United States
Environmental Protection
Agency
Atmospheric Research and Exposure
Assessment Laboratory
Research Triangle Park NC 27711
Research and Development
EPA/600/S3-90/003 Apr. 1990
£EPA Project Summary
Nonoccupational Pesticide
Exposure Study (NOPES)
Frederick W. Immerman and John L. Schaum
The Nonoccupational Pesticide
Exposure Study was the first attempt
to develop a methodology for
measuring the potential exposure of
specified populations to common
pesticides. In this study, as in other
studies utilizing the Total Exposure
Assessment Methodology (TEAM),
the exposures were related to actual
use patterns. A selected list of 32
household pesticides were evaluated
in two different cities during this
study.
Air samples were collected over a
24-hour period in indoor, outdoor and
personal microenvironments. In
addition, limited water and dermal
contact samples were collected for
selected homes. The study
households were selected from
stratified random population samples
in two urbanized areas. The samples
were collected over several seasons
in areas contrasting a relatively high
arid low use of pesticides. Dietary
recall, activity pattern, and pesticide
use data were collected through
survey questionnaires.
The report discusses the results of
the study with an emphasis on the
various routes of exposure (air,
water, dermal, and indirectly, food)
and their relative contribution to total
human exposure.
This Project Summary was
developed by EPA's Atmospheric
Research and Exposure Assessment
Laboratory, Research Triangle Park,
NC, to announce key findings of the
research project that is fully
documented in a separate report of
the same title (see Project Report
ordering information at back).
Introduction
In 1984, Congress appropriated FY85
monies to the U.S. Environmental
Protection Agency (EPA) to assess the
level of pesticide exposure experienced
by the general population. Occupational
exposure of specific groups of pesticide
users, such as farm workers and pest
control operators, had been examined
and characterized by previous studies.
However, little was known about the
general distribution of nonoccupational
exposures to household pesticides. To
begin to overcome this lack of knowl-
edge, NOPES was designed to provide
initial estimates of nonoccupational expo-
sure levels and to address the nature of
the variability in exposures.
NOPES was based on the Total
Exposure Assessment Methodology
(TEAM) approach to exposure estimation.
The Agency began developing the TEAM
approach in 1979 for measuring human
exposure to various environmental
contaminants. In a TEAM study, proba-
bility-based survey sampling procedures
are combined with questionnaire data
collection and modern personal
monitoring techniques to obtain
statistically defensible estimates of
exposure levels in the general population.
The initial application of this innovative
approach (Wallace, 1987) was in the
estimation of exposures to volatile
organic compounds (VOCs).
NOPES had both methodological and
analytical objectives. NOPES sought to
apply the TEAM approach to a class of
chemicals not previously addressed by
-------
TEAM. Therefore, the primary
methodological objective of NOPES was
to develop monitoring instrumentation,
laboratory procedures, and survey
questionnaires for a TEAM study of
pesticides. The overall analytical
objective of NOPES was to estimate the
levels of nonoccupational exposure to
selected household pesticides through
air, drinking water, food, and dermal
contact.
Procedure
Work on the design phase of NOPES
began in 1985. Southwest Research
Institute (SwRI), of San Antonio, Texas,
developed the methodology for collecting
air samples and analyzing them for 32
selected pesticides and pesticide
degradation products. Emphasis was
placed on both identifying and
quantitating the target compounds.
Research Triangle Institute (RTI) of
Research Triangle Park, North Carolina,
developed the probability-based
sampling design and the questionnaires
needed to collect information about
pesticide use and activity patterns. The
questionnaires and monitoring and
analysis procedures were tested in a pilot
study conducted in Jacksonville, Florida
in August and September 1985.
To permit assessment of regional and
seasonal variations in exposure levels,
the mam NOPES data collection was
conducted in three phases:
• Phase I: Summer 1986 in Jacksonville,
Florida.
• Phase II: Spring 1987 in Jacksonville,
Florida, and Springfield and Chicopee,
Massachusetts.
• Phase III: Winter 1988 in Jacksonville,
Florida, and Springfield and Chicopee,
Massachusetts.
