The Linear No-Threshold (LNT) Model

Introduction

The linear no-threshold (LNT) model is the assumption that radiation-induced cancer and heritable risk rise in direct proportion to dose, with no threshold below which risk vanishes — and it is the scientific backbone of essentially every dose limit, ALARA program, and radiation safety decision in medicine. It is also the most debated model in radiation protection, because the very low-dose region where most medical and occupational exposures live is exactly where the human evidence runs out. ^{1, 2}

Understanding LNT matters for anyone who runs a radiation safety program. Occupational dose limits, public dose limits, the entire logic of "keep it As Low As Reasonably Achievable," and the way we counsel patients and staff about imaging risk all rest on this one assumption. If you do not know what LNT does and does not claim, it is easy to either overstate the danger of a chest CT or dismiss the rationale for dose limits — both mistakes a Radiation Safety Officer should avoid. ³

This guide explains what the LNT model is, the evidence for and against it, how ICRP and NCRP treat it today, the arithmetic of risk coefficients, and how it drives real regulatory limits. DRPS supports facilities in applying these principles through its Radiation Safety Officer consulting, radiation safety training, and medical physics consulting across Florida, Maryland, Virginia, Washington DC, California, Nevada, and beyond.

Topic Explanation

What LNT actually says

LNT applies to stochastic effects — effects whose probability, not severity, increases with dose. These are principally cancer and heritable effects. The model makes two linked claims: ¹

1. Linearity — the excess probability of a stochastic effect is proportional to dose. Double the dose, double the excess risk.
2. No threshold — the proportional line passes through the origin, so there is no dose greater than zero that carries exactly zero excess risk.

LNT does not apply to deterministic effects (tissue reactions such as cataracts, skin injury, or organ damage), which do have practical thresholds and increase in severity above them. Distinguishing these two categories is fundamental; see stochastic vs deterministic radiation effects.

Critically, LNT is a radiation protection assumption, not a proven biological law. It is a deliberate, cautious choice about how to manage risk in the face of uncertainty — a distinction that changes how you should communicate it.

Where LNT comes from

The model is anchored in observed data at moderate-to-high doses and then extrapolated downward. The dominant data source is the Life Span Study (LSS) of Hiroshima and Nagasaki atomic bomb survivors, followed for decades by the Radiation Effects Research Foundation. The LSS shows a clear, roughly linear dose-response for solid cancer, and — importantly — a statistically significant excess even in the low-dose subgroups, with formal analysis finding no threshold. ^{4, 5}

Because directly measuring the tiny risks at the doses typical of medical imaging (a few mSv) would require impossibly large study populations, protection bodies extrapolate the measured high-dose slope down to zero, usually with an adjustment for dose rate (discussed below). This extrapolation — reasonable, cautious, but not directly verified — is the heart of both LNT's utility and its controversy. ^{2, 6}

Key Technical Principles

The nominal risk coefficient

ICRP condenses the LNT slope into a single detriment-adjusted nominal risk coefficient — the estimated overall radiation detriment (weighted for cancer incidence, lethality, quality of life, and heritable effects) per unit effective dose. In Publication 103: ¹

The whole-population value combines about $5.5\times 10^{-2}{\text{Sv}}^{-1}$ for cancer with about $0.2\times 10^{-2}{\text{Sv}}^{-1}$ for heritable effects; the adult value is lower because a working-age population has less remaining lifetime for cancers to express. Applied linearly, these coefficients turn a dose into an estimated lifetime detriment.

Worked example

Suppose an interventional cardiologist accrues an occupational effective dose of 10 mSv (0.010 Sv) in a year. Under LNT with the adult-worker coefficient:

That is an estimated added lifetime detriment of about 0.042%, or roughly 4 in 10,000. For a patient receiving a 7 mSv FDG PET/CT, using the whole-population coefficient:

roughly 4 in 10,000 as well. These are small numbers relative to the baseline lifetime cancer risk (on the order of 40% in the general population), and they carry real uncertainty — but LNT lets us estimate them consistently, compare procedures, and justify keeping dose low. For the dose quantities behind these calculations, see radiation dose quantities and units.

