What are the odds of being harmed by a nuclear power plant accident?
Evidence quality 5.0/5
Eight-dimension review score against the quality rubric . Each dimension scored 1–5.
- D1 Source grounding
- 5/5
- D2 Source authority
- 5/5
- D3 Arithmetic
- 5/5
- D4 Uncertainty
- 5/5
- D5 Scope
- 5/5
- D6 Prose
- 5/5
- D7 Perception honesty
- 5/5
- D8 Caveat completeness
- 5/5
Lifetime probability · lifetime, US adult
1 in 833,333
0.0001% lifetime chance
Most people overestimate this.
range 1 in 10,000,000 to 1 in 125,000
● your factors — click this risk ▾ to reveal
≈ As likely as
Perceived
Nuclear power occupies a singular position in public risk perception. Gallup polling consistently finds that a majority of Americans oppose building new nuclear plants, and the word "radiation" triggers dread disproportionate to the dose involved. The fear is anchored to three events — Three Mile Island, Chernobyl, and Fukushima — each of which produced wall-to-wall media coverage and lasting cultural imprints. No rigorous survey isolates the perceived annual probability of harm from a nuclear accident, but intuitive estimates tend to land orders of magnitude above the epidemiological record.
Rough estimate: 42.7% of US adults report being afraid or very afraid of a nuclear accident/meltdown (Chapman Survey 2024)
Actual
~3 major civilian accidents in ~18,500 reactor-years of operation
Global civilian nuclear fleet, 1954–2024
Show derivation
Three major civilian nuclear accidents (TMI 1979, Chernobyl 1986, Fukushima 2011) have occurred in roughly 18,500 cumulative reactor-years of global operation, giving a per-reactor-year major-accident frequency of ~1.6 × 10⁻⁴. However, only Chernobyl produced significant off-site radiation casualties. UNSCEAR and WHO attribute ~30 acute radiation deaths and an estimated 4,000–16,000 excess cancer deaths over decades to Chernobyl; Fukushima produced 1 confirmed radiation fatality among workers. For a US adult living within the 50-mile Emergency Planning Zone of one of ~60 US reactor sites, the NRC's probabilistic risk assessment estimates core-damage frequency at ~2.5 × 10⁻⁵ per reactor-year for the current fleet, with large early release fraction roughly 10× lower. Compounding the individual annual fatality risk (~2 × 10⁻⁸) over 59 adult years yields ~1.2 × 10⁻⁶, or roughly 1 in 830,000. The figure is highly sensitive to reactor design generation and regulatory regime.
Caveats: The normalized figure is driven almost entirely by the Chernobyl precedent and b…
The normalized figure is driven almost entirely by the Chernobyl precedent and by US NRC probabilistic risk assessments for the current fleet. It does not capture the risk profile of RBMK-type reactors (the Chernobyl design), which had no containment structure and a positive void coefficient — features absent from every Western and modern reactor design. The 2,200 evacuation-related deaths at Fukushima, often attributed to "the nuclear accident," were caused by the displacement itself, not radiation — a distinction the headline number does not make. Lifetime risk for someone not living near any reactor is effectively zero. The uncertainty band spans nearly two orders of magnitude because it must accommodate both the possibility that Chernobyl was a non-repeatable design flaw and the possibility that undiscovered common-cause failure modes exist in the current fleet.
Regional breakdown
The headline figure averages across very different populations. Here’s how the probability varies by geography or context:
| Region / context | Lifetime probability | Notes |
|---|---|---|
| US adult within 50-mile EPZ | 1 in 833,333 |
Based on NRC fleet-average CDF and PRA individual fatality risk |
| US adult not near a reactor | 1 in 100,000,000 |
Negligible; only long-range fallout from hypothetical extreme event |
| Global average (near any reactor) | 1 in 500,000 |
Higher than US due to inclusion of older reactor designs worldwide |
Risks at similar odds
Other risks with roughly the same likelihood — useful for calibration.
AMOC collapse
What are the odds of the AMOC experiencing an abrupt collapse before the end of your lifetime?
Drought famine death
What are the odds of dying from drought-induced famine or water scarcity?
Recently viewed on this device
Stored locally — clear anytime.
Pick challenger
Three civilian nuclear power accidents dominate the historical record: Three Mile Island (1979), Chernobyl (1986), and Fukushima Daiichi (2011). Of these, only Chernobyl produced large-scale radiation casualties — roughly 30 acute deaths among emergency workers and an estimated 4,000 to 9,000 excess cancer deaths over subsequent decades, depending on which exposed population you include. Fukushima produced one confirmed radiation fatality among plant workers. TMI produced none. Across approximately 18,500 cumulative reactor-years of global civilian operation, the per-reactor-year frequency of a major accident is about 1.6 × 10⁻⁴, but the per-individual annual fatality risk is far smaller: the NRC’s probabilistic risk assessment puts it at roughly 2 × 10⁻⁸ for someone living within the 50-mile Emergency Planning Zone of a US reactor, yielding a lifetime figure near 1 in 830,000.
The gap between perceived and actual risk is among the largest on this site. Nuclear accidents activate what risk psychologists call “dread risk” — involuntary, unfamiliar, potentially catastrophic, and inequitably distributed. The availability heuristic does the rest: Chernobyl and Fukushima are vivid, emotionally charged reference points that make the risk feel present even though the engineering and regulatory context of those events is largely non-transferable to the current US fleet. The RBMK reactor at Chernobyl had no containment building and a positive void coefficient that amplified the very feedback loop operators were trying to suppress. No Western commercial reactor shares either feature.
