We propose to investigate the hypothesis that various cancers (including lung, breast, colorectal, liver, stomach, esophageal, cervix and leukemia), in addition to their specific causes, have a common nutritional etiology based on the consumption of a diet enriched in animal products. This investigation will use data obtained from a series of cross-sectional surveys in 138 villages (2 in each of 69 counties) in the People's Republic of China (PRC) and 30 villages (2 in each of 15 counties) in Taiwan (funded elsewhere). Four sets of data are, or will be, available from these surveys in the PRC. First, age-, sex-specific mortality rates for 1973-75 for 15 different cancers (and 30 non-cancers) are already available. Second, a more comprehensive and standardized set of mortality data for 1986-88 for about 20 different cancers and 130 non-cancers (3-digit ICD-9 coding) will be available for compilation by early 1991. Third, data are available from a survey of dietary and lifestyle characteristics undertaken in 65 PRC counties in 1983-84 that included a broad range of nutritional variables comprised of dietary intakes (foods and nutrients), blood and urine biochemical indicators of nutritional status, recall information (questionnaire on frequency of food consumption, smoking habits, alcohol consumption and demographic factors) and chemical analysis of foods consumed. Fourth, an updated dietary and lifestyle characteristics survey conducted in 1989 in 69 PRC counties (4 additional counties included with the 65 original counties) will provide information not obtained in 1983 (blood pressure, lung function, childhood anthropometry, socioeconomic characteristics). This latter survey also provided blood and urine samples for analysis of biochemical markers useful for the testing of this hypothesis, as well as others to be specified in other applications. We propose in this application to investigate the above hypothesis by analyzing both the existing data and, after chemical analysis, the blood and urine samples from the 1989 survey for indicators of dietary enrichment. As part of this investigation and in an effort to complete the data processing for the 1989 surveys, we will code, key-punch and computer program data from 1.4 million death certificates and approximately 8280 lengthy questionnaires, and prepare for future storage about 160,000 portions of 8280 blood and urine samples (i.e., prepare county-, age-, sex-specific pools of aliquots of individual samples and store remaining individual samples) and calculate the intakes of nutrients and other dietary constituents from 8280 diet records. (Another 1800 comparable questionnaires and biological specimens will be similarly processed in Taiwan.)