DDREF: adjusting for dose rate

Cells repair some radiation damage, and biology suggests low doses delivered slowly are somewhat less harmful per unit dose than the acute, high-dose exposures of the LSS. To account for this, ICRP applies a dose and dose-rate effectiveness factor (DDREF) of 2 when extrapolating the acute high-dose slope down to the low-dose, low-dose-rate conditions of most protection situations. In other words, the nominal coefficients above already include a factor-of-2 reduction from the raw high-dose slope. ¹ The exact value of DDREF is itself debated, which is one source of the uncertainty band around low-dose risk estimates.

Competing dose-response models

LNT is one of several proposed shapes for the low-dose response. The debate is genuine because the data cannot decisively separate them at low dose.

The most authoritative recent review, NCRP Commentary No. 27 (2018), evaluated 29 epidemiologic studies and concluded that the preponderance of data continues to support LNT (with a possible DDREF) for radiation protection, and that no alternative appears more pragmatic or prudent. ^{2, 3} BEIR VII Phase 2, UNSCEAR, and ICRP reach compatible conclusions. ^{1, 6, 7}

Clinical Impact

It is the reason ALARA exists

If there were a safe threshold, radiation protection would simply be "stay below the threshold." Because LNT assumes no threshold, there is no automatically safe dose, so the operating principle becomes keep exposures As Low As Reasonably Achievable — below the limits, not merely at them. Every shielding decision, every collimation habit, every "do we really need this scan?" conversation is LNT in action. See building an ALARA program.

It shapes how we talk about imaging risk

LNT is easy to misuse in patient and staff communication. The correct framing is that the risk from a single well-justified imaging exam is small, uncertain, and worth accepting when the exam answers a real clinical question — not that the exam is "safe" (which LNT will not let you claim) and not that it is "dangerous" (which the magnitude does not support). A good RSO teaches staff to hold both ideas at once: minimize dose because there is no threshold, and keep the residual risk in honest perspective.

The collective-dose trap

A recurring error is to take a very small individual dose, multiply it across a large exposed population, and announce a projected number of cancer deaths. ICRP explicitly warns against this: collective dose is a tool for optimizing protection, not for projecting deaths from trivial individual exposures, precisely because the low-dose risk estimates are too uncertain to support that arithmetic. ¹ An RSO should recognize and push back on this misuse when it appears in the media or in internal risk discussions.

Practical Optimization Tips

1. Teach the model, not just the limits

Staff who understand why there is no safe threshold follow ALARA more consistently than staff who only memorize dose limits. Build the LNT rationale into radiation safety training.

2. Use risk coefficients honestly

When you estimate a risk for a patient or worker, state the number, the model, and the uncertainty together. "About 4 in 10,000 additional lifetime detriment, under a cautious model that likely overstates low-dose risk" is honest; a bare number is not.

3. Never present LNT-derived population death counts as fact

Resist and correct the collective-dose-times-population calculation. Frame collective dose as an optimization metric only. ¹

4. Anchor dose limits to their basis

When workers ask why the occupational limit is what it is, connect it to the risk coefficient: the limits were set so that occupational radiation risk is comparable to risks accepted in other safe industries. See NRC occupational dose limits (Part 20).

5. Keep current with the science

LNT is periodically re-examined. Track the major reviews (ICRP, NCRP, BEIR, UNSCEAR) so your program reflects the current consensus rather than a decades-old summary. ^{2, 6, 7}

Common pitfalls to avoid

• Calling any dose "safe." LNT does not permit it; say "low risk," not "no risk."
• Claiming low-dose harm is proven. It is assumed, not demonstrated; overstating it erodes credibility.
• Multiplying tiny doses over big populations to predict deaths.
• Applying LNT to deterministic effects. Tissue reactions have thresholds and are a different regime.
• Ignoring DDREF. The nominal coefficients already include a dose-rate adjustment.

