Objectives Studies of underground miners have documented an increased risk of lung cancer mainly linked to radon exposure but possibly influenced by other concurrent exposures.

Methods A cohort study was carried out in 8321 iron ore miners with low exposure to radon, employed in 1923–1998 and followed up for lung cancer in 1958–2000. Historical exposures to radon, crystalline silica and diesel exhaust were assessed. Data including exposure to radon, quartz and diesel exhaust from another mine with higher exposure to radon were reanalysed.

Results Miners had increased risk for lung cancer (SIR 1.48 (95% CI 1.22 to 1.78), based on 112 cases during 227 000 person-years). The increased risk could not be explained by exposure to radon or diesel exhaust but was associated with exposure to crystalline silica: SIR 0.96 (0.53 to 1.62), 1.45 (1.10 to 1.87), 1.99 (1.31 to 2.90) and 1.77 (0.92 to 3.10) in groups with exposure to 0, 0–2, 2–5 and >5 mg years/m3, respectively. Reanalysis of data from the other mine indicated that quartz was a possible confounder in the analysis of relationship between radon and lung cancer. In the highest radon exposed group, the point estimate for the RR decreased from 5.65 to 3.90 when adjusting for concurrent exposure to quartz.

Conclusions Crystalline silica, a known carcinogen, probably affects lung cancer risk in iron ore miners. The main implication of the results is for interpretation of the dose–response curve for radon and lung cancer in underground iron ore miners. Since exposure to radon and quartz is often correlated, quartz exposure can be an important confounder.