EPA-450/5-85-004
The NAAQS Exposure Model (NEM)
Applied to Carbon Monoxide:
Addendum
by
Roy A. Paul and Ted Johnson
PEI Associates, Inc.
505 South Duke Street, Suite 503
Durham, North Carolina 27701
Contract No. 68-02-4309
Work Assignment No 4
PN 3659-4
Task Manager
Thomas R. McCurdy
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711
April 1985 ^ s Environmental pfotectjon
Region 5, Library (PL-12J)
12th
-------
DISCLAIMER
This report was furnished to the U.S. Environmental Protection Agency by
PEI Associates, Inc., in partial fulfillment of Contract No. 68-02-4309, Work
Assignment No. 4. The contents of this report are reproduced herein as
received from the contractor. The opinions, findings, and conclusions
expressed are those of the authors and not necessarily those of the Environ-
mental Protection Agency.
-------
CONTENTS
Tables iv
Acknowledgment vi
1. Introduction 1
2. Exposure Estimates for Adults With Cardiovascular Disease in
Four Urban Areas 4
"Best estimate" results 4
The significance of indoor sources 17
Uncertainty in NEM exposure estimates 17
3. Nationwide Extrapolations 19
Extrapolation results 19
Uncertainty of nationwide estimates 30
Comparing two forms of NAAQS 30
-------
TABLES
Number Page
1-1 Air Quality Indicators for 8-hour Carbon Monoxide Standard
With One Expected Exceedance.of 9 ppm per Year and With
One Observed Exceedance of 9 ppm Per Year 3
2-1 Estimates of Occurrences for Adults With Cardiovascular
Disease of 1-hour CO Exposures Above Selected Concentra-
tion Values Assuming 9 ppm/1 ObEx Standard is Attained .
2-2 Estimates of Adults With Cardiovascular Disease who Have
1-hour CO Exposures Above Selected Concentration Values
Assuming 9 ppm/1 ObEx Standard is Attained
2-3 Estimates of Adults With Cardiovascular Disease Whose Maximum
1-hour CO Exposure Occurs in Selected Concentration Ranges
Assuming 9 ppm/1 ObEx Standard is Attained
2-4 Estimates of Occurrences for Adults With Cardiovascular
Disease of 8-hour CO Exposures Above Selected Concentra-
tion Values Assuming 9 ppm/1 ObEx Standard is Attained ... 8
2-5 Estimates of Adults With Cardiovascular Disease Who Have
8-hour CO Exposures Above Selected Concentration Values
Assuming 9 ppm/1 ObEx Standard is Attained 9
2-6 Estimates of Adults With Cardiovascular Disease Whose Maximum
8-hour CO Exposure Occurs in Selected Concentration Ranges
Assuming 9 ppm/1 ObEx Standard is Attained 10
2-7 Estimates of Occurrences for Adults With Cardiovascular
Disease of COHb Levels Exceeding Selected Values Assuming
9 ppm/1 ObEx Standard is Attained 11
2-8 Estimates of Adults With Cardiovascular Disease Who
Experience COHb Levels Exceeding Selected Values Assuming
9 ppm/1 ObEx Standard is Attained 12
2-9 Estimates of Adults With Cardiovascular Disease Whose Maximum
COHb Level Occurs in Selected Ranges Assuming 9 ppm/1 ObEx
Standard is Attained 13
-------
TABLES (continued)
Number Page
2-10 Percentage of Adults With Cardiovascular Disease Experiencing
COHb Levels Exceeding Selected Values Assuming 9 ppm/1 ObEx
Standard is Attained 16
3-1 Estimates of Occurrences in the Cardiovascular Adult Urban
U.S. Population of 1-hour Average CO Exposures Above
Selected Concentration Values Under Alternative Air
Quality Assumptions 20
3-2 Estimates of Cardiovascular Adults in Urban U.S. With 1-hour
Average CO Exposures Above Selected Concentration Values
Under Alternative Air Quality Assumptions 21
3-3 Estimates of Cardiovascular Adults in Urban U.S. Whose
Maximum 1-hour Average CO Exposure Occurs in Selected Con-
centration Ranges Under Alternative Air Quality Assumptions 22
3-4 Estimates of Occurrences in the Cardiovascular Adult Urban
U.S. Population of 8-hour Average CO Exposures Above
Selected Concentration Values Under Alternative Air
Quality Assumptions 23
3-5 Estimates of Cardiovascular Adults in Urban U.S. With 8-hour
Average CO Exposures Above Selected Concentration Values
Under Alternative Air Quality Assumptions 24
3-6 Estimates of Cardiovascular Adults in Urban U.S. Whose
Maximum 8-hour Average CO Exposure Occurs in Selected Con-
centration Ranges Under Alternative Air Quality Assumptions 25
3-7 Estimates of Occurrences Among Cardiovascular Adults in
Urban U.S. of COHb Levels Exceeding Selected Values Under
Alternative Air Quality Assumptions 26
3-8 Estimates of Cardiovascular Adults in Urban U.S. Experiencing
COHb Levels Exceeding Selected Values Under Alternative Air
Quality Assumptions 27
3-9 Estimates of Cardiovascular Adults in Urban U.S. Whose
Maximum COHb Level Occurs in Selected Concentration Ranges
Under Alternative Air Quality Assumptions 28
3-10 Percentage of Cardiovascular Adult Urban U.S. Population
Experiencing COHb Levels Exceeding Selected Values Under
Alternative Air Quality Assumptions 29
-------
ACKNOWLEDGMENT
This Addendum to the NAAQS Exposure Model Applied to Carbon Monoxide was
largely due to the efforts of Mr. Thomas R. McCurdy, SASD. Appreciation is
extended to him for this administrative and technical supervision. Other
parties who developed and implemented the NEM are acknowledged in the original
report.
vi
-------
SECTION 1
INTRODUCTION
Under the Clean Air Act, the U.S. Environmental Protection Agency (EPA)
is responsible for establishing National Ambient Air Quality Standards (NAAQS)
and for reviewing them periodically to determine their adequacy on the
basis of recent scientific research. In view of these responsibilities, the
Strategies and Air Standards Division (SASD) of the Office of Air Quality
Planning and Standards (OAQPS) is developing quantitative methods for assess-
ing health risks associated with proposed air quality standards.
