Does Radon Really Cause Lung Cancer?
In a recent article on the result of my radon test, I referred to the US EPA’s work on the health effects of radon. They claim that about 21,000 people in the US each year get lung cancer from indoor exposure to radon. In the comments, a few people questioned the EPA claim. How do we know that radon really does cause lung cancer, they ask? The EPA based their conclusion on data from miners working underground. How can radon exposure in homes be related to the higher radiation levels they received?
So I’ve dug in a little bit. I’m going to keep this as brief as I can and relatively simple. If you want to dig into the background work on this yourself, click the links I provide throughout the article and explore the resources at the end.
We have more than a century of data on exposure to all kinds of radiation. When Henri Becquerel discovered radioactivity in 1896, the study of its effect on humans began. It was unintentional at first because the early researchers exposed themselves to high doses without knowing the dangers. Many suffered and died from diseases associated with radiation exposure.
Radium, the parent of radon in the uranium-238 decay chain, became popular for glow-in-the-dark watch faces starting in 1908. That led to the famous case of the “Radium Girls.” They were factory workers who painted the watch faces, using their lips to put a point on their brushes. As a result, they ingested a lot of radium. And they made it worse by painting their fingernails and even their teeth for fun. Many didn’t live long enough to get cancer, though, because the high doses gave them radiation poisoning.Another big batch of data came from the atomic bombs dropped on Hiroshima and Nagasaki at the end of World War II. Then we have all the people who were exposed to radiation during the testing of nuclear weapons. (The documentary Radio Bikini is a scary look at the testing done in 1946 at the Bikini Atoll.) We also have all the monitoring done for occupational exposure.
You may think none of that’s related to radon exposure in homes either, but it is. We need to start with the physics of radioactive decay, though. I covered some of this in the last article already, but let’s review.
A radioactive element has an unstable nucleus that will decay in one of several ways: alpha, beta, gamma, or neutron emission. We describe the rate of decay in terms of a half-life, the time it takes for half of a sample to decay. We know the decay chains of radioactive elements. Radon-222, for example, decays into polonium-218 by alpha emission. Its half-life is 3.82 days. We know these things out to many decimal places.
Alpha particles are helium nuclei, with two protons and two neutrons. Alpha particles are large (compared to beta particles), energetic, and easily stopped. Alpha particles hitting you from outside aren’t a big deal. They don’t get through your clothing. If they hit your skin, they’re absorbed in the outer layers where they don’t cause problems.
Biological effects of alpha particles
The problem with alpha particles arises when they’re emitted inside your body. They have enough mass, energy, and momentum to damage a lot of cells. And that’s where radon comes in. Literally. It comes in through your lungs. Radon is an alpha emitter, but radon’s emissions aren’t the ones thought to cause lung cancer. Radon’s 3.82 day half-life causes it to hang around in your indoor air for a few days before decaying. It’s an electrically neutral atom, though, so it doesn’t stick to your lung tissue when you inhale it. And most radon atoms will leave before decaying when you exhale.
The alpha emissions that occur in your lungs are mainly the ones from radon’s progeny, polonium-218 and polonium-214. They are not neutral, so they stick to the lung tissue when you inhale them. Then they just sit there, waiting to decay and emit their alpha particles. That’s where the trouble begins.
An alpha particle has a lot of energy, so it rips through the cells in your lung tissue like an NFL running back through a middle school football team. It ionizes atoms along the path, which can kill cells or worse, mutate them.
Going from high dose to low dose
Let’s do a quick summary.
- Radioactive elements emit ionizing radiation that can kill or mutate cells.
- Alpha particles are inconsequential with external exposure but damaging from internal exposure.
- Radon has two electrically charged progeny that can be inhaled and release alpha particles in the lungs.
- Much of our data on the health effects of radiation come from high dose samples like the Radium Girls, those exposed during nuclear weapons detonations, and occupational exposures.
Scientists also have tracked enough people who were exposed to lower doses that they have a model for how to extrapolate to lower doses. It’s called the linear no-threshold model. Unlike many scientific terms, you can tell from the name exactly what it means. First, the effects of radiation exposure are linear with dose. If a dose of X results in Y cases of cancer among 100,000 people, a dose of 0.1 X will result in 0.1 Y cases of cancer.
The other part is the threshold. For a long time, scientists have debated whether or not there is a threshold of exposure below which radiation isn’t harmful. The majority of scientists now believe there is no safe threshold. Any exposure to ionizing radiation, they say, can result in cancer, leukemia, or other health problems.
The National Academy of Sciences, the National Council on Radiation Protection and Measurements, and other organizations involved in the health effects on radiation support the linear no-threshold model. Why? Because there’s a lot of data behind it.
