Pm2.5 Epa Cfr50

8/2/2019 Pm2.5 Epa Cfr50

1/102

fed

era

lre

gister

1

FridayJuly 18, 1997

Part II

EnvironmentalProtection Agency40 CFR Part 50

National Ambient Air Quality Standardsfor Particulate Matter; Final Rule

8/2/2019 Pm2.5 Epa Cfr50

2/102

2 Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication

ENVIRONMENTAL PROTECTIONAGENCY

40 CFR Part 50

[ADFRL57252]

RIN 2060AE66

National Ambient Air Quality

Standards for Particulate Matter

AGENCY: Environmental Protection Agency(EPA).ACTION: Final rule.

SUMMARY: This document describes EPAsdecision to revise the national ambient airquality standards (NAAQS) for particulatematter (PM) based on its review of theavailable scientific evidence linkingexposures to ambient PM to adverse healthand welfare effects at levels allowed by thecurrent PM standards. The current primaryPM standards are revised in several respects:Two new PM2.5 standards are added, set at

15 g/m3 , based on the 3-year average ofannual arithmetic mean PM2.5 concentrationsfrom single or multiple community-orientedmonitors, and 65 g/m 3 , based on the 3-year average of the 98th percentile of 24-hourPM2.5 concentrations at each population-oriented monitor within an area; and thecurrent 24-hour PM10 standard is revised tobe based on the 99th percentile of 24-hourPM10 concentrations at each monitor withinan area. The new suite of primary standardswill provide increased protection against awide range of PM-related health effects,including premature mortality and increasedhospital admissions and emergency room

visits, primarily in the elderly and individualswith cardiopulmonary disease; increasedrespiratory symptoms and disease, in childrenand individuals with cardiopulmonary diseasesuch as asthma; decreased lung function,particularly in children and individuals withasthma; and alterations in lung tissue andstructure and in respiratory tract defensemechanisms. The current secondary standardsare revised by making them identical in allrespects to the new suite of primarystandards. The new secondary standards, inconjunction with a regional haze program,will provide appropriate protection againstPM-related public welfare effects includingsoiling, material damage, and visibilityimpairment. In conjunction with the newPM2.5 standards, a new reference method hasbeen specified for monitoring PM as PM2.5.EFFECTIVE DATD: This action is effectiveSeptember 16, 1997.ADDRESSES: A docket containinginformation relating to the EPAs review ofthe PM primary and secondary standards(Docket No. A9554) is available for publicinspection in the Central Docket Section ofthe U.S. Environmental Protection Agency,South Conference Center, Rm. 4, 401 M St.,

SW., Washington, DC. This docketincorporates the docket established for the airquality Criteria Document (Docket No.ECAOCD920671). The docket may beinspected between 8 a.m. and 3 p.m., Mondaythrough Friday, except legal holidays, and areasonable fee may be charged for copying.The information in the docket constitutes thecomplete basis for the decision announced in

this document. For the availability of relatedinformation, see SUPPLEMENTARYINFORMATION.FOR FURTHER INFORMATION CONTACT: JohnH. Haines, MD15, Air Quality Strategiesand Standards Division, Office of Air QualityPlanning and Standards, U.S. EnvironmentalProtection Agency, Research Triangle Park,NC 27711; telephone: (919) 5415533; e-mail: [email protected] INFORMATION:

Related Final Rules on PM Monitoring

In a separate document publishedelsewhere in this issue of the Federal

Register, EPA is amending its ambient airquality surveillance requirements (40 CFRpart 58) and its ambient air monitoringreference and equivalent methods (40 CFRpart 53) for PM.

Availability of Related Information

Certain documents are available from theU.S. Department of Commerce, NationalTechnical Information Service, 5285 PortRoyal Road, Springfield, VA 22161.Available documents include:

(1) Air Quality Criteria for ParticulateMatter (Criteria Document) (three volumes,EPA/600/P95001aF thru EPA/600/P95001cF, April 1996, NTIS #PB96168224,

$234.00 paper copy).(2) Review of the National Ambient Air

Quality Standards for Particulate Matter:Policy Assessment of Scientific andTechnical Information (Staff Paper) (EPA452/R96013, July 1996, NTIS #PB97115406, $47.00 paper copy and $19.50microfiche). (Add a $3.00 handling chargeper order.)

A limited number of copies of otherdocuments generated in connection with thisstandard review, such as technical supportdocuments pertaining to air quality,monitoring, and health risk assessment, canbe obtained from: Environmental Protection

Agency Library (MD35), Research TrianglePark, NC 27711, telephone (919) 5412777.These and other related documents are alsoavailable for inspection and copying in theEPA docket at the address underADDRESSES, at the beginning of thisdocument.

Electronic Availability

The Staff Paper and human health riskassessment support documents are availableon the Agencys Office of Air QualityPlanning and Standards (OAQPS)Technology Transfer Network (TTN) Bulletin

Board System (BBS) in the Clean Air ActAmendments area, under Title I, Policy/Guidance Documents. To access the bulletinboard, a modem and communicationssoftware are necessary. To dial up, set yourcommunications software to 8 data bits, noparity and one stop bit. Dial (919) 5415742and follow the on-screen instructions toregister for access. After registering, proceed

to choice Gateway to TTN TechnicalAreas, then choose CAAA BBS.From the main menu, choose Title I:Attain/Maint of NAAQS, then

PolicyGuidance Documents. To access thesedocuments through the World Wide Web,click on TTN BBSWeb, then proceed tothe Gateway to TTN Technical areas, asabove. If assistance is needed in accessing thesystem, call the help desk at (919) 5415384in Research Triangle Park, NC.

Implementation Strategy for Revised AirQuality Standards

On Wednesday, July 16, 1997, President

Clinton signed a memorandum to theAdministrator specifying his goals for theimplementation of the O3 and PM standards.Attached to the Presidents memorandum isa strategy prepared by an interagencyAdministration group outlining the next stepsthat would be necessary for implementingthese standards. The EPA will prepareguidance and proposed rules consistent withthe Presidents memorandum. Copies of thePresidential document are available in papercopy by contacting the U.S. EnvironmentalProtection Agency Library at the addressunder Availability of Related Informationand in electronic form as discussed above in

Electronic Availability.The following topics are discussed in this

preamble:I. Background

A. Legislative Requirements

B. Related Control Requirements

C. Review of Air Quality Criteria and Standards

for PM

D. Summary of Proposed Revisions to the PM

Standards

II. Rationale for the Primary PM Standards

A. Introduction

B. Need for Revision of the Current Primary PM

Standards

C. Indicators of PM

D. Averaging Time of PM2.5 Standards

E. Form of PM2.5 StandardsF. Levels for the Annual and 24-Hour PM2.5

Standards

G. Conclusions Regarding the Current PM10Standards

H. Final Decisions on Primary PM Standards

III. Rationale for the Secondary Standards

A. Need for Revision of the Current

SecondaryStandards

B. Decision on the Secondary StandardsIV. Other Issues

A. Consideration of CostsB. Margin of Safety

C. Data Availability

8/2/2019 Pm2.5 Epa Cfr50

3/102

3Federal Register / Vol. 62, No. 138 / Friday, July 18, 1997 / Prepublication

1 PM10 refers to particles with an aerodynamic diameterless than or equal to a nominal 10 micrometers. Technicaldetails further specifying the measurement of PM10 arecontained in 40 CFR part 50, Appendices J and M.

2 A more complete history of the PM NAAQS ispresented in section II.B of the OAQPS Staff Paper,Review of National Ambient Air Quality Standards forParticulate Matter: Assessment of Scientific and TechnicalInformation (U.S. EPA, 1996b).

3 A court order entered inAmerican Lung Associationv.Browner, CIV93643TUCACM (D. Ariz.,October 6,1994), as subsequently modified, requires publication ofEPAs final decision on the review of the PM NAAQSby July 19, 1997.

D. 1990 AmendmentsV. Revisions to 40 CFR Part 50, Appendix K

Interpretation of the PM NAAQSA. PM2.5 Computations and Data Handling

ConventionsB. PM10 Computations and Data Handling

ConventionsC. Changes that Apply to Both PM2.5 and PM10

ComputationsVI. Reference Methods for the Determination of

Particulate Matter as PM10 and PM2.5 in theAtmosphere

A. Revisions to 40 CFR Part 50, Appendix JReference Method for PM10

B. 40 CFR Part 50, Appendix LNewReference Method for PM2.5

VII. Effective Date of the Revised PM Standardsand Applicability of the Existing PM10Standards

VIII. Regulatory and Environmental ImpactAnalyses

A. Executive Order 12866B. Regulatory Flexibility AnalysisC. Impact on Reporting RequirementsD. Unfunded Mandates Reform ActE. Environmental JusticeF. Submission to Congress and the Comptroller

GeneralIX. Response to Petition for Administrator

Browners RecusalX. References

I. Background

A. Legislative Requirements

Two sections of the Clean Air Act (Act)govern the establishment, review, andrevision of NAAQS. Section 108 of the Act(42 U.S.C. 7408) directs the Administrator toidentify certain pollutants which mayreasonably be anticipated to endanger publichealth and welfare and to issue air qualitycriteria for them. These air quality criteria are

to accurately reflect the latest scientificknowledge useful in indicating the kind andextent of all identifiable effects on publichealth or welfare which may be expectedfrom the presence of [a] pollutant in theambient air * * *.

