April
20,
2004
EPA
Docket
Center
Environmental
Protection
Agency
Office
of
Environmental
Information
Docket
oei.
docket@
epa.
gov
Mail
Code
2822T
1200
Pennsylvania
Ave
NW
Washington
DC
20460
Office
of
Information
and
Regulatory
Affairs
Office
of
Management
and
Budget
(
OMB)
Attention:
Desk
Officer
for
EPA
725
17th
Street
NW
Washington
DC
20503
RE:
Docket
ID
Number
ORD­
2003­
0011
EPA
ICR
Number
2126.01
EPA
Response
to
Comments
from
the
Phthalate
Esters
Panel
as
presented
in
Document
ID
Number
ORD­
2003­
0011­
0021
The
EPA
has
reviewed
the
comments
of
the
Phthalate
Esters
Panel
(
Panel),
referenced
in
the
document
ID
above,
regarding
EPA's
Information
Collection
Request
(
ICR)
for
the
"
Longitudinal
Study
of
Young
Children's
Exposures
in
Their
Homes
to
Selected
Pesticides,
Phthalates,
Brominated
Flame
Retardants,
and
Perfluorinated
Chemicals
(
A
Children's
Environmental
Exposure
Research
Study
 
CHEERS)."
The
Panel's
comments
focus
on
three
points:

1)
The
Panel
believes
that
the
quantity,
routes,
and
pathways
of
phthalate
exposure
are
characterized
by
existing
information
such
that
the
additional
work
of
this
study
is
not
needed.
2)
Existing
information
derived
from
CDC
biomonitoring
data
demonstrates
that
phthalate
exposures
to
the
general
population
are
well
below
levels
that
could
be
anticipated
to
pose
health
concerns,
and
therefore
that
no
further
regulation
is
needed.
3)
The
study
design
is
inappropriate
for
phthalates.

The
EPA
has
made
a
substantial
effort
to
design
a
study
that
will
provide
both
high
quality
and
high
quantity
data
that
will
have
practical
utility
for
improving
our
understanding
of
children's
exposures
to
chemicals
in
their
homes,
for
developing
exposure
assessment
methodologies
for
non­
persistent
chemicals,
and
for
reducing
the
uncertainty
of
risk
assessments
for
children.
The
study
has
been
designed
with
the
goal
of
collecting
a
large
amount
of
data
of
practical
utility
to
the
Agency
with
lowest
possible
burden
on
the
study
participants.
We
do
not
believe
that
the
comments
submitted
by
the
Panel
adequately
support
their
request
that
OMB
should
not
approve
the
proposed
collection
unless
phthalates
are
dropped
from
the
study
or
unless
the
study
design
is
changed.
On
the
contrary,
we
believe
that
review
of
the
data
that
they
2
cite
and
the
issues
they
raise
further
support
the
need
for
additional
research
on
phthalate
exposures
of
very
young
children
(
under
three
years
of
age),
as
proposed
for
this
study.

The
EPA
disagrees
with
the
Panel's
assertion
that
the
quantity,
routes,
and
pathways
of
phthalate
exposure
are
characterized
by
existing
information
and
are
sufficient.
The
Panel
cites
the
work
of
Clark
et
al.
to
demonstrate
the
quantity
of
available
data,
the
apportionment
of
routes,
and
estimates
of
exposure
for
different
age
groups.
However,
as
discussed
in
detail
below,
the
probabilistic
analysis
by
Clark
et
al.
was
used
to
assess
the
exposure
from
environmental
sources
only;
exposures
from
consumer
product
sources
were
not
included
in
the
estimates.
Clark
et
al.
also
noted
inadequacies
in
the
food,
indoor
air,
and
house
dust
data
used
in
the
exposure
estimates.
The
proposed
study
will
address
inadequacies
in
the
model
estimates
by
collecting
measurement
data
in
the
participant's
home.
These
data
will
be
used
by
the
EPA
to
make
aggregate
exposure
estimates,
which
will
be
related
to
measurements
of
the
phthalate
metabolites
in
urine
samples
collected
from
the
study
participants.

The
Panel
believes
that
because
existing
information
derived
from
CDC
biomonitoring
data
demonstrates
that
phthalate
exposures
to
the
general
population
are
well
below
levels
that
could
be
anticipated
to
pose
health
concerns,
and
therefore
that
no
further
regulation
is
needed.
There
are
three
points
that
should
be
raised
with
regard
to
this
comment:
(
1)
This
study
is
not
designed
to
collect
data
that
will
be
used
directly
to
regulate
phthalates;
this
is
not
one
of
the
objectives.
(
2)
It
should
be
recognized
that
there
is
a
high
level
of
uncertainty
associated
with
the
Brock
et
al.
data
that
the
Panel
cites
to
support
low
levels
of
exposure
for
children
under
six
years
of
age.
For
the
19
children
in
the
Brock
et
al.
study,
the
concentrations
of
monobutyl
phthalate
ranged
from
4.1
to
2540
mg/
mL.
For
one
child,
the
two
urine
samples
were
highly
variable
(
14.4
and
2540
mg/
mL).
Such
a
high
level
of
uncertainty
in
the
urine
metabolite
measurements
reduces
the
confidence
in
comparisons
to
EPA
reference
doses.
Collection
of
additional
data
is
warranted.
(
3)
Currently
risk
assessments
are
performed
for
individual
chemicals.
However,
in
the
past
several
years,
cumulative
risk
assessment,
aggregate
exposure
assessment,
and
research
on
chemical
mixtures
has
taken
on
increased
importance.
Assessing
cumulative
risk
through
complex
exposures
is
one
of
the
Agency's
high
priorities.
One
of
the
primary
objectives
of
this
study
is
to
evaluate
methodologies
for
estimating
aggregate
exposures
for
very
young
children.
The
design
of
this
study
will
address
a
number
of
Agency
needs.
It
is
imperative
that
the
EPA
move
forward
in
developing
methodologies
that
will
improve
the
risk
assessment
process
and
reduce
uncertainty
of
risk
assessments
for
young
children.

