"Flynn, Amy E." <Amy_E._Flynn@omb.eop.gov> 

09/13/2006 02:02 PM

	

To

Lydia Wegman/RTP/USEPA/US@EPA

cc

Subject

comments--PM NAAQS Preamble









Lydia,

 

Please find below comments from DOE on the PM NAAQS Preamble.  Please
let me know if you have any questions.  Thank you, --Amy

1. Pg. 41 (footnote 11)

This footnote is not inaccurate, but could be more complete.  Propose
adding the following language at end of the current footnote 11 as it
runs over to pg. 41:

“The 98th percentile PM level in Fairley et al (2003) was 59 g/m3;
in Burnett and Goldberg (2003) it was 38.9 g/m3; and in Mar et al
(2003) it was 32.2 g/m3.  Burnett et al (2000) found in
multi-pollutant analysis that gases appeared to be more important than
PM measures in effects; the Burnett and Goldberg 2003 reanalysis of this
earlier study didn’t perform multipollutant analysis with gases.  Mar
et al (2003) analyzed many gases separately, but not in multi-pollutant
analyses.”

2.  The Krewski et al (2000) reanalysis of the original ACS study is
mentioned at the bottom of pg. 47, running over to pg. 48, and on pg.
63.  The Pope et al (2002) ACS study update is mentioned on pg. 41 and
pg. 48.   Both are mentioned several other places as well.  In light of
the extensive discussion in the Jan. 17 Preamble to the rule of Krewski
et al (2000) finding that inclusion of SO2 in the model causes fine PM
and sulfate to become insignificant, and the tangential reference to
this issue in the present document, suggest a footnote after the first
mention of Pope et al (2002) on the third line of pg. 48:  

“The Pope et al (2002) update of the original ACS study did not report
findings with regard to inclusion of SO2 in model runs, together with
fine PM or with sulfate, as in Krewski et al (2000).” 

3.  At the bottom of pg. 63, the following statement is not inaccurate,
but relevant information is not included:

For example, while the association was stronger for the subset of the
ACS cohort with the least education, there was an association with
cardiorespiratory mortality in the entire population.

We suggest that the following sentence be substituted for the above
sentence from pg. 63:

For example, there was an association with cardiorespiratory mortality
in the entire population when people from all education levels were
combined; the association was significant and positive for the subset of
the ACS cohort with the least education, but was insignificant and
virtually zero for those with better than a high school education.  

4. Also on pg. 63, mention is made of the findings of Krewski et al
(2000) with regard to spatial correlation.  What this means should be
spelled out.  

5.  On pg. 74, the Mar et al (2003) findings with regard to sulfate are
mentioned in line 3.  Suggest the following clarifying footnote:

The Mar et al (2000) study noted that sulfate alone in a single
pollutant model was not associated with cardiovascular mortality, but
that the sulfate “factor,” which was so associated, contained
elevated levels of lead and bromine.  The authors state that the health
association with the sulfate factor “may be reflective of the
contribution of Pb and Br to the S factor.”  Mar et al (2003) did not
provide information about single pollutant model analysis of sulfate or
about contribution of Pb and Br to the S factor.  

6.  On pg. 96, there is discussion of the Burnett and Goldberg (2003)
study.  Suggest the following clarifying footnote:

Burnett and Goldberg (2003) is a reanalysis of Burnett et al (2000). 
The earlier study analyzed gases and PM in multipollutant models and
found that gases had larger effects than fine PM.  Burnett and Goldberg
(2003) did not analyze gases or do multipollutant analysis with gases.

7.  On pg. 11, we suggest the following footnote to the sentence
discussing the updated ACS study’s findings of lung cancer
associations (the sentence ending in “in the original studies”):

“In the extended ACS study, significant lung cancer associations were
found for those with high school education or less, but not for those
with better than a HS education.  When data is combined for all
educational levels, a significant association is found.  These are the
same patterns found for all-cause and cardiopulmonary mortality.” 
[The last sentence would be informative, but optional.]

8. Pg. 128, in regard to the Children’s Health Study in Southern
California.  Gauderman et al (2004) state the following, after stating
that the lung deficits were associated with a correlated set of
pollutants (NO2, acid vapor, fine PM, and EC):

î

ð

%

&

E

F

U

V

Œ

Wdue to vehicular emissions, with regard to the pollutants discussed at
the bottom of pg. 128.  This appears to buttress the point made on the
next page with regard to the study of Kim et al (2004).

