THE 2018 Health Physics Society (HPS) plenary session in Cleveland titled “Health Physics and the Realm of Low-Dose Radiation” offered invited presentations on low-dose health effects and radiation limits. Presentation titles included “What Have Inappropriate Radiation Limits Done to Our Nuclear Waste Disposal Program?” (Jim Conca), “Prudence and the Hidden Burden of Conservatism” (Roger Coates), and “Low-Dose Radiation Biology Past, Present, and Future” (Antone Brooks). Discussion and debate among health physicists concerning the health effects due to low-dose radiation has been standard fare at HPS meetings for decades.2
The National Council on Radiation Protection and Measurements (NCRP) recently issued Commentary 27, which addressed their conclusions and recommendations on low-dose radiation effects on human health and the environment (NCRP 2018). They concluded that based on current epidemiological data, the linear no-threshold (LNT) model “should continue to be used for radiation protection purposes.” The NCRP points out that “no alternative dose-response relationship appears more pragmatic or prudent for radiation protection purposes than the LNT model.” Thus, NCRP has assessed the latest available epidemiological studies on low-dose radiation health effects and has endorsed the continued use of the LNT model for radiation protection purposes. LNT opponents were quick to articulate their dissatisfactions with this latest endorsement of the LNT model (Ulsh 2018).
While the LNT debate persists, strong consensus among radiation protection professionals may be found on two related fronts. First, health physicists agree that a low-dose radiation research program is needed to develop a long-term national strategy for understanding and addressing the potential health and safety issues associated with low-dose radiation exposures. Second, health physicists agree that regulating very low radiation doses (e.g., on the order of fractions of natural background radiation) is not providing radiation protection and does not benefit society.3 This paper briefly addresses the path forward for low-dose radiation research and proposes a new stopping point for the application of the as low as reasonably achievable (ALARA) principle—by increasing the NCRP’s negligible individual dose (NID) from 0.01 mSv y−1 to 0.1 mSv y−1. If adopted, this new stopping point will certainly not settle the LNT debate, but it is a step in the right direction to alleviate LNT’s more egregious unintended consequences.
Radiation protection standards are based on high dose and dose-rate effects using epidemiological data from the Life Span Study of atomic bomb survivors (Cardarelli and Ulsh 2018). Indeed, dose-response relationships for cancer induction at high dose and dose rate have been well established; however, health effects (risks) at low doses and low-dose rates (LDDR) are uncertain and extrapolated from epidemiologic studies at higher doses. For the purpose of this paper, the NCRP’s definition of LDDR will be used: a low absorbed dose is <100 mGy delivered acutely, and a low absorbed dose rate is <5 mGy h−1 for any accumulated dose (NCRP 2018). Lacking a sufficiently detailed understanding of human biological effects at low doses, standards are established assuming a linear relationship between exposure and effect—resulting in the popular and oft-debated linear no-threshold model. Radiobiological studies at cellular and molecular levels have yielded better understanding of biological mechanisms in response to radiation (Brooks 2018), increasing doubts among health physicists about using the LNT model as the basis of our radiation protection paradigm. Low-dose radiation research is imperative to validate or identify alternatives to the LNT model—which may include hypersensitivity, threshold, and hormesis (Fig. 1).
Road map for low-dose radiation research
The strength of a low-dose radiation research program is the integration of radiation epidemiology and radiobiology to clarify the dose-response relationship at low doses and low dose rates to increase the accuracy and precision of risk estimates. The NCRP published a road map for low-dose research in Commentary 24, Health Effects of Low Doses of Radiation: Perspectives on Integrating Radiation Biology and Epidemiology (NCRP 2015). This commentary advocates use of radiation biology data to augment epidemiological studies to reduce uncertainties in risk estimates of low-dose exposure. Expected research outcomes may include a better understanding of the complex biology associated with radiation-induced stochastic effects that involve protective processes such as elimination of aberrant cells via apoptosis and cancer suppression via immune system functions.