The
Panel's
third
point,
that
the
study
design
is
not
appropriate
for
phthalates
is
incorrect.
The
study
design
was
developed
to
study
current­
use
pesticides
that
are
registered
for
indoor
use.
These
are
primarily
the
pyrethrins
and
synthetic
pyrethroids,
which
are
non­
persistent
chemicals.
In
its
decision
to
include
phthalates
in
this
study,
the
EPA
considered
the
fact
that
the
phthalates
were
also
non­
persistent
chemicals.
The
collection
protocols
for
pesticides
are
applicable
to
the
phthalates.
The
pyrethroid
pesticides
and
the
phthalates
are
metabolized
and
eliminated
from
the
body
at
similar
rates.
Therefore
the
urine
collection
protocol
will
be
appropriate
for
both
classes
3
of
chemicals.
Other
issues
raised
by
the
Panel
regarding
the
study
design
are
not
considered
to
be
valid
within
the
context
of
the
objectives
for
this
study.
Those
issues
are
discussed
below.

The
following
is
a
more
detailed
discussion
of
the
specific
points
raised
by
the
Panel.

Panel
Comment:
EPA
states
that
the
reasons
for
this
study
"
are
to
better
identify
the
exposure
factors,
routes,
and
pathways
of
exposure
for
these
chemicals,
thus
improving
the
Agency's
ability
to
regulate
these
chemicals,
conduct
meaningful
risk
assessments,
and
develop
future
studies
."
69
Fed.
Reg.
at
10034­
35
(
emphasis
added).
In
its
previous
comments
(
pp.
3­
7),
the
Panel
discussed
the
large
quantity
of
data
already
available
for
phthalates
 
an
amount
paralleled
by
few
other
chemicals
 
and
the
many
risk
assessments
already
performed
on
these
chemicals.
The
conclusion
from
those
data
and
risk
assessments
is
that
phthalate
exposures
pose
very
low
health
concerns
and
therefore
that
no
further
regulation
is
needed.
EPA
Response:
The
purpose
of
this
study
is
not
to
conduct
a
regulatory
risk
assessment
for
phthalates
or
to
develop
data
specifically
for
new
regulation
as
implied
by
the
Panel.
The
Office
of
Research
and
Development's
(
ORD)
mission
is
to
provide
the
science
behind
EPA's
decisions
to
protect
health
and
the
natural
environment.
This
study
is
consistent
with
ORD's
mission
in
that
it
will
improve
the
tools
and
data
for
conducting
exposure
assessments
for
very
young
children.
As
stated
in
the
Abstract
of
the
Supporting
Statement,
"
The
study
will
help
the
Agency
reduce
uncertainty
in
exposure
and
risk
assessments
for
children
by
providing
data
on
exposure
factors
and
validated
tools
for
estimating
children's
exposure
to
contaminants,
as
well
as
providing
much
needed
measurement
data
for
model
refinement."
Therefore,
although
the
Panel
states
" 
that
no
further
regulation
is
needed "
(
for
phthalates),
additional
research
by
the
EPA
that
will
generate
data
that
will
reduce
uncertainty
in
exposure
assessments
is
warranted.
As
stated
in
Section
2(
b)
of
the
Supporting
Statement,
one
of
the
primary
objectives
of
this
study
is
to
"
validate
the
systematic
approach
for
evaluating
aggregate
exposure
described
in
the
Draft
Protocol."
However,
the
Agency
is
now
being
asked
to
go
beyond
evaluating
risks
from
aggregate
exposures
(
i.
e.,
exposure
to
a
single
chemical
by
all
routes
and
pathways)
to
evaluating
cumulative
risks
from
exposures
to
multiple
pollutants
and
stressors
over
time.
In
addressing
cumulative
risks,
we
must
move
beyond
quantifying
exposures
and
risk
to
one
chemical
at
a
time
at
a
relatively
high
level
to
understanding
exposures
to
the
mixtures
of
chemicals
that
are
occurring
at
low
levels
in
our
environment.
Because
phthalates
are
commonly
environmental
contaminants,
we
feel
that
it
is
important
to
develop
the
tools
and
data
for
including
phthalates
in
cumulative
risks
assessments.
In
addition,
EPA
uses
epidemiological
studies
as
well
as
risk
assessments
to
evaluate
health
risks
posed
by
environmental
contaminants.
The
study
will
provide
information
on
the
most
important
factors
associated
with
exposures
to
phthalates.
These
results
can
then
be
used
to
develop
simple
methods
and
questionnaires
for
assessing
exposures
in
future
epidemiological
studies
The
EPA
does
not
dispute
that
there
is
a
large
quantity
of
data
available
for
phthalates.
However,
it
is
an
error
to
imply
that
the
quantity
of
data
for
a
chemical
is
directly
correlated
to
our
understanding
of
exposures
to
the
chemical.
As
discussed
below,
the
systematic
approach
that
will
be
used
in
this
study
will
provide
important
data
on
routes
and
pathways
of
exposure
for
both
residential
use
pesticides
and
phthalates.
4
Panel
Comment:
EPA
states
that
existing
data
are
insufficient
"
to
show
how
young
children's
exposure
to
phthalates
changes
as
a
result
of
the
age
and
developmental
stages"
or
to
characterize
"
metabolite
levels
for
children
younger
than
6
years
of
age"
(
Supporting
Statement,
p.
7).
However,
as
discussed
in
the
Panel's
previous
comments,
existing
data
are
sufficient
to
show
that
phthalate
exposures
vary
little
with
age
and
would
in
all
cases
be
well
below
health
concern
levels.
For
example,
the
CDC
data
provide
the
following
exposures
to
dibutyl
phthalate
(
DBP),
in
micrograms
per
kilogram
per
day
(
ug/
kg/
day):

