TABLE
OF
CONTENTS
Page
No.

4.
SOIL
INGESTION
AND
PICA
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
4.1
BACKGROUND
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
4.2.
KEY
STUDIES
ON
SOIL
INTAKE
AMONG
CHILDREN
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
4.3.
RELEVANT
STUDIES
ON
SOIL
INTAKE
AMONG
CHILDREN
.
.
.
.
.
.
.
.
11
4.4.
SOIL
INTAKE
AMONG
ADULTS
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
16
4.5.
PREVALENCE
OF
PICA
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
17
4.6.
DELIBERATE
SOIL
INGESTION
AMONG
CHILDREN
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
18
4.7.
RECOMMENDATIONS
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
20
4.8.
REFERENCES
FOR
CHAPTER
4
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
25
T
i,
e
'
f
i,
e
x
F
i
S
i,
e
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Exposure
Factors
Handbook
Page
August
1997
4­
1
(
Eqn.
4­
1)

where:
T
=
estimated
soil
ingestion
for
child
i
based
on
element
e
i,
e
(
g/
day);

f
=
concentration
of
element
e
in
fecal
sample
of
child
i
i,
e
(
mg/
g);

F
=
fecal
dry
weight
(
g/
day);
and
i
S
=
concentration
of
element
e
in
child
i's
yard
soil
(
mg/
g).
i,
e
4.
SOIL
INGESTION
AND
PICA
4.1.
BACKGROUND
The
ingestion
of
soil
is
a
potential
source
of
human
summer
of
1984
as
part
of
a
larger
study
of
residents
living
exposure
to
toxicants.
The
potential
for
exposure
to
near
a
lead
smelter
in
East
Helena,
Montana.
Soiled
diapers
contaminants
via
this
source
is
greater
for
children
because
were
collected
over
a
3­
day
period
from
65
children
(
42
they
are
more
likely
to
ingest
more
soil
than
adults
as
a
males
and
23
females),
and
composited
samples
of
soil
result
of
behavioral
patterns
present
during
childhood.
were
obtained
from
the
children's
yards.
Both
excreta
and
Inadvertent
soil
ingestion
among
children
may
occur
soil
samples
were
analyzed
for
aluminum,
silicon,
and
through
the
mouthing
of
objects
or
hands.
Mouthing
titanium.
These
elements
were
found
in
soil,
but
were
behavior
is
considered
to
be
a
normal
phase
of
childhood
thought
to
be
poorly
absorbed
in
the
gut
and
to
have
been
development.
Adults
may
also
ingest
soil
or
dust
particles
present
in
the
diet
only
in
limited
quantities.
This
made
that
adhere
to
food,
cigarettes,
or
their
hands.
Deliberate
them
useful
tracers
for
estimating
soil
intake.
Excreta
soil
ingestion
is
defined
as
pica
and
is
considered
to
be
measurements
were
obtained
for
59
of
the
children.
Soil
relatively
uncommon.
Because
normal,
inadvertent
soil
ingestion
by
each
child
was
estimated
based
on
each
of
the
ingestion
is
more
prevalent
and
data
for
individuals
with
three
tracer
elements
using
a
standard
assumed
fecal
dry
pica
behavior
are
limited,
this
section
focuses
primarily
on
weight
of
15
g/
day,
and
the
following
equation:
normal
soil
ingestion
that
occurs
as
a
result
of
mouthing
or
unintentional
hand­
to­
mouth
activity.
Several
studies
have
been
conducted
to
estimate
the
amount
of
soil
ingested
by
children.
Most
of
the
early
studies
attempted
to
estimate
the
amount
of
soil
ingested
by
measuring
the
amount
of
dirt
present
on
children's
hands
and
making
generalizations
based
on
behavior.
More
recently,
soil
intake
studies
have
been
conducted
using
a
methodology
that
measures
trace
elements
in
feces
and
soil
that
are
believed
to
be
poorly
absorbed
in
the
gut.
These
measurements
are
used
to
estimate
the
amount
of
soil
ingested
over
a
specified
time
period.
The
available
studies
on
soil
intake
are
summarized
in
the
following
sections.
Studies
on
soil
intake
among
children
have
been
classified
as
either
key
studies
or
relevant
studies
based
on
their
applicability
to
exposure
assessment
needs.
Recommended
The
analysis
conducted
by
Binder
et
al.
(
1986)
intake
rates
are
based
on
the
results
of
key
studies,
but
assumed
that:
(
1)
the
tracer
elements
were
neither
lost
nor
relevant
studies
are
also
presented
to
provide
the
reader
introduced
during
sample
processing;
(
2)
the
soil
ingested
with
added
perspective
on
the
current
state­
of­
knowledge
by
children
originates
primarily
from
their
own
yards;
and
pertaining
to
soil
intake.
Information
on
soil
ingestion
(
3)
that
absorption
of
the
tracer
elements
by
children
among
adults
is
presented
based
on
available
data
from
a
occurred
in
only
small
amounts.
The
study
did
not
limited
number
of
studies.
This
is
an
area
where
more
data
distinguish
between
ingestion
of
soil
and
housedust
nor
did
and
more
research
are
needed.
Relevant
information
on
the
it
account
for
the
presence
of
the
tracer
elements
in
ingested
prevalence
of
pica
and
intake
among
individuals
exhibiting
foods
or
medicines.
pica
behavior
is
also
presented.
The
arithmetic
mean
quantity
of
soil
ingested
by
the
4.2.
KEY
STUDIES
ON
SOIL
INTAKE
AMONG
CHILDREN
Binder
et
al.
(
1986)
­
Estimating
Soil
Ingestion:
Use
of
Tracer
Elements
in
Estimating
the
Amount
of
Soil
Ingested
by
Young
Children
­
Binder
et
al.
(
1986)
studied
the
ingestion
of
soil
among
children
1
to
3
years
of
age
who
wore
diapers
using
a
tracer
technique
modified
from
a
method
previously
used
to
measure
soil
ingestion
among
grazing
animals.
The
children
were
studied
during
the
children
in
the
Binder
et
al.
(
1986)
study
was
estimated
to
be
181
mg/
day
(
range
25
to
1,324)
based
on
the
aluminum
tracer;
184
mg/
day
(
range
31
to
799)
based
on
the
silicon
tracer;
and
1,834
mg/
day
(
range
4
to
17,076)
based
on
the
titanium
tracer
(
Table
4­
1).
The
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Page
Exposure
Factors
Handbook
4­
2
August
1997
Table
4­
1.
Estimated
Daily
Soil
Ingestion
Based
on
Aluminum,
Silicon,
and
Titanium
Concentrations
Estimation
Mean
Median
Deviation
Range
95th
Percentile
Mean
Method
(
mg/
day)
(
mg/
day)
(
mg/
day)
(
mg/
day)
(
mg/
day)
(
mg/
day)
Standard
Geometric
Aluminum
181
121
203
25­
1,324
584
128
Silicon
184
136
175
31­
799
5,78
130
Titanium
1,834
618
3,091
4­
17,076
9,590
401
Minimum
108
88
121
4­
708
386
65
Source:
Binder
et
al.,
1986.