Section 109 of the Act (42 U.S.C. 7409)directs the Administrator to propose andpromulgate primary and secondaryNAAQS for pollutants identified undersection 108 of the Act. Section 109(b)(1) ofthe Act defines a primary standard as onethe attainment and maintenance of which inthe judgment of the Administrator, based on[the] criteria and allowing an adequate marginof safety, are requisite to protect the publichealth. The margin of safety requirementwas intended to address uncertaintiesassociated with inconclusive scientific andtechnical information available at the time ofstandard setting, as well as to provide areasonable degree of protection againsthazards that research has not yet identified.Both kinds of uncertainties are components ofthe risk associated with pollution at levelsbelow those at which human health effectscan be said to occur with reasonable scientificcertainty. Thus, by selecting primarystandards that provide an adequate margin of

safety, the Administrator is seeking not onlyto prevent pollution levels that have beendemonstrated to be harmful but also toprevent lower pollutant levels that she findsmay pose an unacceptable risk of harm, evenif the risk is not precisely identified as tonature or degree. The Act does not require theAdministrator to establish a primary NAAQSat a zero-risk level, but rather at a level that

reduces risk sufficiently so as to protectpublic health with an adequate margin ofsafety. The selection of any particularapproach to providing an adequate margin ofsafety is a policy choice left specifically tothe Administrators judgment.Lead

Industries Assn v.EPA, 647 F.2d 1130,11611162 (D.C. Cir.1980).

A secondary standard, as defined in section109 (b)(2) of the Act, must specify a levelof air quality the attainment and maintenanceof which in the judgment of theAdministrator, based on [the] criteria, [are]requisite to protect the public welfare fromany known or anticipated adverse effects

associated with the presence of [the] pollutantin the ambient air. Welfare effects asdefined in section 302(h) of the Act (42U.S.C. 7602(h)) include, but are not limitedto, effects on soils, water, crops, vegetation,manmade materials, animals, wildlife,weather, visibility, and climate, damage toand deterioration of property, and hazards totransportation, as well as effects on economicvalues and on personal comfort and well-being.

Section 109(d)(1) of the Act requiresperiodic review and, if appropriate, revisionof existing air quality criteria and NAAQS.Section 109(d)(2) of the Act requiresappointment of an independent scientific

review committee to review criteria andstandards and recommend new standards orrevisions of existing criteria and standards, asappropriate. The committee established undersection 109(d)(2) of the Act is known as theClean Air Scientific Advisory Committee(CASAC), a standing committee of EPAsScience Advisory Board.

B. Related Control Requirements

States are primarily responsible forensuring attainment and maintenance ofambient air quality standards once EPA hasestablished them. Under section 110 of theAct (42 U.S.C. 7410) and related provisions,States are to submit, for EPA approval, Stateimplementation plans (SIPs) that provide forthe attainment and maintenance of suchstandards through control programs directedto sources of the pollutants involved. TheStates, in conjunction with EPA, alsoadminister the prevention of significantdeterioration program (42 U.S.C. 74707479)for these pollutants. In addition, Federalprograms provide for nationwide reductionsin emissions of these and other air pollutantsthrough the Federal Motor Vehicle ControlProgram under Title II of the Act (42 U.S.C.75217574), which involves controls for

automobile, truck, bus, motorcycle, nonroadengine, and aircraft emissions; the new sourceperformance standards under section 111 ofthe Act (42 U.S.C. 7411); and the nationalemission standards for hazardous airpollutants under section 112 of the Act (42U.S.C. 7412).

C. Review of Air Quality Criteria and

Standards for PMParticulate matter is the generic term for a

broad class of chemically and physicallydiverse substances that exist as discreteparticles (liquid droplets or solids) over awide range of sizes. Particles originate froma variety of anthropogenic stationary andmobile sources as well as from naturalsources. Particles may be emitted directly orformed in the atmosphere by transformationsof gaseous emissions such as sulfur oxides(SOx), nitrogen oxides (NOx), and volatileorganic compounds (VOC). The chemical andphysical properties of PM vary greatly withtime, region, meteorology, and source

category, thus complicating the assessment ofhealth and welfare effects.The last review of PM air quality criteria

and standards was completed in July 1987with notice of a final decision to revise theexisting standards published in the FederalRegister (52 FR 24854, July 1, 1987). In thatdecision, EPA changed the indicator for PMfrom total suspended particles (TSP) toPM10.1 Identical primary and secondary PM10standards were set for two averaging times:50 g/m3, expected annual arithmetic mean,averaged over 3 years, and 150 g/m3, 24-hour average, with no more than one expectedexceedance per year.2

The EPA initiated this current review of

the air quality criteria and standards for PMin April 1994 by announcing its intention todevelop a revised Air Quality CriteriaDocument for Particulate Matter (henceforth,the Criteria Document). Thereafter, theEPA presented its plans for review of thecriteria and standards for PM under a highlyaccelerated, court-ordered schedule3 at apublic meeting of the CASAC in December1994. Several workshops were held by EPAsNational Center for EnvironmentalAssessment (NCEA) to discuss importantnew health effects information in November1994 and January 1995. External reviewdrafts of the Criteria Document were made

available for public comment and were

8/2/2019 Pm2.5 Epa Cfr50

4/102


4 The Staff Paper evaluates policy implications of thekey studies and scientific information in the CriteriaDocument, identifies critical elements that EPA staffbelieves should be considered, and presents staffconclusions and recommendations of suggested options forthe Administrators consideration.

5 PM2.5 refers to particles with an aerodynamic diameterless than or equal to a nominal 2.5 micrometers, as furtherspecified in 40 CFR part 50, Appendix L in this document.

6 PM102.5 refers to those particles with an aerodynamicdiameter less than or equal to a nominal 10 micrometersbut greater than 2.5 micrometers. In other words, it refersto the inhalable particles that remain if fine (PM2.5)particles are removed from a sample of PM10 particles.

reviewed by CASAC at public meetings heldin August and December 1995 and February1996. The CASAC came to closure in itsreview of the Criteria Document, advising theAdministrator in a March 15, 1996 closureletter (Wolff, 1996a) that although ourunderstanding of the health effects of PM isfar from complete, a revised CriteriaDocument which incorporates the Panels

latest comments will provide an adequatereview of the available scientific data andrelevant studies of PM. CASAC and publiccomments from these meetings, and fromsubsequent written comments and the closureletter, were incorporated as appropriate in thefinal Criteria Document (U.S. EPA, 1996a).

External review drafts of a Staff Paperprepared by the Office of Air QualityPlanning and Standards (OAQPS), Review ofthe National Ambient Air Quality Standardsfor Particulate Matter: Assessment ofScientific and Technical Information(henceforth, the Staff Paper), were madeavailable for public comment and were

reviewed by CASAC at public meetings inDecember 1995 and May 1996.4 The CASACcame to closure in its review of the StaffPaper, advising the Administrator in a June13, 1996 closure letter (Wolff, 1996b) thatthe Staff Paper, when revised, will providean adequate summary of our presentunderstanding of the scientific basis formaking regulatory decisions concerning PMstandards. CASAC and public commentsfrom these meetings, subsequent writtencomments, and the CASAC closure letterwere incorporated as appropriate in the finalStaff Paper (U.S. EPA, 1996b).

On November 27, 1996, EPA announcedits proposed decision to revise the NAAQS

for PM (61 FR 65638, December 13, 1996)(hereafter proposal) as well as its proposeddecision to revise the NAAQS for ozone(O3)(61 FR 65716, December 13, 1996). Inthe proposal, EPA identified proposedrevisions, based on the air quality criteria forPM, and solicited public comments onalternative primary standards and on theproposed forms of the standards.

To ensure the broadest possible publicinput on the PM and O3 proposals, EPA tookextensive and unprecedented steps tofacilitate the public comment process beyondthe normal process of providing anopportunity to request a hearing and receiving

written comments submitted to therulemaking docket. The EPA established anational toll-free telephone hotline tofacilitate public comments on the proposedrevisions to the PM and O3 NAAQS, and onrelated notices dealing with theimplementation of revised PM and O3standards, as well as a system for the public

to submit comments on the proposalselectronically via the Internet. Over 14,000calls and over 4,000 electronic mail messageswere received through these channels. Thepublic could also access key supportingdocuments (including the Criteria Document,Staff Paper, related technical documents andfact sheets) via the Internet.

The EPA also held several public hearings

and meetings across the country to providedirect opportunities for public comment onthe proposed revisions to the PM and O3NAAQS and to disseminate information tothe public about the proposed standardrevisions. On January 14 and 15, 1997, EPAheld concurrent, 2-day public hearings inBoston, MA, Chicago, IL, and Salt Lake City,UT. A fourth public hearing, which focusedprimarily on PM monitoring issues, was heldin Durham, NC on January 14, 1997. Over400 citizens and organizations testified duringthese public hearings. EPA also held twonational satellite telecasts to answer questionson the standards and participated in meetings

sponsored by the Air and Waste ManagementAssociation on the proposed revisions to thestandards at more than 10 locations across thecountry. Beyond that, several EPA regionaloffices held public meetings and workshopsand participated in hearings that States andcities held around the country.

As a result of this intensive effort to solicitpublic input, over 50,000 written and oralcomments were received on the proposedrevisions to the PM NAAQS by the close ofthe public comment period on March 12,1997. Major issues raised in the commentsare discussed throughout the preamble of thisfinal decision. A comprehensive summary ofall significant comments, along with EPAs

response to such comments (hereafterResponse to Comments), can be found inthe docket for this rulemaking (Docket No.A9554).