Regulatory Considerations

Radiation protection regulation in the United States is built on the LNT model, so an RSO must understand it to explain and defend the program's dose limits and ALARA requirements. Key frameworks:

• ICRP Publication 103 — the 2007 Recommendations that define detriment, the nominal risk coefficients, DDREF, and the formal adoption of LNT for protection. ¹
• NCRP Commentary No. 27 (2018) — the current U.S. expert review concluding that recent epidemiology continues to support LNT for protection. ^{2, 3}
• NCRP Report No. 136 — the earlier detailed evaluation of the LNT dose-response model. ⁸
• BEIR VII Phase 2 (2006) — the U.S. National Academies assessment of health risks from low-level ionizing radiation, endorsing a linear no-threshold risk model for solid cancer. ⁶
• UNSCEAR 2006 Report — the United Nations scientific assessment of low-dose radiation effects. ⁷
• 10 CFR Part 20 — the NRC Standards for Protection Against Radiation, whose occupational limit (50 mSv/year total effective dose equivalent) and public limit (1 mSv/year) are risk-based caps derived from LNT-based detriment, above which ALARA still applies. ⁹

The occupational and public dose limits in 10 CFR Part 20 (and equivalent Agreement State rules) exist because LNT implies proportional risk with no threshold: the limits cap the accepted risk, and ALARA drives exposures below them. Of the states DRPS serves, Florida, Maryland, Virginia, California, Nevada, Pennsylvania, New York, and New Jersey are NRC Agreement States administering equivalent radiation-protection regulations, while Washington DC and Delaware are regulated directly by the NRC. For the public-dose side of the framework, see public dose limits (Part 20), and for how dose is weighted across tissues, see effective dose and tissue weighting factors.

Frequently Asked Questions (FAQs)

What is the linear no-threshold (LNT) model?

The linear no-threshold model is the assumption used in radiation protection that the probability of stochastic effects — mainly cancer and heritable effects — rises in direct, straight-line proportion to radiation dose, with no threshold dose below which the risk is zero. Under LNT, any increment of dose carries a proportional increment of risk, so there is no dose that is provably free of risk.

Why is the LNT model used if low-dose risks are uncertain?

LNT is used because it is prudent and practical for setting protection standards. Human epidemiologic data are reliable above roughly 100 mSv but cannot resolve the very small risks at lower doses. National and international committees have concluded that, given this uncertainty, no alternative dose-response relationship is more pragmatic or prudent for protection purposes than LNT, so it remains the working basis for dose limits and ALARA.

What is the evidence for LNT?

The strongest evidence comes from the Life Span Study of atomic bomb survivors, which shows a linear dose-response for solid cancer with a statistically significant excess risk at low doses and no detectable threshold. Additional support comes from pooled occupational, medical, and environmental low-dose studies reviewed by NCRP, most of which are consistent with continued use of LNT for protection.

Does LNT mean low doses of radiation are proven to cause cancer?

No. LNT is a protection assumption, not a claim that harm at very low doses has been directly measured. At low doses the risk is small and uncertain, and it may never be possible to prove or disprove the linear assumption by epidemiology alone. LNT deliberately errs toward caution rather than asserting that any specific low dose has a demonstrated effect.

What is the nominal risk coefficient?

It is the estimate of overall radiation detriment per unit effective dose used by ICRP. In Publication 103, the detriment-adjusted nominal risk coefficient for stochastic effects is about 5.7 percent per sievert for the whole population and about 4.2 percent per sievert for adult workers. These coefficients, applied linearly, translate a dose into an estimated lifetime risk.

How does LNT relate to ALARA and dose limits?

Because LNT implies risk with no threshold, there is no dose that is automatically "safe," so exposures should be kept As Low As Reasonably Achievable below the regulatory limits. Dose limits such as those in 10 CFR Part 20 cap risk at an accepted level, while ALARA drives exposures lower still. Both concepts follow directly from the no-threshold assumption.

Is it valid to multiply tiny doses across a large population to predict cancers?

No. ICRP specifically cautions against computing hypothetical numbers of cancer deaths by multiplying very small individual doses across large populations (collective dose) over long times. Collective dose is a tool for optimizing protection, not for projecting deaths from trivial individual exposures, because the low-dose risk estimates are too uncertain for that use.

Key Takeaways

• LNT assumes stochastic risk is proportional to dose with no threshold. It applies to cancer and heritable effects, not to deterministic tissue reactions.
• It is a protection assumption, not a proven law. Human data are reliable above ~100 mSv; below that, risks are small and uncertain.
• The evidence base is the atomic bomb survivor Life Span Study, which shows a linear dose-response with no detectable threshold, plus supporting low-dose studies.
• ICRP's nominal risk coefficient is ~5.7 × 10⁻²/Sv (population) and ~4.2 × 10⁻²/Sv (workers), already including a DDREF of 2.
• LNT is the reason ALARA exists and the rationale behind 10 CFR Part 20 dose limits.
• Do not multiply tiny doses across large populations to predict deaths — ICRP forbids this use of collective dose.
• NCRP, ICRP, BEIR VII, and UNSCEAR continue to endorse LNT for radiation protection despite low-dose uncertainty.