The number is most misleading for people who do not live near a reactor — their risk is effectively zero. It is also misleading in the other direction for the roughly 2,200 people who died from Fukushima evacuation stress: they are often counted as “nuclear accident deaths,” but their cause of death was displacement, not radiation. Modern Gen III+ designs (AP1000, EPR) incorporate passive safety systems that reduce core-damage frequency by roughly an order of magnitude below the fleet average, pushing individual fatality risk into the range of asteroid impacts. The uncertainty band remains wide because the sample size of major accidents is three, and because the linear no-threshold dose-response model used to project long-term cancer deaths at low doses is itself contested.
Claim ledger
Every number below is what each source reported, with the verbatim quote we relied on and how we arrived at our figure. Click any link to verify directly.
-
[1] United Nations Scientific Committee on the Effects of Atomic Radiation — UNSCEAR 2008 Report Vol. II: Sources and Effects of Ionizing Radiation — Annex D: Health effects due to radiation from the Chernobyl accident
UNSCEAR 2008 Report Vol. II: Sources and Effects of Ionizing Radiation — Annex D: Health effects due to radiation from the Chernobyl accident- Statistic
~30 acute radiation deaths among emergency workers; up to 4,000 eventual excess cancer deaths in most exposed populations- Excerpt
“"Among the 134 emergency workers who received high doses of radiation and suffered acute radiation syndrome, 28 died in 1986 and two more in the following years. … Among the most exposed populations, an estimated 4,000 additional cancer deaths could occur over the lifetime of the approximately 600,000 persons who received the highest doses." ”
- Source data from
- 2008-12-01
- Accessed
- 2026-04-18 · archived copy
- Calculation
- UNSCEAR's 2008 Chernobyl annex provides the most widely cited authoritative casualty count. The 4,000-death projection applies to the highest-dose cohorts (liquidators, evacuees, residents of strict-control zones). Broader projections covering all of Europe range up to 16,000 (WHO 2006) or higher (Greenpeace), but UNSCEAR notes these are within the statistical noise of baseline cancer incidence. For the normalized estimate, the Chernobyl data informs the severity term but the frequency term comes from the global reactor-year accident rate and US NRC probabilistic risk assessment.
- Independence
- UNSCEAR is a UN General Assembly body independent of the nuclear industry and national regulators. Its Chernobyl assessment draws on separate dosimetric and epidemiological studies from Belarus, Russia, and Ukraine.
-
[2] US Nuclear Regulatory Commission — Probabilistic Risk Assessment (PRA)
Probabilistic Risk Assessment (PRA)- Statistic
Core-damage frequency for US fleet averages ~2.5 × 10⁻⁵ per reactor-year; large early release frequency ~10× lower- Excerpt
“"PRA is a systematic methodology to evaluate risks associated with a complex engineered technology. … The NRC uses PRA results to focus regulatory attention on design and operational issues that pose the greatest risk to public health and safety." ”
- Source data from
- 2024-06-15
- Accessed
- 2026-04-18 · archived copy
- Calculation
- NRC publishes fleet-average core-damage frequency (CDF) estimates derived from plant-specific PRAs. Current fleet CDF ≈ 2.5 × 10⁻⁵/reactor-year. The conditional probability of a large early release given core damage (LERF/CDF) is roughly 0.1, and only a fraction of LER events would produce off-site fatalities at the individual level. Conservative estimate of annual individual fatality risk for a person within the 50-mile EPZ: ~2 × 10⁻⁸. Over 59 adult years: 1 − (1 − 2 × 10⁻⁸)⁵⁹ ≈ 1.2 × 10⁻⁶.
- Independence
- NRC probabilistic risk assessments are regulatory analyses using plant-specific fault and event trees, independent of UNSCEAR's epidemiological dose-response approach.
-
[3] World Health Organization — Health effects of the Chernobyl accident and special health care programmes
Health effects of the Chernobyl accident and special health care programmes- Statistic
Up to 9,000 excess cancer deaths projected among highest-exposure populations in Belarus, Russia, and Ukraine- Excerpt
“"This report concludes that up to 9,000 excess cancer deaths may occur among the approximately 6.8 million most exposed people, although this number may be an overestimate because of the methodology used." ”
- Source data from
- 2006-04-20
- Accessed
- 2026-04-18
- Calculation
- WHO's 2006 Chernobyl health assessment extends the UNSCEAR analysis to broader exposed populations (6.8 million vs 600,000), yielding a higher central estimate of ~9,000 excess deaths. The report notes this may overestimate due to application of linear no-threshold (LNT) dose-response at very low doses. This figure contextualizes the severity of the single worst civilian nuclear accident in history, used here to bound the upper end of the uncertainty range.
- Independence
- WHO assessment was conducted by an international expert group separate from UNSCEAR, though both reference overlapping dosimetric data from the Chernobyl Forum.
-
[4] World Health Organization — Health risk assessment from the nuclear accident after the 2011 Great East Japan earthquake and tsunami
Health risk assessment from the nuclear accident after the 2011 Great East Japan earthquake and tsunami- Statistic
Estimated lifetime excess cancer risks in most affected areas of Fukushima are small and below detectable levels- Excerpt
“"For the general population inside and outside of Japan, the predicted risks are low and no observable increases in cancer rates above baseline rates are anticipated." ”
- Source data from
- 2013-02-28
- Accessed
- 2026-04-18 · archived copy
- Calculation
- WHO's 2013 Fukushima health risk assessment concluded that radiation doses to the general population were low enough that excess cancer incidence would be statistically undetectable. One worker death was attributed to radiation-induced lung cancer in 2018. Approximately 2,200 deaths were attributed to evacuation stress and disruption rather than radiation exposure. This confirms that modern containment and evacuation protocols, even when severely tested, produce radiation harm far below the Chernobyl precedent.
- Independence
- Independent WHO assessment using dose reconstructions by UNSCEAR 2013 and Japanese government monitoring data.