Full text of DOI:10.1080/10473289.2000.10464199

This article was downloaded by: [Stony Brook University]
On: 05 October 2014, At: 01:51
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer
House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of the Air & Waste Management Association
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/uawm20
Exposure Efficiency: Concept and Application to
Perchloroethylene Exposure from Dry Cleaners
J.S. Evans a , K.M. Thompson a & D. Hattis b
a Harvard School of Public Health , Boston , Massachusetts , USA
b Clark University , Worcester , Massachusetts , USA
Published online: 27 Dec 2011.
To cite this article: J.S. Evans , K.M. Thompson & D. Hattis (2000) Exposure Efficiency: Concept and Application to
Perchloroethylene Exposure from Dry Cleaners, Journal of the Air & Waste Management Association, 50:9, 1700-1703,
DOI: 10.1080/10473289.2000.10464199
To link to this article: http://dx.doi.org/10.1080/10473289.2000.10464199
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose
of the Content. Any opinions and views expressed in this publication are the opinions and views of the
authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should
not be relied upon and should be independently verified with primary sources of information. Taylor
and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses,
damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection
with, in relation to or arising out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

Evans, Thompson, and Hattis
TECHNICAL PAPER
ISSN 1047-3289 J. Air & Waste Manage. Assoc. 50:1700-1703
Copyright 2000 Air & Waste Management Association
Exposure Efficiency: Concept and Application to
Perchloroethylene Exposure from Dry Cleaners
J.S. Evans and K.M. Thompson
Harvard School of Public Health, Boston, Massachusetts
D. Hattis
Clark University, Worcester, Massachusetts
ABSTRACT
estimates of emissions; use these in conjunction with in-
Standard approaches for computing population exposures
due to specific sources of air pollutants are relatively com-
plex. In many cases, more simple and approximate meth-
ods would be useful. This paper develops an approach,
based on the concept of exposure efficiency, that may be
used for estimating the impact of a source (or source class)
on the integrated population exposure. The approach is
illustrated by an example, which uses the concept of ex-
posure efficiency to examine the impact of perchloroeth-
ylene emissions from dry cleaners in the United States.
The paper explores the geographic variability of exposure
efficiency by evaluating it for each of 100 randomly se-
lected dry cleaners. For perchloroethylene, which has a
long atmospheric residence time, the site-to-site variabil-
ity in exposure efficiency is found to be relatively small.
This suggests that simple exposure assessments, based on
generic distributional characterizations of exposure effi-
ciency, may be used in risk assessments without intro-
ducing appreciable uncertainty. For many compounds, like
perchloroethylene, the uncertainty inherent in the esti-
mation of cancer potency or source emissions would domi-
nate these small errors.
formation on meteorology to estimate ambient concen-
trations; link these concentration estimates with data on
population density to compute population exposure; and,
finally, multiply this estimate of population exposure by
the appropriate cancer potency factor to estimate popu-
lation risk. This approach is intuitive and is perfectly rea-
sonable when resources, time, and data are available.
However, there are many circumstances (when such de-
tailed site-specific analyses are not feasible) in which crude
estimates of exposure would be quite useful, particularly
if the uncertainty inherent in the crude estimates could
be well characterized.
Here we consider an approach for developing prelimi-
nary exposure estimates, based on the idea of “exposure
efficiency.” Exposure efficiency is defined as the probabil-
ity that a molecule of pollutant emitted from a source
will reach and be inhaled by a human receptor.1 The inte-
grated exposure efficiency, ε (dimensionless), is found by
dividing the rate of population intake, I (g/sec), by the
rate of emissions, Q (g/sec)
(1)
INTRODUCTION
A common approach for estimation of population cancer
risk due to point sources of air pollution is to begin with
The population intake is found by summing the prod-
uct of the breathing zone concentration, Ci (g/m3), and
the breathing rate, B (m3/sec), across the population of
exposed individuals
IMPLICATIONS
(2)
Exposure efficiency calculations may be quite useful in
the early stages of regulatory analysis and risk assess-
ment. Analysts and regulators may benefit by adopting a
sequential approach in which the benefits of developing
improved exposure estimates are judged in the context
of other sources of uncertainty, for example, dose-
response, and in comparison to the likely costs and ben-
efits expected to flow from such improvements.