DBP
Exposure
Levels,
ug/
kg/
day
Age
Group
Mean
95th
percentile
6­
11
0.90
3.71
12­
19
0.52
2.05
20+
0.80
3.86
In
other
words,
mean
exposure
in
each
age
group
was
less
than
1
ug/
kg/
day,
and
the
95th
percentile
in
each
group
was
less
than
4
ug/
kg/
day.
In
contrast,
the
EPA
reference
dose
for
DBP
is
100
ug/
kg/
day
 
far
above
the
exposures
for
any
age
group.
4
The
no
observed
adverse
effect
level
(
NOAEL)
for
DBP
used
by
the
National
Toxicological
Program
Center
for
Evaluation
for
Risks
to
Human
Reproduction
is
50
milligrams
per
kilogram
per
day
(
mg/
kg/
day),
or
50,000
ug/
kg/
day
 
five
orders
of
magnitude
above
the
exposures
of
any
age
group.
There
are
some
variations
among
the
values
for
each
age
group,
but
those
variations
are
dwarfed
by
the
magnitude
of
levels
which
would
pose
any
health
concern.
As
EPA
points
out,
the
CDC
data
do
not
cover
children
below
6
years
of
age.
However,
a
study
by
Brock
et
al.
(
2002)
does
provide
a
median
value
of
2.6
ug/
kg/
day
in
19
infants
of
approximately
one­
year
old.
This
value
is
higher
than
the
values
from
the
CDC
data,
which
is
not
surprising
since
infants
eat,
drink
and
breathe
more
in
comparison
to
their
body
weight
than
older
children.
However,
in
comparison
to
the
EPA
reference
dose
and
the
NTP
CERHR
NOAEL,
the
value
is
near
the
CDC
values.
Detailing
the
precise
variations
in
closely­
grouped
levels
is
unnecessary
when
all
those
levels
are
far
before
levels
of
concern.
There
is
no
reason
to
believe
that
children
at
any
age
or
developmental
stage
would
have
exposures
significantly
different
from
those
observed
in
the
CDC
and
Brock
studies.
Even
if
one
assumes
that
the
Brock
infants
were
for
some
reason
all
at
the
low
end
of
the
exposure
spectrum,
and
that
"
real"
exposures
for
children
less
than
6
years
old
are
10
times
higher
than
seen
in
the
Brock
study
(
26
ug/
kg/
day),
such
values
yet
would
be
well
below
the
EPA
reference
dose
of
100
ug/
kg/
day
and
far
below
the
NOAEL
of
50,000
ug/
kg/
day.
EPA
Response:
The
EPA
disagrees
with
the
Panels
assertion
that
there
are
sufficient
data
"
to
show
how
young
children's
exposure
to
phthalates
changes
as
a
result
of
the
age
and
developmental
stages"
or
to
characterize
"
metabolite
levels
for
children
younger
than
6
years
of
age."
The
data
presented
in
the
table
are
for
children
over
age
6.
It
is
not
scientifically
sound
to
extrapolate
data
from
children
older
than
6
years
to
younger
children
for
whom
the
factors
impacting
exposure
may
be
substantially
different.
The
EPA
Risk
Assessment
Forum
(
RAF),
in
its
report,
Guidance
on
Selecting
the
Appropriate
Age
Groups
for
Assessing
Childhood
Exposures
to
Environmental
Contaminants,
discusses
the
substantial
developmental
changes
that
5
occur
over
the
period
from
birth
to
21
years
of
age
and
the
potential
impact
on
children's
exposures.
The
EPA
has
followed
the
RAF
guidance
by
designing
the
study
so
that
measurements
are
made
at
times
consistent
with
the
RAF
age
bins
of
1­
3
months,
3­
6
months,
6­
12
months,
1­
2
years,
and
2­
3
years,
the
periods
of
substantial
changes
in
development
and
activities
that
may
impact
exposure.
Exposures
of
children
in
this
age
group
to
chemicals
in
their
homes
are
not
well­
characterized.
The
EPA
is
not
aware
of
longitudinal
studies
that
have
taken
a
systematic
approach
to
performing
aggregate
exposure
estimates
for
children
in
this
age
group
to
a
suite
of
phthalates
or
that
collected
data
that
could
be
used
to
assess
temporal
variability
in
exposures.
Although
the
Panel
states
that
"
There
is
no
reason
to
believe
that
children
at
any
age
or
developmental
stage
would
have
exposures
significantly
different
from
those
observed
in
the
CDC
and
Brock
studies,"
the
Panel
fails
to
provide
data
to
support
that
assumption.
This
study
would
provide
the
needed
data.
The
Panel
suggests
that,
although
CDC
does
not
collect
data
for
children
under
6
years
of
age,
the
median
value
of
2.6
µ
g/
kg/
day
in
19
infants
of
approximately
one­
year
old
in
the
Brock
et
al.
study
should
be
sufficient
to
demonstrate
that
the
levels
of
exposure
are
far
below
levels
of
concern.
We
do
not
believe
that
the
Brock
et
al.
study
data
are
sufficient
because
of
the
limited
number
of
samples
collected
and
the
very
high
variability
in
the
urine
measurements.
The
samples
collected
in
the
Brock
et
al.
study
consisted
of
two
urine
samples
from
12
children
and
a
single
urine
sample
from
seven
children.
The
mean
urinary
metabolite
level
for
monbutylphthalate
(
MBP)
was
117.4
ng/
mL.
But
the
standard
deviation
of
the
measurement
for
the
31
samples
was
287.6
ng/
mL,
more
than
twice
the
mean,
reflecting
high
variability
in
the
MBP
concentrations
among
the
children.
The
range
of
MBP
urine
concentrations
was
from
4.1
to
2540
ng/
mL
in
the
31
samples.
Of
greater
concern
was
the
variability
in
the
urine
concentrations
for
the
same
children
with
two
urine
samples.
Although
there
were
cases
where
the
urine
MBP
concentrations
were
reasonably
similar,
child
7
had
measurements
of
14.4
and
2540
ng/
mL,
child
15
had
measurements
of
68.8
and
234
ng/
mL,
and
child
19
had
measurements
of
34.7
and
154
ng/
mL.
Brock
et
al.
concluded
that
"
The
large
variation
between
samples
from
the
same
child
at
different
times
suggests
that
two
spot­
urine
samples
alone
probably
do
not
adequately
describe
phthalate
exposure
for
any
individual
child.
Multiple
urine
samples
from
each
individual
child
may
be
required
to
fully
assess
the
exposure
to
phthalates."
The
EPA
believes
that
the
Panel
has
failed
to
adequately
evaluate
the
available
data
and
has
drawn
conclusions
that
are
not
supported
by
the
data
in
the
Brock
et
al.
study.
Contrary
to
the
opinion
of
the
Panel,
the
results
of
the
Brock
et
al.
study
show
that
it
is
essential
that
we
collect
more
data
in
order
to
be
able
to
understand
children's
exposures
to
phthalates,
the
exposure
levels,
and
the
factors
that
effect
those
exposures.
The
design
of
EPA's
proposed
study
will
address
the
problem
that
Brock
et
al.
has
identified
by
collecting
multiple
urine
samples
(
six
during
each
data
collection
event)
at
multiple
times
during
the
two
year
longitudinal
study.
Furthermore,
because
of
the
very
high
variability
in
the
urinary
concentrations
in
the
Brock
study,
we
do
not
believe
any
statement
can
be
made
about
the
levels
of
young
children's
exposure
to
phthalates.
Because
of
this
uncertainty,
we
should
not
attempt
to
relate
exposure
levels
to
EPA
reference
levels
until
more
data
are
available.

