‘Robust and consistent’ signal: Cancer mortality rates higher near nuclear power plants
When a hypothesis is at odds with data, you don’t discard the data – you modify the hypothesis.
Medical data trump hypothetical estimates of “radiation doses” that are disconnected from reality, not measured nor even measurable.
Internal exposures to alpha emitters like plutonium and pure beta emitters like tritium and carbon-14 are notoriously difficult to measure, especially when it comes to pregnant women and their developing fetuses.Too often, medical data have been mistrusted or even discarded because the estimated radiation doses were “too low” to account for the harmful effects recorded
This happened in the aftermath of the TMI nuclear accident, for example, and following the German KiKK study that found significantly increased leukaemia in children under 5 within 5 kilometres of any one of Germany’s then-operating 17 nuclear power reactors.All 17 reactors are now shut down, as Germany has completely phased out of nuclear power.
——————————————————————————–
‘Robust and consistent’ signal: Cancer mortality rates higher near nuclear power plants
By Mark Leiser, Fact checked by Heather Bile, Healio, March 16, 2026
[from Hematology/Oncology News Today]
Key Takeaways
An analysis of every U.S. county showed higher cancer mortality rates in those located closer to nuclear power plants.
The findings cannot prove causality but warrant further investigation, researchers concluded.
U.S. counties located closer to nuclear power plants have higher cancer mortality rates than those located farther away, results of a national analysis showed.
The study — which accounted for environmental, socioeconomic and other factors — yielded results that remained consistent through multiple sensitivity analyses.
During the 19 year study period, researchers estimated that 115,586 cancer deaths nationwide could be attributed to nuclear power plant proximity.
Data derived from Alwadi A, et al. Nat Commun. 2026;doi:10.1038/s41467-026-69285-4.
In light of increased attention on nuclear power as a low-carbon energy alternative, more research into its potential effects on public health is warranted, according to Yazan Alwadi, PhD, postdoctoral research fellow in the department of environmental health at Harvard T.H. Chan School of Public Health.
“We want to be very clear that we cannot prove causality. However, the signal we observed is very robust and consistent, and it is surprising it has not been shown before,” Alwadi told Healio. “In my opinion, we have all the evidence we need to justify going to the next level of investigation.”
Impact of ‘normal operation’
A majority of studies that examined the effects of routine operation assessed cancer incidence or mortality in a specific region located near one or two plants. The limited setting reduces the statistical power to detect effects, he said.
Alwadi and colleagues launched their study after local public health officials in Plymouth County, Massachusetts — where the now-closed Pilgrim Nuclear Power Station is located — asked them to evaluate what they considered concerning cancer patterns in the region.
“Rather than focusing on a single county, we felt it was scientifically stronger to conduct a national analysis,” Alwadi said.
The researchers used U.S. Energy Information Administration records to identify the locations and operational dates of all nuclear power plants located within 200 km — about 124 miles — from the center of any U.S. county. They obtained county-level cancer mortality data from the CDC, focusing on the period between 2000 and 2018.
Alwadi and colleagues employed what they described as a “spatially resolved, inverse distance-weighted proximity metric.”
They used statistical modeling to calculate cumulative effects of multiple nearby nuclear power plants on people aged 35 years or older, controlling for potential confounders — such as BMI, smoking prevalence, household income and educational attainment — in each county.
A positive association
The results revealed a positive association between proximity to nuclear power plants and cancer mortality.
Investigators estimated 115,586 cancer deaths (95% CI, 56,964-173,326) during the 19-year study period — or approximately 6,400 per year across the country — could be attributed to nuclear power plant proximity.
For men and women in most age groups, results showed considerably higher relative risks when equivalent plant distance was 50 km or less, with risk curves beginning to plateau with greater distance.
Relative risk estimates were lowest among the 35-to-44 age group for both women and men, then began to increase with age.
Investigators estimated 115,586 cancer deaths (95% CI, 56,964-173,326) during the 19-year study period — or approximately 6,400 per year across the country — could be attributed to nuclear power plant proximity.
For men and women in most age groups, results showed considerably higher relative risks when equivalent plant distance was 50 km or less, with risk curves beginning to plateau with greater distance.
Relative risk estimates were lowest among the 35-to-44 age group for both women and men, then began to increase with age.
Among women, those aged 55 to 64 years exhibited the highest relative risk (RR = 1.19), with 2.1% (95% CI, 1.3%-2.9%) of cancer deaths in that age group attributable to nuclear power plant proximity.
Among men, those aged 65 to 74 years had the highest relative risk (RR = 1.2), with an estimated 2% (95% CI, 1.2%-2.7%) of cancer deaths in that age group attributable to nuclear power plant proximity.
Overall results showed the highest attributable cancer mortality burden among individuals aged 65 to 84 years. Researchers estimated 4,266 deaths (95% CI, 3,000-9,112) per year among those aged 65 or older to be attributable to proximity to nuclear power plants.