Effects of radiation exposure
In my last radon article, I discussed the units used to measure radioactivity in your indoor air. A picocurie per liter tells you the concentration of radon in the air. Curies are a measure of how many radioactive decays occur per unit time. But that doesn’t tell you what effect it might have on your body. For that we use other units.
The sievert is a unit for the effective radiation dose. It accounts for what type of radiation you’re exposed to and what part of the body is exposed. It’s the unit that governments set limits for in occupational exposure.
The chart below gives you an idea of the scale of effective radiation dose we get from different sources. Radon makes up a big chunk of the exposure. (Thoron, by the way, is a special name for the radon-220 isotope, so it’s still radon. The primary isotope is radon-222.) Those of us who have had significant doses of medical radiation skew the chart. If that’s not you, you’re mainly getting background radiation, which is mostly radon.
And over the course of a year, the average person in the US gets an effective dose of 2.28 millisievert (mSv) from radon. The chart below puts that in perspective so we can get to the issue of whether or not radon really does cause lung cancer.
So, 2.28 mSv is pretty low on the chart. It’s well below the 100 mSv where scientists can state definitively that you’re likely to get cancer. But remember: The linear no-threshold model says there’s no threshold. And the linear part means that the chances drop linearly with dose.
The EPA’s documentation
The EPA published a 98 page report titled EPA Assessment of Risks from Radon in Homes in 2003. Their report is a fine-tuning of a 592 page report from the National Academy of Sciences titled Biological Effects of Ionizing Radiation (BEIR) VI Report: “The Health Effects of Exposure to Indoor Radon.” (Links to both documents are below.) These are high-level scientific reports done by professional scientists.
Both reports use data on radon exposure to miners working underground and extrapolate down to lower doses that people receive in their homes. That’s where the linear no-threshold model comes in. Both reports discussed that model extensively. They also discussed challenges to the linear no-threshold model and why they rejected them. The EPA report states:
The BEIR VI committee adopted the linear no-threshold assumption based on our current understanding of the mechanisms of radon-induced lung cancer, but recognized that this understanding is incomplete and that therefore the evidence for this assumption is not conclusive.
In other words, if scientists find enough evidence to contradict the linear no-threshold model, they’d have to revise their use of it. Maybe it’s not linear. Maybe there is a threshold. The current status, though, is that the model is valid.
The reports go into great detail describing the model, the data, the history, and the uncertainties. Then the EPA report states definitively, “There is overwhelming evidence that exposure to radon and its decay products can lead to lung cancer.” We know it’s true for miners with high exposures. Scientists have a high level of confidence that it’s true in homes with lower exposures.
Yeah, they’re saying radon really does cause lung cancer.
What’s your risk level?
One thing that’s clear from the studies done is that smokers are at much greater risk than nonsmokers. Actually, they distinguish the two groups as “ever smokers” and “never smokers.” If you smoked for ten years and then quit, your susceptibility to radon-induced lung cancer is higher than if you had never smoked. So what we know is that your chances of getting cancer from radon exposure at home depends on:
- Your smoking history
- The indoor radon level
- The amount of time you’re exposed
If your home has an elevated radon level but you’re rarely there, your risk is lower. A person who spends most of their time at home in a house that has a lower radon level may have a higher total dose and be at greater risk.
I’ve looked at the EPA and BEIR VI documents, spending more time with the EPA report. I don’t have the time to digest even the EPA report completely, but I’ve read enough to have confidence in their conclusions. The only real debate is whether or not the linear no threshold model applies or not. And if not, what threshold should we use?
Is it ironclad that exposure to 4 picocuries per liter of radon in your home will lead to a certain number of cancers? No. Every scientific measurement and result is uncertain. Uncertainty is part of science. Models are part of science, too. But the basic facts are well understood.
- We breathe in radon and its progeny.
- Radon and its progeny emit alpha particles.
- Alpha particles kill and mutate many cells.
- The radiation health science community mostly supports linear extrapolation of dose and cancer risk with no threshold.
So, does radon really cause lung cancer? Yes, the data on miners’ exposure makes that clear. The real question is what happens at lower exposures. Without a better alternative, the linear no-threshold model seems the best way to extrapolate.
But you have to consider the risk assessment, too. According to the EPA, nonsmokers exposed to 4 picoCuries per liter over a lifetime will result in 7 out of 1,000 getting lung cancer. They compare that to the risk of dying in a car crash. For smokers, that number jumps to 62 out 1,000.