Panel
Comment:
While
it
may
be
of
academic
interest
to
detail
the
slight
variations
in
phthalate
exposures
among
children,
it
is
not
of
practical
interest
 
with
all
exposures
well
below
health
6
effect
levels
of
concern,
there
is
no
need
for
agency
intervention.
Therefore,
inclusion
of
phthalates
in
the
proposed
study
would
have
no
practical
utility.
EPA
Response:
As
discussed
above,
the
Brock
et
al.
data
show
large
variations
in
phthalate
exposures,
not
"
slight"
variations.
This
study
is
not
being
performed
out
of
academic
interest
to
detail
slight
variations
in
phthalate
exposures
among
children.
Nor
is
this
study
being
performed
in
response
to
a
"
need
for
Agency
intervention."
The
study
is
intended
to
collect
data
that
will
serve
as
the
scientific
basis
for
EPA's
exposure
and
risk
assessment
methodologies.
There
is
a
need
to
improve
the
risk
assessment
methodologies
and
to
reduce
uncertainty
in
risk
assessments.
This
study
will
provide
data
for
a
suite
of
phthalates
accumulated
over
sources,
over
time,
and
in
concert
with
many
other
chemical
contaminants.
Results
of
the
study
are
expected
to
have
broad
application
to
the
risk
assessment
methodologies
for
non­
persistent
chemicals.
This
study
will
provide
data
that
will
address
the
Agency's
needs
for
improved
methods
for
assessing
both
aggregate
and
cumulative
exposures
and
risks.
These
data
will
be
available
to
the
Panel
so
that
they
can
also
improve
their
risk
assessments.