The findings of EPA's National Urban
Pesticide Applicator Survey arid earlier
studies were used to select two study
areas. Jacksonville was selected as
representative of an area of the country
with relatively high pesticide use, and the
Springfield region was selected to
represent an area of low to moderate
pesticide use. In both study areas, some
sample members were asked to
participate in all seasons of the study,
whereas others were recruited only for a
single season. Monitoring some people in
more than one season permitted
assessment of whether the overall
differences observed between seasons
were due to true seasonal variations or
due to random sampling variations.
Short-term temporal variations were
addressed by monitoring some
respondents twice in the same season.
The following activities were performed
for each sample member who agreed to
participate in the study:
• A study questionnaire was admin-
istered
• A personal air sampler was given to the
participant to wear or keep in close
proximity for 24 h.
• Two or more fixed-site air samplers
were set up and run for 24 h. At least
one sampler was run in the
respondent's home, and at least one
was run outside the home.
• At the end of the 24-h monitoring
period, an activity log questionnaire
was administered.
In some households, drinking water
samples were collected for analyses.
Dermal exposure during pesticide
application events was estimated for a
small number of respondents by
analyzing cotton gloves worn during
typical application events following the
regular monitoring period.
In all phases, RTI recruited the sample
households, administered the
questionnaires, and statistically analyzed
the questionnaire and chemical data.
SwRI performed the environmental
monitoring and laboratory analyses. In
Phases I and II, Environmental Monitoring
and Services, Inc. (EMSI), of Camarillo,
California, provided overall program
management and quality assurance. EPA
assumed these functions in Phase III.
Results and Discussion
The second-stage (household
screening) sample size was 1,501
housing units in Jacksonville and 2,472
housing units in Springfield/Chicopee.
Screening information was obtained from
1,005 Jacksonville households and 1,774
Springfield households. Second-stage
response rates, computed as the number
of respondents divided by the number of
eligible sample members, were relatively
low for face-to-face household screening,
ranging trom 66% for the Jacksonville
spring season to 84% for the
Springfield/Chicopee winter season
(Table 1). Second-stage nonresponse
was due more to inability to contact
household members during the time
period allotted for screening (56% of
nonres p* ending eligible sample
members) than to refusals (32% of
nonresponding eligible sample
members).
Third-stage (personal monitoring)
response rates varied by study area,
season, and whether sample members
were single-season or multiseason
subjects. Nonresponse in the third stage
was primarily due to refusals to
participate (73% of nonresponding
eligible sample members). The two most
commonly cited reasons for refusing to
participate were the amount of time
required and the perceived burden
associated with keeping the personal
sampler nearby.
The overall response rates presented
in Table 1 (45% for Jacksonville and 40%
for Springfield/Chicopee) are comparable
to the 44% response rate experienced in
the New Jersey segment of the TEAM-
VOC study (Wallace, 1987). Although
these response rates are low relative to
those experienced in traditional area-
household surveys, they are typical of the
rates experienced in personal monitoring
studies. Low personal-monitoring
response rates are believed to be
primarily due to the respondent burden
imposed by the monitoring systems and
procedures.
Tables 2 and 3 present estimated
arithmetic means for indoor, outdoor, and
personal air concentrations for each
season in Jacksonville and Springfield/
Chicopee, respectively. Figures 1 and 2
present estimated cumulative frequency
distributions as log-normal probability
plots for personal air exposures for two of
the study pesticides, chlorpyrifos and
propoxur.
Mean outdoor air concentrations were
almost always lower than mean indoor
and personal concentrations. Mean
personal air and indoor air concentrations
were usually similar. Seasonal patterns
were somewhat inconsistent. However,
the pesticides found at higher
concentrations in Jacksonville were
highest in summer, followed by spring
and then winter. For Springfield/
Chicopee, the majority of the pesticides
found at higher levels had higher
concentrations in the spring than in the
winter. For a majority of the pesticides,
indoor and personal air concentrations
were higher in Jacksonville than in
Springfield/Chicopee, as expected.
Differences between the sites were less
consistent for outdoor air concentrations.
