POPULATION-DENSITY AND CANCER MORTALITY DIFFERENTIALS IN NEW-YORK-STATE, 1978-1982

Mahoney M C (Bureau of Cancer Epidemiology, New York State Department of Health, Albany, New York, USA), LaBrie D S, Nasca P C, Wolfgang P E and Burnett W S. Population density and cancer mortality differentials in New York State, 1978-1982. International Journal of Epidemiology 1990;19: 483-490.Patterns of cancer mortality within five population density quintiles in Upstate New York (New York State, exclusive of New York City), were investigated between 1978 and 1982. Sex-specific standardized mortality ratios (SMRs) were calculated within each population density quintile for all cancer deaths combined and for site-specific cancer deaths based on cancer mortality patterns exhibited by the general population of New York State, exclusive of New York City. Areas with the highest population density demonstrated a 12% excess of cancer deaths among males and a 6% excess among females. In contrast, areas with the lowest population density exhibited lower cancer mortality, among both males (9% less) and females (7% less). Males demonstrated a significant linear relationship between increasing population density and deaths for all cancer sites combined and for cancers of the oral cavity and pharynx, oesophagus, stomach, colon, gallbladder, pancreas, lung, prostate, and kidney. Among females, a significant linear relationship was observed between increasing population density and deaths for all cancer sites combined and for deaths due to cancers of the stomach, colon, liver and breast. Deaths due to cancers of the rectum, malignant melanomas of the skin, and cervix also exhibited unusual patterns of mortality across the population density quintiles. These data are most useful in generating hypotheses for further studies to define specific aetiological factors operating within population density groupings. Population density, as measured in this investigation, may represent a surrogate measure for other factors which are related to cancer morbidity and cancer mortality. It is also possible that differences in cancer mortality across population density quintiles may result from an interaction of differences in diagnostic methods, medical treatment, access to care, and survival. © 1990 International Epidemiological Association.