The principal focus of this current reviewof the air quality criteria and standards forPM is on recent epidemiological evidencereporting associations between ambientconcentrations of PM and a range of serioushealth effects. Particular attention has beengiven to several size-specific classes ofparticles, including PM10 and the principalfractions of PM10, referred to as the fine(PM2.5)5 and coarse (PM102.5)6 fractions. Asdiscussed in the Criteria Document, fine and

coarse fraction particles can be differentiatedby their sources and formation processes andtheir chemical and physical properties,including behavior in the atmosphere.Detailed discussions of atmosphericformation, ambient concentrations, and health

and welfare effects of PM, as well asquantitative estimates of human health risksassociated with exposure to PM, can be foundin the Criteria Document and in the StaffPaper.

D. Summary of Proposed Revisions to the PMStandards

For reasons discussed in the proposal, the

Administrator proposed to revise the currentprimary standards for PM (as indicated byPM10), by adding two new primary PM2.5standards set at 15 g/m3, annual mean, and50 g/m3, 24-hour average. The proposedannual PM2.5 standard would be based on the3-year average of the annual arithmetic meanPM2.5 concentrations, spatially averagedacross an area. The proposed 24-hour PM2.5standard would be based on the 3-yearaverage of the 98th percentile of 24-hourPM2.5 concentrations at each population-oriented monitor within an area. The proposalsolicited comment on two alternativeapproaches for selecting the levels of PM2.5

standards. The Administrator also proposed torevise the current 24-hour primary PM10standard of 150 g/m3 by replacing the 1-expected-exceedance form with a 98th

percentile form, averaged over 3 years at eachmonitor within an area, solicited comment onan alternative proposal to revoke the 24-hourPM10 standard, and proposed to retain thecurrent annual primary PM10 standard of 50g/m3. The proposal also solicited commenton proposed revisions to 40 CFR part 50,Appendix K to establish new data handlingconventions for calculating 98th percentilevalues and spatial averages, revisions to 40CFR part 50, Appendix J to modify thereference method for monitoring PM asPM10, and a proposed new reference methodfor monitoring PM as PM2.5 (40 CFR part 50,Appendix L).

With regard to the secondary standards, theAdministrator proposed to revise the currentsecondary standards by making them identicalto the suite of proposed primary standards, inconjunction with the establishment of aregional haze program under section 169A ofthe Act.

II. Rationale for the Primary Standards

A. Introduction

1. Overview. This document presents theAdministrators final decisions regarding theneed to revise the current primary ambient airquality standards for PM, and, morespecifically, regarding the establishment ofnew annual and 24-hour PM2.5 primarystandards and revisions to the form of thecurrent 24-hour PM10 primary NAAQS.These decisions are based on a thoroughreview, in the Criteria Document, of the latestscientific information on known and potentialhuman health effects associated withexposure to PM at levels typically found inthe ambient air. These decisions also take intoaccount:

8/2/2019 Pm2.5 Epa Cfr50

5/102


7 The risk assessment results that appear in the StaffPaper and are summarized in the proposal have beenupdated to include analyses of the particular forms ofstandard alternatives contained in the proposal and tocorrect estimates for one effects category (mortality fromlong-term exposure) to reflect the actual statistics used inthe study upon which they were based (Pope et al., 1995).The corrections, which cumulatively reduce estimates ofmortality associated with long-term exposures by 20 to35%, have no effect on risk estimates for mortality

Continued

(1) Staff Paper assessments of the mostpolicy-relevant information in the CriteriaDocument, upon which staffrecommendations for new and revisedprimary standards are based.

(2) CASAC advice and recommendations,as reflected in discussions of drafts of theCriteria Document and Staff Paper at publicmeetings, in separate written comments, and

in the CASACs closure letters to theAdministrator.

(3) Public comments received during thedevelopment of these documents, either inconnection with CASAC meetings orseparately.

(4) Extensive public comments received onthe proposed decisions regarding the primaryPM standards.

After taking this information andcomments into account, and for the reasonsdiscussed below in this unit, theAdministrator concludes that revisions to thecurrent primary standards to provideincreased public health protection against avariety of health risks are appropriate. Morespecifically, the Administrator has determinedthat it is appropriate to establish new annualand 24-hour PM2.5 standards, to revise thecurrent 24-hour PM10 standard, and to retainthe current annual PM10 standard. Asdiscussed more fully below in this unit, therationale for the final decisions regarding thePM primary NAAQS includes considerationof:

(1) Health effects information, andalternative views on the appropriateinterpretation and use of the information, asthe basis for judgments about the risks topublic health presented by populationexposures to ambient PM.

(2) Insights gained from a quantitative riskassessment conducted to provide a broaderperspective for judgments about protectingpublic health from the risks associated withPM exposures.

(3) Specific conclusions regarding the needfor revisions to the current standards and theelements of PM standards (i.e., indicator,averaging time, form, and level) that, takentogether, would be appropriate to protectpublic health with an adequate margin ofsafety.

As with virtually any policy-relevantscientific research, there is uncertainty in thecharacterization of health effects attributableto exposure to ambient PM. As discussed inthe proposal, however, there is now a greatlyexpanded body of health effects informationas compared with that available during thelast review of the PM standards. Moreover,the recent evidence on PM-related healtheffects has undergone an unusually highdegree of scrutiny and reanalysis over thepast several years, beginning with a series ofworkshops held early in the review processto discuss important new information. Anumber of opportunities were provided forpublic comment on successive drafts of theCriteria Document and Staff Paper, as well as

for intensive peer review of these documentsby CASAC at several public meetingsattended by many knowledgeable individualsand representatives of interestedorganizations. In addition, there have been anumber of important scientific conferences,symposia, and colloquia on PM issues,sponsored by the EPA and others, in the U.S.and abroad, during this period. While

significant uncertainties exist, the review ofthe health effects information has beenthorough and deliberate. In the judgment ofthe Administrator, this intensive evaluation ofthe scientific evidence has provided anadequate basis for regulatory decision makingat this time, as well as for the comprehensiveresearch needs document recently developedby EPA, and reviewed by CASAC and others,for improving our future understanding of therelationships between ambient PM exposuresand health effects.

The health effects information and humanrisk assessment were summarized in theproposal and are only briefly outlined below

in this unit. Subsequent units provide a morecomplete discussion of the Administratorsrationale, in light of key issues raised inpublic comments, for concluding that it isappropriate to revise the current primarystandards (Unit II.B. of this preamble) and torevise the specific elements of the standardsincluding indicator (Unit II.C. of thispreamble); averaging time, form, and level ofnew PM2.5 standards (Units II.D., II.E., andII.F. of this preamble); and averaging time,form, and level of revised PM10 standards(Unit II.G. of this preamble).

2. Summary of the health effects evidence.In brief, since the last review of the PM

criteria and standards, the most significantnew evidence on the health effects of PM isthe greatly expanded body of communityepidemiological studies. The CriteriaDocument stated that these recent studiesprovide evidence that serious health effects(mortality, exacerbation of chronic disease,increased hospital admissions, etc.) areassociated with exposures to ambient levelsof PM found in contemporary U.S. urbanairsheds even at concentrations below currentU.S. PM standard (U.S. EPA, 1996a; p. 13-1). Although a variety of responses toconstituents of ambient PM have beenhypothesized to contribute to the reportedhealth effects, the relevant toxicological andcontrolled human studies published to datehave not identified any acceptedmechanism(s) that would explain how suchrelatively low concentrations of ambient PMmight cause the health effects reported in theepidemiological literature.

Unit II.A. of the proposal further outlineskey information contained in the CriteriaDocument, Chapters 10-13, and the StaffPaper, Chapter V, on the known and potentialhealth effects associated with airborne PM,alone and in combination with otherpollutants that are routinely present in the

ambient air. The information highlightedthere summarizes:

(1) The nature of the effects that have beenreported to be associated with ambient PM,which include premature mortality,aggravation of respiratory and cardiovasculardisease (as indicated by increased hospitaladmissions and emergency room visits,school absences, work loss days, and

restricted activity days), changes in lungfunction and increased respiratory symptoms,changes to lung tissues and structure, andaltered respiratory defense mechanisms.

(2) Sensitive subpopulations that appear tobe at greater risk to such effects, specificallyindividuals with respiratory disease andcardiovascular disease and the elderly(premature mortality and hospitalization),children (increased respiratory symptoms anddecreased lung function), and asthmaticchildren and adults (aggravation ofsymptoms).

(3) An integrated evaluation of the healtheffects evidence, with an emphasis on the key

issues raised in assessing communityepidemiological studies, including alternativeinterpretations of the evidence, both forindividual studies and for the evidence as awhole.

(4) The PM fractions of greatest concernto health.

The summary in the proposal will not berepeated here. EPA emphasizes that the finaldecisions on these standards take into accountthe more comprehensive and detaileddiscussions of the scientific information onthese issues contained in the CriteriaDocument and Staff Paper, which werereviewed by the CASAC and the public.