overall
mean
soil
ingestion
estimate
based
on
the
minimum
children,
aged
2
to
4
years,
attending
a
nursery
school,
and
of
the
three
individual
tracer
estimates
for
each
child
was
for
samples
of
playground
dirt
at
that
school.
Twenty­
seven
108
mg/
day
(
range
4
to
708).
The
95th
percentile
values
daily
fecal
samples
were
obtained
over
a
5­
day
period
for
for
aluminum,
silicon,
and
titanium
were
584
mg/
day,
578
the
18
children
examined.
Using
the
average
soil
mg/
day,
and
9,590
mg/
day,
respectively.
The
95th
concentrations
present
at
the
school,
and
assuming
a
percentile
value
based
on
the
minimum
of
the
three
standard
fecal
dry
weight
of
10
g/
day,
Clausing
et
al.
(
1987)
individual
tracer
estimates
for
each
child
was
386
mg/
day.
estimated
soil
ingestion
for
each
tracer.
Clausing
et
al.
The
authors
were
not
able
to
explain
the
difference
(
1987)
also
collected
eight
daily
fecal
samples
from
six
between
the
results
for
titanium
and
for
the
other
two
hospitalized,
bedridden
children.
These
children
served
as
elements,
but
speculated
that
unrecognized
sources
of
a
control
group,
representing
children
who
had
very
limited
titanium
in
the
diet
or
in
the
laboratory
processing
of
stool
access
to
soil.
samples
may
have
accounted
for
the
increased
levels.
The
The
average
quantity
of
soil
ingested
by
the
school
frequency
distribution
graph
of
soil
ingestion
estimates
children
in
this
study
was
as
follows:
230
mg/
day
(
range
23
based
on
titanium
shows
that
a
group
of
21
children
had
to
979
mg/
day)
for
aluminum;
129
mg/
day
(
range
48
to
362
particularly
high
titanium
values
(
i.
e.,
>
1,000
mg/
day).
The
mg/
day)
for
AIR;
and
1,430
mg/
day
(
range
64
to
11,620
remainder
of
the
children
showed
titanium
ingestion
mg/
day)
for
titanium
(
Table
4­
2).
As
in
the
Binder
et
al.
estimates
at
lower
levels,
with
a
distribution
more
(
1986)
study,
a
fraction
of
the
children
(
6/
19)
showed
comparable
to
that
of
the
other
elements.
titanium
values
well
above
1,000
mg/
day,
with
most
of
the
The
advantages
of
this
study
are
that
a
relatively
remaining
children
showing
substantially
lower
values.
large
number
of
children
were
studied
and
tracer
elements
Based
on
the
Limiting
Tracer
Method
(
LTM),
mean
soil
were
used
to
estimate
soil
ingestion.
However,
the
children
intake
was
estimated
to
be
105
mg/
day
with
a
population
studied
may
not
be
representative
of
the
U.
S.
population
standard
deviation
of
67
mg/
day
(
range
23
to
362
mg/
day).
and
the
study
did
not
account
for
tracers
ingested
via
foods
Use
of
the
LTM
assumed
that
"
the
maximum
amount
of
soil
or
medicines.
Also,
the
use
of
an
assumed
fecal
weight
ingested
corresponded
with
the
lowest
estimate
from
the
instead
of
actual
fecal
weights
may
have
biased
the
results
three
tracers"
(
Clausing
et
al.,
1987).
Geometric
mean
soil
of
this
study.
Finally,
because
of
the
short­
term
nature
of
intake
was
estimated
to
be
90
mg/
day.
This
assumes
that
the
survey,
soil
intake
estimates
may
not
be
entirely
the
maximum
amount
of
soil
ingested
cannot
be
higher
than
representative
of
long­
term
behavior,
especially
at
the
the
lowest
estimate
for
the
individual
tracers.
upper­
end
of
the
distribution
of
intake.
Mean
soil
intake
for
the
hospitalized
children
was
Clausing
et
al.
(
1987)
­
A
Method
for
Estimating
Soil
Ingestion
by
Children
­
Clausing
et
al.
(
1987)
conducted
a
soil
ingestion
study
with
Dutch
children
using
a
tracer
element
methodology
similar
to
that
of
Binder
et
al.
(
1986).
Aluminum,
titanium,
and
acid­
insoluble
residue
(
AIR)
contents
were
determined
for
fecal
samples
from
estimated
to
be
56
mg/
day
based
on
aluminum
(
Table
4­
3).
For
titanium,
three
of
the
children
had
estimates
well
in
excess
of
1,000
mg/
day,
with
the
remaining
three
children
in
the
range
of
28
to
58
mg/
day.
Using
the
LTM
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Exposure
Factors
Handbook
Page
August
1997
4­
3
Table
4­
2.
Calculated
Soil
Ingestion
by
Nursery
School
Children
Child
Number
(
mg/
day)
(
mg/
day)
(
mg/
day)
(
mg/
day)
Sample
Calculated
from
Ti
Calculated
from
Al
Calculated
from
AIR
Limiting
Tracer
Soil
Ingestion
as
Soil
Ingestion
as
Soil
Ingestion
as
1
L3
103
300
107
103
L14
154
211
172
154
L25
130
23
­
23
2
L5
131
­
71
71
L13
184
103
82
82
L27
142
81
84
81
3
L2
124
42
84
42
L17
670
566
174
174
4
L4
246
62
145
62
L11
2,990
65
139
65
5
L8
293
­
108
108
L21
313
­
152
152
6
L12
1,110
693
362
362
L16
176
­
145
145
7
L18
11,620
­
120
120
L22
11,320
77
­
77
8
L1
3,060
82
96
82
9
L6
624
979
111
111
10
L7
600
200
124
124
11
L9
133
­
95
95
12
L10
354
195
106
106
13
L15
2,400
­
48
48
14
L19
124
71
93
71
15
L20
269
212
274
212
16
L23
1,130
51
84
51
17
L24
64
566
­
64
18
L26
184
56
­
56
Arithmetic
Mean
1,431
232
129
105
Source:
Adapted
from
Clausing
et
al.
1987.