An important aspect of health risk assessment is the estimation of popu-
lation exposure. For the past few years, SASD has been engaged in the develop-
ment of an exposure model suitable for evaluating alternative ambient air
standards. The model is known as NEM, an acronym for NAAQS Dcposure Model.
Several versions of NEM have been developed. The exposure district
version of NEM simulates the pollutant concentrations expected to occur in
selected exposure districts within a study area under user-specified regulatory
scenarios. It then adjusts these estimates to account for an exhaustive set
of microenvironments and simulates typical movements of population groups,
called cohorts, through the districts and microenvironments. Outputs of the
simulation program are population exposure estimates at specified pollutant
1 eve!s.
Another version of NEM substitutes "neighborhood types" for exposure
districts. A 1983 report described this neighborhood type (NT) model and
its application to four U.S. urban areas (Chicago, IL, St. Louis, MO-IL,
Philadelphia, PA, and Los Angeles, CA) to estimate population exposures
associated with alternative NAAQS proposed for carbon monoxide (CO).
Computer outputs from NEM provide estimates of population exposure for
various measures of exposure and averaging times. In the case of CO, NEM
also estimates carboxyhemoglobin (COHb) levels, an important indicator of the
physiological effects of CO on the exposed population. In this addendum to
-------
the 1983 report, results of NEM analyses of CO exposure in four study areas
under two air quality assumptions are compared. The first assumption is that
an air quality standard of 9 ppm with one expected exceedance (9 ppm 1 ExEx)
is met in all four study areas. This assumption was used in previous reported
analyses. The second assumption is that an air quality standard of 9 ppm
with one observed exceedance is met in all study areas. The difference is that
the first form of standard is based on the characteristic high, an air quality
indicator statistically derived from daily maximum 8-hour values, while the
second standard is based on the second highest 8-hour running average observed
during the year. Note that a complete year of data contains 365 daily maximum
8-hour values versus nearly 8760 8-hour running average values. Note that
some "observed" values are estimates based on the time series model used in
NEM analyses. Table 1-1 lists the air quality indicators used for roll-back
purposes under each form of the standard.
-------
TABLE 1-1. AIR QUALITY INDICATORS FOR 8-HOUR CARBON MONOXIDE STANDARD
WITH ONE EXPECTED EXCEEDANCE OF 9 ppm PER YEAR AND WITH ONE OBSERVED
EXCEEDANCE OF 9 ppm PER YEAR
Study area
Chicago
Los Angeles
Philadelphia
*
St. Louis
NT
CR
CC
CI
SR
SC
SI
CR
CC
CI
SR
SC
SI
CR
CC
CI
SR
SC
SI
CR
CC
CI
SR
SC
SI
Air quality
Expected
exceedance
standard3
7,2
15.6
10.8
9.5
10.0
8.3
8.8
20.1
20.3
15.5
16.6
20.3
14.3
14.3
7.2
7.9
7.9
10.4
6.1
10.7
14.7
12.5
11.5
14.7
indicator, ppm
Observed
exceedance
standard0
7.5
15.9
10.6
12.2
12.2
9.5
9.0
18.1
21.7
13.6
15.6
21.7
18.3
18.3
11.4
8.3
8.3
11.7
6.7
11.1
14.4
14.0
10.2
14.4
Characteristic largest value estimated from distribution fit to daily maxi-
mum 8-hour values.
DSecond highest observed 8-hour running average value.
-------
SECTION 2
EXPOSURE ESTIMATES FOR ADULTS
WITH CARDIOVASCULAR DISEASE IN FOUR URBAN AREAS
Exposure estimates presented in this report are for adults with cardio-
vascular disease. Adults are defined to be persons at least 18 years old.
Based on the currently available evidence, this subpopulation is judged to be
the most sensitive group of persons with respect to CO-induced adverse health
effects. As indicated on page 7-34 of the 1983 report, the values used for
the percentage of adult females with cardiovascular disease was 4.2 percent
and for adult males 5.8 percent. These values are based on the U.S. Depart-
2
ment of Health, Education, and Welfare data on coronary heart disease and,
contrary to the statement on page 7-1 of the 1983 report, these estimates do
not include individuals with peripheral vascular disease.
2.1 "BEST ESTIMATE" RESULTS
Tables 2-1 through 2-9 contain selected printouts of a NEM analysis of
exposure of adults with cardiovascular disease to CO in the four study areas
under various air quality assumptions. Each table is identified as to CO/COHb
indicator and air quality standard being simulated. CO exposure estimates
are provided for both 1- and 8-hour average CO concentrations. In each case,
the "best-estimate" microenvironment factors developed in Section 6 of the
1983 report were used to simulate the contribution of gas stoves and smoking
to total CO exposure. Tables 2-1 through 2-9 are exactly analagous to Tables
7-1 through 7-9 of the 1983 report (pages 7-2 to 7-10) because the conditions
specified are exactly the same except that a different form of NAAQS is used.