3. Key insights from the risk assessment.The Staff Paper presents the results of aquantitative assessment of health risks for twoexample cities, including risk estimates forseveral categories of health effects associatedwith: existing PM air quality levels, projectedPM air quality levels that would occur uponattainment of the current PM10 standards, andprojected PM air quality levels that wouldoccur upon attainment of alternative PM2.5standards. The risk assessment is intended asan aid to the Administrator in judging whichalternative PM NAAQS would reduce riskssufficiently to protect public health with anadequate margin of safety, recognizing that

such standards will not be risk-free. The riskassessment is described more fully in theStaff Paper and summarized in the proposal.Related technical reports and updates7 have

8/2/2019 Pm2.5 Epa Cfr50

6/102


associated with short-term exposures or the estimates forany other effects. Because the key sensitivity analyses thatprovide additional insights regarding thresholds,copollutants, averaging time and related issues involvedthe short-term exposure studies, none of these results areaffected by changes to the long-term exposure riskestimates.

8 As discussed more fully below in this unit,epidemiological studies alone cannot be used todemonstrate mechanisms of action, but they can provideevidence useful in making inferences with regard to causalrelationships (U.S. EPA, 1996b, p. V-9).

9 As noted in the proposal, the kinds of effects observedin the epidemiological studies are logically related. Forexample, the association of PM with mortality is mainlylinked to respiratory and cardiovascular causes, which iscoherent with observed PM associations with respiratoryand cardiovascular hospital admissions and respiratorysymptoms. Further, similar categories of effects are seenin long- and short-term exposure studies.

been placed in the docket (Abt Associates,1996a,b; 1997a,b).

EPA emphasizes that it places greaterweight on the overall conclusions derivedfrom the studiesthat PM air pollution islikely causing or contributing to significantadverse effects at levels below thosepermitted by the current standardsthan on

the specific concentration-response functionsand quantitative risk estimates derived fromthem. These quantitative risk estimatesinclude significant uncertainty and, therefore,should not be viewed as demonstrated healthimpacts. EPA believes, however, that they dorepresent reasonable estimates as to thepossible extent of risk for these effects giventhe available information. Keeping in mindthe important uncertainties inherent in anysuch analyses, the key insights from the riskassessment that are most pertinent to thecurrent decision include:

(1) Fairly wide ranges of estimates of theincidence of PM-related mortality and

morbidity effects and risk reductionsassociated with attainment of alternativestandards were calculated for the twolocations analyzed when the effects of keyuncertainties and alternative assumptionswere considered. Significantly, the combinedanalysis for these two cities alone found thatthe risk remaining after attaining the currentPM10 standards was on the order of hundredsof premature deaths each year, hundreds tothousands of respiratory-related hospitaladmissions, and tens of thousands ofadditional respiratory related symptoms inchildren.

(2) Based on the results from the sensitivityanalyses of key uncertainties and theintegrated uncertainty analyses, the singlemost important factor influencing theuncertainty associated with the risk estimatesis whether or not a threshold concentrationexists below which PM-associated healthrisks are not likely to occur.

(3) Over the course of a year, the few peak24-hour PM2.5 concentrations appear tocontribute a relatively small amount to thetotal health risk posed by the entire air qualitydistribution as compared to the aggregatedrisks associated with the low to mid-rangeconcentrations.

(4) There is greater uncertainty about boththe existence and the magnitude of estimatedexcess mortality and other effects associatedwith PM exposures as one considersincreasingly lower concentrationsapproaching background levels.

B. Need for Revision of the Current PrimaryPM Standards

1.Introduction. The overarching issue inthe present review of the primary NAAQS iswhether, in view of the advances in scientificknowledge reflected in the Criteria Documentand Staff Paper, the existing PM standardsshould be revised and, if so, what revised or

new standards would be appropriate. Theconcluding section of the integrative synthesisof health effects information in the CriteriaDocument, which CASAC characterized asEPAs best ever example of a trueintegrative summary of the state ofknowledge about the health effects ofairborne PM, (Wolff, 1996b) provides thefollowing summary of the science withrespect to this issue:

The evidence for PM-related effects fromepidemiological studies is fairly strong, with moststudies showing increases in mortality, hospitaladmissions, respiratory symptoms, and pulmonaryfunction decrements associated with several PMindices. These epidemiological findings cannot bewholly attributed to inappropriate or incorrectstatistical methods, misspecification ofconcentration-effect models, biases in study designor implementation, measurement errors in healthendpoint, pollution exposure, weather, or othervariables, nor confounding of PM effects witheffects of other factors. While the results of theepidemiological studies should be interpretedcautiously, they nonetheless provide ample reasonto be concerned that there are detectable healtheffects attributable to PM at levels below thecurrent NAAQS. [U.S. EPA, 1996a, p. 13-92]

Given the nature of the health effects inquestion, this finding, which is based on alarge number of studies that used PM10

measurements, as well as studies using otherindicators of PM, clearly indicates thatrevision of the current PM NAAQS isappropriate. Quite apart from the issue ofwhether PM10 should be the sole indicator forthe PM NAAQS, the extensive PMepidemiological data base provides evidenceof serious health effects (e.g., mortality,exacerbation of chronic disease, increasedhospital admissions) in sensitive populations(e.g., the elderly, individuals withcardiopulmonary disease), as well assignificant adverse health effects (e.g.,increased respiratory symptoms, schoolabsences, and lung function decrements) inchildren. Moreover, these effects associationsare observed in areas or at times when thelevels of the current PM10 standards are met.Although the increase in relative risk is smallfor the most serious outcomes, EPA believesit is significant from an overall public healthperspective, because of the large number ofindividuals in sensitive populations that areexposed to ambient PM, as well as thesignificance of the health effects involved(U.S. EPA, 1996a, p. 1-21). The results of thetwo-city PM risk assessment reinforce theseconclusions regarding the significance of thepublic health riskeven under a scenario in

which the current PM10 standards areattained.

While the lack of demonstratedmechanisms that explain the extensive bodyof epidemiological findings is an importantcaution, which presents difficulties inproviding an integrated assessment of PMhealth effects research, a number of potentialmechanisms have been hypothesized in the

recent literature (U.S. EPA, 1996b; p. V-5 toV-8; appendix D). Moreover, qualitativeinformation from laboratory studies of theeffects of particle components at highconcentrations and dosimetry considerationssuggest that the kinds of effects observed incommunity studies (e.g., respiratory- andcardiovascular-related responses) are at leastplausibly related to inhalation of PM.8

Indeed, as discussed in the Criteria Documentand section V.E of the Staff Paper, theconsistency of the results of theepidemiological studies from a large numberof different locations and the coherent natureof the observed effects9 are suggestive of a

likely causal role of ambient PM incontributing to the reported effects.2. Comments on scientific basis for

revision. A majority of the public commentsreceived on the proposal agreed that, basedon the available scientific information, thecurrent PM10 standards are not of themselvessufficient to protect public health and itwould be appropriate to revise them. Includedin those calling for revisions to the currentstandards are many public healthprofessionals, including numerous medicaldoctors and academic researchers. Forexample, a group of 27 members of thescientific and medical community recognizedas having substantial expertise in conductingresearch on the health effects of air pollutionstated:

Health studies conducted in the U.S. and aroundthe world have demonstrated that levels ofparticulate and ozone air pollution below thecurrent U.S. National Air Quality Standardsexacerbate serious respiratory disease andcontribute to early death. A large body of scientificand medical evidence clearly indicates that thecurrent NAAQS are not sufficiently protective ofpublic health. [Thurston, 1997]

Similar conclusions were reached in a lettersigned by more than 1,000 scientists,clinicians, researchers, and other health careprofessionals (Dickey, 1997). The cosigners

to this letter argued that tens of thousands ofhospital visits and premature deaths could be

8/2/2019 Pm2.5 Epa Cfr50

7/102


prevented with the proposed air qualitystandard revisions. In fact, these commentersargued that even stronger standards than thoseproposed by EPA are needed to protect thehealth of the most vulnerable residents of ourcommunities.

A number of State and local governmentauthorities also submitted comments insupport of adopting new air quality standards

for fine particulate matter. The commentersconcurred with conclusions reached throughthe EPAs peer review process that the PMstandards should be revised to protect publichealth. A number of these commenterssuggested that the standards proposed by EPAshould be even stronger, while several otherState agencies recommended that EPA adoptPM2.5 standards, but at less stringent levels.A number of the comments from statessupporting even stronger standardsacknowledged the lack of demonstratedmechanism(s) and other uncertainties butstressed the strength of the other evidence inurging EPA to set protective standards.

Many comments were also received fromrepresentatives of environmental orcommunity health organizations thatsupported the adoption of air qualitystandards for PM2.5. These commentersagreed with EPAs finding that a large bodyof compelling evidence demonstrates thatexposure to particulate matter pollution, ingeneral, is associated with premature death,aggravation of heart and lung diseases,increased respiratory illness and reduced lungfunction. They agreed with EPA that thesestudies present a consistent and coherentrelationship between exposure to PM andboth mortality and various measures ofmorbidity. However, the majority of these

commenters argued that EPAs proposedstandards for PM2.5 were inadequate andrecommended adoption of more stringentlevels of the 24-hour and/or annual air qualitystandards for PM2.5. Many of thesecommenters also urged EPA to revise theNAAQS for PM10 to be more protective ofpublic health. These commenters based theirrecommendations on the findings of thestudies that were reviewed in the preparationof the Criteria Document and Staff Paper.One commenter used results from five ofthese studies as the basis for recommendingPM2.5 standards of 10 g/m3 (annual) and 18g/m3 (24-hour) (Dockery et al., 1993; Pope

et al., 1995; Schwartz et al., 1996; Schwartzet al., 1994; Thurston et al., 1994). Thecommenters agreed with EPA on thesignificance of these studies results and theneed to revise the PM standards, whilediffering with EPAs interpretation of thefindings for purposes of developing theproposed PM standards.