To assess the magnitude of short-term
variability relative to measurement error
and seasonal variations, absolute
differences between pairs of indoor air
-------
Table 1. Response Rates
Jacksonville
SpringfieldtChicopee
Second Stage
Sample Size
Eligible
Respondents
Response rate
Third Stage
First-time sample:
Selected
Eligible
Respondents
Response rate
Overall Response
Rate3
Followup sample:
Selected
Eligible
Respondents
Response rate
Total:
Selected
Eligible
Respondents
Summer
'86
401
363
267
74%
125
120
65
54%
40%
—
-
-
--
725
720
65
Spring
'87
550
510
336
66%
79
73
53
73%
48%
29
29
19
66%
108
102
72
Winter
'88
550
499
402
81%
95
90
55
61%
49%
79
79
76
84%
774
709
77
Total
1501
1372
1005
73%
299
283
773
67%
45%
48
48
35
73%
347
337
208
Spring
'87
1422
'367
956
70%
92
89
49
55%
39%
--
--
92
89
49
Winter
'88
7050
978
878
84%
73
72
37
57%
43%
20
20
75
75%
93
92
52
Total
2472
2339
1774
76%
165
161
86
53%
40%
20
20
15
75%
185
181
101
aOverall response rate = (Second-stage response rate)" (third-stige response rate) for first
time members of the sample
measurements were computed for the
five most prevalent pesticides. The mean
absolute differences in replicate indoor
air concentrations were computed for
each study area and season and
compared to fhe mean absolute
differences between duplicate indoor air
readings (Table 4) The mean absolute
differences between seasons in multi
season respondent indoor an
concentrations were also computed and
are presented in Table 4. The magnitude
of the differences between estimated
measure m e n t error van a b i I i t y
(duplicates;, estimated short-term
variability (replicates), and seasonal
variability /multisoason respondents)
varied considerably both within and
between analyles Because of the small
sample size devoted to this aspect of the
study and the magnitude of the variability
observed, only qualitative conclusions are
supported regarding the relative
magnitudes of these components of
variation. Measurement error variability ss
generally less than short-term variability,
which itself is usually less than seasonal
variability. Moreover, short-term and
seasonal variability are generally more
comparable than short-term and
measurement error variability. The fact
that the short-term and seasonal
variations were generally comparable in
magnitude suggests that the factors
contributing to short-term variations may
also be major components of seasonal
variations.
Conclusions and
Recommendations
Water sampling was by design only a
small component of NOPES Routine
sampling of public water supplies by
Jacksonville inci Springfield prior to
NOPES h a 1 i o 1 identified any
contamination by tne target compounds,
and watei ;. amplos collected and
analyzed during the NOPES pilot study
also din not contain detectable levels of
any analyte;-. Therefore, a minimal
sampling effort was believed to be
sufficient for estimating water exposure to
the target compounds.
The small sample sizes prevent
estimation < f weighted population
exposure est mates from these data.
However, the ;ack of detectable levels for
most anal/tes and the relatively low
lovels occasionally detected for others
suggest that exposure to the NOPES
target compounds from water is minimal
in the two study areas.
The dermal exposure component of
NOPES was primarily a pilot study of
a method for quantifying dermal
exposure levels during acute exposure
events. Chronic dermal exposure was not
addressed. The number of events
monitored was small, and events were
not randomly selected, so estimated
population exposure levels cannot be
developed. However, analysis of the
glove data does permit assessment of
the method, and provides an initial
impression of the relative importance of
acute dermal exposure.
Dermal dose was estimated for all 16
target compound applications monitored
in NOPES. It was computed by
multiplying the glove concentration by
the appropriate absorption factor and
ranged from 0.02 jig to 16,000 pg. Daily
air exposure doses were calculated as
the mean personal air concentration
estimates (ng/m3) from Tables 2 and 3
multiplied by 20 rn3 per day of respired
air. In only three of the 16 cases was the
dermal dose less than the estimated daily
air dose. The dermal dose was more than
an order of magnitude greater than the
daily air dose in more than half the cases.
