The EPA attributes about 21,000 US lung cancer deaths a year to radon, making it the leading cause among people who never smoked. Risk builds with concentration and years of exposure, and it falls once the level falls.
More detail
The mechanism is physical: radon's short-lived decay products attach to dust and lodge in lung tissue when inhaled, where their radiation can damage cells over time. The EPA's lifetime risk tables make the dose-response concrete. Among 1,000 people who never smoked and lived at 4 pCi/L, about 7 could develop lung cancer from it; at 8 pCi/L, roughly double. Smokers face several times those figures at every level because the exposures compound. Two readings of the numbers keep people sane. First, this is a chronic risk, not an acute one: no one is harmed by a high week, and a high result is a this-season project, not a tonight evacuation. Second, the risk is genuinely reducible, which cannot be said of every environmental hazard: bring an 8 down to 1.5 and the ongoing exposure drops with it. In a state averaging about 8.5 pCi/L, that reducibility is the whole argument for testing.