Clearly, the exposure efficiency depends on the loca-
tion (and density) of populations relative to the source of
interest, the various meteorological factors (e.g., wind
speed and direction, atmospheric stability, and mixing
height), and the properties of the pollutant (e.g., atmo-
spheric residence time) that influence pollutant transport
1700 Journal of the Air & Waste Management Association
Volume 50 September 2000

Evans, Thompson, and Hattis
and dispersion. Thus, spatial and temporal variations in
exposure efficiency are expected.
pact of local and global contributions to exposure effi-
ciency.
From the perspective of risk assessment, the concept
of exposure efficiency is most useful for pollutants where
risk is proportional to cumulative exposure and there are
no strong dependencies of risk on dose rate. In these cases,
the integrated exposure efficiency is of interest, and the
day-to-day variations in exposure efficiency due to
changes in meteorology can be ignored. The proposed
approach involves computing site-specific estimates of
exposure efficiency for a random sample from the popu-
lation of sources of interest and using the distribution of
site-specific exposure efficiency estimates to provide a
probabilistic characterization of the exposure efficiency
for other members of the source class.
Recently, there has been interest in the application
of approaches closely related to exposure efficiency in
pollution prevention and life-cycle analysis. For example,
Sheringer evaluated the spatial range and persistence of
organic compounds using multicompartmental “unit
world” models.4 Similarly, several researchers interested
in life-cycle analysis are currently evaluating exposure
efficiency as a method of accounting for chemical-to-
chemical variation in fate and transport.5,6 Most recently,
Lai et al. have estimated “inhalation transfer factors” for
particles from indoor and ambient sources.7 Wolff et al.
provide a more complete history of work related to expo-
sure efficiency.8
The concept of exposure efficiency is related to, and
in some sense derivative of, the compartmental model-
ing approaches proposed in the early 1970s for evaluat-
ing urban air pollution. For example, Gifford and Hanna
noted that on an urban scale, the average concentration
of an air pollutant is proportional to emissions, and that,
therefore, concentration per unit emissions, χ/Q, is a con-
stant that is determined largely by the wind speed and
the extent of vertical mixing.2 Exposure efficiency extends
this idea by integrating ambient concentrations across
exposed populations and accounting for breathing rate.
The population integrated exposure efficiency, ε = I/Q,
exhibits this same proportionality with emissions, but the
constant of proportionality reflects the combined impact
of geographic variation in air pollution concentrations
and in population density integrated across the relevant
spatial domain.
Harrison et al.1 were the first to discuss exposure effi-
ciency, defining it as “the fraction of total production that
is likely to reach people or the ratio of human intake to
the amount emitted.” In a technical report developed for
the U.S. Environmental Protection Agency (EPA), they
examined the variation in the exposure efficiency across
source classes and conditions of release (e.g., emissions in
private residences, emissions in the workplace, emissions
to the ambient environment) and developed point esti-
mates of “the ratio of exposure to emissions” for releases
of 35 compounds from a hypothetical point source in the
United States. Somewhat later, Smith defined “exposure
effectiveness” as “the fraction of released material that
actually enters someone’s breathing zone as measured in
exposure units (person - µg/m3)” and demonstrated the
power of exposure effectiveness in elucidating large dif-
ferences in the potential cost-effectiveness of alternative
pollution control strategies.3 While these studies provided
prototypical values of exposure efficiency for various
source classes, they did not examine the site-to-site varia-
tions in exposure efficiency or explore the relative im-
METHODS
To explore the potential utility of this idea, we computed
exposure efficiencies for 100 point sources of perchloro-
ethylene in the United States. The locations chosen for
analysis were the sites of 100 commercial dry cleaners se-
lected randomly from a representative list of 600 U.S. dry
cleaners provided by a national trade association. Because
perchloroethylene has a long atmospheric residence time,
people both near and far from the source may be exposed.