From an epidemiological perspective, the Million Person Study (MPS) of low-dose health effects holds much promise. This is an epidemiologic study of US radiation workers that is 12 times larger than the Life Span Study (86,000 people)—the MPS includes 115,000 atomic veterans, 360,000 US Department of Energy (DOE) workers, 150,000 nuclear utility workers, 250,000 radiologists and medical workers, and 130,000 industrial radiographers (Boice et al. 2018). Radiation doses to these populations are more characteristic of occupationally exposed workers today than radiation doses in the Life Span Study, largely because radiation workers received radiation exposures gradually over time.
Low-dose radiation research has been a topic of keen interest on Capitol Hill due to expected outcomes of this program that will inform the regulatory environment with regards to worker exposures, as well as environmental cleanup requirements. A recent US Government Accountability Office (GAO) report (GAO 2017) found that the nation will benefit from a low-dose radiation program with DOE leading a collaborative research effort among federal agencies. In 2018, the Senate passed HR 589, US DOE Research and Innovation Act (US Congress 2018) with unanimous consent, directing the US DOE Office of Science to “carry out a low-dose radiation research program to enhance the scientific knowledge of, and reduce uncertainties associated with, the effects of exposure to low-dose radiation to inform improved risk-management methods.” While this program is not across the finish line, restarting the low-dose research program has momentum and Congressional interest.
LNT, ALARA, and unintended consequences
For now, the LNT model reigns supreme in the realm of radiation protection. The no-threshold feature of the LNT model provides the foundation for the ALARA construct. Absent scientific consensus on the existence of a threshold, prudent regulatory application of LNT has resulted in radiation protection optimization—a paradigm that requires that the likelihood of incurring exposures, the number of people exposed, and the magnitude of individual exposures be as low as reasonably achievable, considering economic and other societal factors. To the public, regulators, and interested stakeholders, the no-threshold attribute of the LNT model often conveys a clear and concerning message: “There is no safe level of radiation dose.” While some may quibble about the meaning of “safe,” the LNT/ALARA paradigm requires that radiation doses be lowered as long as it is reasonable to do so.
It’s important to give LNT/ALARA its due credit. The ALARA optimization approach for radiation protection has been quite effective at maintaining radiation doses below occupational dose limits through the judicious application of engineering and administrative controls, as well as personal protective equipment. The US Nuclear Regulatory Commission (NRC) has defined ALARA in 10 CFR 20.1003 as “making every reasonable effort to maintain exposures to radiation as far below the dose limits in this part as is practical consistent with the purpose for which the licensed activity is undertaken, taking into account the state of technology, the economics of improvements in relation to state of technology, the economics of improvements in relation to benefits to the public health and safety, and other societal and socioeconomic considerations, and in relation to utilization of nuclear energy and licensed materials in the public interest” (US NRC 1991). Health physicists and the radiation protection community in general have exceled at lowering radiation doses by practicing ALARA.
In fact, according to the most recent (2016) US NRC data, reported in NUREG-0713 (US NRC 2018), not one individual worker of the 121,129 monitored radiation workers exceeded the occupational limit radiation dose limit (50 mSv), and 99.9% of these workers received less than 20 mSv. Note that these occupational exposure data do not include all monitored radiation workers (e.g., medical radiation workers); the five categories of US NRC licensees subject to reporting requirements include commercial nuclear power reactors and test reactor facilities; industrial radiographers; fuel processors (including uranium enrichment facilities), fabricators, and reprocessors; manufacturers and distributors of byproduct material; and independent spent fuel storage installations. For these workers, the average measurable dose was 1.4 mSv—this value can be compared with the 3.1 mSv that the average person in the United States receives annually from natural background radiation (NCRP 2009). Furthermore, over the past three decades the US DOE- and NRC-licensed radiation safety communities have lowered the average dose delivered to radiation workers by a factor of three.4 Future occupational radiation doses may be further reduced if ALARA programs deem that reasonable efforts can be made to maintain exposures to radiation as far below the dose limits as is practical—this is the nature of LNT/ALARA-based radiation protection.
