A closer look at NRC’s proposed rule eliminating ALARA

On July 1, the Nuclear Regulatory Commission proposed removing the "as low as reasonably achievable" (ALARA) principle from its radiation protection regulations and replacing it with a graded approach. The agency also proposed increases to effluent dose limits and adjusting how it handles allowances for exceeding dose limits.

According to the proposed rule, the NRC expects that by complying with the existing rules, current licensees will already be compliant with the updated rule. They would have the option to-but not be required to-change their existing programs.

"We're raising the standard for regulatory clarity, not lowering the standard for safety," NRC Chairman Ho Nieh said in a press release. "Our radiation dose limits remain unchanged-what we're eliminating is unnecessary ambiguity."

Executive pressure: In May 2025, executive order (EO) 14300 called for a slew of changes at the NRC. Among these was an order to "adopt science-based radiation limits," specifically, to reconsider ALARA and the linear no-threshold (LNT) model for radiation exposure on which it's based, which the EO describes as "flawed."

In response, the NRC's proposal reexamines 10 CFR Part 20, which sets radiation protection standards for all NRC licensees and Agreement States.

The NRC states the revisions aim to "address unnecessary conservatism and excessive subjectivity" associated with regulating low doses of radiation and "incorporate flexibility and acceptable alternatives."

The rule change also discusses the limits of the LNT model but finds that "no consensus-supported, regulation-ready alternative model exists at this time."

The proposal's only mention of hormesis by name does not explicitly state the agency's current position on the concept, but it quotes the agency's 2021 position, stating that there is no compelling evidence for hormesis. However, in the section describing effluents, the release of radioactive material into the environment, the NRC states without citing a source that "there is evidence for adaptive cellular response, which would mitigate the health effects of exposures at low doses, especially those resulting from effluents."

Amir Bahadori, a professor at Kansas State University and certified health physicist, said he thinks the NRC's response to the EO is "generally consistent with current scientific knowledge, the collective experience in the field of radiation protection, and the NRC's regulatory experience," but notes that the agency doesn't acknowledge key studies that support the continued use of the LNT model.

Low-dose radiation science: While the health impacts of higher-dose radiation are clear, it has proven difficult to measure the effect of radiation at low doses.

Many factors impact cancer risk, including age, weight, diet, cigarette smoking, alcohol consumption, physical inactivity, and asbestos exposure, and there is a high baseline rate of cancer occurrence. As a result, small population low-dose radiation studies have lacked sufficient statistical power to draw strong conclusions, and current dosimetric models are partially reliant on extrapolation from high-dose data.

The International Nuclear Workers Study (INWORKS) and the Million Person Study (MPS) aim to address this gap by studying large groups of radiation workers. Both studies are ongoing, though so far published results support a linear association between some types of cancer risk and cumulative dose, even at low dose levels.

So far, there is no established non-zero threshold dose level below which risk is known to become zero, and this has driven the interpretation of ALARA that the NRC is aiming to move away from.

ALARAin practice: The NRC has proposed removing references to the ALARA principle from its regulations, moving instead to a graded approach for managing radiation doses below regulatory limits.

It would be a significant shift-ALARA has been a cornerstone of U.S. radiation safety for more than 50 years.

In the proposed rule change, the NRC said the issue at the regulatory level is more about implementation than the standards themselves. The agency argues that ALARA has been interpreted as requiring dose minimization "without a reasonable stopping point," which has led to an expectation that dose reduction methods be applied whenever available, regardless of the amount of reduction offered, compared with total dose.

Bahadori said he approves of ALARA at its core but agrees it has been "consistently misinterpreted."

Nicole Martinez, a professor at Clemson University and certified health physicist, acknowledged that there is a tension between wanting to enable workers to use their professional judgment and needing regulatory mechanisms that support workers in getting the protection they deserve.

As a member of the International Commission on Radiological Protection (ICRP), she said she tends to think about ALARA as the ICRP does, applying the principle as a consideration for optimization of dose limits decisions rather than at the application step.

"I see ALARA as that sweet spot that's supposed to be able to account for prevailing circumstances," she said.

In the proposed rule, the NRC said that retiring the ALARA terminology supports a return to its original intent, "utilizing selected aspects of the linear dose response model in favor of a strict adherence to the LNT model."

Replacing ALARA: The NRC has proposed moving to a graded approach to dose management, establishing radiation protection actions that are triggered at particular levels of prospective or actual exposure to provide "reasonable assurance" that regulatory dose limits aren't exceeded.

The agency describes a "go/no-go" regulatory approach that "would deemphasize radiation protection precautions in lieu of verifications that a licensee is maintaining dose values below the applicable dose limit."

Despite her appreciation of ALARA, Martinez said she's in favor of changing the regulatory language to a graded approach.

"I like reasonable assurance of adequate protection," she said. "But I also understand that just saying something is adequate doesn't necessarily make it adequate."

The NRC already has an established annual dose limit, requiring that radiation exposure is monitored and that a radiation worker's total effective dose equivalent for the whole body does not exceed 5 rem in a year.

The current rule requires an exemption request to exceed this limit, such as during maintenance periods when exposure may be higher. One of the proposed changes adds some flexibility to this policy, allowing doses of up to twice the annual limit as long as a five-year average remains below the annual limit.

In the rule change proposal, the agency said it is still developing an explanation of what will be considered an acceptable approach but gave as an example "expected doses of 100 mrem/year, 500 mrem/year, and 5 rem/year," as a set of thresholds where corresponding actions such as radiation training, individual dose monitoring, and use of a dose limit extension, respectively, may be required.

