We studied cancer mortality in a cohort of 5,573 women with scoliosis and other spine disorders who were diagnosed between 1912 and 1965 and were exposed to frequent diagnostic X-ray procedures. Patients were identified from medical records in 14 orthopedic medical centers in the United States and followed for vital status and address through December 31, 2004, using publicly available regional, state and nationwide databases. Causes of death were obtained from death certificates or through linkage with the National Death Index (NDI). Statistical analyses included standardized mortality ratios (SMR = observed/expected) based on death rates for U.S. females and internal comparisons using Cox regression models with attained age as the time scale. Diagnostic radiation exposure was estimated from radiology files for over 137,000 procedures; estimated average cumulative radiat on doses to the breast, lung, thyroid and bone marrow were 10.9, 4.1, 7.4 and 1.0 cGy, respectively. After a median follow-up period of 47 years, 1527 women died, including 355 from cancer. Cancer mortality was 8% higher than expected (95% CI = 0.97-1.20). Mortality from breast cancer was significantly elevated (SMR = 1.68; 95% CI: 1.38- 2.02), whereas death rates from several other cancers were below expectation, in particular lung (SMR = 0.77), cervical (SMR = 0.31), and liver (SMR = 0.17). The excess relative risk (ERR) for breast cancer mortality increased significantly with 10-year lagged radiation dose to the breast (ERR/Gy = 3.9; 95% CI: 1.0-9.3).