Ecological Statistical Studies concerning Environmental Pollution and Chronic Disease

Abstract

Ecological statistical studies employing methods of multivariate analysis, based on a radiomimetic or mutagenic hypothesis, have yielded a number of statistically significant multiple regression equations in which concentrations of environmental chemicals, largely air pollutants, predict annual mortality rates for major categories of cancer and heart disease, as well as for congenital malformations, for populations of 38 metropolitan areas of the United States. Median age of these populations was also predicted statistically. Squares of multiple correlation coefficients R2 in excess of 0.5 were frequently obtained for these equations along with related t and F statistics of suitable magnitude. A new computer program for optimal regression analysis was employed in the studies. Among the chemical predictors whose atmospheric concentrations are frequently found positively correlated with mortality rates are SO2, NO2, and particulate sulfate. Among frequently recurring negatively correlated predictors are Cu, Cd, and Ti. Evidence regarding whether SO2 and NO2 may be considered as mutagenic hazards to life is discussed, as are some potentially relevant biochemical functions of the metals.