This Final Report is the product of the scheduled Provision of Service for the Study of Impact of Ambient Air Pollution on Public Health (Tender Ref. AS 96-60). The service is provided for the Environmental Protection Department, of the Hong Kong Government by the Department of Community and Family Medicine, The Chinese University of Hong Kong.

Recent international developments in the field of environmental health have contributed to the renewed recognition of air quality as a major public health issue (Guidotti, 1995). The World Health Organization estimates that at least a hundred million people world-wide are potentially affected by respiratory diseases associated with air pollution (Briggs, Corvalan & Nurminen, 1996). Early epidemiological studies clearly demonstrated the short-term health effects of high pollution levels and provided the basis for stricter air pollution controls in the 1970's. The lower levels of pollution that often resulted, redirected attention to chronic health effects in the 1980's and led to some degree of complacency about acute effects. However, the most recent research into short-term effects of air pollution have shown that there are significant health effects associated with these lower ambient air pollution levels, largely due to the ubiquity of exposure (Katsouyanni, 1996).

Studies of aggregate exposures to mixed air pollutants in urban settings have shown clear associations with respiratory diseases such as bronchitis and asthma, (Pope, Dockery & Schwartz, 1995; Roemer, Hoek & Brunekreef, 1993) with allergic diseases, (Rusznak, Devalia & Davies, 1994) cardiovascular diseases (Schwartz 1995; Schwartz & Morris, 1995) and with overall mortality (Dockery et al., 1993; Bates & Sizto, 1983; Touloumi, Pocock, Katsouyanni & Trichopoulos 1994; Samet, Zeger & Berhane, 1995). Similarly, adverse health outcomes associated with environmental exposure to specific air pollutants have been reasonably well-established in the epidemiological, toxicological and clinical experimental literature. Pollutants of concern include those being monitored routinely in Hong Kong by the Environmental Protection Department (Environmental Protection Department, 1994), namely, particulates, sulphur dioxide, nitrogen dioxide, carbon monoxide and ozone.

The actual disease burden attributable to air pollution exposure in Hong Kong is not known. As in other large cities, the pollution levels vary considerably with season. Environment and health data sets are poorly matched and methods for analyzing the relationships between them (which are still being developed and refined) have not been applied locally. Nevertheless, it is vital that attempts are made to link the data that are available in Hong Kong, to better understand the relationships between measured levels of ambient air pollution and actual health outcomes. Such an understanding is central to the task of refining the programmes of the Hong Kong Government that are designed to control exposures and protect the public's health. Local studies aimed at characterizing the complex associations between air pollution levels and public health outcomes serve mutually compatible public policy goals. First, they address the growing levels of awareness and concern by the general public about perceived health risks, over which they have no direct control. Second, they assist in the evaluation of local air pollution standards and guidelines as well as the relative costs associated with exposure to air pollution compared to the costs of air quality management. Overseas studies cannot be extrapolated and used for these purposes.

The linkage of routinely collected air quality and health outcome data (in the form of hospital admissions) has not been possible previously for several reasons. Foremost among these is that computerized medical records have only become available since 1994, after the formation of the Hospital Authority.

Time series analysis has been extensively applied in studies of air pollution and health (Schwartz et al., 1996; Katsouyanni et al., 1996). Standardized methods have been developed in the United States (Dockery et al., 1993; Samet et al., 1995) and Europe (Katsouyanni et al., 1996). Recent time series studies have found positive associations between individual air pollutants and acute health effects (Dockery et al., 1993; Brunekreef, Dockery & Krzyzanowski, 1995). To ascertain whether such associations can be found in the matched (air pollution / hospital admission) time series data in Hong Kong, the Environmental Protection Department has commissioned this baseline study. This Report represents the findings by the research team from the Department of Community and Family Medicine, The Chinese University of Hong Kong. As temporal data series are extended and improved in Hong Kong, the analytical power and usefulness of this approach will inevitably increase. This study thus forms a scientific basis for an on-going health and environmental surveillance system for improved decision-making and policy support.