An invisible radioactive gas already known to cause lung cancer may also be associated with heart failure. A large U.S. study raises new questions about residential radon exposure, although it does not establish that radon causes heart disease.
Key Takeaways
- A study of 43,188 postmenopausal women found that living in areas with estimated indoor radon concentrations above 4 picocuries per liter was associated with a 19% higher adjusted hazard of hospitalization for heart failure compared with areas below 2 picocuries per liter.
- During an average follow-up of 16.2 years, researchers identified 3,446 first hospitalizations for definite or possible acute decompensated heart failure.
- The unadjusted cumulative incidence was 68 heart failure events per 1,000 women in the lowest radon category, compared with 101 per 1,000 in the highest.
- The association was more apparent among women who had ever smoked. Researchers also reported associations with heart failure involving reduced pumping function and with heart failure mortality.
- Radon exposure was estimated using residential addresses and geographic data, not measured inside individual homes. The findings do not prove causation or establish that radon mitigation prevents heart failure.
Introduction: Could Radon Affect More Than the Lungs?
A study of 43,188 postmenopausal women found that higher estimated residential radon exposure was associated with a 19% greater adjusted hazard of heart failure hospitalization. The observational findings do not establish causation.
Radon is an invisible, odorless radioactive gas that can accumulate inside homes. Its best-established health consequence is lung cancer, but researchers are increasingly investigating whether exposure might also affect the cardiovascular system.
A study published November 5, 2026, in JAMA Network Open found that postmenopausal women living in areas with higher estimated indoor radon concentrations had a greater hazard of hospitalization for heart failure during long-term follow-up.
The association persisted after researchers accounted for numerous factors, including age, smoking, blood pressure, body mass index, existing medical conditions, socioeconomic characteristics, and fine-particle air pollution.
The findings raise an important question: Could an environmental exposure already linked to lung cancer also be associated with heart failure?
The answer remains uncertain. Although the investigation was large and carefully conducted, it relied on geographic estimates of radon rather than measurements inside each participant’s home.
What the Study Examined
Researchers led by Shu Chen analyzed data from 43,188 participants in the Women’s Health Initiative, a major U.S. research program involving postmenopausal women.
Participants were between 50 and 79 years old when they entered the program from 1993 through 1998. Their average age was 63.
The women came from 40 clinical centers across the United States and had no self-reported history of heart failure at enrollment.
Researchers followed them for an average of 16.2 years, using medical records to identify first hospitalizations for acute decompensated heart failure, a worsening of heart function serious enough to require hospital treatment.
The study included both definite and possible heart failure events. Professional abstractors reviewed hospital records, and physicians classified events using symptoms, examination findings, diagnostic imaging, and treatment response.
To estimate residential radon exposure, researchers linked participants’ home addresses to geographic classifications based on the Environmental Protection Agency’s radon-zone maps. These maps incorporate information including indoor radon measurements, geology, soil characteristics, and building-related factors to estimate the potential for elevated radon concentrations across different areas.
Participants were assigned to three categories based on estimated area-level indoor radon concentrations:
- Less than 2 picocuries per liter (pCi/L)
- 2 to 4 pCi/L
- More than 4 pCi/L
These categories did not represent measurements taken inside participants’ homes. Actual indoor radon concentrations can differ substantially between properties within the same geographic area.
Approximately 36% of participants lived in the lowest estimated exposure category, 41% in the middle category, and 23% in the highest.
The researchers used statistical models to examine whether the hazard of heart failure differed across these categories after accounting for other measured risk factors.
The findings were reported by Chen et al. (2026) in JAMA Network Open.
What Researchers Found
Higher Estimated Radon Exposure Was Associated With More Heart Failure Hospitalizations
During follow-up, researchers identified 3,446 first hospitalizations for definite or possible acute decompensated heart failure.
Of these events, 1,946, or 56%, were classified as definite heart failure, while 1,500, or 44%, were classified as possible heart failure.
