In Response to the PM Debate
by George T. Wolff
As chairman of the EPAs Clean Air Scientific Advisory Committee during the recent review of the scientific basis for new fine particulate matter (PM) standards, I would like to comment on Crandall et al. and Brenner and Morgensterns interpretations of CASACs conclusions (last issue). The CASAC panels recommendations are summarized in two closure reports on the Criteria Document and Staff Paper, and they are available from the EPA. My comments here are based on those reports.
Crandall et al. accurately report that many of the CASAC panel members had "substantial doubts" about the causality issue. Four of the twenty-one members completely accepted a cause-and-effect relationship between fine particulate matter (PM2.5) and mortality at concentrations well below the present PM10 standards. They argued that the consistency and coherence of a large number of studies were compelling. Other members, to varying degrees, were influenced by numerous unanswered questions. Some felt that the EPA did not make a convincing case that PM2.5 was the causative fraction but accepted a PM10 causal effect. Because most air pollutants exhibit some degree of temporal correlation, some felt it was a general air pollution/mortality signal. Some felt that it may be some toxic component of PM or air pollution that would not necessarily be reduced if the total PM were targeted for reductions. Some were troubled by the lack of a plausible biological mechanism that could explain a PM/mortality relationship at such low concentrations. Others were concerned that there was a harvesting effect on terminally ill individuals. Some were concerned about the influence of exposure misclassification, measurement errors, and unaccounted confounders. Some felt that the reanalysis studies, many of which do not support a PM/mortality relationship, undermined the consistency and coherence arguments.
Brenner and Morgenstern state that it is incorrect that the CASAC panel rejected the broad scientific literature on PM mortality. If all of the panel members were convinced that the reported PM2.5/mortality relationship was causal, I believe we would have come to consensus on PM standards at the low end of the EPAs recommended range. The fact that we did not is because some members doubted the PM2.5/mortality causality while others were not convinced of an air pollution/mortality causality at the low concentrations reported in many of the studies. No one doubts the existence of a mortality/air pollution relationship in Donora in 1948 or London in the 1950s, but some panel members question a causality relationship at levels well below the present standards.
Brenner and Morgenstern state that CASAC "did reach agreement" on the Criteria Document. To be more accurate, CASAC reached "closure" on the Criteria Document. Closure means that the CASAC panel was convinced that the Criteria Document contained a comprehensive summary of the scientific studies relevant to setting a PM standard. This by no means implies that the panel members agreed with all of the cited studies or the EPAs interpretation of the studies. Clearly, there are panel members who do not agree with the quotation cited by Brenner and Morgenstern.
Brenner and Morgenstern mention the "strong longitudinal correlation between indoor and outdoor levels of fine particles." This was unconvincing to many CASAC panel members because of the numerous studies that showed poor or no correlation between personal exposure and ambient PM measured at the centrally located monitor that was used in most of the studies. Many also question how the outdoor concentrations relate to exposure in hospitals where a significant number of the deaths have occurred.
Finally, Brenner and Morgenstern state: "Crandall et al. are correct that no scientific consensus exists on the biological mechanism(s) by which fine particles cause premature deaths; however, this fact did not deter CASAC from reaching an agreement." It is not clear what they say CASAC agreed upon. CASAC did agree that a PM2.5 standard is appropriate, but there was no agreement on the level because of all of the unanswered questions that were raised. One of those questions was the lack of a plausible biological mechanism.
Crandall et al. reasonably interpreted CASACs recommendations. Brenner and Morgenstern, on the other hand, misinterpreted CASACs recommendations on a number of points.
Regulation is published four times a year by the Cato Institute. Editorial and business offices are located at 1000 Massachusetts Avenue, N.W., Washington, D.C., 20001. For subscription information, please write to Circulation Department, Cato Institute, same address, or call (202) 842-0200. Send email inquiries to firstname.lastname@example.org, or subscribe online via the World Wide Web at: http://www.cato.org/pubs/regulation/reg-ordr.html
| Regulation | Home | Publications |