Researchers put a week-long winter radon test and a year-long detector side by side in 158 basements. The short test was usually right. Every home it wrongly cleared had read between 3 and 4.
If a short-term radon test in your home came back between 3 and 4 pCi/L, follow it with a long-term test. In 158 Iowa homes, about 4 in 10 with a winter short test in that range averaged 4 pCi/L or more over a full year, the level where the EPA says to fix a home.
The EPA’s testing steps call for a follow-up test when the first result is 4 or higher. Below 4, the advice is to consider testing again sometime in the future. So a 3.6 on a home inspection report usually gets filed as fine.
The study
Nirmalla Barros and William Field of the University of Iowa, with Daniel Steck of St. John’s University in Minnesota, published the study in Health Physics in 2014. The authors declare no conflicts of interest.
The homes came from the Iowa Radon Lung Cancer Study, run from 1992 to 1997, in which trained staff placed and collected every detector. The authors call Iowa well suited to the question: EPA surveys found it had the highest average home radon of any state surveyed.
Each basement got two measurements about 7 inches apart. One was a winter short-term test lasting about a week (8.4 days on average, median 7.5). The other was a detector left in place for a full year, which the authors treat as the reference. That gave 158 matched pairs.
The paper reports radon in becquerels per cubic meter (Bq/m3). We converted at 37 Bq/m3 per pCi/L, so the paper’s 148 Bq/m3 is the EPA’s 4 pCi/L, 111 is 3.0, 122 is about 3.3, 214 is about 5.8 and 74 is 2.0. Those pCi/L figures are our conversions.
What they found
Overall the short test did well. Short-term and year-long results were strongly correlated (r = 0.87), and the short test put 83% of homes on the correct side of 4 pCi/L.
The misses clustered near the line.
Of the 28 homes whose short test read between 3.0 and 4.0 pCi/L, 11 averaged 4 or more over the year, about 40%. Of the 30 homes that read 3.0 or lower, none did.
Counted from the other side: among homes that truly averaged 4 pCi/L or more for the year, 12% (95% CI 6 to 20%) had a short test under 4. That is the result that, under the usual steps, tells a homeowner no follow-up is needed.
The authors worked out a band for decisions. A winter short test under about 3.3 pCi/L gave more than 95% confidence that the yearly average was under 4. A result above about 5.8 gave more than 95% confidence it was 4 or over. In between, the authors write, “additional radon testing would be advised.”
Errors ran the other direction as well. Of 100 homes with a short test at 4 or higher, 16 averaged under 4 over the year. On average the short tests read a little high: 61% came in above their year-long match.
At a lower cutoff of 2 pCi/L, the short test almost never missed. Its false negative rate there was 2% (95% CI 0 to 6%).
Why it happens
Radon levels move with weather, season and how a house is used. The EPA’s Citizen’s Guide to Radon says a short-term test “is less likely than a long-term test to tell you your year-round average radon level.” A week is one sample of a year. Far from the cutoff, that does not matter. Near it, ordinary swings decide which side a single reading lands on.
The limits, stated plainly
This was Iowa, in winter, in basements. The authors say their findings are most generalizable to high-radon Midwestern homes with similar climate and construction. Portland and Vancouver have a different climate and many crawlspace homes, so the exact percentages may not carry over here.
The 3-to-4 group was only 28 homes, so “about 40%” is a rough figure with wide uncertainty. The measurements were made in the 1990s.
These were careful tests: about a week long, in the heating season, placed by trained staff. The authors point out that the usual screening test is often a 2-day kit in any season, and they say those may miss more often. That is their expectation. This study did not measure it.
What survives: a short-term result just under 4 pCi/L is weaker evidence than it looks, and the fix is a longer test.
What happens when you fix it
The EPA’s guide says some radon reduction systems “can reduce radon levels in your home by up to 99%.” It also advises testing again after a system is installed, and every two years after that. Every one of those decisions depends on a number you can trust.
What gets lost in the radon test marketing
Short-term kits are sold as a quick answer, and home sales run on them. The authors note that US short-term tests usually last 2 to 5 days because real estate deals need fast results.
The EPA’s own guide says short-term tests are “less definitive about whether or not your home is above 4 pCi/L” when results are close to 4. The EPA’s action level page, last updated November 12, 2025, also says the agency “recommends that Americans consider fixing their home for radon levels between 2 pCi/L and 4 pCi/L,” because there is no known safe level.
What the data supports
- A short test between 3 and 4 pCi/L: follow up with a long-term test, which the EPA defines as more than 90 days, in the lowest lived-in level of the home.
- Buying a house: ask for the actual number on the radon report. A 3.8 deserves a second test.
- During a short test: keep windows and outside doors closed, as the EPA and the kit direct. Every short test in this study was done in the heating season.
- After a remodel or a new heating or cooling system: test again. The Oregon Health Authority says changes like these can alter the air pressure in a house and with it the radon level.
Bottom line
A short-term radon test that reads just under 4 pCi/L is a weak all-clear. In this study, about 4 in 10 of those homes averaged 4 or more for the year. Follow a 3-point-something with a long-term test.
First step: find your last radon result and read the number itself, then decide whether it needs a second test.
illumenair does indoor air quality testing in Portland and Vancouver homes.
If you have a test result and are not sure what it tells you, we are glad to help you read it.
971-363-5626
Sources
- Barros NG, Steck DJ, Field RW. “A comparison of winter short-term and annual average radon measurements in basements of a radon-prone region and evaluation of further radon testing indicators.” Health Physics, 2014;106(5):535-544. doi:10.1097/HP.0000000000000004.
- US EPA. “A Citizen’s Guide to Radon.” EPA 402/K-12/002.
- US EPA. “What is EPA’s Action Level for Radon and What Does it Mean?” Last updated November 12, 2025.
- Oregon Health Authority. “How to Test A Home for Radon.” Accessed September 28, 2026.