The unadjusted cumulative incidence increased across the three radon categories.
| Estimated indoor radon category | Heart failure events per 1,000 women |
|---|---|
| Less than 2 pCi/L | 68 |
| 2 to 4 pCi/L | 79 |
| More than 4 pCi/L | 101 |
In other words, approximately 68 of every 1,000 women in the lowest estimated radon category experienced a heart failure hospitalization, compared with 101 of every 1,000 in the highest category.
That represents an unadjusted difference of 33 hospitalizations per 1,000 women, or 3.3 percentage points, over the study’s observation period.
However, these raw figures do not account for differences in age, smoking history, blood pressure, or other factors that could influence heart failure risk.
After adjusting for measured demographic, social, behavioral, clinical, and environmental factors, researchers found that women in the highest estimated radon category had a hazard ratio of 1.19, with a 95% confidence interval of 1.07 to 1.33.
This represents a 19% higher adjusted relative hazard of heart failure hospitalization compared with women in the lowest category.
The finding describes a relative difference in the rate at which heart failure hospitalizations occurred during follow-up. It does not mean that radon increased each woman’s absolute probability of developing heart failure by 19 percentage points, nor does it establish that radon caused the additional events.
For women in the middle category, the fully adjusted hazard ratio was 1.09, with a 95% confidence interval of 0.99 to 1.20. Because the interval included 1.00, this comparison did not meet the study’s threshold for statistical significance.
The overall association between radon category and heart failure was statistically significant in the fully adjusted model, with a reported P value of .008.
Heart Failure Incidence Rates Also Increased
The researchers examined incidence rates that accounted for the amount of time participants contributed to follow-up.
They reported 439 heart failure events per 100,000 woman-years in the lowest radon category, 488 in the middle category, and 568 in the highest category.
These rates describe events across accumulated observation time, not an individual woman’s annual probability of developing heart failure.
By age 90, the estimated cumulative incidence was 9.0%, 10.0%, and 11.9% across the three categories, respectively.
Together, the absolute and relative results show a consistent pattern of more heart failure events among women assigned to higher radon categories. They do not establish that radon was responsible for those differences.
The Association Was Also Observed for Heart Failure With Reduced Ejection Fraction
Researchers examined whether the association differed according to the heart’s pumping function.
Heart failure with reduced ejection fraction occurs when the heart’s main pumping chamber does not contract effectively enough to eject a normal proportion of blood.
In the highest radon category, the adjusted hazard ratio for this form of heart failure was 1.24, with a 95% confidence interval of 1.04 to 1.49.
This corresponds to a 24% higher adjusted relative hazard compared with the lowest category.
However, the statistical test comparing associations across heart failure subtypes had a P value of .07, above the study’s prespecified significance threshold of .05.
Therefore, although the association was numerically stronger for heart failure with reduced ejection fraction, the study did not establish a statistically significant difference between heart failure subtypes.
Of the 3,446 heart failure events, 1,723 involved preserved ejection fraction, 1,152 involved reduced ejection fraction, 554 had unknown ejection fraction, and 17 were classified as improved ejection fraction.
The substantial number of events with unknown ejection fraction adds uncertainty to comparisons between subtypes.
Heart Failure Mortality Was Also Associated With Higher Radon Exposure
Researchers reported an adjusted hazard ratio of 1.37 for heart failure mortality in the highest radon category compared with the lowest.
The 95% confidence interval was 1.12 to 1.69.
This corresponds to a 37% higher adjusted relative hazard of death attributed to heart failure.
The finding requires caution because heart failure mortality was identified using underlying causes of death recorded on death certificates, and physician adjudication was not uniform throughout follow-up.
The study cannot establish that radon caused these deaths.
Smoking History May Influence the Association
One of the more notable findings involved smoking.
Researchers reported that the relationship between estimated radon exposure and heart failure differed according to whether women had ever smoked.
