The objectives of this project are to identify and describe environmental and host determinants of cancer through analytical, biostatistical, and epidemiologic approaches, including studies in areas at high risk of cancer, particularly case-control studies of specific cancers. Continuing investigation in several areas of the United States of risk factors for oral and esophageal cancer revealed that higher rates among blacks were linked to alcohol consumption, and for oral cancer little racial or sex difference in incidence was found among nonsmokers and nondrinkers. Data collection continued in a multicenter study of risk factors for adenocarcinoma of the esophagus, and analyses of existing case-control data revealed associations of esophageal adenocarcinoma with tobacco and alcohol (though not as strong as for squamous cell cancers) and with obesity. Work continued on assessing determinants of kidney cancer in Minnesota, where rates are exceptionally high. Data collection is ongoing for a large national case-control study evaluating risk of childhood leukemia associated with electromagnetic field exposures. Several studies of rare tumors were completed, with oral contraceptive use confirmed as a risk factor for liver tumors in young women. International collaborations were active during the year. Results from the randomized nutrition intervention trials in Linxian, China were reported. Supplementation with beta-carotene, vitamin E, and selenium resulted in a small but significant reduction in cancer mortality, the first report from a large clinical trial of cancer prevention from vitamin supplements. In another area of China, low levels of serum vitamin C and beta-carotene were associated with advanced precancerous lesions of the stomach. Occupational studies of radon and benzene continued in China; and cancer, census, and hospital discharge registries in Sweden and Denmark were linked to provide a series of clues to medical conditions and procedures that may predispose to cancer.