Qualitative comparisons of the relative
exposure contributions of air and food
were possible for some of the target
compounds. The relative air and food
contributions were computed for daily
exposures. Mean daily exposure from
inhalation was estimated by multiplying
the mean personal air concentration
estimates (ng/m3) for each season
(Tables 2 and 3) by 20 m3 air respired
per day. These daily air exposure
estimates were then compared to daily
dietary exposure estimates. Only
qualitative comparisons were supported
by the data.
The NOPES air exposure data were
evaluated with regard to potential chronic
health effects. Bom cancer and non-
cancer risks were evaluated No risks of
major concern were identified.
Evaluation o! NOPES results, in
addition to providing important insights
about the nature and magnitude of
nonoccupationa! pesticide exposure,
suggests a number of possible avenues
for further research. Specific recom-
mendations are;
1. Develop guidance for conducting
exposure monitoring studies and
associated methodologies for
assessing human non-dietary
exposure to pesticides in residential
settings. These follow-up studies
will be designed to permit a more
comprehensive analysis of the
health risks associated with
-------
Table 2. Weighted Arithmetic
Analyte
Dichlorvos
alpha-BHC
Hexachlorobenzene
gamma-BHC
Chlorothalonil
Heptachlor
Ronnel
Chlorpyrifos
Aldrin
Dacthal
Heptachlor epox/de
Oxychlordane
Captan
Folpet
2,4-D ester*1
Dieldrin
Methoxychlor
Dicofol
cis-Permethhn
trans-Permethrin
Chlordane
4,4' -DDT
4,4' -ODD
4, 4' -DDE
ortho-Phenylphenol
Propoxur
Bendiocarb
Atrazine
Diazinon
Carbaryl
Malathion
Resmethrin
Mean Concentrations in Jacksonville Aira (ng/m*)
Indoor
Summer
134.5
1.2
1.3
20.2
5.3
163.4
0.2
366.6
31.3
0.2
0.5
5.2
1.9
0.5
1.8
14.7
0.2
0
0.5
0.4
324.0
-
-
-
96.0
528.5
85.7
0
420.7
68.1
20.8
0.1
Spring
86.2
1.2
0.4
13.4
2.2
154.9
0
205.4
6.8
0
0.8
0
2.2
0.7
0
8.3
0.3
11.0
1.9
1.1
245.5
1.0
0
0.6
70.4
222.3
5.5
0
109.2
0.4
14.9
0
Winter
24.5
1.1
0.3
6.0
6.7
72.2
0
120.3
6.9
0.3
0.8
6.5
0.1
0.6
25
72
0.2
0
1.3
0.8
220.3
0.5
0
0.2
59.0
162.5
3.4
0
85.7
0
20.4
0
Summer
0
0.0
0.2
1.3
0.2
30.2
0.1
16.7
0.2
0
0.7
0
0
0.3
0.0
0.7
0
0
0
0
38.4
-
-
-
1.2
10.2
0
0
12.6
0.2
0.3
0
Outdoor
Spring
0
0
0
0.5
0.3
10.7
0
3.5
0
0
0.1
0
0
0.4
0
0.0
0
0
0
0
9.5
0
0
0
0.0
0.8
0
0
1.1
0
0
0
Winter
3.2
0.0
0
0.6
0.6
2.8
0
2.5
0.1
0
0
0
0
0
0.8
0.8
0.1
0
0
0
27.3
0
0
0
0.1
2.5
0
0
13.8
0
0.2
0
Summer
147.6
0.9
0.9
22.1
0.5
129.1
0.1
280.4
19.9
0.6
0.6
0
0
0.4
0.7
10.1
0.3
0
0.1
0.1
212.0
..
„
„
79.7
375.6
51.4
0.3
321.6
28.3
9.2
0.4
Personal
Spring
40.2
0.8
0.4
7.0
0.0
133.7
0
182.8
38.5
0
0.5
0
0.1
0.4
0
5.4
0.1
0
1.3
0.3
190.7
0.5
0
0.5
55.6
141.1
4.4
0
112.7
0.8
10.1
0
Winter
21.4
0.7
0.4
8.5
2.5
64.2
0.0
118.2
6.9
0.2
0.1
0
0.1
0.8
3.5
4.8
0.6
0
0.8
0.5
194.8
0.4
0
0.8
39.7
142.8
3.5
o
89.0
0
16.8
0
a A weighted mean of "0
0.05.
b Methyl ester in summer,
means no detectable levels were observed. A
butoxyethyl ester in spring and winter
weighted mean of "0.0" means that the weighted mean was less than
exposure to pesticides from
different routes.