2.1.1 Attainment of 9 ppm/1 QbEx Standard
NEM estimates in Tables 2-1 through 2-9 were developed by adjusting the
air quality data for each study area using the roll-back formula described
in Section 5.1 of the 1983 report so that the most polluted neighborhood type
just meets a 9 ppm - 8 hour average - 1 observed exceedance (9 ppm/1 ObEx)
-------
TABLE 2-1. ESTIMATES OF OCCURRENCES FOR ADULTS WITH CARDIOVASCULAR
DISEASE OF 1-HOUR CO EXPOSURES ABOVE SELECTED CONCENTRATION
VALUES ASSUMING 9 PPM/1 OBEX STANDARD IS ATTAINED
CONCENTRATION
EXCEEDED
(PPM)
60.0
55.0
50.0
45.0
40.0
35.0
30.0
25.0
20.0
15.0
12.0
9.0
7.0
0.0
MAX. CONCENTRATION
ENCOUNTERS AT MAX.
CHICAGO
524
24,600
212,000
746,000
2,960,000
9,760,000
1,070.000,000
25.1
523
LOS ANGELES
5,790
97,400
849,000
4,330,000
14,800,000
2,670,000,000
20.5
5,790
PHILADELPHIA
1,280
1,230
33,100
141,000
441,0^0
1,320,000
1,020,000,000
28.0
1,270
ST LOUIS
709
6,910
16,600
41,800
241,000
381,000
3,070,000
416,000,000
32.5
707
-------
TABLE 2-2. ESTIMATES OF ADULTS WITH CARDIOVASCULAR DISEASE WHO HAVE
1-HOUR CO EXPOSURES ABOVE SELECTED CONCENTRATION VALUES ASSUMING
9 PPM/1 OBEX STANDARD IS ATTAINED
CONCENTRATION
EXCEEDED
(PPM)
60.0
55.0
50.0
45.0
40.0
35.0
30.0
25.0
20.0
15.0
12.0
9.0
7.0
0.0
MAX. CONCENTRATION
PEOPLE AT MAXIMUM
CHICAGO
523
11,400
22.300
53,200
109,000
121,000
122,000
25.1
523
LOS ANGELES
5,790
49,300
167,000
243,000
298,000
305,000
20.5
5,790
PHILADELPHIA
1,270
1,270
23,000
36,300
69,300
91,100
116,000
28.0
1,270
1
ST LOUIS
707
6,200
10,400'
21,000
28,800
36,600
44,800
47,500
32.5
707
-------
TABLE 2-3. ESTIMATES OF ADULTS WITH CARDIOVASCULAR DISEASE WHOSE MAXIMUM
1-HOUR CO EXPOSURE OCCURS IN SELECTED CONCENTRATION RANGES
ASSUMING 9 PPM/1 OBEX STANDARD IS ATTAINED
CONCENTRATION
RANGE
(PFM)
60.0 < C <= 100.0
55.0 < C <= 60.0
50.0 < C <= 55.0
45.0 < C <= 50.0
40.0 < C <= 45.0
35.0 < C <= 40.0
30.0 < C <= 35.0
25.0 < C <= 30.0
20.0 < C <= 25.0
15.0 < C <= 20.0
12.0 < C <= 15.0
9.0 < C <= 12.0
7.0 < C <= 9.0
0.0 < C <= 7.0
MAX. CONCENTRATION
PEOPLE AT MAXIMUM
.
CHICAGO
524
10,900
10,900
30,900
55,500
12,400
494
25.1
523
LOS ANGELES
5,790
43,500
118,000
80, add
50,500
6,390
20.5
5,790
PHILADELPHIA
1,280
21,700
13,800
32,500
21,800
25,200
28.0
1,270
1
1
1
ST LOUIS 1
1
1
1
709
5,490
4,190
10,700
7,730
7,850
3,160
2,730
32.5
707
-------
TABLE 2-4. ESTIMATES OF OCCURRENCES FOR ADULTS WITH CARDIOVASCULAR DISEASE
OF 8-HOUR CO EXPOSURES ABOVE SELECTED CONCENTRATION VALUES ASSUMING
9 PPM/1 OBEX STANDARD IS ATTAINED
CONCENTRATION
EXCEEDED
(PPM)
60.0
55.0
50.0
45.0
40.0
35.0
30.0
25.0
20.0
15.0
12. 0
9.0
7.0
0.0
MAX. CONCENTRATION
ENCOUNTERS AT MAX.
CHICAGO
78,700
1,840,000
1,070,000,000
11.6
120
LOS ANGELES
80,100
1,370,000
2,670,000,000
10.1
2,000
PHILADELPHIA
66,700
422,000
1,020,000,000
11.0
29
ST LOUIS
81,100
473,000
416,000,000
11.6
11
-------
TABLE 2-5. ESTIMATES OF ADULTS WITH CARDIOVASCULAR DISEASE WHO HAVE
8-HOUR CO EXPOSURES ABOVE SELECTED CONCENTRATION VALUES ASSUMING
9 PPM/1 OBEX STANDARD IS ATTAINED
CONCENTRATION
EXCEEDED
(PPM)
60.0
55.0
50.0
45.0
40.0
35.0
30.0
25.0
20.0
15.0
12.0
9.0
7.0
1 0.0
1
1
1 MAX. CONCENTRATION
i PEOPLE AT MAXIMUM
I
CHICAGO
5,260
67,100
122,000
11.6
120
LOS ANGELES
5,790
65,100
305,000
10.1
2,000
PHILADELPHIA
20,300
35,400
116,000
11.0
29
ST LOUIS
15,000
24,400
47,500
11.6
11
-------
TABLE 2-6. ESTIMATES OF ADULTS WITH CARDIOVASCULAR DISEASE WHOSE MAXIMUM
8-HOUR CO EXPOSURE OCCURS IN SELECTED CONCENTRATION RANGES ASSUMING
9 PPM/1 OBEX STANDARD IS ATTAINED '
1
CONCENTRATION
RANGE
(PPM)
'
60.0 < C <= 100.0
55.0 < C < = 60.0
50.0 < C <= 55.0
45.0 < C <= 50.0
40.0 < C <= 45.0
35.0 < C <= 40.0
30.0 < C <= 35.0
25.0 < C <= 30.0
20.0 < C <= 25.0
15.0 < C <= 20.0
12.0 < C <= 15.0
9.0 < C <= 12.0
7.0 < C <= 9.0
0.0 < C <= 7.0
MAX. CONCENTRATION
PEOPLE AT MAXIMUM
CHICAGO
5,270
61,900
54,400
11.6
120
LOS ANGELES
5,790
59,300
240,000
10.1
2,000
PHILADELPHIA
20,300
15,200
80,900
11.0
29
ST LOUIS
15,000
9,430
23,100
11.6
11
10
-------
TABLE 2-7. ESTIMATES OF OCCURRENCES FOR ADULTS WITH CARDIOVASCULAR DISEASE
OF COHb LEVELS EXCEEDING SELECTED VALUES ASSUMING
9 PPM/1 OBEX STANDARD IS ATTAINED
COHB LEVEL
EXCEEDED
(PERCENT)
3.70
3.50
3.30
3.10
3.00
2.90
2.70
2.50
2.30
2.10
2.00
1.50
1.00
0.00
MAX. COHB CONC.