Conclusion

The linear no-threshold model is the quiet foundation under almost everything a radiation safety program does. It is not a claim that a chest X-ray is dangerous, and it is not a claim that any dose is safe. It is a disciplined, cautious way to manage uncertainty: assume proportional risk down to zero, cap exposures with risk-based limits, and drive them lower with ALARA. The recent expert consensus — NCRP Commentary 27, BEIR VII, UNSCEAR, and ICRP — is that this remains the most prudent basis for protection, even as the low-dose region stays scientifically unsettled.

For an RSO, the practical task is twofold: apply the model correctly in dose limits and ALARA, and communicate it honestly — small, uncertain, worth minimizing, never "zero," never "proven harm." A program that understands the reasoning behind its limits protects people better than one that only follows the numbers.

How DRPS Can Help

Diagnostic Radiation Physics Services helps facilities build radiation safety programs grounded in the science behind the regulations. Our support includes Radiation Safety Officer consulting, ALARA program design and audits, dose-limit and risk-communication guidance, radiation safety training that teaches the LNT rationale, and medical physics consulting — all delivered by board-certified medical physicists.

DRPS supports facilities across our service locations, including Florida, Maryland, Virginia, Washington DC, California, Nevada, New York, Pennsylvania, New Jersey, and Delaware.

A radiation safety program is stronger when its people understand not just the dose limits, but the model those limits are built on.

Related Resources

• Stochastic vs deterministic radiation effects
• Effective dose and tissue weighting factors
• NRC occupational dose limits (Part 20)
• Public dose limits (Part 20)
• Building an ALARA program
• Radiation dose quantities and units
• Radiation Safety Officer consulting
• Radiation safety training

References

1. International Commission on Radiological Protection. ICRP Publication 103: The 2007 Recommendations of the International Commission on Radiological Protection. Annals of the ICRP. 2007;37(2-4). icrp.org
2. National Council on Radiation Protection and Measurements. NCRP Commentary No. 27: Implications of Recent Epidemiologic Studies for the Linear-Nonthreshold Model and Radiation Protection. 2018. ncrponline.org
3. Shore RE, Beck HL, Boice JD, et al. Recent epidemiologic studies and the linear no-threshold model for radiation protection — considerations regarding NCRP Commentary 27. Health Physics. 2019;116(2):235-246. doi:10.1097/HP.0000000000001015. PubMed
4. Ozasa K, Shimizu Y, Suyama A, et al. Studies of the mortality of atomic bomb survivors, Report 14, 1950-2003: an overview of cancer and noncancer diseases. Radiation Research. 2012;177(3):229-243. doi:10.1667/RR2629.1. PubMed
5. Preston DL, Ron E, Tokuoka S, et al. Solid cancer incidence in atomic bomb survivors: 1958-1998. Radiation Research. 2007;168(1):1-64. doi:10.1667/RR0763.1. PubMed
6. National Research Council (US) Committee to Assess Health Risks from Exposure to Low Levels of Ionizing Radiation. Health Risks from Exposure to Low Levels of Ionizing Radiation: BEIR VII Phase 2. Washington, DC: The National Academies Press; 2006. doi:10.17226/11340. nationalacademies.org
7. United Nations Scientific Committee on the Effects of Atomic Radiation. UNSCEAR 2006 Report: Effects of Ionizing Radiation, Volume I, Annex A — Epidemiological Studies of Radiation and Cancer. unscear.org
8. National Council on Radiation Protection and Measurements. NCRP Report No. 136: Evaluation of the Linear-Nonthreshold Dose-Response Model for Ionizing Radiation. 2001. ncrponline.org
9. U.S. Nuclear Regulatory Commission. 10 CFR Part 20: Standards for Protection Against Radiation. ecfr.gov
10. International Commission on Radiological Protection. ICRP Publication 99: Low-dose Extrapolation of Radiation-related Cancer Risk. Annals of the ICRP. 2005;35(4). icrp.org
11. U.S. Nuclear Regulatory Commission. Backgrounder on Biological Effects of Radiation. nrc.gov