Among women, those aged 55 to 64 years exhibited the highest relative risk (RR = 1.19), with 2.1% (95% CI, 1.3%-2.9%) of cancer deaths in that age group attributable to nuclear power plant proximity.
Among men, those aged 65 to 74 years had the highest relative risk (RR = 1.2), with an estimated 2% (95% CI, 1.2%-2.7%) of cancer deaths in that age group attributable to nuclear power plant proximity.
Overall results showed the highest attributable cancer mortality burden among individuals aged 65 to 84 years. Researchers estimated 4,266 deaths (95% CI, 3,000-9,112) per year among those aged 65 or older to be attributable to proximity to nuclear power plants.
The associations between proximity and cancer mortality persisted in multiple sensitivity analyses, Alwadi said. In one, researchers adjusted the distance from nuclear power plants to county centers, changing by increments of 10 km until it reached a 100-km radius. In another, investigators varied the average proximity window across five intervals, ranging from 2 years to 20 years.
The consistency of the results demonstrate that they “are not driven by arbitrary choices in model variables or parameters,” the researchers wrote.
The investigators acknowledged study limitations.
The analysis assumed equal impact of all nuclear power plants rather than incorporating direct radiation measurements, and it assessed all malignancies combined even though radiation sensitivities and latency periods vary by cancer type.
Also, the standard formula investigators used to calculate attributable fraction assumes a causal relationship between the outcome and exposure without accounting for potential exposure misclassification or residual confounding.
‘We need to dig deeper’
The study is the first to the authors’ knowledge that uses a continuous proximity metric to examine nuclear power plant proximity and cancer mortality on a national level.
The use of 19 years of national cancer mortality data and a 10-year average nuclear power plant proximity window allowed for a “robust temporal assessment” of proximity’s long-term effects, the researchers wrote.
However, the findings have been the subject of some public criticism.
The Breakthrough Institute — a California-based research center that seeks to identify technological solutions to environmental challenges — published a post on its website challenging the accuracy of the paper, as well as another that Alwadi’s research group previously published that showed an association between residential proximity to nuclear power plants and elevated cancer incidence among people in Massachusetts.
The Breakthrough Institute — a California-based research center that seeks to identify technological solutions to environmental challenges — published a post on its website challenging the accuracy of the paper, as well as another that Alwadi’s research group previously published that showed an association between residential proximity to nuclear power plants and elevated cancer incidence among people in Massachusetts.
“The two papers make the fundamental mistake of confusing correlation with causation,” the online post reads.
The post authors point to the lack of a control group in the Massachusetts state-level analysis and use of “an improperly sampled group” in the national study. Distance from a nuclear plant is not a substitute measure of radiation dose, they argued, noting factors such as wind direction, shielding or monitored emissions had not been taken into account. Researchers also could not demonstrate that people who live nearby receive “any incremental dose beyond natural background radiation,” they added.
Consequently, the research is “fundamentally dangerous” and increases public health risks by “fueling efforts” to close existing nuclear plants and prevent new ones from coming online, the post authors wrote, suggesting this would compound the health risks associated with fossil-fueled electricity generation.
Alwadi said he is aware of the criticisms but believes many of them result from “lack of knowledge of statistics or epidemiology.”
Many of the concerns expressed in the online post already have been acknowledged by researchers in the manuscript as study limitations or addressed in sensitivity analyses performed to answer questions raised by peer reviewers prior to publication, Alwadi said. The methodology “has been put through the wringer and checked step by step,” he added.
“Anyone can write what they want on their own website,” Alwadi said. “If they have a legitimate criticism, they can submit it to the journal. If the editors determine it is valid, we would have to respond to it. We haven’t received anything like that.
“We have, however, received emails from so many people asking to collaborate with us or to investigate this more closely in specific regions,” Alwadi added. “People are very interested in this. They want to know if there is an effect. We want to know, too.”
Alwadi and colleagues are continuing to analyze additional datasets and perform cohort analyses. They have hypotheses that they hope will serve as the foundation for exposure pathway-specific analyses. Alwadi also emphasized the need for additional research into latency effects and impacts on risks for specific malignancies.
“The best data we get comes from randomized clinical trials, but that design is not applicable to the study of environmental exposures,” Alwadi said.
“Epidemiological studies progress in stages. If you find a signal, you keep going. We certainly did not want to see an effect, but we observed a systematic association that is robust to sensitivity analyses and observed across multiple datasets and geographic aggregations.
“We acknowledge that does not establish causality,” he added. “But what if you lived in a town and noticed that everybody who drank from a specific well got sick? If you didn’t know the exact mechanism, would you still drink from that well or would you investigate it? That’s all we’re saying. We need to dig deeper.”
nuclear-news 
Leave a comment