ENCOUNTERS AT MAX.
1
CHICAGO
66,400
7,820,000
1,070,000,000
1.89
54
LOS ANGELES
49,400
13,900,000
2,670,000,000
1.78
1,560
PHILADELPHIA
43,400
1,070,000
1,020,000,000
1.86
15
1
1
ST LOUIS 1
21
58,700
2,140,000
416,000,000
2.05
5
11
-------
TABLE 2-8. ESTIMATES OF ADULTS WITH CARDIOVASCULAR DISEASE WHO EXPERIENCE
COHb LEVELS EXCEEDING SELECTED VALUES ASSUMING
9 PPM/1 OBEX STANDARD IS ATTAINED
COHB LEVEL
EXCEEDED
(PERCENT)
3.70
3.50
3.30
3.10
3.00
2.90
2.70
2.50
2.30
2.10
2.00
1.50
1.00
0.00
MAX. COHB CONC.
PEOPLE AT MAXIMUM
CHICAGO
9,160
94,200
122,000
1.S9
54
LOS ANGELES
5,790
232,000
305,000
1.78
1,560
PHILADELPHIA
15,500
40,300
116,000
1.86
15
ST LOUIS
21
16,700
35,000
47,500
2.05
5
12
-------
TABLE 2-9. ESTIMATES OF ADULTS WITH CARDIOVASCULAR DISEASE WHOSE MAXIMUM
COHb LEVEL OCCURS IN SELECTED RANGES ASSUMING
9 PPM/1 OBEX STANDARD IS ATTAINED
COHB LEVEL
RANGE
(PERCENT)
3.70 < C <= 10.00
3.50 < C <= 3.70
3.30 < C <= 3.50
3.10 < C <= 3.30
3.00 < C <= 3.10
2.90 < C <= 3.00
1
2.70 < C <= 2.90
2.50 < C <= 2.70
1
2.30 < C <= 2.50
2.10 < C <= 2.30
2.00 < C <= 2.10
1.50 < C <= 2.00
1.00 < C <= 1.50
0.00 < C <= 1.00
MAX. COHB CONG.
PEOPLE AT MAXIMUM
CHICAGO
9,160
85,000
27,400
1.89
54
LOS ANGELES
5,790
227,000
72,800
1.78
1,560
PHILADELPHIA
15,500
24,800
76,000
1.86
15
1
ST LOUIS 1
1
1
21
16,700 1
18,300
1
12,500 1
2.05 1
5
1
13
-------
standard, i.e., one specifying that the number of 8-hour running average CO
values exceeding 9 ppm shall not be greater than 1 per year. Table 2-1 pro-
vides estimates of the number of occurrences for adults with cardiovascular
disease of 1-hour exposures to CO concentrations exceeding selected values.
(Exposures exactly equal to zero are counted as exceeding zero.) Thus, each
column in Table 2-1 presents a cumulative frequency distribution in which the
number of 1-hour exposures increases as CO concentration decreases; the dis-
tribution reaches a maximum at a CO concentration of zero. This maximum is
the number of adults with cardiovascular disease used in the simulation times
the number of possible occurrences in a year (8760). Although NEM yields
individual frequency distributions for cohorts who are at low, medium, and
high activity levels when a given CO concentration is encountered, only the
total frequency distribution for all activity levels is presented in Table
2-1. According to these estimates, none of the four study areas would have
more than 6,910 occurrences of 1-hour CO exposures above 25 ppm if a 9 ppm/1
ObEx standard were just attained. This statement is also true for the 9 ppm/1
ExEx standard, as shown in Table 7-1 in the 1983 report.
Table 2-2 uses an alternative exposure indicator, adults with cardio-
vascular disease with 1-hour exposures. This is the number of adults with
cardiovascular disease in the study area that experience one or more 1-hour
exposures per year to CO concentrations that exceed a specified value. This
exposure indicator is also expressed as a cumulative frequency distribution.
The number of adults with cardiovascular disease exposed at zero concentra-
tion (or above) is the total population of the study area.
Table 2-3 provides estimates of the number of adults with cardiovascular
disease who experience their peak exposure of the year within selected inter-
vals of 1-hour CO concentrations. These estimates are not cumulative; each
peak exposure falls within a single interval.
Tables 2-4 through 2-6 are similar to Tables 2-1 through 2-3 except that
exposures are estimated in terms of 8-hour running average CO concentrations.
Because the average of any eight successive hourly concentrations is less than
or equal to the highest value in the series, pollutant exposures usually occur
at lower concentrations for 8-hour running averages than for 1-hour averages.
