The technical team of scientists
associated with A.S.L. & Associates completed its review
of the EPA Report: A Review of the Ambient Air Quality Standards
for Particulate Matter: Policy Assessment of Scientific and Technical
Information - OAQPS Staff Paper - First Draft. The comments
were submitted to EPA in October 2003.
The draft of the OAQPS
Staff Paper summarizes the fourth draft of the EPA PM Criteria
Document. The Staff Paper has attempted to translate the CD and
associated research on the health effects of PM into recommendations
for new regulatory limits for ambient PM. The Staff Paper, like
the CD, justifies its conclusions based on what it perceives
to be a certain coherence among recent epidemiologic studies
pertaining to PM. While identifying assorted deficiencies and
limitations of these studies and substantial unresolved uncertainties,
the EPA is apparently not concerned with these limitations and
uncertainties. In addition to drawing attention not only to the
limitations described in the Staff Paper and to other fundamental
unresolved problems associated with the corpus of PM epidemiology,
the technical team's comments question the logic of the EPA's
methodology for establishing the form and level of the PM2.5
standards.
In using time-series data for proposing
a form and level of the PM standard, the Staff Paper mainly focused
on the PM concentrations for selected cities where statistically
significant effects were observed in the epidemiological studies.
The minimum of these selected values was considered for the proposed
annual and short-term regulatory standards. The EPA assumed that
the city with the minimum annual PM average that showed a statistically
significant PM effect would presumably reduce this effect if
that city reduced its PM concentrations. Applying this methodology
for defining a regulatory limit ignores the (1) implications
of the actual magnitudes of the PM effect estimates, (2) implications
of no-threshold modeling, and (3) implications of city-to-city
heterogeneity of effect estimates.
In illustrating the problems in applying
the Agency's methodology for proposing the form and level of
the PM-2.5 standards based on the time-series data, the technical
team integrated the data from Appendix A of the Staff Paper with
air characterization information on PM-2.5 concentrations from
the Ross (2003) memo. Using the data provided for Total (nonaccidental)
Mortality and Total Cardiovascular Mortality, no relationship
was evident between the concentrations as listed in Ross (2003)
and human health endpoint data as provided in Appendix A.
In order to establish a scientifically
defensible set of standards for PM-2.5, the Agency should have
first established dose-response relationships upon which the
proposal for the form and level of the two PM-2.5 standards would
have been based. Unfortunately, the Agency has chosen to utilize
a methodology that has no biological connection to the epidemiology
studies. Therefore, serious questions exist concerning the methodology
utilized by the Agency to define the form and level of the PM-2.5
annual mean and 24-hour standards.