Table
4­
3.
Calculated
Soil
Ingestion
by
Hospitalized,
Bedridden
Children
Child
Sample
(
mg/
day)
(
mg/
day)
(
mg/
day)
Soil
Ingestion
as
Calculated
Soil
Ingestion
as
Calculated
from
Ti
from
Al
Limiting
Tracer
1
G5
3,290
57
57
G6
4,790
71
71
2
G1
28
26
26
3
G2
6,570
94
84
G8
2,480
57
57
4
G3
28
77
28
5
G4
1,100
30
30
6
G7
58
38
38
Arithmetic
Mean
2,293
56
49
Source:
Adapted
from
Clausing
et
al.
1987.

method,
the
mean
soil
ingestion
rate
was
estimated
to
be
49
suggest
a
major
nonsoil
source
of
titanium
for
some
mg/
day
with
a
population
standard
deviation
of
22
mg/
day
children,
and
may
suggest
a
background
nonsoil
source
of
(
range
26
to
84
mg/
day).
The
geometric
mean
soil
intake
aluminum.
However,
conditions
specific
to
hospitalization
rate
was
45
mg/
day.
The
data
on
hospitalized
children
(
e.
g.,
medications)
were
not
considered.
AIR
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Page
Exposure
Factors
Handbook
4­
4
August
1997
measurements
were
not
reported
for
the
hospitalized
determine
which
tracer
elements
provided
the
most
reliable
children.
Assuming
that
the
tracer­
based
soil
ingestion
data
on
soil
ingestion,
known
amounts
of
soil
(
i.
e.,
300
mg
rates
observed
in
hospitalized
children
actually
represent
over
three
days
and
1,500
mg
over
three
days)
containing
background
tracer
intake
from
dietary
and
other
nonsoil
eight
tracers
were
administered
to
six
adult
volunteers
(
i.
e.,
sources,
mean
soil
ingestion
by
nursery
school
children
was
three
males
and
three
females).
Soil
samples
and
feces
estimated
to
be
56
mg/
day,
based
on
the
LTM
(
i.
e.,
105
samples
from
these
adults
and
duplicate
food
samples
were
mg/
day
for
nursery
school
children
minus
49
mg/
day
for
analyzed
for
tracer
elements
to
calculate
recovery
rates
of
hospitalized
children)
(
Clausing
et
al.
1987).
tracer
elements
in
soil.
Based
on
the
adult
validation
study,
The
advantages
of
this
study
are
that
Clausing
et
al.
Calabrese
et
al.
(
1989)
confirmed
that
the
tracer
(
1987)
evaluated
soil
ingestion
among
two
populations
of
methodology
could
adequately
detect
tracer
elements
in
children
that
had
differences
in
access
to
soil,
and
corrected
feces
at
levels
expected
to
correspond
with
soil
intake
rates
soil
intake
rates
based
on
background
estimates
derived
in
children.
Calabrese
et
al.
(
1989)
also
found
that
from
the
hospitalized
group.
However,
a
smaller
number
of
aluminum,
silicon,
and
yttrium
were
the
most
reliable
of
the
children
were
used
in
this
study
than
in
the
Binder
et
al.
eight
tracer
elements
analyzed.
The
standard
deviation
of
(
1986)
study
and
these
children
may
not
be
representative
recovery
of
these
three
tracers
was
the
lowest
and
the
of
the
U.
S.
population.
Tracer
elements
in
foods
or
percentage
of
recovery
was
closest
to
100
percent
medicines
were
not
evaluated.
Also,
intake
rates
derived
(
Calabrese,
et
al.,
1989).
The
recovery
of
these
three
from
this
study
may
not
be
representative
of
soil
intake
over
tracers
ranged
from
120
to
153
percent
when
300
mg
of
soil
the
long­
term
because
of
the
short­
term
nature
of
the
study.
had
been
ingested
over
a
three­
day
period
and
from
88
to
94
In
addition,
one
of
the
factors
that
could
affect
soil
intake
percent
when
1,500
mg
soil
had
been
ingested
over
a
threerates
is
hygiene
(
e.
g.,
hand
washing
frequency).
Hygienic
day
period
(
Table
4­
4).
practices
can
vary
across
countries
and
cultures
and
may
be
Using
the
three
most
reliable
tracer
elements,
the
more
stringently
emphasized
in
a
more
structured
mean
soil
intake
rate
for
children,
adjusted
to
account
for
environment
such
as
child
care
centers
in
The
Netherlands
the
amount
of
tracer
found
in
food
and
medicines,
was
and
other
European
countries
than
in
child
care
centers
in
estimated
to
be
153
mg/
day
based
on
aluminum,
154
the
United
States.
mg/
day
based
on
silicon,
and
85
mg/
day
based
on
yttrium
Calabrese
et
al.
(
1989)
­
How
Much
Soil
do
Young
Children
Ingest:
An
Epidemiologic
Study
­
Calabrese
et
al.
(
1989)
studied
soil
ingestion
among
children
using
the
basic
tracer
design
developed
by
Binder
et
al.
(
1986).
However,
in
contrast
to
the
Binder
et
al.
(
1986)
study,
eight
tracer
elements
(
i.
e.,
aluminum,
barium,
manganese,
silicon,
titanium,
vanadium,
yttrium,
and
zirconium)
were
analyzed
instead
of
only
three
(
i.
e.,
aluminum,
silicon,
and
titanium).
A
total
of
64
children
between
the
ages
of
1
and
4
years
old
were
included
in
the
study.
These
children
were
all
selected
from
the
greater
Amherst,
Massachusetts
area
and
were
predominantly
from
two­
parent
households
where
the
parents
were
highly
educated.
The
Calabrese
et
al.
(
1989)
study
was
conducted
over
eight
days
during
a
two
week
period
and
included
the
use
of
a
mass­
balance
methodology
in
which
duplicate
samples
of
food,
medicines,
vitamins,
and
others
were
collected
and
analyzed
on
a
daily
basis,
in
addition
to
soil
and
dust
samples
collected
from
the
child's
home
and
play
area.
Fecal
and
urine
samples
were
also
collected
and
analyzed
for
tracer
elements.
Toothpaste,
low
in
tracer
content,
was
provided
to
all
participants.
In
order
to
validate
the
mass­
balance
methodology
used
to
estimate
soil
ingestion
rates
among
children
and
to
(
Table
4­
5).
Median
intake
rates
were
somewhat
lower
(
29
mg/
day
for
aluminum,
40
mg/
day
for
silicon,
and
9
mg/
day
for
yttrium).
Upper­
percentile
(
i.
e.,
95th)
values
were
223
mg/
day
for
aluminum,
276
mg/
day
for
silicon,
and
106
mg/
day
for
yttrium.
Similar
results
were
observed
when
soil
and
dust
ingestion
was
combined
(
Table
4­
5).
Intake
of
soil
and
dust
was
estimated
using
a
weighted
average
of
tracer
concentration
in
dust
composite
samples
and
in
soil
composite
samples
based
on
the
timechildren
spent
at
home
and
away
from
home,
and
indoors
and
outdoors.
Calabrese
et
al.
(
1989)
suggested
that
the
use
of
titanium
as
a
tracer
in
earlier
studies
that
lacked
food
ingestion
data
may
have
significantly
overestimated
soil
intake
because
of
the
high
levels
of
titanium
in
food.
Using
the
median
values
of
aluminum
and
silicon,
Calabrese
et
al.
(
1989)
estimated
the
quantity
of
soil
ingested
daily
to
be
29
mg/
day
and
40
mg/
day,
respectively.
It
should
be
noted
that
soil
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Exposure
Factors
Handbook
Page
August
1997
4­
5
Table
4­
4.
Mean
and
Standard
Deviation
Percentage
Recovery
of
Eight
Tracer
Elements
Tracer
Element
300
mg
Soil
Ingested
1500
mg
Soil
Ingested
Mean
SD
Mean
SD
Al
152.8
107.5
93.5
15.5
Ba
2304.3
4533.0
149.8
69.5
Mn
1177.2
1341.0
248.3
183.6
Si
139.3
149.6
91.8
16.6
Ti
251.5
316.0
286.3
380.0
V
345.0
247.0
147.6
66.8
Y
120.5
42.4
87.5
12.6
Zr
80.6
43.7
54.6
33.4
Source:
Adapted
from
Calabrese
et
al.,
1989.