To account for this, we computed the total exposure effi-
ciency for each site as the sum of the local exposure effi-
ciency, reflecting the exposure of persons within 50 km
of the source, and the global exposure efficiency, reflect-
ing the exposure of more distant populations.
To compute the exposure efficiency, we used an emis-
sions rate, Q, of 1 g/sec, an effective release height of 1 m,
relevant population data from the census, and a nominal
breathing rate of 20 m3/day. The local exposure efficiency
for each site was computed using the EPA’s Human Expo-
sure Model (HEM) version II.9 HEM is a sector-averaged
Gaussian dispersion model coupled with block-level popu-
lation data from the 1990 U.S. Census. It relies on meteo-
rological data from the nearest stability array site.
Because the exposure efficiency is an integrated mea-
sure, the limits of integration are potentially important.
In our basic calculations, the population exposure was
evaluated over the range from 20 m to 50 km from the
source. Additional calculations were performed to exam-
ine the sensitivity of our local exposure efficiency esti-
mates to the scale of analysis. In these calculations, the
outer radius was varied from 25 and 100 km from the
source.
An estimate of the global exposure efficiency of per-
chloroethylene was developed using the simple one-com-
partment model developed by Thompson and Evans10
based on the concepts discussed by Turekian, Nozaki, and
Benniger.11 In this model, the global exposure efficiency,
Volume 50 September 2000
Journal of the Air & Waste Management Association 1701

Evans, Thompson, and Hattis
εg, of perchloroethylene is given by
limits of integration, the mean exposure efficiency shifted
only slightly—dropping to 92% of its original value when
integrated to 25 km and rising to 108% of its original value
when integrated to 100 km. Our best estimate of the glo-
bal exposure efficiency of perchloroethylene is 7 × 10–6,
with an uncertain range from 3.5 × 10–6 to 1.4 × 10–5. This
suggests that total exposure efficiencies, obtained by sum-
ming the local and global exposure efficiency, are on the
order of 1 × 10–5.
(3)
where Ng is the number of exposed people, B is the breath-
ing rate (m3/sec), V is the volume of the compartment
(m3), and L is the globally averaged rate constant (1/sec).
Following Thompson and Evans,10 our calculations as-
sumed that on a global scale, mixing occurs throughout
the full vertical extent of the troposphere, and used a
mixing height of 11 km. Since we were concerned with
emissions in the northern hemisphere, and there is lim-
ited atmospheric mixing between the hemispheres, we
used a box volume, V, of 1.6 × 109 km3 and a population,
Ng, of 3.8 × 109, both representative of the region between
15 and 55° N. The atmospheric residence time of perchlo-
roethylene is long but uncertain. Wang, Blake, and
Rowland provided a point estimate of L of two and one-
half per year.12 To characterize the overall uncertainty in
our estimates of global exposure efficiency, we treated the
atmospheric residence time as uncertain—allowing the
globally averaged rate constant, L, to vary from 1 per year
to 5 per year.
DISCUSSION
The concept of exposure efficiency is potentially useful in
preliminary calculations of risk for carcinogenic compounds
that do not involve mechanisms that strongly depend on
dose rate. For emissions of such compounds from point
sources in the United States, our calculations indicate that
when good estimates of emissions are available, popula-
tion exposures can be estimated to within a factor of 4
(95% confidence interval) without the need for detailed
site-specific transport and dispersion calculations.
