Major efforts included: (1) Publishing an analysis of risk of lung cancer from radon and smoking and using this model to estimate risk attributable to radon in the U.S. and Germany; (2) publishing case- control data on risks from exposure to radon in homes and completing a pilot assessment of radon exposures in underground dwellings in Gansu, China; (3) publishing data from six cohorts that demonstrate persistent risk of thyroid cancer forty years after exposure; (4) completing validation studies to relate """"""""gold standard"""""""" personal dosimetry to other measurements of electromagnetic radiation; (5) publishing papers on the design, exposure assessment and gender differences in risk of hematopoietic disease following benzene exposure; (6) establishing dose- response relationships for bladder cancer from cyclophosphamide treatment for non-Hodgkin's lymphoma; (7) demonstrating increased levels of P45012A after ingestion of pan-fried meats and an inverse relation between P45012A blood levels and amounts of unmetabolized heterocyclic aromatic amines in urine; (8) showing that DNA ploidy adds little prognostic information in ovarian cancer; (9) showing that vitamin supplementation is associated with fewer strokes and lower blood pressure in men in an intervention trial in Linxian, China; (10) publishing two papers on a large-scale community intervention trial that show slightly increased rates of smoking cessation among light or moderate smokers but not among heavy smokers in intervention communities; (11) publishing data on survivors of childhood leukemia showing that those with cranial irradiation needed more special education training but graduated from high school at normal rates; (12) developing a model to project breast cancer that incorporates mammographic density, and devising convenient graphs for risk projection; (13) publishing attributable risk calculations for esophageal cancer in Shanghai, mesothelioma in the U.S. and stomach cancer in Northern Italy; (14) showing that rates of loss of CD4+ lymphocytes have not changed with calendar time of HIV infection; and (15) using back-calculation to estimate age-, race- and gender- specific prevalence and patterns of HIV in the U.S.