Several commenters made reference to theconclusions of a number of internationalscientific panels regarding the health effectsof exposure to airborne particulate matterthe British Expert Panel on Air QualityStandards, the British Committee on the

Medical Effects of Air Pollutants, the WorldHealth Organization, the Canadian Ministryof Environment, Lands and Parks, and theHealth Council of the Netherlands -- andargued that all these panels found that PMconcentrations equivalent to the current U.S.standards for PM10 are not protective ofhuman health and made recommendations forgreater protection. One commenter noted that

the findings of the British Health Panel haveresulted in a British proposal to adopt a 24-hour PM10 standard of 50 g/m3, which isone-third the level of the current U.S.NAAQS.

In these comments, some toxicologicalstudies were cited as providing evidence fortoxicity of particulate pollution. Thesecommenters disagreed with arguments thatPM standards cannot be adopted due to a lackof a sufficient understanding of the biologicalmechanism of injury. The commenters arguedthat there is sufficient evidence thatparticulate pollution is associated withadverse health effects to make it inappropriate

to delay the establishment of standards whilefurther studies are undertaken. This group ofcommenters was also critical of argumentsagainst the establishment of additional PMstandards based on the possibility ofconfounding by other pollutants, and urgedthat more attention be paid instead to thepossible additive or synergistic effects ofmultiple pollutant exposures.

In general, the EPA agrees with thesecommenters arguments regarding the need torevise the PM standards. The scientificstudies cited by these commenters were thesame studies used in the development of theCriteria Document and the Staff Paper, andthe EPA agrees that there is a sufficient body

of evidence that the current NAAQS for PMare not adequately protective of the publichealth. For reasons detailed in Unit II.F. ofthis preamble and in the Response toComments, EPA disagrees with aspects ofthese commenters views on the level ofprotection that is appropriate and supportedby the available scientific information.

Another body of commenters, includingalmost all commenters representingbusinesses and industry associations, manylocal governmental groups and privatecitizens, and some States opposed revising thestandards. Many of these commenters arguedthat the available scientific evidence does not

provide an adequate basis for revising thecurrent standards. The central argumentsmade by these commenters can be dividedinto two categories: (1) General comments onthe appropriateness of relying on theepidemiological evidence for makingregulatory decisions, and (2) more specificcomments challenging EPAs appraisal of theconsistency and coherence of the availableinformation, EPAs conclusions regardingcausality, and the use of these studies for riskassessment and decisions on whether to revisethe standards. While EPA has includedcomprehensive responses to these comments

in the Response to Comments, certain keypoints are summarized below in this unit.

a. General comments on the use ofepidemiological studies. The first category ofcomments was largely derived from ad hocpanels of occupational and otherepidemiological experts, consulting groups,and individual consultants. Most of theseindividuals and groups commented on the use

of epidemiology in reaching scientific andpolicy conclusions primarily from anoccupational or hazard assessmentperspective, in contrast to the perspective ofthe review of ambient PM criteria andstandards, where the use of community airpollution epidemiological studies are central.Citing accepted criteria used in evaluatingepidemiological studies to assess thelikelihood of causality (most notably those ofSir Austin Bradford Hill, 1965), thesecommenters argued that in the absence of ademonstrated biological mechanism, therelative risks of effects in the PMepidemiological studies are too low (less than

values variously cited as 1.5 to 2.0) to reachany conclusions regarding causality or toform the basis for regulations. In general, thecommenters applied these criteria to a subsetof studies evaluated in the Criteria Document,including as few as two long-term exposurestudies (EOP Group) (API, 1997), a group of9 selected studies (Greenland panel) (API,1997), those studies cited in the proposal(AIHC, 1997), or as many as 23 selectedshort-term exposure studies examined in arecently published review paper (Gamble andLewis, 1996).

Based on a careful review of thesecomments, EPA notes a number of limitationsin these commenters evaluations of the

epidemiological studies that they considered,as discussed in detail in the Response toComments. In summary, EPA notes that thesecommenters provided scientific advice andconclusions that are in substantialdisagreement with the conclusions of thereview reflected in the Criteria Document andStaff Paper. EPA stands behind the scientificconclusions reached in these documentsregarding the appropriate use of the availablecommunity epidemiological studies. Thesedocuments were the product of an extendedpublic process that included conductingpublic workshops involving the leadingresearchers in the field, drafts of the Criteria

Document and Staff Paper providingopportunities for public scrutiny andcomment on, and, not least, receiving theadvice of an independent panel of airpollution experts, including epidemiologists.

EPA clearly specified the key criteria bywhich it evaluated the availableepidemiological studies in section 12.1.2 ofthe Criteria Document, with substantialreliance on those specified by Hill (1965). Inrejecting results with relative risks less than1.5 to 2 as meaningful absent demonstratedbiological mechanisms, the commenters failto note that Hill and other expert groups (U.S.

8/2/2019 Pm2.5 Epa Cfr50

8/102


DHEW, 1964) have emphasized that no onecriterion is definitive by itself, nor is itnecessary that all be met in order to supporta determination of causality (U.S. EPA,1996a, p. 12-3).

With respect to biological plausibility, Hillnoted that this is a feature I am convincedwe cannot demand. What is biologicallyplausible depends upon the biological

knowledge of the day (Hill, 1965). Thisstatement is clearly pertinent to thetoxicological and mechanistic understandingof the effects of PM and associated airpollutants, especially at lower concentrations.It is also important to stress that while themechanistic evidence published as of the timethe Criteria Document closed does notprovide quantitative support for theepidemiological results, neither can suchlimited evidence refute these findings. It isalso important to stress that our understandingof biological mechanisms for PM pollutioneffects is not sufficient to explain the effectsobserved at much higher concentrations in air

pollution episodes, for which causality isgenerally accepted. Moreover, thetoxicological literature has only recentlybegun to examine animal models (orcontrolled human studies) that might reflectthe sensitive populations in question (theelderly, individuals with chronic respiratoryand cardiovascular disease) or that adequatelyreproduce all of the physico-chemicalproperties of particles in the ambientatmosphere. In short, the absence of evidenceof a particular mechanism is hardly proof thatthere are no mechanisms that could explainthe effects observed so consistently in theepidemiological studies. The absence ofbiological mechanisms did not deter CASAC

from recommending revisions to the PMstandards in 1982, 1986, and again in 1996.

While Hill appropriately emphasized thestrength of the association as important (e.g.,size of the relative risk), he also pointed outthat We must not be too ready to dismissa cause-and-effect hypothesis merely on theground that the observed association appearsto be slight. There are many occasions inmedicine when this in truth is so (Hill,1965). EPA believes that the effects of airpollution containing PM is such a case.Unlike the textbook examples of unlikelysignificant associations provided by somecommenters (e.g., ice cream consumption

correlated with heat stroke), the abundantepidemiological literature on combustionparticles documents numerous occasions inwhich single short-term episodes of high airpollution produced unequivocally elevatedrelative risks. For the week of the welldocumented 1952 London air pollutionepisode, for example, the relative risk ofmortality for all causes was 2.6, while therelative risk for bronchitis mortality was ashigh as 9.3 (Ministry of Health, 1954).Hospital admissions also increased by morethan a factor of two. British epidemiologistsin the 1950s concluded that increased

mortality was likely when PM (as masscalibrated British Smoke

8/2/2019 Pm2.5 Epa Cfr50

9/102


10 The term negative studies, as used in thesecomments, should not be construed to mean those in whichthere is a negative effects estimate (either significant ornon-significant) for the nominal cause. As used by thesecommenters, the term also includes statistically non-significant positive effect estimates. In other words, thecommenters define positive studies as including onlythose in which the effect estimate is both positive andstatistically significant.

11 Data sets were those used in the original studies byDockery et al. (1992) for St. Louis and Eastern Tennessee;Pope et al. (1992) for Utah Valley; Schwartz and Dockery(1992a) for Philadelphia; Schwartz (1993) forBirmingham; and a portion of the Santa Clara data fromFairley (1990). The data set from the Moolgavkar et al.(1995a) Philadelphia reanalysis was also included (Sametet al., 1995).

12 The HEI Board of Directors appointed an eightmember Oversight Committee consisting of leadingscientists in several disciplines relevant to air pollutionepidemiology to oversee key aspects of the project and toprepare HEIs assessment of the results.

consistent and coherent as EPA has claimed,and, in particular, charged that EPA ignoredor downplayed a number of studies that thecommenters argue contradict the evidence theAgency cited as supporting the consistencyand coherence of PM effects. The studies, allof which commenters contend do a better jobof addressing one or more key issues, suchas confounding pollutants, weather, exposure

misclassification, and model specification,than earlier studies, include several that wereavailable during preparation of the CriteriaDocument, and a number that appeared afterthe Criteria Document and Staff Paper werecompleted. Because the status of the laterstudies differ from that of the earlier ones forpurposes of decisions under section 109 ofthe Act, the two categories are discussedseparately below in this unit. Additionalresponses to comments relating to both setsof studies have been included in the Responseto Comments. In addition to the inclusion ofspecific studies, commenters also raised otherissues regarding the limitations of the

epidemiological information and the use ofthese studies in EPAs two-city riskassessment. Both of these topics are alsodiscussed below in this unit.