In contrast, estimates of carcinogenic potency are fre-
quently quite uncertain. For example, McKone and Bogen
developed a probabilistic estimate of the low-dose human
carcinogenic potency of perchloroethylene in which 95%
confidence intervals spanned more than 4 orders of magni-
tude.13 Hattis and Goble asserted that the uncertainty for
genetically acting carcinogens is smaller than this, but it can
be characterized using a gsd of ~5.14 Cox gave a distribu-
tional estimate of the low-dose carcinogenic potency of ben-
zene in which more than 50% of the values were negative
(implying that at low doses benzene is anticarcinogenic) and
the positive values ranged from 0 to 2 × 10–2.15
Evans et al. developed
RESULTS
Figure 1 shows the cumulative frequency distribution of
values of local exposure efficiency obtained for the 100
sites. As the plot suggests, the distribution is approximately
lognormal. The median value of local exposure efficiency
for these sites is ~4 × 10–6, and the geometric standard
deviation of the distribution of values is ~2. When the
calculations were repeated using 25 and 100 km as outer
a probabilistic estimate of the
low-dose carcinogenic potency
of chloroform (in drinking
water) in which 95% confi-
dence intervals spanned more
than 5 orders of magnitude.16
Similarly, estimates of emis-
sions may be imprecise and
may depend on the approach
used for computing them.
Considering the large uncer-
tainties in estimates of carcino-
genic potency and the
potentially substantial errors in
Z-Score of Cumulative Density Function
emissions estimates, these
rather approximate estimates
of exposure efficiency may not
significantly influence the
overall uncertainty in esti-
mates of population risk.
For distances of 20 m–50 km from the source.
For perchloroethylene emitted 1 m above the ground.
Line is log (exposure efficiency) = –5.41 + 0.348 * z
Figure 1. Exposure efficiency—100 sites in the United States.
1702 Journal of the Air & Waste Management Association
Volume 50 September 2000

Evans, Thompson, and Hattis
7. Lai, A.C.K.; Thatcher, T.L.; Nazaroff, W.W. Inhalation Transfer Fac-
tors for Health-Risk Assessment; J. Air & Waste Manage. Assoc. 2000,
50, 1688-1699.
8. Wolff, S.; Kanchanasak, P.; Levy, J.; Smith, K.; Evans, J.S. Exposure
Efficiency: An Idea Whose Time Has Come? Atmos. Environ., in press.
9. HEM II User’s Guide; EPA/450/V-91-001-0; U.S. Environmental Pro-
tection Agency; Office of Air Quality Planning and Standards: 1991.
10. Thompson, K.M.; Evans, J.S. The Value of Improved National Expo-
sure Information for Perchloroethylene (Perc): A Case Study for Dry
Cleaners; Risk Anal. 1997, 17, 253-271.
11. Turekian, K.K.; Nozaki, Y.; Benniger, L.K. Geochemistry of Atmospheric
Radon and Radon Products; Annu. Rev. Earth Planet Sci. 1977, 5, 227-
255.
12. Wang, J.C.L.; Blake, D.R.; Rowland, F.S. Seasonal Variations in the
Atmospheric Distributions of a Reactive Chlorine Compound,
Tetrachloroethene (CCl2 = CCl2); Geophys. Res. Lett. 1995, 22 (9), 1097-
1100.
13. McKone, T.E.; Bogen, K.T. Uncertainties in Health-Risk Assessment:
An Integrated Case Study Based on Tetrachloroethylene in California
Groundwater; Regul. Toxicol. Pharmacol. 1992, 15, 86-193.
14. Hattis, D.; Goble, R. Expected Values for Projected Cancer Risks from
Putative Genetically Acting Agents; Risk Anal. 1991, 11, 359-363.
15. Cox, T.A., Jr. More Accurate Dose-Response Estimation Using Monte-
Carlo Uncertainty Analysis: The Data Cube Approach; Hum. Ecol. Risk
Assess. 1996, 2 (1), 150-174.
16. Evans, J.S.; Gray, G.M.; Sielken, R.L., Jr.; Smith, A.E.; Valdez-Flores,
C.; Graham, J.D. Use of Probabilistic Expert Judgment in Uncertainty
Analysis of Carcinogenic Potency; Regul. Toxicol. Pharmacol. 1994,
20, 15-36.
17. Finkel, A.M.; Evans, J.S. Evaluating the Benefits of Uncertainty Re-
duction in Environmental Health Risk Management; J. Air Pollut.
