Background on health effects of acid aerosols.

This introduction to the 1987 NIEHS-EPA Symposium on the Health Effects of Acid Aerosols reviews the state of our knowledge on this topic as of the close of the 1984 NIEHS Conference on the Health Effects of Acid Precipitation (Environmental Health Perspectives, Volume 63) and the results of some key studies completed since that time. These studies, together with the results of the studies presented in the papers that follow, provide a substantial increment in our knowledge of the health effects of acid aerosols.


Introduction
This Symposium on the Health Effects of Acid Aerosols is taking place 3 years after the NIEHS Conference on Health Effects of Acid Precipitation. The earlier conference, also convened at the NIEHS, covered the effects of acid precipitation on mobilization, transport, and speciation of toxic metals, the impact of acid precipitation on metals toxicity, and the health effects of inhaled acid aerosols. Its proceedings were published in Environmental Health Perspectives, Volume 63. This symposium was organized to help the sponsors (NIEHS and EPA), the larger scientific community, and Congress address and assess the expanding data base on the effects of inhaled acid aerosols.
To the extent that we can place the emerging exposure and effects data into a broader perspective on the public health consequences of current and future exposures, we can help the sponsors identify data gaps, research needs, and the possible need for a primary national ambient air quality standard for acid aerosols. To this end, we depend on the skills and insights of the rapporteurs of our half-day sessions on Exposure, Epidemiology, Toxicology, and Acute H u m a n Responses to present integrations and interpretations of the presentations in their sessions. We will also benefit from an evaluation of the potential uses and limitations of the current data base on acidic aerosols for risk assessment by Roy E. Albert (1). His broad perspectives, combined with his experience in experimental studies on the effects of inhaled irritants, environmental epidemiology, and quantitative risk assessment, enable him to evaluate the data base from a broad perspective. *New York University Medical Center, Institute of Environmental Medicine, Tuxedo, NY 10987.

Historical Background
Industrial pollution from steel mills and sulfuric acid (H2SO4) plants were implicated in excess deaths and respiratory illness among elderly people in The Meuse Valley in Belgium in December 1930, and in Donora, PA, in October 1948. Coal smoke containing H2SO4 was implicated in excess deaths, primarily in elderly people in a series of episodes in London from 1952 to 1962, with approximately 4000 excess deaths from the December 1952 episode alone. During that period, deaths attributable to chronic bronchitis were higher in London than in other, less polluted areas in the U.K., and much higher than in other countries in northern Europe. Aggravation of bronchitis among patients in London with chronic lung disease was also demonstrated to occur as pollution levels increased. Unfortunately, levels of exposure to H2SO4 were not determined in the era when excessive mortality and morbidity were clearly evident, and exposure-response relationships were made, if at all, with pollutant indices referred to as British Smoke (BS) and sulfur dioxide (SO2). The actual measures were the blackness of sampled particles (BS) and the net acidity of sampled vapor (SO2).
It is now clear that excess mortality in London was much more closely associated with BS than with SO2. However, BS is primarily a measure of the carbon content of the aerosol, an unlikely causal factor. It seems much more likely that another aerosol constituent, e.g., the free hydrogen ion (H+), was a causal factor.
Many of the reasons for the interest in and concern about H+, or H2SO4 as one dominant source of ambient H+, were discussed at the 1984 NIEHS conference. My assignment at that conference was rapporteur for the session on the health effects of inhaled acid aerosols. Table 1 from the proceedings of that conference indicates the questions I had formulated prior to the conference. It also presents the answers I could come up with by the end of the conference, the conditions under which they were applicable, and the remaining knowledge gaps at that time. My responses were based, in large measure, on material presented at the conference, but were supplemented, as appropriate, with knowledge of other relevant work (2).

Recent Progress
Three years is not a long time for progress in research on so broad and complex an issue as the health effects of acid aerosols, especially when there have been so few active investigators and resources available to address the knowledge gaps. Most of the active research groups are represented at this symposium and will be presenting their recent work to us. It is therefore too early to fully gauge the progress that has been made, although much of it is significant and moves our knowledge base well past that of 3 years ago, as summarized in Table 1. Most of the recent research provides exposure-response data involving directly measured acid aerosol exposures. However, we are not yet at the stage where we can afford to ignore studies with surrogate measures of exposure. Reports from Germany and Poland at this symposium are good examples of papers of great interest that still rely on surrogate metrics of exposure.
Other recent work not being reported at this sym-  posium should also be noted. In particular, I would like to call your attention to papers by Dassen et
(3) measured pulmonary function in primary school children in the Netherlands before, during, and after the January 1985 acidic pollution event that caused elevated exposures throughout northwestern Europe. Reports of studies examining population responses in England and West Germany will be presented in Session II of this Symposium. During this episode, the Dutch children were exposed to respirable particulate and SO2 concentrations in the range of 200 to 250 gg/m3; their baseline functions were measured about 4 to 6 weeks earlier when the pollutants were below 100 ,ug/m3. During the episode, function indices were significantly lower by 3 to 5%, and decrements were still present at 16 days later, but not at 25 days. Some of the results are summarized in   U   II I I I I I I I I I I I I I I I I  Metropolitan Statistical Areas (SMSAs) and four measures of particulate air pollution. These were total suspended particulate matter (TSP); inhalable particulate matter, i.e., particulate < 15 gm in aerodynamic median diameter (IHP); fine particulate matter, i.e., particulate < 2.5 ,um in aerodynamic median diameter (FP); and sulfate (SO2-), a major component of FP. This was a final report on the preliminary analyses reported by Ozkaynak and Spengler (5) at the 1984 NIEHS Conference. They found that FP and SO 4were most consistently and significantly associated with the reported SMSA-specific total annual mortality rates, whereas TSP and IP were often nonsignificant predictors of mortality. These results are summarized in Table 3 and Figure 1. Figure 2 is my adaptation of the results of the SMSA-mortality analyses, showing a hy-pothetical extrapolation of the analytical results based upon the hypothesis that the acid component of the particulate, a greater fraction as one goes from TSP to IP to FP to S02-, is the causal factor. The results reported at this symposium provide a basis for examining the validity of this hypothesis, and/or for helping to establish a better one.