2. Conduct prospective studies to
estimate pesticide concentrations in
household dust in order to explore
the relationship between pesticide
use and exposure, and the relative
importance of the dust pathway to
total human exposure, especially
for infants and toddlers.
3. Refine the dermal exposure
sampling and analytical methods
requited for quantifying dermal
exposures and the estimation of
acute and chronic pesticide
exposures. These studies will
attempt to estimate transfer
coefficients between surface
applications and the dermal and
inhalation routes of exposure.
4. Improve the PUF sampling
technique to reduce variability in
matrix spike recoveries, evaluate
analytical methodology for new
compounds of interest, and prepare
quality assurance standards on
PUF media.
5. Conduct similar NOPES studies
following revision of the population
survey instruments. These revisions
would incorporate improvements to
the original survey design, develop
more appropriate stratification
variables, and permit the
development of a survey data base
with a larger regional or national
-------
Table 3. Weighted Arithmetic Mean Concentrations in Springfield'/Chicopee Air* (ng/m3)
Indoor
Analyte
Dichlorvos
alpha-BHC
Hexachlorobenzene
gamma-BHC
Chlorothalonil
Heptachlor
Ronnel
Chlorpyrifos
Aldrin
Dacthal
Heptachlor epoxide
Oxychlordane
Captan
Folpet
2,4-D butoxyethyl ester
Dieldrin
Methoxychlor
Dicofol
cis-Permethrin
trans-Permethrin
Chlordane
4,4'-DDT
4,4' -ODD
4,4 '-DDE
ortho-Phenylphenol
Propoxur
Bendiocarb
Atrazine
Diazinon
Carbaryl
Malathion
Resmethrin
Spring
4.3
0.2
0
0.5
0.1
31.3
0.2
9.8
0
1.6
0
0
0.1
0.7
2.1
1.0
0
0
0
0
199.3
0.0
0
0.9
44.5
26.7
0.2
0
48.4
0.3
5.0
0
Winter
1.5
0
0.1
9.5
0.1
3.6
0.0
5.1
0.3
0.3
0
0
0.0
0
0
4.2
0
0
0
0
34.8
0.5
0.0
0.6
22.8
17.0
0.4
0
2.5
0
0
0
Outdoor
Spring
0
0
0
0
0.4
0.3
0
13.9
0
0.9
0
0
0
0.5
0
0
0
0
0
0
3.1
0
0
0
1.6
0.8
0
0
8.2
0
0.8
0
Winter
0
0
0
0
0.8
0.1
0
0.0
0
0
0
0
0
0
0
0
0
0
0
0
2.0
0.2
0
0
0
0.1
0
0
9.2
0
0
0
Personal
Spring
3.7
0.0
0
0.7
0.8
34.7
0.1
7.5
0
2.6
0
0
01
07
0
0.8
0
7.0
0
0
252.9
0.9
0
4.9
43.4
16.2
0.3
0
10.1
0.1
0.5
0
Winter
2.1
0
0.0
5.4
0.1
4.6
0.0
5.9
0.2
0.3
0
0
0
0.0
0
0.7
0
0
0
0
35.9
0.7
0
0.5
27.3
11.3
0.2
0
1.4
0
0
0
*A weighted mean of "0" means no detectable levels were observed. A weighted mean of
"0.0" means that the weighted mean was less than 0.05.
application. The survey instruments
would incorporate more detailed
activity pattern information and
pesticide use applications. The data
would be combined with limited
monitoring data and used to
validate a proposed human
exposure model specifically
designed to estimate exposures to
several of the NOPES pesticides.
References
Wallace, L. A., 1987, The Total Exposure
Assessment Methodology (TEAM)
Study: Summary and Analysis: Volume
1. EPA/600/6-87/002. U.S. Environ-
mental Protection Agency, Washington,
DC 192pp.