For example, the maximum 8-hour running average concentration experienced in
Chicago is 11.^ ppm (Table 2-4), while the maximum 1-hour concentration is
14
-------
25.1 ppm (Table 2-1). Similarly, the number of 8-hour running average expo-
sures above 9 ppm is 78,700 in Table 2-4, compared with 2,960,000 1-hour
average exposures in Table 2-1.
The 1983 report lists the general algorithm used by NEM to estimate COHb
levels in the exposed populations. Full documentation of the rationale for
the choice of these values is provided in an EPA memorandum referenced in
that report. In essence, the carboxyhemoglobin algorithm estimates the COHb
levels of an individual at the end of every hour of the year. Although COHb
levels are, strictly speaking, the result of CO exposure, they can be described
using concepts similar to those used for CO exposure. For example, Table 2-7
lists the number of occurrences of COHb levels that exceed selected values.
Table 2-8 lists the number of adults with cardiovascular disease that experience
COHb levels which exceed selected values. Table 2-9 lists the number of adults
with cardiovascular disease who experience their highest COHb level within
selected ranges of COHb values. As would be expected, Tables 2-7 and 2-8
present cumulative distributions, while Table 2-9 lists results in discrete
intervals.
The relative frequencies of high COHb levels among the four study areas
can be compared by normalization, i.e., by converting the estimates of adults
with cardiovascular disease experiencing different COHb levels to the correspond-
ing percentage of total adults with cardiovascular disease in the study area
population. Table 2-10 shows that none of the study areas have adults with
cardiovascular disease with COHb levels exceeding 2.1 percent under the 9
ppm/1 ObEx standard. Approximately 13.4 percent of the Philadelphia adults
with cardiovascular disease experience COHb levels exceeding 1.50 percent.
Maximum COHb levels are 1.89 percent for Chicago, 1.78 percent for Los
Angeles, 1.86 percent for Philadelphia, and 2.05 percent for St. Louis.
As previously noted, the estimates presented in the tables are for
adults with cardiovascular disease. The value used for the percentage of
adult females with cardiovascular disease was 4.2 percent; for adult males
the value was 5.8 percent. These values are identical to those used in the
1983 report. In this application of NEM, estimates for the whole population
were ratioed down to the estimates for adults with cardiovascular disease by
using these two values in conjunction with estimates of the percentages of
adults who are male and female in each of the four cities (52% female and 48%
15
-------
TABLE 2-10. PERCENTAGE OF ADULTS WITH CARDIOVASCULAR DISEASE EXPERIENCING
COHb LEVELS EXCEEDING SELECTED VALUES ASSUMING
9 PPM/1 OBEX STANDARD IS ATTAINED
COHb level
exceeded
(percent)
3.00
2.90
2.80
2.70
2.60
2.50
2.40
2.30
2.10
2.00
1.50
1.00
0.00
Max. COHb
cone.
Percent
at maximum
Chicago
7.51
77.21
100.00
1.89
0.20
Los Angeles
1.90
76.07
100.00
1.78
2.14
Philadelphia
13.36
34.74
100.00
1.86
0.02
St. Louis
0.04
35.16
73.68
100.00
2.05
0.04
16
-------
male). The fact that all married women are females was accounted for in the
calculation, but the fact that the male/female percentage breakdown varies
from one occupation to another was not.
The estimates use 1980 census data for the four cities but are projected
to 1987 by using the multiplicative factor 1.195. The development of this
factor is explained in the 1983 report.
2.2 THE SIGNIFICANCE OF INDOOR SOURCES
Exposure estimates discussed in Section 2.1 assumed gas stoves and smok-
ing contribute to total CO exposure. The significance of these CO sources
was analyzed in Section 7.3 of the 1983 report. Maximum 1-hour CO exposures
were found to be less than 1.0 percent higher when indoor sources were included
in the analysis. Maximum 8-hour CO exposures were found to be 1.0 to 7.7
percent higher when indoor sources were included. In the NEM model, peak CO
levels are generally experienced in transportation vehicles or along roadways—
microenvironments with "best-estimate" multiplicative factors of 2.10 and
1.20, respectively.
2.3 UNCERTAINTY IN NEM EXPOSURE ESTIMATES
Any method used to estimate exposure of large, diverse groups of people
must deal with uncertainties in data and in assumptions used. The exposure
model can only represent major features. Because the relevant data bases are
often incomplete and/or inaccurate, professional judgment plays a significant
role in selecting monitors to represent neighborhood types, in validating air
quality data, in estimating cohort populations, and in determining cohort
movements. Lower, best, and upper estimates of microenvironmental factors
were presented in Section 6.0 of the 1983 report. The results of analyzing
these factors indicate that the difference between the lower estimates and
the upper estimates is appreciable. This large variation primarily results
from the large differences between lower and upper estimates of multiplicative
microenvironment factors, particularly those for transportation vehicles and
roadsides.
A limited sensitivity analysis was conducted on two of the physiological
variables which determine COHb levels in the blood resulting from given
17
-------
patterns of CO exposure. Variations in the Haldane constant and ventilation
rates have a significant effect, but not as large an effect as the variation
in estimated microenvironment factors. Details of these analyses were given
in the 1983 report.1
18
-------
SECTION 3
NATIONWIDE EXTRAPOLATIONS
Section 2 summarizes the results of applying NEM to four study areas,
namely, Chicago, Los Angeles, Philadelphia, and St. Louis. Rough estimates
of national exposure for adults with cardiovascular disease were made by
extrapolating these exposure and COHb estimates. The extrapolation procedure
is described in Section 8 of the 1983 report.