Table
4­
5.
Soil
and
Dust
Ingestion
Estimates
for
Children
Aged
1­
4
Years
Tracer
Element
N
Intake
(
mg/
day)
a
Mean
Median
SD
95th
Percentile
Maximum
Aluminum
soil
64
153
29
852
223
6,837
dust
64
317
31
1,272
506
8,462
soil/
dust
combined
64
154
30
629
478
4,929
Silicon
soil
64
154
40
693
276
5,549
dust
64
964
49
6,848
692
54,870
soil/
dust
combined
64
483
49
3,105
653
24,900
Yttrium
soil
62
85
9
890
106
6,736
dust
64
62
15
687
169
5,096
soil/
dust
combined
62
65
11
717
159
5,269
Titanium
soil
64
218
55
1,150
1,432
6,707
dust
64
163
28
659
1,266
3,354
soil/
dust
combined
64
170
30
691
1,059
3,597
Corrected
for
Tracer
Concentrations
in
Foods
a
Source:
Adapted
from
Calabrese
et
al.,
1989.

ingestion
for
one
child
in
the
study
ranged
from
used
was
larger
than
for
other
studies.
A
relatively
large
approximately
10
to
14
grams/
day
during
the
second
week
population
was
studied,
but
they
may
not
be
entirely
of
observation.
Average
soil
ingestion
for
this
child
was
5
representative
of
the
U.
S.
population
because
they
were
to
7
mg/
day,
based
on
the
entire
study
period.
selected
from
a
single
location.
The
advantages
of
this
study
are
that
intake
rates
were
corrected
for
tracer
concentrations
in
foods
and
medicines
and
that
the
methodology
was
validated
using
adults.
Also,
intake
was
observed
over
a
longer
time
period
in
this
study
than
in
earlier
studies
and
the
number
of
tracers
Davis
et
al.
(
1990)
­
Quantitative
Estimates
of
Soil
Ingestion
in
Normal
Children
Between
the
ages
of
2
and
7
years:
Population­
Based
Estimates
Using
Aluminum,
Silicon,
and
Titanium
as
Soil
Tracer
Elements
­
Davis
et
S
i,
e
'
(
DW
f
%
DW
p
)
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Page
Exposure
Factors
Handbook
4­
6
August
1997
(
Eqn.
4­
2)

where:
S
=
soil
ingested
for
child
i
based
on
tracer
e
(
g);
i,
e
DW
=
feces
dry
weight
(
g);
f
DW
=
feces
dry
weight
on
toilet
paper
(
g);
p
E
=
tracer
amount
in
feces
(
F
g/
g);
f
E
=
tracer
amount
in
urine
(
F
g/
g);
u
DW
=
food
dry
weight
(
g);
fd
E
=
tracer
amount
in
food
(
F
g/
g);
and
fd
E
=
tracer
concentration
in
soil
(
F
g/
g).
soil
al.
(
1990)
also
used
a
mass­
balance/
tracer
technique
to
The
soil
intake
rates
were
corrected
by
adding
the
amount
estimate
soil
ingestion
among
children.
In
this
study,
104
of
tracer
in
vitamins
and
medications
to
the
amount
of
tracer
children
between
the
ages
of
2
and
7
years
were
randomly
in
food,
and
adjusting
the
food
quantities,
feces
dry
weights,
selected
from
a
three­
city
area
in
southeastern
Washington
and
tracer
concentrations
in
urine
to
account
for
missing
State.
The
study
was
conducted
over
a
seven
day
period,
samples.
primarily
during
the
summer.
Daily
soil
ingestion
was
Soil
ingestion
rates
were
highly
variable,
especially
evaluated
by
collecting
and
analyzing
soil
and
house
dust
those
based
on
titanium.
Mean
daily
soil
ingestion
samples,
feces,
urine,
and
duplicate
food
samples
for
estimates
were
38.9
mg/
day
for
aluminum,
82.4
mg/
day
for
aluminum,
silicon,
and
titanium.
In
addition,
information
on
silicon
and
245.5
mg/
day
for
titanium
(
Table
4­
6).
Median
dietary
habits
and
demographics
was
collected
in
an
attempt
values
were
25
mg/
day
for
aluminum,
59
mg/
day
for
silicon,
to
identify
behavioral
and
demographic
characteristics
that
and
81
mg/
day
for
titanium.
Davis
et
al.
(
1990)
also
influence
soil
intake
rates
among
children.
The
amount
of
evaluated
the
extent
to
which
differences
in
tracer
soil
ingested
on
a
daily
basis
was
estimated
using
the
concentrations
in
house
dust
and
yard
soil
impacted
following
equation:
estimated
soil
ingestion
rates.
The
value
used
in
the
denominator
of
the
mass
balance
equation
was
recalculated
to
represent
a
weighted
average
of
the
tracer
concentration
in
yard
soil
and
house
dust
based
on
the
proportion
of
time
the
child
spent
indoors
and
outdoors.
The
adjusted
mean
soil/
dust
intake
rates
were
64.5
mg/
day
for
aluminum,
160.0
mg/
day
for
silicon,
and
268.4
mg/
day
for
titanium.
Adjusted
median
soil/
dust
intake
rates
were:
51.8
mg/
day
for
aluminum,
112.4
mg/
day
for
silicon,
and
116.6
mg/
day
for
titanium.
Davis
et
al.
(
1990)
also
observed
that
the
following
demographic
characteristics
were
associated
with
high
soil
intake
rates:
male
sex,
non­
white
racial
group,
low
income,
operator/
laborer
as
the
principal
Table
4­
6.
Average
Daily
Soil
Ingestion
Values
Based
on
Aluminum,
Silicon,
and
Titanium
as
Tracer
Elementsa
Element
Mean
Median
Mean
Range
(
mg/
d)
(
mg/
d)
(
mg/
d)
(
mg/
d)
Standard
Error
of
the
b
Aluminum
38.9
25.3
14.4
279.0
to
904.5
Silicon
82.4
59.4
12.2
­
404.0
to
534.6
Titanium
245.5
81.3
119.7
­
5,820.8
to
6,182.2
Minimum
38.9
25.3
12.2
­
5,820.8
Maximum
245.5
81.3
119.7
6,182.2
Excludes
three
children
who
did
not
provide
any
samples
(
N=
101).
a
Negative
values
occurred
as
a
result
of
correction
for
nonsoil
sources
of
the
tracer
elements.
b
Source:
Adapted
from
Davis
et
al.,
1990.