(i) Studies available for inclusion in thecriteria review. With some exceptions, mostof the above commenters cited somewhatsimilar lists of negative studies that theyargue EPA ignored or downplayed in arrivingat conclusions on consistency and coherence.Of the most commonly cited studies, thefollowing were available for inclusion in theCriteria Document: daily mortality studies byStyer et al. (1995), Lyon et al. (1995), Li andRoth (1995), Moolgavkar (1995a,b), Wyzgaand Lipfert (1995), Lipfert and Wyzga(1995), and Samet et al. (1995, 1996a,b); thelong-term exposure mortality study by Abbeyet al. (1991); and the re-examination of theSix-City mortality results (Dockery et al.,1993) by Lipfert (1995).

The written record of EPAs evaluations ofthese studies effectively refutes the claim thatthe Agency ignored any of these studies andsupports the treatment the Agency accordedto each of them. All of the studies availableto EPA at the time of CASAC closure on thePM Criteria Document (March 1996) wereexamined for inclusion in the CriteriaDocument and Staff Paper, which form thebasis for the PM proposal. Negative10

studies were evaluated in detail along withpositive studies when they were found tohave no critical methodological deficiencies,or to point out strengths and limitations.Studies that had more serious problems were

generally discussed in less detail, whetherpositive or negative, than studies with feweror small deficiencies. The EPA assessmentswere evaluated by peer reviewers, byCASAC, and by the public.

Most of the short-term exposure studiescited above in this unit are reanalyses andextensions of PM/mortality studies that hadbeen published by other investigators. In

general, the Criteria Document concluded thatthe most comprehensive and thoroughreanalyses were those in the series conductedfor the HEI, which reanalyzed data sets usedin studies from six urban areas in Phase I.A(Samet et al., 1995)11, with extended analysesfor Philadelphia in Phase I.B (Samet et al.,1996a,b). The most important finding in theHEI Phase I.A reanalyses of the six areas isthe confirmation of the numerical results ofthe earlier analyses of all six data sets (HEI,1995)12. After replicating the originalinvestigators analyses, Samet et al. (1995)also found similar results analyzing the datausing an improved statistical model. The HEI

Oversight Committee found[I]t is reasonable to conclude that, in these six

data sets, daily mortality from all causes combined,and from cardiovascular and respiratory causes inparticular, increases as levels of particulate airpollution indexes increase. [HEI, 1995]

It is important to note that these reanalysesby respected independent scientists confirmthe reliability and reproducibility of the workof the original investigators, particularly inview of the concerns some commenters haveexpressed about EPAs reliance on a numberof PM studies published by these authors.

The Phase I.A HEI results for Philadelphiaalso found that it was difficult to separate theeffects of PM from those of co-occurring

SO2, in agreement with the Moolgavkar etal.(1995a) analysis. Subsequent HEI work,and several of the other so-called negativestudies cited above in this unit, furtherexamined this issue in terms of confoundingor effects modification by one or more co-occurring gaseous pollutants or weather.Contrary to commenters claims, this issueand these studies received considerableattention in the Criteria Document and StaffPaper, and the overall implications andconclusions from these assessments weresummarized in the proposal. In particular, theso-called negative and other findings ofMoolgalvkar et al. (1995a,b) in their

Philadelphia and Steubenville studies werediscussed in great detail in section 12.6 of the

PM Critera Document and compared to thoseof the original investigators (Schwartz andDockery, 1992a,b) and other investigators (Liand Roth, 1995; Wyzga and Lipfert, 1995).Further analytical studies of the Philadelphiadata set were carried out by HEI (Samet etal., 1996a,b) and have largely resolved manyof the uncertainties in the earlier analyses; inEPAs opinion, these studies supersede the

results of the original investigators (Schwartzand Dockery, 1992a) and the several earlierreanalyses, including Moolgavkar (1995a),Moolgavkar and Luebeck (1996), Li and Roth(1995), Wyzga and Lipfert (1995), and Sametet al. (1995). Even though TSP is not the bestPM indicator for health effects, since itincludes a substantial fraction of non-thoracicparticles, the extended Criteria Documentassessment (U.S. EPA, 1996a, pp. 12-291 to-299; 12-327) of the Phase I.B HEI analysesin Philadelphia (Samet et al., 1996a,b) servesto support the following findings:

(1) The mortality effects estimates for TSPdo not depend heavily on statistical methods

when appropriate models are used.(2) Estimated PM effects are not highlysensitive to appropriate methods for adjustingfor time trends and for weather.

(3) Air pollution has significant healtheffects above and beyond those of weather.

(4) Copollutants such as ozone, CO, andNO2 may be important predictors ofmortality, but their effects can besubstantially separated from those of TSP andSO2.

(5) The health effects of TSP inPhiladelphia cannot be completely separatedfrom SO2, which is itself a precursor of fineparticles, based solely on the epidemiologicalanalyses in this single city.

The most recent HEI Oversight Committeecomments on these studies (HEI, 1997),which were submitted to the docket by HEI,state that:

Although individual air pollutants (TSP, SO2,and ozone) are associated with increased dailymortality in these data, the limitations of thePhiladelphia data make it impossible to establishthat particulate air pollution alone is responsible forthe widely observed associations between increasedmortality and air pollution in that city. All we canconclude is that it appears to play a role. [HEI,1997; p.38.]

While recognizing the limitations in theconclusions that can be made based on

studies in a single city, the OversightCommittee endorses the approach taken byEPA in evaluating a broader set ofepidemiological studies:

Consistent and repeated observations in localeswith different air pollution profiles can provide themost convincing epidemiological evidence tosupport generalizing the findings from thesemodels. This has been the approach reported by theEPA in its recent Criteria Document and StaffPaper. [HEI, 1997; p. 38.]

As noted in the proposal, based on thisapproach, EPAs assessment of numerousmortality studies concludes that when studies

8/2/2019 Pm2.5 Epa Cfr50

10/102


13 Their March 20, 1996 letter to the Administratorconcludes that the HEI analysis of Philadelphia supersedesearlier analyses, specifically Moolgavkar et al. (1995a),Lipfert and Wyzga (1995), and Li and Roth (1995), andpoints out the limitations of Styer et al. (1995).

14 In response to comments on this rulemaking, somepapers submitted by industry commenters make statementsthat are in substantial agreement with these staffconclusions with respect to the likelihood of SO2penetrating to indoor environments and the lesserlikelihood of affecting sensitive populations indoors(Lipfert and Wyzga, 1997; Lipfert and Urch, 1997).

are evaluated on an individual basis, the PM-effects associations are valid and, in a numberof studies, not seriously confounded by co-pollutants (U.S. EPA, 1996a; p. 13-57); andwhen a collection of studies from multipleareas with differing concentrations of PM andco-pollutants are examined together, theassociation with PM10 remains reasonablyconsistent across a wide range of

concentrations of these potentially influentialpollutants (U.S. EPA, 1996a; p. 12-33; U.S.EPA, 1996b; p. V-55).

In addition to relying on the mostcomprehensive and best analyses inevaluating the reanalysis in Philadelphia andother areas, the Criteria Document gave lessweight to both so-called negative andpositive studies with methodogicallimitations. In particular, EPA agreed with theepidemiological experts on CASAC(Lippmann et al., 1996; Samet, 1995) that theLi and Roth (1995) study approach of usinga panoply of different modeling strategiesto produce seemingly conflicting findings

provides little useful insight and is supersededby the HEI report. The attempt by Lipfert andWyzga (1995) to address relative effects ofdifferent pollutants was consideredinconclusive (Lippmann et al., 1996) andflawed by the use of a metric (elasticity) thatignores the absolute concentrations of thepollutants being compared (see Response toComments).

Further, the Steubenville studies andreanalyses (Schwartz and Dockery, 1992b;Moolgavkar, 1995b) were discussed in detailto examine methodologies, and thedifferences in relative risks between the twowere regarded as small (U.S. EPA, 1996a, p.12-280 to 283). Both studies used TSP as the

PM indicator variable, and they areaugmented by the more recent findings ofSchwartz et al. (1996) that examine PM10 andits components. The mixed results by Lyonet al. (1995) in Utah Valley are compromisedby loss of information related to themethodology (U.S. EPA, 1996a, p. 12-58). Asnoted above, subsequent reanalyses of theUtah Valley study by HEI (Samet et al.,1995) as well as by Pope and Kalkstein(1996) confirmed the original findings ofPope et al. (1992) using different modelspecifications. The Salt Lake City study byStyer et al. (1995) was mentioned in the PMCriteria Document, but received little

discussion because aspects of themethodological approach limited its statisticalpower to detect effects. The analysis ofChicago mortality data in the same papershared these problems, particularly forseasonal analyses; in this larger city, theynonetheless found significant associations onan annual basis between PM10 and mortalitythat are consistent with other studies. In short,the record shows that EPA did not ignorethese short-term exposure studies cited bycommenters; moreover, EPAs assessment ofthese studies is consistent with the views offour researchers on the CASAC panel who

have extensive involvement in conductingpopulation studies of air pollution (Lippmannet al., 1996).13

Similarly, EPA believes that appropriatetreatment and weight were given to studies oflong-term exposure and mortality. EPAconcluded that the lack of associations in theAbbey et al. (1991) prospective cohort studywere not inconsistent with two other such

studies because the use of days of peak TSPlevels as the PM indicator (instead of PM10or PM2.5) is inappropriate for Californiacohorts exposed to both urban smog andfugitive dust episodes, and the overall samplesize may have been too small to detectsignificant effects (U.S. EPA, 1996b; pp. V-17 to -18). The inadequacy of Lipferts(1995) application of state-wide averagesedentary lifestyle data to adjust mortality forthe six cities studied by Dockery et al. (1993),in which superior subject-specific body massindex data had already been considered, wasalso noted and addressed in the Staff Paper(U.S. EPA, 1996b; p. V-16). Again, EPA did

not ignore these studies; the rationale forgiving them less weight was clearlyarticulated in the documents reviewed byCASAC and judged appropriate for use instandard setting.

