Data from the Comprehensive Epidemiology Data Resource (CEDR) allowed me to study patterns of cancer mortality in a cohort of 4,014 uranium-processing workers. Employing risk-set analysis for cohort data, I estimated the effects of external (gamma) and internal (alpha) radiation on cancer mortality. My results indicate that Fernald workers exposed to ionizing radiation experienced an increase in mortality from total cancer (per 100 mSv external dose rate ratio (RR) = 1.92; 95% confidence interval (CI) = 1.11-3.32), radiosensitive solid cancer (RR = 2.00; 95% CI = 1.02-3.94), and lung cancer (RR = 2.77; 95% CI = 1.29-5.95). Effects were strongest when exposure had occurred at older ages (>40 years). In addition, I observed an increase in lung-cancer mortality for workers exposed to > or =200 mSv of internal (alpha) radiation (RR = 1.92; 95% CI = 0.53-6.96). Furthermore, my results demonstrate the importance of a long follow-up time when studying solid cancers, the potential for bias due to worker selection associated with concomitant chemical exposures, problems of exposure measurement, confounding, and effect modification due to age at exposure. Owing to lack of data, a previous pooled analysis of uranium-processing workers could only partially address these issues.