The bottom line is that it takes a long exposure time for low levels of radon. And it’s far worse if you’re a smoker. Having seen both of my parents die of lung cancer, though, I’m going to do what I can to avoid it. I’ve never been a smoker, and I’m doing what I can to reduce the radon level in my house.
EPA Assessment of Risks from Radon in Homes (pdf)
Biological Effects of Ionizing Radiation (BEIR) VI Report: “The Health Effects of Exposure to Indoor Radon.”
Allison A. Bailes III, PhD is a speaker, writer, building science consultant, and the founder of Energy Vanguard in Decatur, Georgia. He has a doctorate in physics and is the author of a popular book on building science. He also writes the Energy Vanguard Blog. You can follow him on Twitter at @EnergyVanguard.
What to Know About a High Radon Test Result
What Percent of Time Do You Spend Indoors?
Which Indoor Air Pollutants Matter Most?
Comments are welcome and moderated. Your comment will not appear below until approved.
This Post Has 26 Comments
I live in a high radon area and my new home was testing in the 40’s before turning the radon fan on. However, I’ve read that many of us get the majority of our radon from water. That makes sense to me. Our water source is a community well, and if there’s this much in the ground I can see how it is also in the water. But while everyone here has radon systems, nobody I’ve talked to has ever tested their water. I have just purchased a radon water test kit.
Did you come across anything in your reading about radon in the water affecting health?
My high radon level was a surprise too because we meticulously sealed the crawlspace and have an ACH=1.
Airtight homes can have higher negative pressures, especially in cold climates where stack effect is stronger in winter. Higher negative pressure at the bottom of the house can pull more radon into the house.
Allison, do you prefer radon mitigation methods of removing radon that is pulled into homes or preventing radon from being able to enter the home by keeping the house at a neutral or slightly positive pressure through outside air infiltration such as ERV/HRV’s or even just barometric damper controlled outside venting? I too live in a high radon area, the area where high levels of radon were first discovered in homes. I keep my home slightly pressurized, have an open to ground French drain system in a portion of my basement, and have not had a radon issue. If any home should have a radon issue, it would be mine.
Robin: I prefer any method that works.
Granite countertops may contain a certain amount of radioactive uranium, which decays (eventually) through the radon decay chain. This causes certain issues, such as interfering with the radon tests, because each decay has a potential to pass through the radon device. But does it get into your lungs? It is pretty unlikely. The half-life of uranium 238 (which decays eventually to radon) is 4.5 billion years. The seven decays from uranium to radon are over 300 thousand years. All of those decays are from solid to solid, so the progeny remain in the granite. The only decay that matters is the radon decay (followed by polonium decays), because radon is a gas that can be inhaled and can potentially escape the granite. But since most countertops are sealed at three sides, the only escape path is the underside into your cabinets. So if you have an unlikely decay to radon and the gas molecule makes it out of the countertop into your cabinet and within 3 days (on average) you inhale the gas from your cabinet and are unlucky enough for the radon gas to decay during your inhalation…it’s possible, but extremely unlikely. (Somewhat more likely is beta emission impacting your eye from one of the beta decays, so don’t stare at the granite too long!)
Bobby: Thanks for the excellent explanation. I knew most of those facts but had never seen them put together that way and hadn’t done it myself either. My intuition was that granite countertops probably weren’t a major contributor, though.
Cindi: I haven’t seen anything to indicate that waterborne radon is as bad as airborne. If it’s in the water, the water, stomach acids, and processing food will absorb a lot of the alpha particles released. It certainly won’t cause lung cancer that way. But the electrically charged radon progeny could attach to tissue and cause problems in the digestive or urinary system.
Radon in water is indeed a major concern. However, the risk isn’t from drinking the water. Radon is a gas, and so radon exists as a dissolved gas in water. Like most dissolved gases, it is released very quickly by turbulence or somewhat more slowly by diffusion. Once water goes through an aerator, radon gas is released into the air. For exposure, normally the two main concerns (where dissolved gas concentrations are highest) are in the shower and immediately after running the tap. However, since radon persists for 3 days (half-life) or up to 10 days in reasonably measurable concentrations, it becomes entrained in the household air.
You can generally measure your risk from radon in water by a normal radon test. For curiosity’s sake, radon in water tests are available. The main use of these, in my opinion, is to see if fixing soil gas will be effective in further reducing indoor concentrations. If water is the main source. local exhaust (kitchen and bath) and increased ventilation (ACH) are the best strategies.
Bobby: More thanks for this excellent explanation! It’s good to have an industrial hygienist chiming in on this topic.
