Critique challenges method behind nuclear plant cancer studies
Breakthrough Journal authors say applying a proximity-based analysis to unrelated landmarks also produced positive cancer associations, raising questions about what the method can establish.
A Breakthrough Journal article challenges Harvard-led studies that reported associations between proximity to nuclear power plants and cancer outcomes. Its authors argue that the studies’ ecological design cannot show that nearby plants caused cancer.
The Harvard studies, described in the article as published from December 2025 onward, used distance-weighted scores for ZIP codes or counties within 120 kilometres of plants in Massachusetts and up to 200 kilometres in national analyses. The researchers examined cancer incidence, cancer mortality and other mortality, according to the critique.
The Breakthrough Journal authors say they replicated the national cancer-mortality analysis and applied its method to unrelated landmarks, including Costco, universities, state capitals and sports venues. They report positive associations across these comparisons and argue that the method may produce results without identifying a causal exposure.
The article says the replication drew on eight lines of code and responses to questions about covariates, with much of the work reconstructed from first principles and trial and error. Its findings and methods have not been independently assessed in the material available for this report.
Ecological studies compare patterns at a population level; they do not, on their own, establish causation. The critique argues that distance from a landmark is not evidence of exposure, and that the analysis does not identify a mechanism linking nuclear plants to cancer outcomes.
The article’s large “attributable death” estimates for landmarks are outputs of its analysis, not evidence that those landmarks caused the deaths. The supplied material does not establish the validity of those estimates or independently verify the replication.