-------
5,000
3,000
7,000
300
700
30
70
Legend
—•— JAX Summer
—m—JAX Spring
-+-JAX Winter
-m-SP/CH Spring
-A-SP/CH Winter
25% 50% 75% 90% 9'.% 99%
JAX 72,500 145,000 217,500 261.000 275.500287.100
SPCH 33,750 67,500 101.250 121,500 128,250 133650
Percent of Population tie low Concentration Shown
Figure 1. Chiorpynfos weighted cumulative frequency distribution for personal
an concentrations.
r
Legend
—»— JAX Summer
• jAX Spring
A -JAX Winter
m SP CH Sprir.c;
-A--SP OH Wirter
m »
25% 50% 7,!% 90% 95% 9S%
JAX 72500 145.000 217.^.00 261.000 27-5.500 287,700
SPCH 3?.750 67,500 101.250 121.500 12H.250 133,650
Percent of Population Below Concentration Shown
Figure 2. Propoxur weighted cumulative frequency distribution for personal air
concentrations.
-------
Table 4. Duplicate, Replicate and Seasonal Indoor Air Concentration Differences (ng/m3)
Chlordane
Jacksonville
Summer
Spring
Winter
Springfield
Spring
Winter
Chlorpyrifos
Jacksonville
Summer
Spring
Winter
Springfield
Spring
Winter
Heptachlor
Jacksonville
Summer
Spring
Winter
Springfield
Spring
Winter
ortho-Phenylphenol
Jacksonville
Summer
Spring
Winter
Springfield
Spring
Winter
Propoxur
Jacksonville
Summer
Spring
Winter
Springfield
Spring
Winter
Mean
Cone.*
55
505
145
51
54
247
268
187
63
18
13
142
43
5
7
81
101
51
107
54
142
378
92
48
10
Duplicates
Mean
Abs.
Diff.b
2
40
60
38
12
38
8
17
16
1
3
14
3
4
< 1
29
33
6
39
12
28
13
10
36
4
No. of
Pairs
6
10
9
8
7
6
10
9
8
7
6
10
9
8
7
4
10
9
8
7
4
10
9
8
7
Mean
Cone.8
277
249
129
64
140
362
162
152
34
5
157
114
64
20
26
91
96
82
26
46
289
168
51
64
17
Replicates
Mean
Abs.
Diff.*>
98
55
22
43
32
169
101
198
14
2
41
75
22
11
3
46
145
87
22
23
138
137
30
18
12
Multiseason Respondents
No. of
Pairs
8
10
9
10
10
8
10
9
10
10
8
10
9
10
10
5
10
9
10
10
5
10
9
10
10
Mean
Cone.
Over
Seasons0
369
242
32
259
122
13
218
124
10
75
80
34
529
197
52
Mean
Abs. Diff.
Between
Seasons''
343
114
29
276
114
11
223
108
15
72
117
38
629
184
77
No. of
Pairs
19
16
15
19
16
15
19
16
15
17
16
15
17
16
15
a Unweighted mean of all matched pair data.
bUnweighted mean of the absolute differences between matched oairs.
c Unweighted mean of data for two seasons from mjltiseason respondents. Values on the rows labelled 'Spring' are means for combined summer
and spring data: rows labelled 'Winter' are for combined spring and winter data.
dValues on rows labelled 'Soring' are the unweighted mean absolute differences between summer and spring concentrations, values on rows
labelled 'Winter' are for mean absolute differences between sp^ng and winter concentrations.
-------
Frederick W. Immerman is with Research Triangle Institute, Research Triangle
Park, NC 27709 and John L. Schaum is with the Office of Health and
Environmental Assessment, U.S. Environmental Protection Agency, Washington,
DC 20460.
Andrew E. Bond is the EPA Project Officer (see below).
The complete report, entitled "Nonoccupational Pesticide Exposure Study
(NOPES)," (Order No. PB 90-152 2241 AS; Cost: $31 00, subject to change) will be
available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Atmospheric Research and Exposure Assessment Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States Center for Environmental Research
Environmental Protection Information
Agency Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S3-90/003
------- |