Air pollution and acute respiratory illness: Evidence from Taiwan and Los Angeles

Developing and newly industrialized countries are suffering from severe air pollution problems. Growing concerns about both the effects of air pollution on health and the rising costs of continued efforts to improve air quality imply that these countries will increasingly need quantitative estimates of the health benefits and costs of reducing air pollution. The health benefits of improved air quality are usually obtained by combining epidemiologic evidence linking pollution levels to health outcomes with the value of avoiding such outcomes (Alberini and Krupnick, Cropper and Freeman). Unfortunately, very few original epidemiologic or willingness-to-pay studies have been conducted in developing countries, leading analysts to offer recommendations based on extrapolating both concentration-response functions and the value of avoiding illness from US studies (Krupnick et al., Ostro). However, this approach neglects differences between the US and the target country in pollution levels, baseline health, the age distribution of the population, medical care systems, sick-leave policies, and cultural factors that might affect perceptions of illness and pollution and behavioral responses. In this paper, we explore the appropriateness of concentration-response function transfers by comparing two health studies conducted following a similar format - but years apart - in Los Angeles and Taiwan. Daily records of acute respiratory symptoms, headaches, and eye irritation experienced by study participants were matched with pollution concentrations at multiple monitoring sites in Taiwan and a site in Azuza in Los Angeles (all participants were less than 1.5 miles from the Azuza site). The Los Angeles respondents report much higher 'background' illness rates and appear much more sensitive to fluctuations in population levels, even at pollution levels comparable to those for Taiwan, calling into question the validity of concentration-response function transfers. We believe that the different makeup and concentration of air pollutants in the two areas only partly explains the different responsiveness to air pollution.