BACKGROUND This study investigates confounding by cigarette smoking of associations between occupational exposure to ionizing radiation and lung cancer mortality among workers at the Savannah River Site (SRS). METHODS Thirteen thousand two hundred sixty-five white males hired at SRS between 1950 and 1986 were followed through 2002 to ascertain causes of death. Estimates of radiation doses from external sources and internal tritium uptakes were derived from dosimetry records. Logistic regression methods were used to derive discrete-time estimates of rate ratios. An indirect approach to control for unmeasured confounding by smoking was employed that involves joint modeling of lung cancer and chronic obstructive pulmonary disease (COPD) mortality. RESULTS Prior to indirect adjustment for smoking, there was minimal evidence of association between lung cancer mortality and cumulative radiation dose under a 10-year lag assumption (RR at 100 mSv = 0.90; 90% CI: 0.80-1.01). Subsequent to indirect adjustment for smoking, the association between lung cancer mortality and cumulative radiation dose under a 10-year lag was positive (RR at 100 mSv = 1.33; 90% CI: 1.01-1.77). CONCLUSIONS In this cohort, there is evidence of negative confounding of radiation dose–lung cancer mortality associations by cigarette smoking.