I live in a house built of Wissahickon schist from the Reading Prong which contains trace amounts of radium which as you know decays into radon. Our testing has consistently shown very low levels so we’re fortunate. We also have a “leaky” house. Based on my reading (Bill Rose’s “Water in Buildings”) and various papers my late restoration architect friend shared, it’s a close call how much sealing we should do in our stone house w/cedar shingle roof, radon notwithstanding.
I love your regular newsletters and these occasional posts. Also your book, where I’m reading pages by fits and starts.
Steven: I’d be interested to see what papers your friend shared with you. Leaving your house intentionally leaky doesn’t guarantee good indoor air quality. You may have read that part in my book by now.
Glad you like the newsletter, articles, and book!
This analysis should be mandatory reading and study for everyone in today’s mis and dis-information society – especially journalists of all stripes. A perfect example of the thorough and proper examination of a question. You set a good standard Dr. Bailes.
Thank you Allison for taking the time to read the reports, something I should have read but never have. For those that still doubt the dangers of radon exposure in homes, there is still wisdom in providing a pathway out of the home for soil gases for methane, water vapor and who knows what else. It simply helps assure better air quality.
Thank you for a clear, concise description of this complex issue. My son bought a house with a high(ish) radon reading, and I supported his installation of a remediation system, and it’s nice to read a detailed (but not as detailed as the 100+ reports!) explanation of the science behind the issue. Thanks!
Nah, if I can’t see it, it won’t hurt me.
It doesn’t appear that lung cancer rates by state correlate very closely with natural radon exposure
Is that because of broader differences in other factors (e.g., smoking rates)?
Jim: Looking for correlation lung cancer between and the EPA’s radon map tells you next to nothing. Yes, lung cancer from smoking is way more prevalent that radon-induced lung cancer. Second, the EPA radon map gives you a rough idea of how much is underground in each county. But it can change a lot in short distances so that doesn’t provide enough granularity to know how much is in the soil under a particular house. And then there’s the issue of how much gets into a house. As I mentioned in my first article, a house can be in EPA zone 3 (lowest radon) and still have a high indoor level. Or you can have a block full of low-radon houses with one that has elevated radon.
Well written article, I always love to read your articles thanks for sharing.
I don’t suppose that higher radon levels could aid in mitigating mold issues, eh?
Thank you for another thoughtful article.
I am in complete agreement with you up until the last section (“My Take”) for three reasons.
First: “When a non-scientist dismisses the link between radon and lung cancer, it tells me they don’t understand how science works.” With respect, this comes off as elitist and, in my view, is a plainly incorrect view of the scientific method. Your next paragraph explains why: scientific models address uncertainty. The null hypothesis in any scientific analysis is that there is no relationship. The most scientific starting point when considering a relationship between radon and lung cancer is to assume there is no relationship unless the evidence is overwhelming enough to conclude otherwise. To claim that a person (non-scientist or otherwise) is being unscientific by expressing skepticism is, in itself, unscientific.
Second: “The radiation health science community supports linear extrapolation of dose and cancer risk with no threshold.” The scientific community has justifiably used linear extrapolation modelling because of a lack of other options. But there is now plenty of reason to believe such models may be incorrect. For example, machine learning modelling (a branch of artificial intelligence) addresses the known limitations of linear extrapolation. I fully expect future radon studies to use more sophisticated modelling strategies and better data to arrive at more nuanced conclusions. This does not necessarily mean that household radon exposure does NOT cause cancer, but rather that the story may be much more nuanced (who, where, why).
In terms of so-called scientific consensus: the modelling strategies upon which the linear extrapolation radon studies were based has increasingly been determined to be all but useless for causal modelling. For example, the flagship journal of the American Statistical Association recently devoted an entire issue (nearly 50 separate articles) wherein medical researchers were implored to “abandon” the modelling strategies they have overwhelmingly relied on to determine causal relationships. See The American Statistician, Volume 73, Issue sup1 (2019).
Third: “So, does radon really cause lung cancer? The scientific community has come to consensus on the issue.” With respect, I am surprised to read this statement coming from someone that normally writes so objectively. The notion of “scientific consensus” as a good thing is very strange indeed. Who would even decide a consensus had been achieved? And regardless, why should a consensus be assumed to be correct? The history of science is clear on this issue: consensus has often been wrong. Many major scientific breakthroughs have occurred when a dissenting voice challenged the status quo – often at great personal risk. The history of science is, in fact, the history of overturning previously held scientific consensus. Even modern, small, incremental improvements in scientific knowledge come at the expense of overturning things that were considered true just yesterday. Thomas Kuhn wrote an excellent book on this topic (The Structure of Scientific Revolutions, 1962).
As you have pointed out so many times (often humorously): the consensus by “expert” home builders has been (continues to be?) that they “need to breathe.”