Panel
Comment:
The
Study
Design
Does
Not
Acknowledge
the
Non­
Persistence
of
Phthalates.
The
proposed
study
was
designed
and
peer­
reviewed
with
respect
to
persistent
chemicals.
Phthalates
are
not
persistent.
EPA
has
acknowledged
this
by
changing
the
title
of
the
study,
but
it
has
not
otherwise
addressed
the
important
ramifications
of
this
distinction
for
the
appropriateness
of
the
study
design
with
respect
to
phthalates.
Phthalates
are
quickly
metabolized
and
eliminated
from
the
body,
so
that
accurate
measurement
of
phthalate
exposures
during
a
longitudinal
study
would
require
many
urinary
samples.
Chapin
et
al.
(
2004)
discuss
the
issue
as
follows:
Phthalates
.
.
.
are
cleared
from
the
body
with
relative
speed
(
several
days
to
a
few
weeks)
and
do
not
bioaccumulate
significantly
(
Tanaka
et
al.
1978).
From
the
public
health
and
environmental
perspectives,
this
is
incontrovertibly
good,
although
it
means
that
actually
determining
which
levels
of
which
compounds
are
adverse
requires
repeated
biological
sampling
and
analysis.
More
recent
data
indicates
that
metabolism
is
even
faster.
Anderson
et
al.
(
2001)
found
in
human
volunteer
studies
that
urinary
recovery
of
ingested
phthalates
was
complete
in
24
hours.
A
well­
designed
longitudinal
exposure
study
of
phthalates
would
take
human
metabolism
into
account
and
plan
for
numerous
samples.
EPA's
study
design,
however,
involves
only
5
or
6
visits
over
a
2
year
period.
Phthalate
values
for
each
visit
will
represent
phthalate
exposure
over
the
day
or
two
immediately
preceding
the
sample
visit,
which
may
or
may
not
be
representative
of
phthalate
exposures
over
the
course
of
the
two
years.
6
It
therefore
is
questionable
whether
the
exposure
factors
from
the
study
design
will
be
applicable
to
phthalates
 
especially
given
the
questions
discussed
below
concerning
potential
bias
in
the
sampling
and
lack
of
validated
questionnaires.
EPA
Response:
The
Panel
either
has
not
adequately
reviewed
the
study
design
or
does
not
understand
the
study
design.
The
study
was
designed
and
peer­
reviewed
to
address
study
objectives
related
to
current­
use
pesticides,
which
are
non­
persistent
chemicals.
The
pesticides
that
will
be
measured
in
the
study
include
chlorpyrifos
and
diazinon,
which
although
discontinued
for
indoor
use
may
still
be
on
homeowners
shelves
and
available
for
use,
and
the
current­
use
pesticides
including
the
pyrethrins
and
the
synthetic
pyrethroids.
These
pesticides
are
non­
persistent.
The
EPA
recognizes
that
phthalates
are
non­
persistent.
The
original
title
of
the
study
included
the
term
"
and
Persistent
Pollutants"
because
we
proposed
to
measure
7
brominated
flame
retardants
in
the
study
(
See
Study
Design,
Section
2.1,
Objectives).
The
title
was
changed
to
state
exactly
what
chemicals
were
being
measured
in
the
study.
The
study
design
is
appropriate
to
address
objectives
related
to
phthalate
exposures
because
it
was
originally
developed
to
address
objectives
related
to
exposures
to
non­
persistent
pesticides.
The
decision
by
the
EPA
to
add
phthalates
recognized
that
both
classes
of
chemicals
were
nonpersistent
The
Panel's
criticism
of
the
EPA
study
design
with
respect
to
the
timing
of
the
collection
of
urine
samples
is
in
error.
EPA
researchers
are
knowledgeable
about
phthalate
metabolism.
At
the
time
that
phthalates
were
considered
for
inclusion
in
the
study,
available
data
were
reviewed
to
determine
if
urine
samples
collected
for
measurement
of
pesticide
metabolites
could
also
be
used
for
measurements
of
phthalate
metabolites.
This
review
indicated
that
the
metabolism
rates
were
similar
enough
for
the
pyrethroid
pesticides
and
the
phthalates
that
the
sample
collection
protocol
in
the
original
study
design
was
acceptable
for
both
classes
of
chemicals.
To
address
concerns
of
the
Panel,
urine
samples
will
be
collected
both
prior
to,
and
following,
the
pesticide
application.
Prior
to
a
pesticide
application
in
the
home,
two
urine
samples
will
be
collected
(
one
in
the
morning
and
one
in
the
evening)
in
order
to
measure
the
background
metabolite
concentrations
prior
to
a
pesticide
application
in
the
home.
Within
24
hours
or
less
following
a
pesticide
application
in
the
home,
urine
sample
collection
will
begin.
The
urine
sample
collection
will
be
timed
to
be
coincidental
with
collection
of
air
samples,
diet
samples,
and
other
environmental
or
personal
samples
in
the
study
home.
A
series
of
four
urine
samples
will
be
collected
at
approximately
12,
24,
36,
and
48
hours
after
the
start
of
collection
of
air
and
diet
samples,
so
that
the
metabolite
concentrations
in
the
urine
can
be
related
to
aggregate
exposures
estimated
from
the
measurements
of
the
chemicals
in
the
environmental
media.
Therefore,
for
each
data
collection
event,
six
urine
samples
will
be
collected
from
each
child
in
the
study,
providing
background
data
and
temporal
variability.
Over
the
two­
year
period,
data
collection
will
be
performed
six
times
for
each
child
enrolled
into
the
study
as
newborns,
for
a
total
of
36
urine
samples
per
child
during
the
study.
The
six
urine
samples
to
be
collected
over
three
days
for
each
child
during
each
sample
collection
event
are
substantially
more
than
Brock
et
al.
or
CDC
collected.
The
protocol
has
been
designed
to
address
deficiencies
in
urine
sample
collection
protocols
used
in
previous
studies.