3.1 EXTRAPOLATION RESULTS
The results of the nationwide extrapolation are presented in Tables 3-1
through 3-10. The first nine tables can be divided into three sets of three
tables. Tables 3-1, 3-2, and 3-3 present exposure estimates for a 1-hour
averaging time. Estimates of occurrences during 1987 among adults with
cardiovascular disease of 1-hour average CO exposures above selected concen-
tration values under four alternative air quality assumptions are presented
in Table 3-1. Estimates of the number of adults with cardiovascular disease
in the urban U.S. who would incur 1-hour average CO exposures above the same
set of selected concentrations under the same assumptions are presented in
Table 3-2. Estimates of the number of urban U.S. adults whose maximum 1-hour
average CO exposure would occur in various concentration ranges are presented
in Table 3-3.
Analogous estimates for 8-hour average CO exposures are presented in
Tables 3-4, 3-5, and 3-6, respectively. Similar estimates for COHb levels
resulting from CO exposure are presented in Tables 3-7, 3-8, and 3-9. The
absolute numbers presented in Table 3-9 are presented in percentage form in
Table 3-10.
Estimates of the number of adults with cardiovascular disease who would
have their blood COHb levels elevated above selected concentrations for
various numbers of days if an 8-hour average 9 ppm/1 ExEx standard were just
met in all urban areas are presented in Table 3-8. The table indicates the
19
-------
TABLE 3-1. ESTIMATES OF OCCURRENCES IN THE CARDIOVASCULAR ADULT URBAN U.S.
POPULATION OF 1-HOUR AVERAGE CO EXPOSURES ABOVE
SELECTED CONCENTRATION VALUES UNDER ALTERNATIVE
AIR QUALITY ASSUMPTIONS
CONCENTRATION
EXCEEDED
(PPM)
60.0
55.0
50.0
45.0
40.0
35.0
30.0
25.0
20.0
15.0
12.0
9.0
7.0
0.0
MAX. CONCENTRATION
ENCOUNTERS AT MAX.
9 PPM 8HR 1EXEX
10,200
21,900
168,000
915,000
6,620,000
27,700,000
102,000,000
370,000,000
45,900,000,000
36.0
10,200
9 PPM 8HR 108EX
11,700
134,000
778,000
5,410,000
22,700,000
93,700,000
306,000,000
45,900,000,000
32.5
11,700
12 PPM 8HR 1EXEX
10,200
21,900
124,000
359,000
1,700,000
6,570,000
31,300,000
88,500,000
296,000,000
873,000,000
45,900,000,000
49.0
10,200
1
15 PPM 8HR 1EXEX 1
10,200
21,900
21,900
33,600
359,000
849,000
2,140,000
6,310,000
21,300,000
78,700,000
185,000,000
602,000,000
1,640,000,000
45,900,000,000
61.5
10,200
20
-------
TABLE 3-2. ESTIMATES OF CARDIOVASCULAR ADULTS IN URBAN U.S. WITH 1-HOUR
AVERAGE CO EXPOSURES ABOVE SELECTED CONCENTRATION
VALUES UNDER ALTERNATIVE AIR QUALITY ASSUMPTIONS
1
1 CONCENTRATION
EXCEEDED
(PPM)
60.0
55.0
50.0
45.0
40.0
35.0
30.0
25.0
20.0
15.0
12.0
9.0
7.0
0.0
MAX. CONCENTRATION
PEOPLE AT MAXIMUM
9 PPM 8HR 1EXEX
10,200
21,900
157,000
535,000
1,290,000
2,820,000
4,390,000
5,140,000
5,240,000
36.0
10,200
9 PPM 8HR 10BEX
11,700
123,000
423,000
1,150,000
2,440,000
4,210,000
4,960,000
5,240,000
32.5
11,700
12 PPM 8HR 1EXEX
10,200
21,900
113,000
343,000
691,000
1,290,000
2,900,000
4,290,000
5,030,000
5,240,000
5,240,000
49.0
10,200
15 PPM 8HR 1EXEX
10,200
21,900
21,900
21,900
348,000
531,000
872,000
1,270,000
2,510,000
4,290,000
4,800,000
5,220,000
5,240,000
5,240,000
61.5
10,200
21
-------
TABLE 3-3. ESTIMATES OF CARDIOVASCULAR ADULTS IN URBAN U.S. WHOSE MAXIMUM
1-HOUR AVERAGE CO EXPOSURE OCCURS IN SELECTED CONCENTRATION
RANGES UNDER ALTERNATIVE AIR QUALITY ASSUMPTIONS
CONCENTRATION
RANGE
(PPM)
60.0 < C <= 100.0
55.0 < C <= 60.0
50.0 < C <= 55.0
45.0 < C <= 50.0
40.0 < C <= 45.0
35.0 < C <= 40.0
30.0 < C <= 35.0
25.0 < C <= 30.0
20.0 < C <= 25.0
15.0 < C <= 20.0
12.0 < C <= 15.0
9.0 < C <= 12.0
7.0 < C <= 9.0
0.0 < C <= 7.0
MAX. CONCENTRATION
PEOPLE AT MAXIMUM
9 PPM 8HR 1EXEX
10,200
11,700
135,000
378,000
753,000
1,530,000
1,570,000
747,000
107,000
36.0
10,200
9 PPM 8HR 10BEX
11,700
111,000
300,000
729,000
1,290,000
1,770,000
745,000
284,000
32.5
11,700
12 PPM 8HR 1EXEX
10,200
11,700
90,600
235,000
344,000
603,000
1,610,000
1,390,000
738,000
211,000
4,730
49.0
10,200
15 PPM 8HR 1EXEX
10,200
11,700
326,000
184,000
341,000
401,000
1,240,000
1,770,000
510,000
426,000
18,600
61.5
10,200
22
-------
TABLE 3-4. ESTIMATES OF OCCURRENCES IN THE CARDIOVASCULAR ADULT URBAN U.S.