While the proposal presents only asummary discussion of key CriteriaDocument and Staff Paper findings, EPAbelieves that discussion is fully consistentwith the state of the science. Furthermore, theproposal highlights the nature of alternativeviewpoints on the epidemiology in aquotation from the Criteria Document (61 FR65644, December 13, 1996) and citesexplicitly the views of most of the authorsnoted above in this unit (Moolgavkar et al.,

1995b; Moolgavkar and Luebeck, 1996; Liand Roth, 1995; Samet et al., 1996; Wyzgaand Lipfert, 1995). The proposal alsosummarizes EPA conclusions based on all ofthe literature as assessed in the CriteriaDocument and Staff Paper with respect toissues raised in these and other studies,including potential confounding byindependent risk factors such as weather andother pollutants, choice of statistical models,use of outdoor monitors, and exposuremisclassification.

More specifically, in the proposal EPA hasnot ignored the view advanced by some thatthe results of individual studies of multiple

pollutants, such as the HEI Philadelphiastudies, are more suggestive of an airpollution effect than an effect of PM alone.Indeed, the proposal notes that it is reasonableto expect that other pollutants may play a rolein modifying the magnitude of the estimatedeffects of PM on mortality, either throughpollutant interactions or independent effects(61 FR 65645, December 13, 1996). Based on

the large body of evidence at hand, however,EPA cannot accept the suggestion that suchmulti-pollutant studies are in any waynegative with respect to EPAsconclusions that PM, alone or in combinationwith other pollutants, is associated withadverse effects at levels below those allowedby the current standards. This conclusion isbased not only on the consistency of PM

effects across areas with widely varyingconcentrations of potentially confoundingcopollutants, but also on the extendedanalyses of the Philadelphia studies in theCriteria Document and Staff Paper.

Because commenters have tended to ignorethe latter analyses, it is appropriate tosummarize them here briefly. As noted abovein this unit, the Criteria Document assessmentof the Philadelphia studies finds that PM canreasonably be distinguished from potentialeffects of all pollutants except SO2. The StaffPaper builds on this analysis through anintegrated assessment that draws oninformation from atmospheric chemistry,

human exposure studies, and respiratory tractpenetration results to provide insight as towhich of these two pollutants is more likelyto be responsible for mortality in the elderlyand individuals with cardiopulmonary disease(U.S. EPA 1996b; pp. V-46 to -50). Thatassessment notes that the inhalable (PM10),including the fine (PM2.5), components ofTSP are more likely than SO2 to penetrateand remain indoors where the sensitivepopulation resides most of the time.14 Inaddition, these PM components, especiallyPM2.5, penetrate far more effectively to theairways and gas exchange regions of the lungthan does SO2. Furthermore, in Philadelphia,it is possible that SO2 is a surrogate for fine

particulate acid sulfates. For these reasons,even though statistical analyses of thePhiladelphia data set cannot fully distinguishbetween these two highly correlatedpollutants, EPA believes that the weight ofthe available evidence from an integratedassessment more strongly supports the notionthat PM is playing an important direct rolein the observed mortality effects associationsin Philadelphia. Moreover, as noted above inthis unit, in some other locations withsignificant PM-mortality associations,ambient SO2 levels are too low to confoundPM.

(ii)Recent studies available after

completion of criteria review. As noted abovein this unit, other studies cited by somecommenters as so-called negativeevidence ignored by EPA were published orotherwise made available only aftercompletion of the PM Criteria Document.EPA agrees that it did not rely on these

8/2/2019 Pm2.5 Epa Cfr50

11/102


15 Since the 1970 amendments, the EPA has taken theview that NAAQS decisions are to be based on scientificstudies that have been assessed in air quality criteria [seee.g., 36 FR 8186 (April 30, 1971) (EPA based originalNAAQS for six pollutants on scientific studies discussedin the air quality criteria and limited consideration ofcomments to those concerning validity of scientific basis);38 FR 25678, 25679-25680 (September 14, 1973) (EPArevised air quality criteria for sulfur oxides to provide basisfor reevaluation of secondary NAAQS)]. This longstandinginterpretation was strengthened by new legislativerequirements enacted in 1977 (section 109(d)(2) of theAct; section 8(c) of the Environmental Research,Development, and Demonstration Authorization Act of1978) for CASAC review of air quality criteria andreaffirmed in EPAs decision not to revise the ozonestandards in 1993. 58 FR 13008, 13013-13014 (March 9,

1993). Some of the commenters now criticizing EPA fornot considering the most recent PM studies stronglysupported the Agencys interpretation in the 1993 decision(UARG, 1992).

16 As discussed in EPAs 1993 decision not to revisethe NAAQS for ozone, new studies may sometimes be ofsuch significance that it is appropriate to delay a decisionon revision of NAAQS and to supplement the pertinentair quality criteria so the new studies can be taken intoaccount. 58 FR at 13014, March 9, 1993. In the presentcase, EPAs provisional examination of recent studiessuggests that reopening the air quality criteria reviewwould not be warranted even if there were time to do sounder the court order governing the schedule for thisrulemaking. Accordingly, EPA believes that theappropriate course of action is to consider the newlypublished studies during the next periodic review cycle.

17 For example, commenting on the Roth examinationof alternative model specifications, Dr. Stolwijk noted Ifyou select out of his [Roths] matrix the things that otherpeople have done, he comes to a different conclusion thanwhen he takes his whole matrix * * *. [Y]ou are goingto get a random effect that shows that there is no effect.He [Roth] did this, I think, on purpose in this case. Mostepidemiologists, I think, have been trained to limit theirobservations to something that they can state or wouldhave stated before they started and observe that and basetheir conclusions on it [U.S. EPA 1996(c); May 17, 1996Transcript, pages 45-46].

18 CASAC panelists recommended a discussion of thisissue in the Staff Paper. The Staff Paper notes: Whilegreater measurement error for the coarse fraction coulddepress a potential coarse particle effect, this would notexplain the results in Topeka relative to other cities. Evenconsidering relative measurement error, these resultsprovide no clear evidence implicating coarse particles inthe reported effects. (U.S. EPA, 1996b p. V-64). EPAsprovisional examination of the Lipfert and Wyzga (1997)paper in the Response to Comments, finds that it isimplausible that most of the effect attributed to PM2.5could in fact be due to PM10-2.5, since differentialmeasurement error cannot make a weaker effect appearstronger than a stronger one, except under extremely

unusual circumstances.19 The APHEA (Air Pollution and Health: a EuropeanApproach) project was supported by the European UnionEnvironment 1991-1994 Programme to investigate thepossible short-term health effects of exposure to low ormoderate levels of ambient air pollutants. Eleven Europeanresearch groups carried out studies in 15 cities(Amsterdam, Athens, Barcelona, Bratislava, Cracow,Helsinki, Koln, Lodz, London, Lyon, Milan, Paris, Poznan,Rotterdam and Wroclaw) in which air pollutantconcentration data had been collected for at least 5 years.Initial findings of studies on mortality and hospitaladmissions were published in a series of papers inSupplement 1 to the Journal of Epidemiology andCommunity Health in 1996 and a meta-analysis of themortality data from 12 cities is currently in press(Katsouyanni et al., 1997).

studies, based on its long-standing practice ofbasing NAAQS decisions on studies andrelated information included in the pertinentair quality criteria and available for CASACreview.15 Although EPA has not relied onsuch studies in this review and decisionprocess, the Agency nevertheless hasconducted a provisional examination of theseand other recent studies to assess their general

consistency with the much larger body ofliterature evaluated in the CriteriaDocument.16 EPA has placed its examinationof recent studies in the rulemaking docket.

Among the most frequently cited newstudies relied on by commenters were Daviset al. (1996), Moolgavkar et al. (1997), andRoth and Li (1997). Davis et al. (1996)conducted a reanalysis of the Birminghammortality data set originally investigated inSchwartz (1993). At the time of the close ofthe public comment period, the paper basedon this manuscript had not been accepted forpublication in a peer reviewed journal (Sacks,1997). Commenters nevertheless highlight the

authors claim that when humidity isincluded among the meteorological variables(it is excluded in the analysis by Schwartz[1993]), we find that the PM10 effect is notstatistically significant. EPAs review foundimportant factual errors in this study.Contrary to Davis et al., Schwartz did includehumidity in his 1993 study, and his findingof a hot-and-humid-day effect was reportedthere. In addition, the PM-related variablesused by Davis et al. in their manuscript werenot, as the authors claimed, the same as thosein Schwartz (1993). Davis et al. also used adifferent humidity indicator, specifichumidity. Reanalysis by one of the co-authors(R. Smith, personal communication, February

8, 1997) showed that when Schwartzs PMmetric was used, the estimated PM10 effectwas of about the same magnitude, andstatistically significant at the 0.05 level, evenusing the characterization of humidity effectproposed by Davis et al. It therefore appearsthat the Davis et al. PM10 result was, in fact,consistent with that of Schwartz, and robustagainst a very different weather model

specification.Based on its examination of both the

content and the publication status of thisstudy, EPA believes the heavy reliance andattention given to it are misguided. In contrastto commenters assertions, this study does notcontradict EPAs conclusions with respect toconsistency of the epidemiological evidenceand confounding by weather variables;indeed, the consideration of the correctedresults would actually support EPAsconclusions. EPA believes this examplereinforces the importance of relying on peerreviewed studies and also conducting the kindof critical examination of such studies that

takes place in the criteria and standardsreview process.Several commenters note that Roth and Li

(1997) also reexamined the Birminghammortality data, as well as hospital admissionsdata from Schwartz (1994), and produced anumber of negative and inconsistent resultsthat depend on temperature effects and choiceof statistical model. Preliminary findings fromthis study were presented by Roth at the May1996 CASAC meeting. CASACepidemiologists and statisticians at themeeting pointed out a number ofshortcomings, both in the analytical strategyand in details of the models beingevaluated.17 As discussed in more detail in

the Response to Comments, the materialsfrom Roth and Li (1997) recently provided toEPA as attachments to public comments showthat the deficiencies pointed out at the May1996 CASAC meeting have not beenadequately addressed. EPA concludes thatthis study does not support commentersclaims.