Panel
Comment:
The
study
design
does
not
address
potential
bias
due
to
timing
of
sampling
with
pesticide
use.
The
timing
for
the
samples
to
be
analyzed
for
phthalates
is
after
a
pesticide
use,
but
the
study
design
provides
no
evidence
of
consideration
of
whether
such
timing
might
bias
estimation
of
exposures
to
the
other
substances
under
study,
including
phthalates.
For
example,
will
pesticide
use
affect
behavioral
activity
(
different
rooms,
different
parental
control
over
crawling,
mouthing
behavior)
that
could
affect
phthalate
exposure?
Are
phthalates
present
in
the
pesticides
being
used?
With
respect
to
the
last
question,
EPA's
response
is
that
"
EPA
is
not
attempting
to
identify
the
source
of
the
phthalates,
but
the
young
children's
aggregate
exposure
to
them"
(
Supporting
Statement,
p.
9).
However,
if
all
samples
are
taken
after
pesticide
use,
and
the
pesticide
contains
phthalates,
then
the
resulting
aggregate
exposure
value
is
going
to
be
artificially
exaggerated.
EPA's
Response:
It
is
true
that
the
timing
of
sample
collection
is
related
to
a
pesticide
application
in
the
residence.
This
was
an
essential
element
of
the
study
design
in
order
for
the
8
study
to
meet
the
objectives
related
to
measurements
of
exposure
factors,
evaluation
of
the
impact
of
activity
and
developmental
lifestage
on
exposure,
etc.
We
considered
this
in
the
design
of
the
phthalate
component
of
the
study.
This
element
of
the
study
design
will
not
bias
the
estimates
of
phthalate
exposures.
Urine
samples
will
be
collected
both
before
and
following
the
pesticide
application.
Therefore,
there
will
be
data
on
the
background
level
of
phthalate
metabolites
in
the
urine
prior
to
the
pesticide
application.
Even
if
there
are
phthalates
in
the
pesticides
used
by
the
occupant
of
the
study
home,
this
could
not
be
considered
a
bias
because
exposure
to
phthalates
in
this
consumer
product
(
the
pesticide)
should
be
considered
part
of
the
child's
total
exposure
to
phthalates
in
the
home.
With
regard
to
the
impact
of
a
pesticide
application
on
a
child's
behavior
and
activities,
we
will
collect
time/
activity
data
during
both
the
pre­
and
post­
application
periods
and
will
be
able
to
evaluate
whether
the
application
affected
the
behavioral
activity
of
the
child.
We
are
not
attempting
to
"
exaggerate"
the
exposure
to
phthalates,
but
to
measure
the
exposures
in
real­
life
situations,
which
include
use
of
everyday
consumer
products,
including
pesticides.

Panel
Comment:
It
appears
there
will
not
be
appropriate
questionnaires
for
phthalates.
In
its
previous
comments,
the
Panel
asked
whether
the
environmental
questionnaires
were
appropriately
designed
to
collect
exposure
information
relevant
to
sources
of
phthalates.
EPA's
response
is
thus:
After
consultation
with
many
government
and
non­
government
researchers,
the
conclusion
was
reached
that
there
are
no
validated
questionnaires
currently
being
used
for
exposure
assessments
to
phthalates.
However,
EPA
believes
it
critically
important
to
use
questionnaires
that
have
been
evaluated
or
used
by
other
researchers.
This
helps
in
consistency
of
the
analysis
and
interpretation
of
the
data.
Therefore,
the
questions
relating
to
potential
phthalate
exposures
have
been
taken
from
other
researchers.
(
Supporting
Statement,
p.
9)
This
response
appears
to
be
an
admission
that
the
questionnaires
may
not
be
appropriate
for
phthalates
 
