Concentration-response functions for short-term exposure and air pollution health effects

Abstract

In this work, we propose to use a new class of variable coefficient risk functions to represent the health effects of short-term exposure to air pollution. The presented concentration-response functions can adapt many forms of potentially nonlinear associations. These functions are suitable to more accurately represent risk in health impact assessments. The concentrations of air pollutants are transformed by a linear or a logarithmic function of concentrations multiplied by a logistic weighting function. A minimization process for nonlinear functions is applied to determine the model parameters. The proposed methodology is illustrated using four databases; among them are the Milan mortality (Italy) and the Chicago mortality (USA) databases. The results indicate the adequacy and reliability of the proposed methodology.