The paper recently accepted for publicationby Moolgavkar et al. (1997) examineshospital admissions and air pollution inMinneapolis and Birmingham and comes todifferent conclusions than earlier investigatorswith respect to the role of PM10. While thepaper is a useful addition to the literature, the

authors clearly do not attempt to replicate theoriginal studies, making the kind of directcomparisons suggested by commenters

difficult. The paper finds an air pollutioneffect in one city that implicates ozone butis unable to separate effects of PM from agroup of other pollutants. EPAs provisionalexamination of this study raises somequestions about the methodology, whichmight usefully be supplemented to furtherseparate pollutants as was done by Samet etal. (1996a,b) in Philadelphia, and about the

authors interpretation of the results in bothcities. In any event, EPA does not believe thisstudy negates the PM associations withhospital admissions reported in a number ofother studies cited in the Criteria Document.

Another recent paper by Lipfert and Wyzga(1997) provides analyses suggesting thatdifferential measurement error might accountfor some or all of the observation bySchwartz et al. (1996) that daily mortality ismore strongly associated with fine (PM2.5)than with coarse (PM10-2.5) PM. EPA staffand CASAC accounted for this possibility,however, and it was factored into both theStaff Paper and CASAC recommendations.18

Some commenters have highlightedselected individual papers or summaries fromthe APHEA19 project conducted in Europe,and from Roth (1996), calling attentionparticularly to negative results found inheavily polluted regions of Eastern Europe.EPA notes that a number of the recentAPHEA and other studies in Western Europehave shown significant associations betweenmortality and air pollution including PM, andthat a meta-analysis of 12 Western andCentral-eastern European studies issupportive of a causal association betweenPM and SO2 exposure and all-causemortality (Katsouyanni et al., 1997). The

Eastern and Western European studies useddiffering measurement methods for PM,including PM10, gravimetric suspended

8/2/2019 Pm2.5 Epa Cfr50

12/102


20 The Roth et al. (1997) study in Prague used ameasurement termed suspended particles that appearsto be close to TSP. The relation of this indicator to PM10or PM2.5 in this city is not reported. Moreover, this studyuses a variant of the problematic methodology in the Rothanalyses cited above.

21 These concerns are consistent with EPAs treatmentof a number of European and South American studies thatare included in the Criteria Document and contributed tothe evaluation of the epidemiology in Chapter 12. Becauseof differences in aerometry methods and characteristicsource classes between North America and other regionsof the world, however, the integrative assessment chapterreported results only from studies conducted in the U.S.and Canada (cf. Tables 13-3 to 13-5) in reachingquantitative conclusions for effects estimates.

22 See, for example, the United Kingdom Air QualityStrategy, 1997; Swiss Federal Commission of Air Hygiene,1996; World Health Organization Revised Air QualityGuidelines for Europe, In Press).

23 Paradoxically, some commenters have argued (e.g.,Valdberg, 1997) that the PM results are confounded

because the weather and other factors that cause dailyvariations in outdoor pollution will cause similar dailyvariations in indoor generated air pollution. For this to betrue, outdoor ambient pollution concentrations would haveto be correlated with personal exposure to indoor generatedair pollution such as that from smoking, cleaning, andcooking. This argument is logically inconsistent with theother comments on the lack of any such correlation withpersonal exposure, and these commenters have offered noscientific evidence to support their claim. In response, EPAhas performed and included in the Response to Commentsa numerical analysis of the relevant information from thePTEAM exposure study that finds no evidence for sucha correspondence in the actual data.

24 As documented in Chapter 7 of the CriteriaDocument, time-series community studies observe theeffects of varying levels of ambient air pollution; thereforethe effects of indoor-generated air pollution would beindependent of and in addition to the effects found in these

epidemiological studies. Commenters apparently believeEPA is claiming such studies are detecting the effects ofdaily variations in total PM personal exposure from indoorand outdoor sources. This misunderstanding is evidenced,for example, by Wyzga and Lipferts (1995) treatment ofthe difference between ambient monitors and actualpersonal exposures as exposure errors and Brownscomment for API that if (ambient) PM is causally relatedto mortality/morbidity, then it is personal PM exposure thatmust be reduced to have an effect. On the contrary, itis personal exposure to ambient PM that must be reducedto address the risk identified in community air pollutionstudies. Any lack of significant correlation betweenoutdoor PM concentrations and personal exposure to totalPM from all sources is irrelevant, except to the extent itmay decrease the power of time-series studies to detect theeffects of ambient pollution.

particles, and the British Smoke method.20

The differences in aerometry and thesubstantial differences in location andstrength of primary PM emissions sources incentral and eastern Europe as compared towestern Europe or the U.S. might wellexplain the different results in these uniqueareas. Consequently, integration of theseresults would involve comprehensive

examination of the various PM instrumentsused, monitor siting in relation to sources,mass calibration procedures and other aspectsof these studies.21 EPA notes that a numberof European authorities, who are familiar withthis recent literature, have proceeded withrecommendations to strengthen their healthguidelines, risk assessments, or regulationsfor PM.22

Aside from the recent literature cited bythese commenters, there are a number ofother recent epidemiological studies that, ifconsidered in todays decision, would tend tosupport EPAs conclusions about the effectsof PM at lower concentrations, assuming theirresults were accepted following a full reviewin the criteria and CASAC process. Forexample, in addition to the APHEA studies,several other recent epidemiologic studieshave reported significant positive associationsbetween PM and health effects (Lipsett et al.,1997; Peters et al., 1997; Borja-Aburto et al.,1997; Delfino et al., 1997; Scarlett et al.,1996; Woodruff et al., 1997; Wordley et al.,1977). In addition, a number of recenttoxicologic papers have been accepted orappear in proceedings (Costa and Dreher,1997; Killingsworth et al., 1997; Godleski etal., 1997) that involve exposure toconcentrated ambient fine particles or PM

constituents and appear to provide supportiveevidence as to the plausibility of the effectsthat have been reported epidemiologically. Ifconsidered in this decision, these studieswould also provide biological support for theepidemiological observation that certainsusceptible groups (notably those withcardiopulmonary disease) are most likely tobe affected by PM, again assuming the resultswere sustained in the full criteria and CASACreview process.

In summary, EPA has conducted aprovisional assessment of the more recentscientific literature. Based on this provisionalassessment, EPA disagrees with commentersassertion that full consideration of selectednew studies in this decision would materiallychange the Criteria Document and Staff Paperconclusions on the consistency and coherenceof the PM data, or on the need to revise the

current standards.(iii) Other specific comments on the

epidemiological studies. Aside from theirassertion that EPA ignored or downplayedparticular studies, this second group ofcommenters raise additional objections, basedon the statistical modeling strategies used andthe potential importance of personal exposuremisclassification, to EPAs conclusionsregarding the consistency of theepidemiological evidence. EPA conclusionson these topics were summarized in theproposal and supported by extensivetreatments in the Criteria Document and StaffPaper. With respect to the first issue,

commenters argued that sufficient flexibilityexists in the analyses of large data sets thatit may be possible to obtain almost any resultdesired through choice of statistical method.Analytical choices include the specificstatistical model; methods used to adjust forseasonal variation and the trends in the data;treatment of other variables (e.g., otherpollutants, weather, and day of week); lagstructure; and study population.

A more detailed discussion of this issue,which expands on the assessment summarizedin the Criteria Document, is included in theResponse to Comments. In summary, EPAmust reject commenters contention thatlegitimate alternative analyses can obtain

almost any result. As outlined above inthis unit, EPAs detailed reviews of individualstudies have shown that not all methods areequally valid or legitimate. Moreover, strongarguments can be made that the methods andanalytical strategies in the studies EPA reliedupon are more appropriate approaches thanthose cited by commenters (e.g., Li and Roth,1995; Lipfert and Wyzga, 1995; Davis et al.,1996; Roth and Li, 1997). While not allstudies have addressed each of the aboveissues in this unit equally well, the mostcomprehensive analyses of these issues (e.g.,Samet et al., 1995, 1996a,b; Pope andKalkstein, 1996), as well as the EPA analyses

comparing study results for each issue (U.S.EPA, 1996a, pp. 12-261 to 12-305) found thatthe authors of studies on which EPA chieflyrelied made appropriate modeling choices.The Criteria

Pm2.5 Epa Cfr50

Documents

Transcript of Pm2.5 Epa Cfr50