certainly
that
they
are
not
validated
for
phthalates.
The
EPA
concludes
that
since
no
validated
questionnaires
are
currently
being
used
for
exposure
assessments
to
phthalates,
the
next
best
thing
is
to
take
questionnaires
used
in
studies
for
other
chemical
exposures.
This
approach
seems
to
be
based
on
expediency
and
a
lack
of
understanding
of
exposure
to
phthalates.
Detailed
information
on
use
patterns,
including
brands,
and
timing
in
relation
to
exposure
measurements
would
be
needed
to
place
results
in
a
meaningful
context.
This
is
not
being
done
in
the
proposed
study.
EPA's
Response:
The
EPA's
approach
for
this
study
is
based
on
the
best
available
information
on
exposure
to
phthalates.
The
Panel's
insinuation
that
the
EPA's
approach
is
based
on
expediency
and
a
lack
of
understanding
is
incorrect.
The
fact
that
there
is
not
a
validated
questionnaire
for
phthalates
suggests
that
information
is
needed
to
improve
our
understanding
of
exposures
to
phthalates
and
the
factors
that
affect
these
exposures.
The
EPA
reviewed
a
number
of
questionnaires
that
are
currently
being
used,
or
have
recently
been
used,
in
studies
involving
exposure
to
phthalates.
The
predictive
value
of
questionnaires
being
used
in
current,
or
very
recently
completed,
studies
has
not
been
validated
because
data
analyses
are
not
completed.
Other
questionnaires
used
in
previous
studies
could
not
be
validated
due
to
inadequate
study
design.
Knowledgeable
researchers
in
the
exposure
assessment
field
are
aware
that
there
are
few
questionnaires
available
that
have
been
validated
to
accurately
predict
exposure
to
chemicals
in
the
environment.
This
is
particularly
true
for
chemicals
with
multiple
and
complex
intermittent
sources
such
as
phthalates.
We
have
developed
a
questionnaire
that
is
based
on
the
best
9
available
information
from
academic
and
government
researchers
at
this
time.
The
fact
that
the
questionnaire
is
not
"
validated"
does
not
mean
it
is
not
"
appropriate"
for
use
in
the
study.
We
will
be
able
to
evaluate
its
predictive
value
from
the
data
collected
in
this
study.
The
Panel
is
correct
that
detailed
information
on
use
patterns,
including
brands
and
timing
in
relation
to
exposure
measurements,
is
not
being
collected
with
the
questionnaires.
Collection
of
such
detailed
information
would
place
too
heavy
of
a
burden
on
the
participants
in
this
study
and
was
not
considered
to
have
practical
utility
for
the
study.
Although
this
information
may
be
useful,
it
is
not
required
to
meet
the
objectives
of
the
study,
which
include
estimating
aggregate
exposures
of
children
to
phthalates,
evaluating
the
relationship
between
the
biomarkers
of
exposure
and
the
aggregate
exposure
estimates,
and
evaluating
and
apportioning
the
exposure
pathways
for
selected
phthalates.
It
is
not
necessary
to
identify
the
specific
sources
or
use
patterns
to
meet
the
objectives
of
the
study.
The
questions
included
in
Form
#
4:
Monitoring
Period
Questionnaire,
will
provide
information
of
practical
utility
to
address
the
study
objectives,
without
placing
a
heavy
burden
on
the
participant.

Panel
Comment:
The
study
cannot
significantly
increase
our
understanding
of
phthalate
exposure
pathways.
EPA's
Supporting
Statement
(
p.
5)
asserts
that
the
data
generated
by
the
proposed
study
will:
°
"
estimate
the
aggregate
exposures
of
a
cohort
of
very
young
children
to
the
selected
phthalates
in
their
residential
environments,"
°
"
evaluate
the
relationship
between
biomarker
concentrations
and
exposure
estimates
based
on
measurements
of
the
selected
phthalates
in
environmental
media
and
diet,"
and
°
"
evaluate
and
apportion
the
exposure
pathways
for
the
selected
phthalates."
However,
EPA
elsewhere
states
"
EPA
is
not
attempting
to
identify
the
source
of
the
phthalates,
but
the
young
children's
exposure
to
them"
(
Supporting
Statement,
p.
9).
If
EPA
does
not
identify
the
sources
of
phthalates,
it
is
not
clear
how
it
can
evaluate
the
relationship
between
biomarker
concentrations
and
exposure
estimates
or
evaluate
and
apportion
the
exposure
pathways.
The
only
answer
apparent
to
the
Panel
is
that
EPA
will
use
its
measurements
of
phthalates
in
air,
dust
and
food
to
estimate
the
exposure
from
each
route,
and
will
compare
those
estimates
with
the
aggregate
exposure
indicated
by
the
urinary
metabolite
measurements.
As
described
in
the
Panel's
previous
comments
(
p.
4),
such
work
has
already
been
accomplished
by
Clark
et
al.
(
2003a,
b),
who
conducted
a
thorough
literature
search
of
information
on
phthalate
environmental
concentrations
and
used
that
information
to
estimate
exposures
of
various
age
groups.
As
designed,
EPA's
proposed
study
would
not
be
expected
to
provide
additional
significant
information
about
exposure
pathways.
EPA's
Response:
The
Panel
is
correct
that
the
EPA
will
use
the
measurement
data
for
the
various
media
collected
to
assess
the
different
routes
of
exposure
to
estimate
the
aggregate
exposure
for
each
child
during
each
of
the
five
or
six
data
collection
events
over
the
two­
year
study.
The
data
collected
to
perform
these
estimates,
and
the
algorithms
used
for
the
calculations
are
described
in
the
Draft
Protocol
for
Measuring
Children's
Non­
Occupational
Exposure
to
Pesticides
by
all
Relevant
Pathways
and
shown
below.
10
Summary
of
Data
Collection
Requirements
by
Exposure
Route
Parameter
Measurement
How
Collected
Units
Inhalation
Exposure
Eime/
ma
=
Came
x
Tme/
ma
x
IRma
Came
Air
concentration
in
me
Measured
with
active
sorbent
collection
µ
g/
m3
Tme/
ma
Time
spent
in
each
me/
ma
Time­
activity
diary,
questionnaire
h/
d
IRma
Inhalation
rate
Estimated
from
size,
age,
and
activity
data
collected
with
diaries
and
questionnaires
using
reference
values
m3/
h
Dermal
Exposure
­
Macroactivity
Approach
Edme/
ma
=
Csurf
x
TCme/
ma
x
ADme/
ma
Csurf
Surface
loading
(
total
or
transferrable)
in
each
me
Measured
by
wipe,
press,
or
roller
methods
µ
g/
cm2
TCme/
ma
Transfer
coefficienta
Empirically
determined
for
each
me/
ma
from
laboratory
experiments
or
field
studies
cm2/
h
ADme/
ma
Activity
duration
for
ma
in
a
specific
me
Time­
activity
diary,
questionnaire
h/
d
Dermal
Exposure
­
Microactivity
Approach
Edmi
=
Csurf
x
TE
x
SA
x
EF
Csurf
Surface
loading
(
total
or
transferable)
in
me
Measured
by
wipe,
press,
or
roller
method
µ
g/
cm2
TE
Transfer
efficiencya
Empirically
determined
from
laboratory
experiments
unitless
SA
Surface
area
contacted
Visual
observation
or
videotape
cm2/
event
EF
Frequency
of
contact
events
Visual
observation
or
videotape
events/
d
Dietary
Ingestion
Exposure
Ef
=