Health physicists are charged with protecting workers, the public, and the environment from potential radiation hazards. Health physicists do not employ a practical threshold below which radiation levels no longer pose a hazard to people; rather, they apply ALARA as part and parcel of the radiation protection optimization process. The ALARA process should prevent the expenditure of an inordinate amount of financial resources to eliminate or mitigate vanishingly small radiation hazards. How do we justify the expenditure of resources to drive radiation doses lower and lower—or have we stopped counting the costs? Indeed, even the cost of performing an evaluation for the demonstration of ALARA—let alone implementing some ALARA measure—may be nonnegligible at very low doses. This is the unintended consequence of the LNT/ALARA paradigm—various stakeholders do not always agree when it’s reasonable to stop going lower. Some believe that if reducing radiation doses by 10% is good, then 15 or 20% is even better.
The US NRC has issued regulatory guidance that allows economic considerations when evaluating the cost of “every reasonable effort.” Draft NUREG-1530 proposes a reassessment of the cost per collective dose averted (i.e., $5,100 per person rem) that is used in cost-benefit assessments for establishing reasonableness under the ALARA program (US NRC 2015). This value is based on the cost of a statistical life ($9 million in the latest draft NUREG-1530); thus, the cost per collective dose is dynamic, varying due to a number of social and economic factors. The calculation of ALARA is not just for an individual but also relies on the concept of collective dose; thus, whereas implementation of some ALARA action may not be justified to reduce dose by 0.05 mSv for 1 person, it may meet the threshold if it reduces the dose by 0.05 mSv for 100 people. This collective dose reduction does not achieve any measurable improvement to health and safety while incurring significant real costs to achieve that dose reduction.
From the perspective of radiation protection, applying ALARA at increasingly lower radiation doses eventually reaches levels that are consistent or lower than background levels, and background radiation levels are not usually viewed as potential radiation hazards (elevated radon levels being the notable exception). While ALARA is an effective paradigm for optimizing radiation protection, the approach breaks down when the radiation doses being considered (background levels) are not radiation hazards; and ubiquitous background is not a hazard. This is an unintended consequence of LNT/ALARA paradigm. An analysis of economic factors alone would make it virtually impossible to justify lowering radiation doses that are already fractions of background levels.
Consider the environmental cleanup and nuclear decommissioning industry. The US NRC’s license termination rule specifies an unrestricted release criterion of 0.25 mSv y−1 and ALARA. NUREG-1757 (Appendix N, volume 2) provides guidance on the ALARA analyses needed to demonstrate that the ALARA requirement of the license termination rule (LTR) has been met (US NRC 2006). The US NRC’s ALARA analysis example for soil-contaminated sites demonstrates that complying with the annual dose limit (0.25 mSv) is more limiting than the ALARA analysis outcome by more than a factor of 100. In other words, remediating soil to achieve additional cleanup margins below the regulatory limit of 0.25 mSv y−1 (due to ALARA requirements) is clearly not justified (and it’s not even close). Yet there are tens of decommissioning sites across the United States that drive cleanup levels significantly below 0.25 mSv y−1 and incur millions of dollars in additional costs to achieve these marginally lower residual radioactivity levels. This is a blatant unintended consequence of LNT/ALARA paradigm and highlights how the “reasonably” in ALARA is essentially absent (ALArA).
An ALARA stopping point—a reintroduction of negligible individual dose
Here’s a straightforward proposition for radiation protection paradigms: question the application of ALARA in the range of background levels. That is, recognize that at background levels it is unreasonable to expend resources or to increase nonradiological risks (e.g., worker health, environmental impacts, and unwarranted evacuations) in an effort to further reduce radiation doses that are already at background levels. Fig. 2 shows the annual gamma effective dose in the United States from background radiation—the mean is 3.1 mSv, with a standard deviation of 3.6 mSv (NCRP 2009). Note that 2.5 million people are estimated to receive more than 20 mSv y−1 from background.
Further, the terrestrial gamma component of annual background radiation alone ranges 0.4 mSv from 5th to 95th percentile (NCRP 2009). What is the value of practicing radiation protection for annual doses in the range of 0.1 mSv, when background terrestrial gamma dose alone has more variability? Now consider that background radiation doses for US residents can easily vary by 0.1 mSv y−1 simply by relocating to different regions of the United States or by taking a job that requires increased air travel. Furthermore, NCRP 160 estimated the effective dose per individual in the US population from consumer products and activities (e.g., commercial air travel, cigarette smoking, and radionuclide exposure from common building materials) as 0.13 mSv y−1 (NCRP 2009). Would the public consider it reasonable to spend resources to lower radiation doses below 0.1 mSv y−1 in light of the comparable doses received from background sources?