occupation
of
the
parent,
and
city
of
residence.
However,
The
advantages
of
the
Davis
et
al.
(
1990)
study
are
none
of
these
factors
were
predictive
of
soil
intake
rates
that
soil
intake
rates
were
corrected
based
on
the
tracer
when
tested
using
multiple
linear
regression.
content
of
foods
and
medicines
and
that
a
relatively
large
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Exposure
Factors
Handbook
Page
August
1997
4­
7
number
of
children
were
sampled.
Also,
demographic
and
of
these
children
were
sampled
during
both
periods
(
i.
e.,
at
behavioral
information
was
collected
for
the
survey
group.
the
beginning
and
near
the
end
of
the
summer
of
1986).
A
However,
although
a
relatively
large
sample
population
was
total
of
78
children
were
sampled
at
campgrounds,
and
15
surveyed,
these
children
were
all
from
a
single
area
of
the
hospitalized
children
were
sampled.
The
mean
values
for
U.
S.
and
may
not
be
representative
of
the
U.
S.
population
these
groups
were:
162
mg/
day
for
children
in
daycare
as
a
whole.
The
study
was
conducted
over
a
one­
week
centers,
213
mg/
day
for
campers
and
93
mg/
day
for
period
during
the
summer
and
may
not
be
representative
of
hospitalized
children.
Van
Wïjnen
et
al.
(
1990)
also
long­
term
(
i.
e.,
annual)
patterns
of
intake.
reported
geometric
mean
LTM
values
because
soil
intake
Van
Wïjnen
et
al.
(
1990)
­
Estimated
Soil
Ingestion
by
Children
­
In
a
study
by
Van
Wïjnen
et
al.
(
1990),
soil
ingestion
among
Dutch
children
ranging
in
age
from
1
to
5
years
was
evaluated
using
a
tracer
element
methodology
similar
to
that
used
by
Clausing
et
al.
(
1987).
Van
Wïjnen
et
al.
(
1990)
measured
three
tracers
(
i.
e.,
titanium,
aluminum,
and
AIR)
in
soil
and
feces
and
estimated
soil
ingestion
based
on
the
LTM.
An
average
daily
feces
weight
of
15
g
dry
weight
was
assumed.
A
total
of
292
children
attending
daycare
centers
were
sampled
during
the
first
of
two
sampling
periods
and
187
children
were
sampled
in
the
second
sampling
period;
162
rates
were
found
to
be
skewed
and
the
log
transformed
data
were
approximately
normally
distributed.
Geometric
mean
LTM
values
were
estimated
to
be
111
mg/
day
for
children
in
daycare
centers,
174
mg/
day
for
children
vacationing
at
campgrounds
(
Table
4­
7)
and
74
mg/
day
for
hospitalized
children
(
70­
120
mg/
day
based
on
the
95
percent
confidence
limits
of
the
mean).
AIR
was
the
limiting
tracer
in
about
80
percent
of
the
samples.
Among
children
attending
daycare
centers,
soil
intake
was
also
found
to
be
higher
when
the
weather
was
good
(
i.
e.,
<
2
days/
week
precipitation)
than
when
the
weather
was
bad
(
i.
e.,
>
4
days/
week
precipitation
(
Table
4­
8).
Van
Wïjnen
et
al.
(
1990)
suggest
that
the
mean
LTM
value
for
hospitalized
infants
represents
background
intake
of
tracers
and
should
be
used
to
correct
the
soil
intake
rates
based
on
LTM
values
for
other
sampling
groups.
Using
mean
values,
corrected
soil
intake
rates
were
69
mg/
day
(
162
mg/
day
minus
93
mg/
day)
for
daycare
children
and
120
mg/
day
(
213
mg/
day
minus
93
mg/
day)
for
campers.