POPULATION OF 8-HOUR AVERAGE CO EXPOSURES ABOVE SELECTED CONCENTRATION
VALUES UNDER ALTERNATIVE AIR QUALITY ASSUMPTIONS
CONCENTRATION
EXCEEDED
(PPM)
60.0
55.0
50.0
45.0
40.0
35.0
30.0
25.0
20.0
15.0
12.0
9.0
7.0
0.0
MAX. CONCENTRATION
ENCOUNTERS AT MAX.
9 PPM 8HR 1EXEX
200,000
6,340,000
63,500,000
45,900,000,000
14.0
232
9 PPM 6HR 10BEX
3,690,000
51,900,000
45,900,000,000
11.5
2,480
12 PPM SHR 1EXEX
429,000
4,290,000
54,900,000
322,000,000
45,900,000,000
18.5
232
15 PPM 8HR 1EXEX
104,000
3,830,000
29,200,000
206,000,000
835,000,000
45,900,000,000
23.0
232
23
-------
TABLE 3-5. ESTIMATES OF CARDIOVASCULAR ADULTS IN URBAN U.S. WITH 8-HOUR
AVERAGE CO EXPOSURES ABOVE SELECTED CONCENTRATION VALUES UNDER
ALTERNATIVE AIR QUALITY ASSUMPTIONS
CONCENTRATION
EXCEEDED
(PPM)
60.0
55.0
50.0
45.0
40.0
35.0
30.0
25.0
ao.o
15.0
12.0
9.0
7.0
0.0
MAX. CONCENTRATION
PEOPLE AT MAXIMUM
9 PPtl 8HR 1EXEX
98,000
618,000
2,650,000
5,240,000
14.0
232
9 PPM 8HR 10BEX
534,000
2,230,000
5,240,000
11.5
2,480
12 PPM 8HR 1EXEX
176,000
538,000
2,410,000
4,090,000
5,240,000
18.5
232
15 PPM 8HR 1EXEX
49,400
529,000
1,770,000
3,660,000
4,950,000
5,240,000
23.0
232
24
-------
TABLE 3-6. ESTIMATES OF CARDIOVASCULAR ADULTS IN URBAN U.S. WHOSE MAXIMUM
8-HOUR AVERAGE CO EXPOSURE OCCURS IN SELECTED CONCENTRATION
RANGES UNDER ALTERNATIVE AIR QUALITY ASSUMPTIONS
CONCENTRATION
RANGE
(PPM)
60.0 < C <= 100.0
55.0 < C <= 60.0
50.0 < C <= 55.0
45.0 < C <= 50.0
40.0 < C <= 45.0
35.0 < C <= 40.0
30.0 < C <= 35.0
25.0 < C <= 30.0
20.0 < C <= 25.0
15.0 < C <= 20.0
12.0 < C <= 15.0
9.0 < C <= 12.0
7.0 < C < = 9.0
0.0 < C <= 7.0
MAX. CONCENTRATION
PEOPLE AT MAXIMUM
9 PPM 8HR 1EXEX
98,100
520,000
2,030,000
2,590,000
14.0
232
9 PPM 8HR 10BEX
534,000
1,700,000
3,010,000
11.5
2,480
12 PPM 8HR 1EXEX
176,000
362,000
1,870,000
1,680,000
1,160,000
18.5
232
15 PPM 8HR 1EXEX
49,400
480,000
1,240,000
1,890,000
1,290,000
291,000
23.0
232
25
-------
TABLE 3-7. ESTIMATES OF OCCURRENCES AMONG CARDIOVASCULAR ADULTS IN URBAN U.S.
OF COHb LEVELS EXCEEDING SELECTED VALUES UNDER ALTERNATIVE
AIR QUALITY ASSUMPTIONS
COHB LEVEL
EXCEEDED
(PERCENT)
3.70
3.50
3.30
3.10
3.00
2.90
2.70
2.50
2.30
2.10
2.00
1.50
1.00
0.00
MAX. COHB CONC.
ENCOUNTERS AT MAX.
9 PPM 8HR 1EXEX
570
12,000
41,700
4,730,000
296,000,000
45,900,000,000
2.30
120
9 PPM 8HR 10BEX
350
2,780,000
248,000,000
45,900,000,000
2.04
82
12 PPM 8HR 1EXEX
570
2,650
14,800
76,700
473,000
1,580,000
2,730,000
36,300,000
1,070,000,000
45,900,000,000
3.02
120
15 PPM 8HR 1EXEX
637
6,690
15,300
51,800
106,000
250,000
810,000
1,880,000
4,100,000
8,580,000
13,900,000
151,000,000
2,270,000,000
45,900,000,000
3.75
120
26
-------
TABLE 3-8. ESTIMATES OF CARDIOVASCULAR ADULTS IN URBAN U.S. EXPERIENCING
COHb LEVELS EXCEEDING SELECTED VALUES UNDER ALTERNATIVE
AIR QUALITY ASSUMPTIONS
COHB UEVEL
EXCEEDED
(PERCENT)
3.70
3.50
3.30
3.10
3.00
2.90
£.70
2.50
2.30
2.10
2.00
1.50
1.00
0.00
MAX. COHB CONC.