The statistical test for this interaction produced a P value of .02.
Positive associations were observed among women who had ever smoked but not among women who had never smoked.
This pattern is noteworthy because smoking is already known to increase the lung cancer risk associated with radon exposure.
However, the cardiovascular interpretation is less certain.
The study’s ever-versus-never smoking classification did not fully capture how many cigarettes participants smoked, how long they smoked, or how much time had passed since quitting.
These differences could influence the observed results.
The findings therefore do not prove that smoking and radon act together to cause heart failure. Nor do they establish that radon exposure is harmless to the hearts of people who have never smoked.
How Does Radon Enter Homes?
Radon forms naturally as uranium breaks down in soil and rock.
The radioactive gas can move through the ground and enter buildings through foundation cracks, openings around pipes, gaps in floors, and other pathways.
Because radon cannot be seen or smelled, testing is necessary to determine its concentration inside a home.
Levels can vary considerably between neighboring houses, even when they share similar geographic conditions.
Radon is an established cause of lung cancer, and its risk is especially important among people who smoke.
The Environmental Protection Agency recommends fixing homes with measured radon concentrations of 4 pCi/L or higher. It also recommends considering mitigation when concentrations are between 2 and 4 pCi/L because no level of radon exposure is known to be completely without risk (U.S. Environmental Protection Agency, 2026).
These recommendations are based on established radon health risks, particularly lung cancer.
The new study does not establish a separate cardiovascular threshold or demonstrate that lowering radon prevents heart failure.
How the Findings Compare With Earlier Research
Scientists have investigated possible cardiovascular effects of radon for decades, but the evidence remains mixed.
A previous study involving the Women’s Health Initiative examined radon exposure and stroke.
In that investigation, Buchheit et al. (2024) reported an association between higher estimated residential radon exposure and incident stroke among postmenopausal women.
Stroke and heart failure are different conditions, but the findings raise related questions about whether environmental radiation exposure may be associated with cardiovascular disease.
Other research has produced less supportive results.
For example, Turner et al. (2012) examined residential radon and nonrespiratory mortality in the American Cancer Society cohort and did not identify a significant association with heart disease mortality.
These contrasting findings are important.
Differences in participant age, exposure assessment, cardiovascular outcomes, smoking patterns, and statistical methods may help explain why studies have not reached consistent conclusions.
The new investigation contributes evidence about a more specific outcome: first hospitalization for medically reviewed heart failure.
However, one large observational study cannot resolve the broader uncertainty about radon and cardiovascular disease.
What Might Explain a Connection Between Radon and Heart Failure?
The researchers discussed several possible biological mechanisms.
Radon releases ionizing radiation as it decays. Its radioactive decay products can be inhaled, and exposure may contribute to cellular injury.
The authors proposed that radiation-related damage could affect blood vessels, heart muscle cells, and certain blood cells.
Potential pathways include oxidative stress, inflammation, tissue injury, and changes in blood-cell populations.
One process discussed in the paper is clonal hematopoiesis, in which certain blood-forming cells acquire mutations and expand into populations of related cells. Some forms of clonal hematopoiesis have been associated with cardiovascular disease.
The researchers also considered whether radiation-related inflammation and tissue damage might contribute to changes in heart muscle function.
These explanations remain hypotheses.
The study did not directly measure radiation damage in participants’ hearts, establish a biological pathway from radon exposure to heart failure, or demonstrate that reducing exposure reverses cardiovascular injury.
Biological plausibility makes the association worth investigating, but it is not proof of causation.
How Strong Is the Evidence?
The investigation has several important strengths.
It included more than 43,000 women, followed participants for an average of 16.2 years, and used physician-reviewed medical records rather than relying entirely on self-reported heart failure diagnoses.
The researchers also accounted for numerous established risk factors and conducted extensive sensitivity analyses.