Table
4­
7.
Geometric
Mean
(
GM)
and
Standard
Deviation
(
GSD)
LTM
Values
for
Children
at
Daycare
Centers
and
Campgrounds
Age
(
yrs)
Sex
Daycare
Centers
Campgrounds
n
GM
LTM
GSD
LTM
n
GM
LTM
GSD
LTM
(
mg/
day)
(
mg/
day)
(
mg/
day)
(
mg/
day)

<
1
Girls
3
81
1.09
­
­
­
Boys
1
75
­
­
­
­

1­<
2
Girls
20
124
1.87
3
207
1.99
Boys
17
114
1.47
5
312
2.58
2­<
3
Girls
34
118
1.74
4
367
2.44
Boys
17
96
1.53
8
232
2.15
3­
4
Girls
26
111
1.57
6
164
1.27
Boys
29
110
1.32
8
148
1.42
4­<
5
Girls
1
180
­
19
164
1.48
Boys
4
99
1.62
18
136
1.30
All
girls
86
117
1.70
36
179
1.67
All
boys
72
104
1.46
42
169
1.79
Total
162
111
1.60
78
174
1.73
a
b
Age
and/
or
sex
not
registered
for
eight
children.
a
Age
not
registered
for
seven
children.
b
Source:
Adapted
from
Van
Wijnen
et
al.,
1990.
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Page
Exposure
Factors
Handbook
4­
8
August
1997
Table
4­
8.
Estimated
Geometric
Mean
LTM
Values
of
Children
Attending
Daycare
Centers
According
to
Age,
Weather
Category,
and
Sampling
Period
Weather
Category
Age
(
years)
First
Sampling
Period
Second
Sampling
Period
n
(
mg/
day)
n
(
mg/
day)
Estimated
Geometric
Mean
Estimated
Geometric
Mean
LTM
Value
LTM
Value
Bad
<
1
3
94
3
67
(>
4
days/
week
precipitation)
1­<
2
18
103
33
80
2­<
3
33
109
48
91
4­<
5
5
124
6
109
Reasonable
<
1
1
61
(
2­
3
days/
week
precipitation)
1­<
2
10
96
2­<
3
13
99
3­<
4
19
94
4­<
5
1
61
Good
<
1
4
102
(<
2
days/
week
precipitation)
1­<
2
42
229
2­<
3
65
166
3­<
4
67
138
4­<
5
10
132
Source:
Van
Wijnen
et
al.,
1990.

