TL;DR
Get oils, diffusers and self-care delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
A commentary in The Ecomodernist says a proximity-based method used in several studies linking nuclear power plants with cancer can produce striking associations for unrelated landmarks, including Costco and colleges. The authors’ demonstration challenges how the results should be interpreted; it does not establish that those places cause cancer or independently resolve whether nearby residents face any risk.
A commentary in The Ecomodernist argues that a statistical method used in recent studies of cancer rates near nuclear power plants can also produce large associations for unrelated places, including Costco warehouses and private colleges. The authors say this exposes limits in drawing conclusions from distance-based comparisons; the examples do not show that those landmarks cause cancer, and the commentary does not settle whether nearby communities face any radiation-related risk.
The commentary describes studies by researchers led by Yazan Alwadi at Harvard’s T.H. Chan School of Public Health. It says a December 2025 paper in Environmental Health examined cancer incidence in Massachusetts, followed by a nationwide study of cancer mortality in Nature Communications. The report also cites later studies of selected cancers and total mortality. The supplied material does not include the underlying papers’ full methods or responses from their authors.
As described in the commentary, the studies assign locations a proximity score based on distance from nuclear plants, using radii of up to 120 kilometers in Massachusetts and 200 kilometers in national analyses. Nearby ZIP codes or counties receive greater weight, and locations within range of multiple plants can accumulate higher scores. Researchers then compare those scores with health outcomes while adjusting for other variables, and calculate estimates of relative risk and deaths attributable to proximity.
The critique says its authors reproduced the approach and applied it to other landmarks. It reports estimates such as 120,687 annual cancer deaths associated with proximity to Costco and 83,782 associated with private colleges, as well as a comparison between locations near Major League Baseball and Major League Soccer fields. These are figures presented by the commentary as outputs of its analysis, not verified counts of deaths caused by those places. The report argues that the implausible-looking results illustrate why statistical association alone cannot establish causation.
Why Proximity Estimates Need Caution
The dispute matters because studies associating location with disease can shape public understanding and policy, even when they do not demonstrate a cause. A statistical estimate of deaths “attributable” to a proximity score may sound like a count of confirmed deaths caused by a facility, but that interpretation depends on whether the exposure measure and model support it. The commentary’s examples challenge readers and decision-makers to distinguish modeled associations from direct evidence of exposure and harm.
At the same time, the critique’s landmark comparisons are not evidence that the nuclear-plant studies are necessarily wrong in every respect. Applying a method to other places can reveal sensitivity to model choices or confounding, but it does not by itself identify which assumptions failed or determine the true health effect. Resolving that requires scrutiny of the original data, methods, and plausible exposure pathways.
Products Worth Considering
Levoit Eleva 300X HEPA Air Purifier for Home and Bedroom
OMRON VIVA 360+ Body Composition Monitor and Smart Scale
OMRON Complete 2-in-1 Upper Arm Blood Pressure Monitor with EKG
As an affiliate, we earn on qualifying purchases.
How the Plant Studies Measured Proximity
According to the commentary, the cited studies did not measure each resident’s radiation dose. Instead, they used distance from a plant as a population-level proxy and compared location-level scores with cancer incidence or mortality. The authors of the studies, as characterized in the report, have described their work as ecological and said it identifies associations rather than proving causality.
The commentary contrasts this approach with monitoring of nuclear workers, asserting that most closely monitored U.S. nuclear workers receive no measurable annual dose and that average doses among exposed workers are a small fraction of occupational limits. Those claims are made by the commentary; they do not constitute measurements of radiation exposure among residents near plants. Worker dose data and residents’ exposures are not interchangeable without evidence about the specific pathways and conditions involved.
The report says its authors received eight lines of code from the researchers and answers to some questions about covariates, but those materials did not reproduce the results. It says the rest of their replication relied on reconstructing the analysis. That account makes the availability of complete methods, data, and independent replication relevant to evaluating the disagreement.
“Proximity is not exposure.”
— The Ecomodernist commentary
What the Replication Can Establish
The supplied material does not provide the original studies’ full results, datasets, model specifications, or a response from Alwadi and co-authors. It is therefore not possible from this report alone to independently assess whether the commentary reproduced the analyses accurately or whether the landmark comparisons used equivalent assumptions and data.
It also remains unclear what, if any, radiation exposure residents near the plants experienced and whether exposure could explain the reported patterns. The commentary questions that link but does not present direct measurements of residents’ doses. Its Costco and college figures should be read as model-generated associations reported by the critique, not as established cancer-death counts or evidence that shopping or attending college increases cancer risk.
Methods and Responses to Watch
The next step for readers assessing the dispute is to compare the commentary’s replication with the original papers’ methods, data, and statistical results. Full code and data access, clear explanations of covariate choices, and independent replication could help determine whether the reported associations are robust or sensitive to modeling decisions.
The source material does not identify a scheduled response, correction, or follow-up study. Until further evidence is available, the careful conclusion is limited: the commentary argues that proximity-based models can generate striking associations, while the causal significance of the nuclear-plant findings and the reported landmark comparisons remains unresolved.
Key Questions
Did the report show that Costco causes cancer?
No. The Costco figure is presented as an association generated by the commentary’s application of a statistical method. It is not evidence that Costco causes cancer or that the reported number represents deaths confirmed to result from shopping there.
What did the nuclear-plant studies measure?
As described by The Ecomodernist, the studies assigned ZIP codes or counties a score based on their distance from nuclear plants and compared that score with cancer or mortality outcomes. The commentary says this was a proximity measure, not a direct measurement of residents’ radiation exposure.
Do the studies prove that nuclear plants cause cancer?
The commentary says the studies report population-level associations and do not prove causality. The supplied source does not establish whether nearby residents faced a measurable cancer risk; that question requires evidence beyond proximity-based associations.
What is needed to evaluate the disagreement?
Readers would need to examine the original studies and the critique, including their data, code, exposure assumptions, statistical models, and replication results. Responses from the study authors and independent analyses could clarify whether the findings are robust.
Source: hn
