Abstract

We conducted a health impact assessment of the government's proposed new air quality objectives (AQO) in Hong Kong, a high pollution environment. We based this on the World Health Organization (WHO) 2005 Air Quality Guidelines (AQG) and used a lognormal probability density function to model possible changes in annual mean pollutant levels resulting from the new AQO. All of the proposed short-term AQO were based on WHO interim targets (IT) or AQG, but allowed additional exceedances of these single limit values. Compliance with these short-term AQO may reduce (18-30%) the current annual mean concentrations but the distribution proportions exceeding the annual or annualized AQG remain high (83-100%). For SO2, the proposed 24-hr AQO of 125 µg m-3, with 3 days exceedances, cannot ensure reduction of the current annual mean but may legally permit an increase of the pollutant concentration distribution. If the proposed legal limits of AQO are fully exploited by polluters, we estimated the annual number of avoidable deaths at 1860, and avoidable health care events at 5.2 million doctor visits and 92745 hospital bed-days with a total annual community cost of US$2.6 billion. The proposed AQO may only reduce the current air pollution health impacts by 17% but could achieve 41% reduction if additional exceedances were not permitted. An epidemiologic approach should be adopted to assess external costs arising from modifications of WHOAQG and support accountability in air quality management. This analysis of the WHOAQG in a high pollution setting demonstrates problems arising from the absence of annual limits for some pollutants and the discordance between the short-term and annual AQG, suggesting that revisions based on a lognormal probability model should be considered.