Control Assoc. 1987, 37, 1164-1171.
18. Hammitt, J.K.; Cave, J.A.K. Research Planning for Food Safety: A Value
of Information Approach; R-3946-ASPE/NCTR; Rand Corporation: Santa
Monica, CA, 1991.
In view of this, the approach that we propose for
analysis of population cancer risks from point sources of
air pollution would involve probabilistic estimation of
risks based on distributional estimates of emissions, ex-
posure efficiencies, and cancer potencies. Value of infor-
mation calculations, following an approach similar to that
outlined by Finkel and Evans or Hammitt and Cave, would
then be used to determine whether better estimates of
emissions, exposure, or potency were necessary.17,18
Thompson and Evans have illustrated this approach for
risk assessment and management of perchloroethylene
exposures from dry cleaners in the United States.10
Regulators with responsibilities for local or regional
air pollution control may focus on the estimates of local
exposure efficiency. However, those with interests in na-
tional or international pollution prevention and control
will need to rely on the estimates of total exposure effi-
ciency, which accounts for contribution of exposures of
populations at great distance from the sources. If we are
to achieve the full benefits of the proposed approach, fur-
ther work will be needed to develop default distributions
of exposure efficiency appropriate for a variety of com-
pounds, emissions scenarios, and locations, and regula-
tors will need to move toward more sequential approaches
for risk assessment and risk management, which reflect
the basic principles of value of information analysis.
ACKNOWLEDGMENTS
This work was funded partially by a cooperative agree-
ment with the EPA (CR 81 809 0-01-0), which supported
Kimberly Thompson’s doctoral work that formed the ba-
sis for this paper. We thank Mike Dusetzina, George
Duggan, and Warren Peters of the EPA’s Office of Air Qual-
ity Planning and Standards for providing us with infor-
mation about and access to the EPA’s Human Exposure
Model. Drs. Thompson and Evans have also received sup-
port from the Harvard Center for Risk Analysis.
REFERENCES
1. Harrison, K.; Hattis, D.; Abbat, K. Implications of Chemical Use for Ex-
posure Assessment: Development of an Exposure-Estimation Methodology
for Application in a Use Clustered Priority Setting System; Prepared for
U.S. Environmental Protection Agency; Report No. CTPID 86-2; MIT
Center for Technology Policy and Industrial Development: Cam-
bridge, MA, 1986.
About the Authors
2. Gifford, F.A.; Hanna, S.R. Modelling Urban Air Pollution; Atmos.
Environ. 1972, 7, 131-136.
Dr. Evans (corresponding author) is senior lecturer in Envi-
ronmental Health at the Harvard School of Public Health in
Boston, MA, where he co-directs the Program in Environ-
mental Science and Risk Management. Dr. Thompson is
assistant professor of Risk Analysis and Decision Science
at the Harvard School of Public Health. Dr. Hattis is research
professor at the Center for Technology, Environment and
Development in the George Perkins Marsh Institute at Clark
University in Worcester, MA.
3. Smith, K.R. Fuel Combustion, Air Pollution Exposure, and Health:
The Situation in Developing Countries; Annu. Rev. Energy Environ.
1993, 18, 529-566.
4. Scheringer, M. Persistence and Spatial Range of Endpoints of an Ex-
posure-Based Assessment of Organic Chemicals; Environ. Sci. Technol.
1996, 30, 1652-1659.
5. Jolliet, O.; Crettaz, P. Critical Surface-Time 95: A Life Cycle Impact As-
sessment Methodology including Fate and Exposure; Swiss Federal Insti-
tute of Technology; Institute of Soil and Water Management:
Lausanne, Switzerland, 1996.
6. Hertwich, E.; McKone, T.; Pease, W. A Systematic Uncertainty Analy-
sis of an Evaluative Fate and Exposure Model; Risk Anal., in press.
Volume 50 September 2000
Journal of the Air & Waste Management Association 1703