PEOPLE AT MAXIMUM
1
9 PPM 8HR 1EXEX
281
5,500
22,200
7*2,000
4,140,000
5,240,000
2.30
120
9 PPM 8HR 10BEX
346
598,000
3,620,000
5,240,000
2.04
82
12 PPM 8HR 1EXEX
281
1,250
6,230
42,000
217,000
452,000
660,000
2,100,000
5,090,000
5,240,000
3.02
120
15 PPM 8HR 1EXEX
345
3,200
6,600
25,800
58,200
132,000
308,000
506,000
732,000
1,060,000
1,350,000
3,480,000
5,180,000
5,240,000
3.75
120
27
-------
TABLE 3-9. ESTIMATES OF CARDIOVASCULAR ADULTS IN URBAN U.S. WHOSE MAXIMUM
COHb LEVEL OCCURS IN SELECTED CONCENTRATION RANGES
UNDER ALTERNATIVE AIR QUALITY ASSUMPTIONS
1
COHB LEVEL
RANGE
(PERCENT)
1
3.70 < C <= 10.00
3.50 < C <= 3.70
3.30 < C <= 3.50
3.10 < C <= 3.30
3.00 < C <= 3.10
2.90 < C <= 3.00
2.70 < C <= 2.90
2.50 < C < = 2.70
2.30 < C <= 2.50
2.10 < C <= 2.30
2.00 < C <= 2.10
1.50 < C <= 2.00
1.00 < C <= 1.50
0.00 < C <= 1.00
MAX. COHB CONC.
PEOPLE AT MAXIMUM
9 PPM 8HR 1EXEX
285
5,230
16,700
720,000
3,400,000
1,100,000
2.30
120
9 PPM 8HR 10BEX
350
598,000
3,020,000
1,630,000
2.04
82
12 PPM 8HR 1EXEX
285
972
4,990
35,900
175,000
235,000
208,000
1,440,000
2,990,000
149,000
3.02
120
15 PPM 8HR 1EXEX
351
2,860
3,400
19,200
32,400
74,100
175,000
198,000
227,000
327,000
291,000
2,130,000
1,700,000
63,000
3.75
120
28
-------
TABLE 3-10. PERCENTAGE OF CARDIOVASCULAR ADULT URBAN U.S. POPULATION
EXPERIENCING COHb LEVELS EXCEEDING SELECTED VALUES
UNDER ALTERNATIVE AIR QUALITY ASSUMPTIONS
COHb level
exceeded
(percent)
3.70
3.50
3.30
3.10
3.00
2.90
2.70
2.50
2.30
2.10
2.00
1.50
1.00
0.00
9 ppm 8 h
1 ExEx
0.01
0.10
0.42
14.16
79.01
100.00
9 ppm 8 h
1 ObEx
0.01
11.41
69.08
100.00
12 ppm 8 h
1 Ex Ex
0.01
0.02
0.12
0.80
4.14
8.63
12.60
40.08
97.14
100.00
15 ppm 8 h
1 ExEx
0.01
0.06
0.13
0.49
1.11
2.52
5.88
9.66
13.97
20.23
25.76
66.41
98.85
100.00
29
-------
frequency of repeated peak COHb levels. The table indicates, for example,
that of the 5,500 adults with cardiovascular disease who are estimated to have
their blood COHb level exceed 2.1 percent under the 9 ppm/1 ExEx standard,
none would have it occur more than one day.
3.2 UNCERTAINTY OF NATIONWIDE ESTIMATES
The uncertainty in the CO exposure and COHb estimates for the four base
cities was discussed in Section 2. The nationwide estimates are even more
uncertain because of the additional uncertainty introduced by the extrapola-
tion of exposure estimates for these four cities to all urban areas in the
U.S. The analyses discussed in the 1983 report indicated that uncertainty
is already great at the city level. That even greater uncertainty exists in
the nationwide estimates should be recognized when considering the estimates
presented in Tables 3-1 through 3-10.
3.3 COMPARING TWO FORMS OF NAAQS
Two forms of NAAQS are listed in Tables 3-1 through 3-10, both defined
at the 9 ppm level. The first form is the expected exceedance (ExEx) form
whereby one exceedance is expected to occur during the year, according to a
distribution fitting procedure. The second form is the observed exceedance
(ObEx) form whereby one observed exceedance means one 8-hour value at a
CO monitor exceeded the standard during the year.
Comparing exposure estimates under two forms of NAAQS in Tables 3-1 to
3-9 shows a rough similarity between the two distributions of exposure. For
example, Table 3-1 shows 102,000,000 occurrences of 1-hour average exposure at
9 ppm or above under the 1 ExEx form of NAAQS and 93,700,000 occurrences under
the 1 ObEx form. Table 3-6 shows 520,000 cardiovascular adults whose maximum
exposure occurred in the range 9 ppm to 12 ppm under the 1 ExEx standard and
534,000 cardiovascular adults whose maximum exposure occurred in the same
range under the 1 ObEx form.
Comparing exposure estimates from the two forms of standard also shows
a consistent downward shift in the distribution of exposures when the 1 ObEx
form is employed. Each table shows a small number of people at the tail of
the distribution who receive higher exposure under the 1 ExEx form of the
30
-------
NAAQS. Table 3-2 shows that an estimated 10,200 cardiovascular adults are
exposed to CO above 35 ppm (1-hour average) under the 1 ExEx form, whereas none
are exposed above this level under the 1 ObEx form. Table 3-5 shows that
98,000 cardiovascular adults are exposed above 12 ppm CO (8-h average) under
the 1 ExEx form whereas none are exposed above this level in the 1 ObEx form.
Comparing estimates of COHb levels of cardiovascular adults in Table
3-8 shows that 5,500 adults experience COHb above 2.10 ppm under the 1 ExEx
form whereas none experience this level under the 1 ObEx form. While these
distributions are different, both sets of estimates show no COHb levels above
2.50.
REFERENCES
1. Johnson, T., and R. Paul. The NAAQS Exposure Model (NEM) Applied to
Carbon Monoxide. Prepared by PEDCo Environmental, Inc. for Office of
Air Quality Planning and Standards, U.S. Environmental Protection Agency.
EPA-450/5-83-004, December 1983.
2. U.S. Department of Health, Education, and Welfare, Public Health Service.
Coronary Heart Disease in Adults. United States: 1960-1962, Vital and
Health Statistics Series 11, No. 10, December 1975.
31
------- |