Institutes of Health – National Cancer Institute (NCI) report, "Cancer in Populations Living Near Nuclear Facilities." The NRC has used the 1990 NCI report as a primary resource when communicating with the public about cancer risks in counties that contain or are adjacent to nuclear sites.

The work follows a feasibility study conducted by a 20-member NAS expert committee, which produced a report in 2012: "Analysis of Cancer Risk in Populations near Nuclear Faculties — Phase 1".

According to an October 2012 summary of the NRC’s plans at that time, the NAS report had three findings and recommendations.
The first finding was essentially that proving cancer risks will be very difficult. It identified four limitations for performing epidemiological studies around nuclear facilities:

  • uneven availability and quality of data on cancer mortality and incidence at geographic levels smaller than a county
  • uneven availability and quality of data on nuclear facility effluent releases based on an initial review of a limited set of records (In the summary, NRC staff disputed these claims)
  • inability to reliably capture information on population mobility, risk factors, and potential confounding factors (e.g. smoking, diet, or other lifestyle factors that would affect cancer risk)
  • low expected statistical power.

For example, the NAS committee declined to give a sample size required to detect health effects at the low offsite doses from these facilities (less than 0.01 mSv/year) because that number would be ‘truly enormous’ (greater than hundreds of millions) given the low doses involved. NRC staff said that this conclusion "confirms the staff position that at the low offsite doses from these facilities, researchers would not expect to observe any increased cancer risks in the populations surrounding these facilities attributed to the regulated release of radioactive effluents."

However, it went on to say, "Nevertheless, the staff believes that despite these potential limitations and expected outcomes, the studies would be helpful to address public health concerns and are therefore still worthwhile to pursue."
The second finding was that despite these challenges there are possible study designs that could be used. It went on to recommend two types of epidemiology studies: an ecologic study of multiple cancer types of populations living near nuclear facilities and a case-control study of cancers in children born near nuclear facilities.

The third finding was that that facility data on radioactive effluent release, direct exposure, and meteorology can be used to obtain estimates of annual variations in dose as a function of distance and direction from nuclear facilities. (Although it noted that effluent releases cannot necessarily predict absorbed dose because most results are below lab detection limits).

Another recommendation was for the pilot study which is currently beginning.

The last recommendation called for a communication and engagement plan to be developed before initiating data gathering and analysis.