For example, the main association remained generally consistent when investigators changed statistical assumptions, excluded early heart failure events, considered competing mortality, updated exposure estimates after residential moves, and used alternative geographic radon measures.
The alternative exposure models also produced positive associations.
For every 1 pCi/L increase in an alternative county-level estimate from Lawrence Berkeley National Laboratory, the adjusted hazard ratio was 1.10, with a 95% confidence interval of 1.02 to 1.19.
Using a ZIP-code-level estimate generated by a random-forest model, the adjusted hazard ratio was 1.11 per 1 pCi/L increase, with a 95% confidence interval of 1.05 to 1.18.
These results strengthen confidence that the main association was not dependent on a single geographic exposure model.
Nevertheless, all the exposure measures were estimates rather than direct long-term measurements inside participants’ homes. They may also share sources of error.
Consistency across statistical analyses cannot eliminate those limitations.
The Study’s Most Important Limitations
Radon Exposure Was Estimated, Not Measured in Individual Homes
The greatest limitation concerns how radon exposure was assessed.
Researchers assigned county-level radon categories to participants based on their residential addresses.
They did not measure radon inside each woman’s home throughout the study.
This matters because radon concentrations can vary considerably within the same county.
Home construction, ventilation, foundation conditions, local geology, and mitigation systems can all influence indoor levels.
The researchers described the investigation as partially ecologic because it combined individual-level health outcomes and risk factors with area-level exposure estimates.
This approach can introduce exposure misclassification and residual geographic confounding. The direction of any resulting bias is uncertain.
The Study Population Was Not Representative of Everyone
All participants were postmenopausal women aged 50 to 79 at enrollment.
The findings therefore cannot automatically be generalized to men, younger women, or children.
The highest estimated radon category also differed from lower categories in geography and demographic composition.
For example, women in the highest category were disproportionately located in the Midwest and Northeast.
Although the researchers adjusted for census region and examined other geographic factors, unmeasured differences between locations could still have influenced the results.
Other Heart Failure Risk Factors Could Remain Unaccounted For
At baseline, approximately 69% of participants had hypertension, 10% had diabetes, and 48% had ever smoked.
The statistical models accounted for these and many other measured characteristics.
However, adjustment cannot completely eliminate the possibility that unmeasured or imperfectly measured factors contributed to the association.
Smoking intensity and duration were particular concerns in the smoking subgroup analysis.
The investigators also lacked information about some cardiovascular conditions that developed during follow-up, including incident atrial fibrillation, which could potentially influence the pathway to heart failure.
No Radon-Reduction Intervention Was Tested
The study did not compare heart failure outcomes between households randomly assigned to radon mitigation and households without mitigation.
It therefore cannot determine whether installing a radon reduction system would lower the incidence of heart failure.
That question remains open.
What the Results Mean for Homeowners
The study raises questions about whether radon exposure could contribute to cardiovascular disease, but it does not change existing recommendations for household radon safety.
The strongest established reason to test for radon remains its proven connection to lung cancer.
Because radon is invisible and odorless, the only reliable way to determine its concentration inside a particular home is to test the indoor air.
Homeowners can take several practical steps:
- Test the home. Use an appropriate radon testing kit or hire a qualified radon measurement professional. Testing is worthwhile even in areas classified as having relatively low radon potential.
- Understand the results. The EPA recommends fixing homes with radon concentrations of 4 pCi/L or higher and considering mitigation when levels are between 2 and 4 pCi/L.
- Confirm results when appropriate. Radon concentrations can fluctuate over time. Depending on the initial result and testing method, additional short-term or long-term testing may help establish the home’s typical exposure level.
- Seek professional mitigation when needed. Qualified radon mitigation specialists can assess the property and recommend systems designed to reduce radon entering indoor living spaces.
- Test again after mitigation. Follow-up measurements help determine whether the installed system has successfully reduced indoor radon concentrations.