Corrected
geometric
mean
soil
intake
was
estimated
to
structured
environment
such
as
child
care
centers
in
The
range
from
0
to
90
mg/
day
with
a
90th
percentile
value
of
Netherlands
and
other
European
countries
than
in
child
care
190
mg/
day
for
the
various
age
categories
within
the
centers
in
the
United
States.
daycare
group
and
30
to
200
mg/
day
with
a
90th
percentile
value
of
300
mg/
day
for
the
various
age
categories
within
the
camping
group.
The
advantage
of
this
study
is
that
soil
intake
was
estimated
for
three
different
populations
of
children;
one
expected
to
have
high
intake,
one
expected
to
have
"
typical"
intake,
and
one
expected
to
have
low
or
background­
level
intake.
Van
Wïjnen
et
al.
(
1990)
used
the
background
tracer
measurements
to
correct
soil
intake
rates
for
the
other
two
populations.
Tracer
concentrations
in
food
and
medicine
were
not
evaluated.
Also,
the
population
of
children
studied
was
relatively
large,
but
may
not
be
representative
of
the
U.
S.
population.
This
study
was
conducted
over
a
relatively
short
time
period.
Thus,
estimated
intake
rates
may
not
reflect
long­
term
patterns,
especially
at
the
high­
end
of
the
distribution.
Another
limitation
of
this
study
is
that
values
were
not
reported
element­
by­
element
which
would
be
the
preferred
way
of
reporting.
In
addition,
one
of
the
factors
that
could
affect
soil
intake
rates
is
hygiene
(
e.
g.,
hand
washing
frequency).
Hygienic
practices
can
vary
across
countries
and
cultures
and
may
be
more
stringently
emphasized
in
a
more
Stanek
and
Calabrese
(
1995a)
­
Daily
Estimates
of
Soil
Ingestion
in
Children
­
Stanek
and
Calabrese
(
1995a)
presented
a
methodology
which
links
the
physical
passage
of
food
and
fecal
samples
to
construct
daily
soil
ingestion
estimates
from
daily
food
and
fecal
trace­
element
concentrations.
Soil
ingestion
data
for
children
obtained
from
the
Amherst
study
(
Calabrese
et
al.,
1989)
were
reanalyzed
by
Stanek
and
Calabrese
(
1995a).
In
the
Amherst
study,
soil
ingestion
measurements
were
made
over
a
period
of
2
weeks
for
a
non­
random
sample
of
sixtyfour
children
(
ages
of
1­
4
years
old)
living
adjacent
to
an
academic
area
in
western
Massachusetts.
During
each
week,
duplicate
food
samples
were
collected
for
3
consecutive
days
and
fecal
samples
were
collected
for
4
consecutive
days
for
each
subject.
The
total
amount
of
each
of
eight
trace
elements
present
in
the
food
and
fecal
samples
were
measured.
The
eight
trace
elements
are
aluminum,
barium,
manganese,
silicon,
titanium,
vanadium,
yttrium,
and
zirconium.
The
authors
expressed
the
amount
of
trace
element
in
food
input
or
fecal
output
as
a
"
soil
equivalent,"
which
was
defined
as
the
amount
of
the
element
in
average
daily
food
intake
(
or
average
daily
fecal
output)
divided
by
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Exposure
Factors
Handbook
Page
August
1997
4­
9
the
concentration
of
the
element
in
soil.
A
lag
period
of
28
Specifically,
among
elements
that
may
be
more
useful
for
hours
between
food
intake
and
fecal
output
was
assumed
for
estimation
of
ingestion,
the
mean
estimates
decreased
for
Al
all
respondents.
Day
1
for
the
food
sample
corresponded
(
153
mg/
d
to
122
mg/
d)
and
Si
(
154
mg/
d
to
139
mg/
d),
to
the
24
hour
period
from
midnight
on
Sunday
to
midnight
but
increased
for
Ti
(
218
mg/
d
to
271
mg/
d)
and
Y
(
85
on
Monday
of
a
study
week;
day
1
of
the
fecal
sample
mg/
d
to
165
mg/
d).
The
"
overall"
mean
estimate
from
this
corresponded
to
the
24
hour
period
from
noon
on
Monday
reanalysis
was
179
mg/
d.
Table
4­
9
presents
the
empirical
to
noon
on
Tuesday
(
Stanek
and
Calabrese,
1995a).
Based
distribution
of
the
the
"