CfWf
Cf
Concentration
of
pesticide
in
the
food
item
(
s)
Measurement
in
individual
food
items
or
composite
duplicate
diet
samples
µ
g/
kg
Wf
Weight
of
food
item
consumed
Measured
in
duplicate
diet
sample
kg/
d
11
Indirect
Ingestion
Exposure
­
Microactivity
Approach
Eingmi
=
Cx
x
TEx
x
SAx
x
EF
Csurfx
Surface
loading
(
total
or
transferable)
on
object
x
Measure
by
a
wipe
or
press
method
µ
g/
cm2
TEx
Transfer
efficiencya
Empirically
determined
from
laboratory
experiments
unitless
SAx
Surface
area
contacted
Visual
observation
or
videotape
cm2/
event
EF
Frequency
of
mouthing
events
Visual
observation
or
videotape
events/
d
a
This
parameter
must
be
calculated
using
the
same
surface
loading
measurement
method
as
used
to
measure
Csurf
The
Panel
suggests
that
this
work
has
already
been
done
by
Clark
et
al.
(
2003a,
b).
The
work
by
Clark
et
al.
is
a
probabilistic
modeling
analysis;
it
was
not
a
measurement
study
for
predicting
exposures,
nor
were
measurements
conducted
to
verify
the
analysis.
Physical
and
behavioral
characteristics
of
receptors
in
various
age
groups
were
first
specified.
Phthalate
concentrations
were
obtained
from
various
databases
that
consisted
of
concentrations
of
six
phthalate
esters
measured
in
a
number
of
independent
research
and
testing
programs/
projects
in
Europe,
the
United
States,
and
Japan.
According
to
Clark
et
al.,
"
The
reported
concentrations
vary
widely
as
a
result
of
several
factors
including
analytical
error,
sample
contamination,
and
proximity
to
a
variety
of
past
and
present
sources."
The
model
was
used
to
assess
the
exposure
from
environmental
sources
and
to
apportion
the
pathways.
Exposures
from
consumer
products
were
not
included
in
the
estimates.
Although
food
was
the
major
route
of
exposure
for
all
six
phthalates,
indoor
air
and
dust
ingestion
were
also
important
routes
for
infants
and
toddlers.
In
their
discussion
of
the
exposure
assessment
results,
Clarke
et
al.
made
the
following
observations:
 
"
Estimates
of
phthalate
ester
exposure
would
be
improved
by
obtaining
recent
measurements
of
phthalate
esters
in
a
variety
of
foods."
 
"
Ingestion
of
dust
and
inhalation
of
indoor
air
represent
the
most
important
nonfood
sources
of
exposure
to
phthalate
esters.
These
estimates
rely
heavily
on
European
data,
due
to
a
lack
of
North
American
data."
 
"
A
link
between
human
exposure
and
multimedia
modeling
of
phthalate
esters
has
not
yet
been
established.
This
link
has
been
difficult
to
estimate,
for
several
reasons,
including:


Phthalate
ester
exposure
may
arise
from
industrial
or
consumer
use
in
addition
to
environmental
sources;


As
discussed
in
Chapter
5,
phthalate
esters
appear
to
be
metabolized
as
they
progress
up
the
food
chain
(
i.
e.,
they
biodilute),
making
it
difficult
to
estimate
concentrations
in
food
from
abiotic
media;


Some
foods
that
contribute
significantly
to
human
exposure
are
imported
from
outside
the
region
being
modeled.
As
a
result,
human
exposure
estimates
will
not
be
substantially
improved
by
measuring
phthalate
ester
concentrations
in
foods
from
different
regions.
Regional
differences
in
exposure
are
more
likely
to
be
due
to
dietary
differences
(
i.
e.,
differences
12
in
ingestion
rates
of
different
foods)
than
in
the
concentrations
of
the
phthalate
esters)."

The
proposed
study
will
collect
samples
that
address
some
of
the
issues
identified
by
Clark
et
al.
We
will
collect
duplicate
diet
samples
from
the
study
participants.
The
measurement
results
for
the
actual
foods
that
were
eaten
during
the
measurement
period
will
be
used
in
the
aggregate
exposure
estimates.
Indoor
air
samples
and
house
dust
samples
will
be
collected
during
the
study,
filling
a
critical
data
gap
identified
by
Clark
et
al.
Measurements
will
be
performed
in
residences
where
consumer
products
are
used,
addressing
the
potentially
largest
source
of
exposure,
which
the
Clark
et
al.
model
did
not
include.
Based
on
Clark
et
al.'
s
assessment,
we
believe
that
the
data
collected
by
EPA
in
this
study
will
provide
additional
significant
information
and
will
address
important
issues
raised
by
Clark
et
al.
The
measurement
data
collected
in
this
study
is
important
because
it
will
be
available
to
Clark
and
others
for
evaluating
the
probabilistic
models
used
to
estimate
exposure.