Please don’t interpret my comments as a personal attack. I very much appreciate your work.
Ken: Good points, and no, I don’t take your comment as a personal attack. Here’s my response:
1. You’re right about the sounding elitist part. I’ve changed “it tells me they don’t understand how science works” to “it tells me they may not understand how science works.” Some non-scientists have done their homework and understand the issues. But I’ve had plenty of conversations with people who know only a tiny bit of the whole story. With radon, for example, there are plenty of people who say, “The EPA is wrong because they based their conclusion on high-doses to miners.” And that’s all they know. They’ve never heard of the LNT model. They don’t how radioactivity works and what alpha particles can do.
2. Yes, there’s debate about the LNT model. Maybe it will be replaced with something better someday, but it’s the tool we have now. Based on the damage we know alpha particles can do inside a lung, I think it’s a reasonable model.
3. If scientific consensus weren’t possible, what good would science do? It could never say anything definite. Rather than science having a history of overturning things previously considered true, it more often modifies the boundaries of where the previously truth applies. Newtonian mechanics, for example, still applies to most mechanical engineering projects, but relativity and quantum mechanics take over in the realms of the very fast and very small.
You ask, “why should a consensus be assumed to be correct?” and say, “The notion of ‘scientific consensus’ as a good thing is very strange indeed.” How can science do anything if there’s no consensus? Consensus grows as the data supporting an idea grows, and with the absence of data disproving the idea. Surely you wouldn’t say consensuses about gravity, planetary motion, or the germ theory are counterproductive or irrelevant.
I’ve read Kuhn’s book. It’s a great look at the big picture of science and how it has developed.
This is a great conversation with really good points from all, but it’s important to keep in mind that that this is a blog – not a scientific journal. Cut some slack here. There is a practical limit as to how many words, graphics and photos one can include before the audience becomes bored and turns away. I am in awe of how Allison can deal with such complicated subjects in his very concise posts. It takes an amazing amount of discipline. I have just retired from 25 years of writing monthly magazine articles, always feeling my hands were tied by the editor’s requirement of less than 1,200 words. Allison consistently deals with seriously complicated topics in posts of less than 800 words, WITH references and photos. Of course there are points that can be challenged, but keep in mind that there are always going to be unlimited nits to pick.
I guess that I am still a skeptic on the LNT model as I indicated in the previous blog on this subject. But since there seems to be nothing better at this point, let’s accept it and dig into it deeper. The EPA risk table in the previous blog is interesting if you look at it closely. It indicates that if your house tests at 4 pCi/L, you should “fix your home”. It also says that reducing indoor levels below 2 pCi/L is “difficult”. So if you install a mitigation system that reduces your exposure from 4 to 2 pCi/L, you would expect to reduce your cancer risk over your lifetime by 50%. That sounds significant. But look at the actual risk reduction in that table which goes from 0.7% to 0.4% (7/1000 to 4/1000). So you have reduced your risk of cancer during your lifetime by 0.3 percentage points. Is it worth it? I don’t know. If you look at Allison’s house (almost 8 pCi/L), mitigation could reduce his risk by about a factor of 4 or from 1.5% to 0.4% which is a reduction of 1.1 percentage points. Is that worth it? I wonder how EPA picked 4 pCi/L as the threshold. Is it because that is about the same risk as dying in a car crash during your lifetime? We always need to consider the true risk/reward for various situations. I am not going to quit using cars just to reduce my chance of dying during my lifetime by 0.7 percentage points.
I have issues with all IAQ exposure limits. They imply that if you are below the limit, you are safe and if you are above it, you are at risk. Regardless of the dose/response characteristic of any contaminant, that is just not the case.
As a three time home renovator of 100+ year old houses, I truly appreciate Allison’s blog and view his opinions/judgement as valuable.
As a retired, would-be scientist of sorts, I truly appreciate Ken’s exposition just above.
As a retired emergency physician, I look back over 40+ years trying to hew to that way of thinking while simultaneously–as a matter of (sometimes urgent) necessity making decisions in individual instances. Yes, I did soothe myself with thoughts of what the consensus advised, though I also recognized then and still do, “The art of medicine is long, Hippocrates tells us, “and life is short; opportunity fleeting; the experiment perilous; judgment flawed.”
While the stakes are different in home building and renovation in the greater scheme of things life is short, opportunity is fleeting, experiment is perilous and judgement flawed. Yet fulfillment of the responsibility remains necessary or my family and I would be living in different circumstances.
Modern day “risk management” was never forseen by Hippocrates and the ancient Greeks but must not be eschewed by modern day home builders, contractors and homeowners.