Consider a thought exercise regarding radiation protection practices. For each of the following annual doses, how much effort (and financial resources) should be expended to further reduce individual annual dose?
- 100 mSv—radiation level is above occupational dose limit; dose level must be reduced.
- 50 mSv—radiation level is at occupational dose limit; ALARA needed to reduce dose.
- 10 mSv—radiation level is a few multiples of background; ALARA should be applied.
- 1 mSv—radiation level is consistent or lower than background; ALARA may be applied but many times will not be reasonable due to economic factors.
- 0.125 mSv—radiation level is not protective under the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA); remediation is necessary.
- 0.1 mSv—radiation level is a fraction of background variability; ALARA will never be justified if accepted economic factors for person rem averted are considered.
- 0.04 mSv—US EPA drinking water standard for beta gamma emitters.
NCRP 116 introduced the negligible individual dose with a value of 0.01 mSv y−1 (NCRP 1993). A NID of 0.01 mSv y−1 is of course truly negligible, but it’s not of any practical value. How many health physicists have referenced the NID when justifying radiation program decisions? It’s time to “NID-pick.” NCRP should increase the NID to a more practical value of 0.1 mSv y−1. This would be the new ALARA stopping point; Fig. 3 illustrates how small 0.1 mSv is relative to background variation. Radiation protection should not include trying to protect people from background radiation doses (elevated radon levels being the exception).
Coincidentally, and consistent with the proposed ALARA stopping point, an increase of the International Atomic Energy Agency (IAEA) basic safety standard (BSS) minimum dose constraint to 0.1 mSv y−1 was proposed 10 y ago (Hattori 2008). Hattori shows through probabilistic analyses that additional exposure to man-made radiation up to 0.5 mSv y−1 (as a dose constraint) does not significantly change the distribution of total public doses, using the lognormal natural background mean of 2.4 mSv y−1, with geometric mean and standard deviation of 2.0 mSv y−1 (UNSCEAR 2000). Hattori concludes that “additional dose resulting from setting the dose constraint to the order of 0.1 mSv y−1 can be regarded as trivial” (Hattori 2008). Let’s not wait another 10 y to increase the minimum dose constraint/negligible individual dose.
It is recognized that adoption of a NID of 0.1 mSv y−1 may challenge and conflict with existing US Environmental Protection Agency guidance and regulations. These include US EPA’s National Emission Standards for Hazardous Air Pollutants (NESHAP), which requires that emissions of radionuclides in air not exceed 0.1 mSv y−1; US EPA’s public drinking water standard of 0.04 mSv y−1; and recent CERCLA guidance that states that radiation standards that allow doses above 0.125 mSv y−1 are not protective and may not be used for cleanup of CERCLA sites. An ALARA stopping point of 0.1 mSv y−1 would indeed be a discussion point for how these public exposures regulated by EPA should be addressed.
Uncertainties in low-dose radiation health effects continue to fuel the LNT debate and support the need for a low-dose radiation research program. Recent radiation biology and epidemiology studies have greatly increased our knowledge in the low-dose region (less than 100 mSv), but more research is needed. The recent Radiation Protection Research Needs Workshop focused on the research needs facing federal agencies in meeting their radiation protection mission needs (Davis et al. 2019), not the least of which was a greater understanding of low-dose radiation health effects.
ALARA is an important pillar of radiation protection programs, so let’s apply it where it adds value, not within the range of background variations. Radiation safety professionals overwhelmingly agree that applying ALARA at very low doses, such as those consistent with background radiation levels, is not improving radiation safety of the public or radiation workers. To the contrary, this practice has significant financial and social consequences. Not only are we spending vast resources to reduce trivial doses in the name of LNT/ALARA, the mere fact that we do this sends the message that radiation exposure at these low levels must be harmful; why else would we spend millions of dollars to reduce levels that low? It’s a reasonable question for the public to ask.