Martinez said she thinks 500 mrem/year as the cutoff for receiving an individual dosimeter is too high and that 100 mrem/year would be more appropriate. She also noted that the NRC does not establish a lifetime limit, claiming it's not practical to track.

"One hundred millisieverts [10 rem] [lifetime risk] is cited up, down, left, right as where our uncertainties really get high in the dose response models, but there are studies that show that there is attributable risk lower than that," she said. "The extension of the dose limits over 50 mSv in a year doesn't necessarily concern me on its own, but . . . I don't see how you can have an extension of the annual dose limit without considering lifetime exposure and be consistent with what we know about radiogenic risk."

Under the proposed rule change, licensees would be permitted to conduct cost-benefit analyses to determine if they should implement a radiation protection measure or allow the dose to accrue, as long the choice complies with applicable dose limits.

More than radiation workers: ALARA currently extends across all radiation management contexts, and the NRC has also proposed a graded approach to public dose limits and effluents.

The NRC's public dose limit is 100 mrem/year, 2 percent of the limit for radiation workers. It applies in contexts such as medical waiting rooms, where members of the public may be exposed to low levels of radiation. Under the graded approach, licensees would not be required to analyze projected public doses less than 25 mrem/year, and they would be permitted to conduct cost-benefit analyses when considering radiation management options for doses greater than 25 mrem/year.

For effluents, the NRC has proposed increasing its limit from 10 mrem/year to 25 mrem/year, which it said changes excess fatal cancer risk over 70 years from 0.04 cases per 100 people to 0.09 cases per 100 people.

Regardless of the effluent level measured, licensees will only be required to collect and retain effluent data on an annual basis. If a licensee measures above 25 mrem/year, it would be required to "evaluate and consider implementing cost-justified corrective actions to restore effluent levels to below the 25 mrem/year constraint."

If the licensee's cost analysis determines that corrective actions are not justified, the licensee is given the option to propose a new constraint for itself above 25 mrem/year, as long as it's lower than 100 mrem/year. New facilities would also be allowed to propose an effluent constraint level in that range with cost justification.

"I am supportive of establishing [a public dose optimization] threshold, but I think the appropriate dose value should be determined only after extensive stakeholder engagement," Bahadori said. "The very low effluent ALARA objectives and emissions limits currently in the regulations make it very straightforward to rebut claims that nuclear power plants cause measurable adverse health effects for the general public. Even small increased doses to the public make it more difficult to defend against such claims."

Justifications: Part of the NRC's justification for the changes is a cost-benefit analysis it conducted that found the changes would save the nuclear industry, Agreement States, and the NRC $9.53 million, $244,000, and $704,000 per year, respectively.

Bahadori is said these savings are "quite small" relative to annual operating costs and could easily become a wash.

"If doses to radiation workers increase even slightly, or if the workers' perception of their risk changes, workers might demand greater compensation, potentially offsetting these cost savings," he said.

Does this achieve the stated goals? The NRC cites addressing aspects of the low-dose radiation standards that are susceptible to inconsistent reinforcement or unnecessary conservatism as a key goal.

"I think the NRC did a good job of responding to the president's executive order," said Bahadori. "There are, however, some fairly definitive statements in the proposed rule regarding conservatism built into the LNT model that do not fully align with current ICRP and NCRP [National Council on Radiation Protection and Measurements] recommendations."

Martinez said her primary worry is unintended consequences that the language in the proposed rule change may not cover, such as employers offering rewards for workers receiving a higher dose, which muddies informed and voluntary consent.

"I worry about workers maybe having access to less information, maybe having less autonomy when it comes to exposures," said Martinez. "The rule also argues that there's a problem of subjectivity. I don't necessarily disagree with that, but I'm not convinced that some of the proposed changes aren't just a different type of subjectivity."

Recent history: ALARA was introduced in 1971, and the last major revision to 10 CFR Part 20 was in 1991, where the most significant change was adopting the concept of "effective dose," which provides a combined assessment of internal and external radiation exposures and relies on the LNT model.

In 2015, the NRC received petitions for rulemaking to discontinue use of the LNT model as the primary basis for its standards, including removing ALARA. Some of the petitions' arguments overlapped with those in EO 14300, claiming that the NRC's implementation of ALARA has led to excessive costs.

The NRC denied the petitions, arguing that "compliance with the ALARA requirement is based on whether the licensee has incorporated reasonable measures to track and, if necessary, to reduce exposures-not whether exposures and doses represent an absolute minimum or whether the licensee has used all possible methods to reduce exposures."

According to the new proposed rule change, the agency said it maintains that position while also saying that its lack of clarity, excessive subjectivity, and inconsistent enforcement of ALARA justifies removing use of the principle.

In a recent article, George Joslin, Arden Rowell, Seyed Reihani, and Zahra Mohaghegh outline the shifting legal framework around nuclear regulation since the passing of the 2024 Advanced Nuclear for Clean Energy (ADVANCE) Act and how it is driving changes to how radiation dose is considered.

According to the article, what has changed "is not the existence of uncertainty in radiation science but the institutional requirement that such uncertainty be addressed explicitly within a legal framework that demands justification rather than reliance on judicial deference."

Voice your opinion: The NRC will accept public comments on the proposed changes for 45 days following publication in the Federal Register, which has yet to occur. It will hold a public meeting during the comment period.