The EPA’s geographic radon maps cannot establish whether an individual home has elevated radon levels. Two neighboring houses may have substantially different concentrations.
For people concerned about heart failure, established preventive measures remain important, including controlling blood pressure, avoiding smoking, managing diabetes, maintaining physical activity, and following appropriate medical advice.
Although the new study identified an association between higher estimated radon exposure and heart failure hospitalization, it did not demonstrate that radon mitigation prevents heart failure.
Reducing elevated household radon remains a sensible health measure because of its established lung cancer risk, regardless of whether future research confirms additional cardiovascular benefits.
Final Thoughts
A large U.S. study involving 43,188 postmenopausal women found that living in areas with higher estimated indoor radon concentrations was associated with a greater hazard of hospitalization for heart failure.
Women in the highest estimated exposure category had a 19% higher adjusted relative hazard than those in the lowest category. Researchers also observed associations with heart failure involving reduced pumping function and with deaths attributed to heart failure.
These findings add to growing interest in the potential cardiovascular effects of residential radon exposure. However, the study was observational, and researchers estimated exposure using geographic data rather than direct measurements inside individual homes.
Further investigations involving verified household radon measurements and more diverse populations are needed to determine whether radon contributes directly to heart failure and whether mitigation can reduce cardiovascular risk.
For homeowners, the immediate recommendation remains unchanged: test for radon and address elevated concentrations because radon is an established cause of lung cancer.
Whether protecting homes from radon also helps protect the heart remains an important question for future research.
Funding and Disclosures
The study was supported by the Women’s Health Initiative Radon Risk Investigation, funded through grant R01ES034050 from the National Institute of Environmental Health Sciences.
The Women’s Health Initiative also received funding through contracts from the National Heart, Lung, and Blood Institute, part of the National Institutes of Health.
The National Heart, Lung, and Blood Institute supported the conduct of the Women’s Health Initiative, including data collection and management and review and approval of the manuscript. According to the authors, it otherwise had no role in the specific radon investigation’s design, analysis, interpretation, manuscript preparation, or submission decision.
Several authors disclosed federal research funding associated with the study.
Coauthor Bernhard Haring reported receiving personal fees from Bristol Myers Squibb, Boehringer Ingelheim, Pfizer, Daiichi Sankyo, Inari, and Novartis outside the submitted work.
No other disclosures were reported.
References
Buchheit, S. F., Collins, J. M., Anthony, K. M., Love, S.-A. M., Stewart, J. D., Gondalia, R., Huang, D. Y., Manson, J. E., Reiner, A. P., Schwartz, G. G., Vitolins, M. Z., Schumann, R. R., Smith, R. L., & Whitsel, E. A. (2024). Radon exposure and incident stroke risk in the Women’s Health Initiative. Neurology, 102(4), e209143. https://doi.org/10.1212/WNL.0000000000209143
Chen, S., Shi, C., Collins, J. M., Ekanayaka, A. K., Mowrer, C., Stewart, J. D., Huang, D. Y., Allison, M. A., Eaton, C. B., Haring, B., LaMonte, M. J., Nudy, M. S., Reiner, A. P., Schwartz, G. G., Shou, D. Z., Tinker, L. F., Williamson, M. R., Smith, R. L., & Whitsel, E. A. (2026). Indoor radon and incident heart failure among postmenopausal women. JAMA Network Open, 9(11), e2642391. https://doi.org/10.1001/jamanetworkopen.2026.42391
Turner, M. C., Krewski, D., Chen, Y., Pope, C. A., III, Gapstur, S. M., & Thun, M. J. (2012). Radon and nonrespiratory mortality in the American Cancer Society cohort. American Journal of Epidemiology, 176(9), 808–814. https://doi.org/10.1093/aje/kws198
U.S. Environmental Protection Agency. (2026). What is EPA’s action level for radon and what does it mean? https://www.epa.gov/radon/what-epas-action-level-radon-and-what-does-it-mean
