Is SO2 a causative factor for the PM-associated mortality risks in the Netherlands

Associations between serious health risks and PM have been found in numerous studies, including studies in the Netherlands (Verhoeff et al., 1996). More recent European studies have also found associations with gaseous components (Katsouyanni er ai., 1997; Hoek et ai., 1997), of which SO2 is one of the gasses. A recent report in the United Kingdom (COMEAP, 1998) concludes that in ambient air SO2 leads to an increase in total mortality of 0.6% per 10 mu g/m(3). Although these statistical associations have been found it remains questionable as to whether or not the associations are causal. A careful analysis of a 9-yr Dutch time series (Hoek et al., 1997) by successive exclusion of the highest concentrations indicates that SO2 is probably not causally associated with the health effects, but that it is correlated. A separate analysis of the mortality over different 3-yr periods indicates that in the first 3 yr SO2 led to a significantly lower relative risk than in the last 3 yr, which had the lowest SO2 concentrations. The conclusion that in the Netherlands SO2 does not seem to be a causative factor for PM associated health effects is substantiated by further circumstantial evidence, in combination with biological arguments, indicating that a factor correlating with SO2 (probably PM) might explain the observed associations with total mortality.