In the meantime, we should apply what we already know—that regulating very low doses (on the order of fractions of background radiation) is not radiation protection. To move forward, the NCRP should increase the negligible individual dose to a more practical value of 0.1 mSv y−1; this is the new ALARA stopping point.
Boice JD Jr, Ellis ED, Golden AP, Girardi DJ, Cohen SS, Chen H, Mumma MT, Shore RE, Leggett RW. The past informs the future: an overview of the million worker study and the Mallinckrodt chemical works cohort. Health Phys 114:381–385; 2018. DOI .
Brooks AL. Low dose radiation: the history of the US Department of Energy research program. Pullman, WA: WSU Press; 2018.
Cardarelli JJ, Ulsh BA. It is time to move beyond the linear no-threshold theory for low-dose radiation protection
. Dose Response 16:1–24; 2018. DOI 10.1177/1559325818779651.
Davis J, Dewji SA, Abelquist E, Hertel N. Synopsis of the Oak Ridge radiation protection
research needs workshop. Health Phys 116:69–80; 2019.
Hattori T. Reconsideration of the minimum dose constraint for public exposures in radiological protection. Radiat Protect Dosim 131:340–345; 2008. DOI .
National Council on Radiation Protection
and Measurements. Health effects of low doses of radiation: perspectives on integrating radiation biology and epidemiology. Bethesda, MD: NCRP; Commentary 24; 2015.
National Council on Radiation Protection
and Measurements. Implications of recent epidemiologic studies for the linear-nonthreshold model. Bethesda, MD: NCRP; Commentary 27; 2018.
National Council on Radiation Protection
and Measurements. Ionizing radiation exposure of the population of the United States. Bethesda, MD: NCRP; Report 160; 2009.
National Council on Radiation Protection
and Measurements. Limitation of exposure to ionizing radiation. Bethesda, MD: NCRP; Report 116; 1993.
Ulsh BA. A critical evaluation of the NCRP commentary 27 endorsement of the linear no-threshold model of radiation effects. Environ Res 167:472–487; 2018. DOI .
United Nations Scientific Committee on the Effects of Atomic Radiation. Sources and effects of ionizing radiation. New York: United Nations; UNSCEAR 2000 report; 2000.
US Government Accountability Office. Low dose radiation: interagency collaboration on planning research could improve information on health effects. Washington, DC: US GAO; GAO-17-546; 2017.
US Nuclear Regulatory Commission. Consolidated decommissioning guidance: characterization, survey, and determination of radiological criteria. Washington, DC: US Government Printing Office; NUREG-1757, vol. 2, rev. 1; 2006.
US Nuclear Regulatory Commission. Reassessment of NRC’s dollar per person-rem conversion factor policy. Draft report for comment. Washington, DC: US Government Printing Office; NUREG-1530, rev. 1; 2015.
US Nuclear Regulatory Commission. Occupational radiation exposure at commercial nuclear power reactors and other facilities 2016; 49th annual report. Washington, DC: US Government Printing Office; NUREG-0713, Vol. 38; 2018.
2 In addition to the HPS annual meeting in Cleveland, health effects from low-dose radiation was the topic at the American Nuclear Society and HPS jointly sponsored Applicability of Radiation-Response Models to Low-Dose Protection Standards Conference in Pasco, Washington, US, from 30 September to 3 October 2018.
3 Support for these conclusions regarding health physicist consensus is based on my HPS president-elect chapter visits in 2016 and 2017 and on the 2016 HPS Future Directions Membership Survey [online]; available at https://hps.org/membersonly/documents/survey_future_directions.pdf.
4 US DOE’s Radiation Exposure Monitoring System (REMS) database of occupational radiation exposures reports for individuals with measurable dose that the average dose equivalent in 1986 was 1.79 mSv and in 2016 it was 0.59 mSv. Similarly, US NRC’s Radiation Exposure Information and Reporting System (REIRS) reports the average dose in 1986 was 4 mSv and in 2016 it was 1.4 mSv. The occupational dose limit of 50 mSv has not changed during this period.