overall"
mean
daily
soil
ingestion
on
these
definitions,
the
food
soil
equivalent
was
subtracted
estimates
for
the
8­
day
study
period
(
not
based
on
from
the
fecal
soil
equivalent
to
obtain
an
estimate
of
soil
lognormal
modeling).
The
estimated
intake
based
on
the
ingestion
for
a
trace
element.
A
daily
"
overall"
ingestion
"
overall"
estimates
is
45
mg/
day
or
less
for
50
percent
of
estimate
was
constructed
for
each
child
as
the
median
of
the
children
and
208
mg/
day
or
less
for
95
percent
of
the
trace
element
values
remaining
after
tracers
falling
outside
children.
The
upper
percentile
values
for
most
of
the
of
a
defined
range
around
the
overall
median
were
excluded.
individual
trace
elements
are
somewhat
higher.
Next,
Additionally,
estimates
of
the
distribution
of
soil
ingestion
estimates
of
the
respondents
soil
intake
averaged
over
a
projected
over
a
period
of
365
days
were
derived
by
fitting
period
of
365
days
were
presented
based
upon
the
log­
normal
distributions
to
the
"
overall"
daily
soil
ingestion
lognormal
models
fit
to
the
daily
ingestion
estimates
(
Table
estimates.
4­
10).
The
estimated
median
value
of
the
64
respondents'
Table
4­
9
presents
the
estimates
of
mean
daily
soil
daily
soil
ingestion
averaged
over
a
year
is
75
mg/
day,
while
ingestion
intake
per
child
(
mg/
day)
for
the
64
study
the
95th
percentile
is
1,751
mg/
day.
participants.
(
The
authors
also
presented
estimates
of
the
A
strength
of
this
study
is
that
it
attempts
to
make
median
values
of
daily
intake
for
each
child.
For
most
risk
full
use
of
the
collected
data
through
estimation
of
daily
assessment
purposes
the
child
mean
values,
which
are
ingestion
rates
for
children.
The
data
are
then
screened
to
proportional
to
the
cumulative
soil
intake
by
the
child,
are
remove
less
consistent
tracer
estimates
and
the
remaining
needed
instead
of
the
median
values.)
The
approach
values
are
aggregated.
Individual
daily
estimates
of
adopted
in
this
paper
led
to
changes
in
ingestion
estimates
ingestion
will
be
subject
to
larger
errors
than
are
weekly
from
those
presented
in
Calabrese
et
al.
(
1989).
average
values,
particularly
since
the
assumption
of
a
constant
lag
time
between
food
intake
and
fecal
output
may
be
not
be
correct
for
many
subject
days.
The
aggregation
approach
used
to
arrive
at
the
"
overall"
ingestion
estimates
rests
on
the
assumption
that
the
mean
Table
4­
9.
Distribution
of
Average
(
Mean)
Daily
Soil
Ingestion
Estimates
Per
Child
for
64
Children
(
mg/
day)
a
Type
of
Estimate
Overall
A1
Ba
Mn
Si
Ti
V
Y
Zr
Number
of
Samples
(
64)
(
64)
(
33)
(
19)
(
63)
(
56)
(
52)
(
61)
(
62)

Mean
179
122
655
1,053
139
271
112
165
23
25th
Percentile
10
10
28
35
5
8
8
0
0
50th
Percentile
45
19
65
121
32
31
47
15
15
75th
Percentile
88
73
260
319
94
93
177
47
41
90th
Percentile
186
131
470
478
206
154
340
105
87
95th
Percentile
208
254
518
17,374
224
279
398
144
117
Maximum
7,703
4,692
17,991
17,374
4,975
12,055
845
8,976
208
For
each
child,
estimates
of
soil
ingestion
were
formed
on
days
4­
8
and
the
mean
of
these
estimates
was
then
evaluated
for
each
child.
The
values
a
in
the
column
"
overall"
correspond
to
percentiles
of
the
distribution
of
these
means
over
the
64
children.
When
specific
trace
elements
were
not
excluded
via
the
relative
standard
deviation
criteria,
estimates
of
soil
ingestion
based
on
the
specific
trace
element
were
formed
for
108
days
for
each
subject.
The
mean
soil
ingestion
estimate
was
again
evaluated.
The
distribution
of
these
means
for
specific
trace
elements
is
shown.
Source:
Stanek
and
Calabrese,
1995a.
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Page
Exposure
Factors
Handbook
4­
10
August
1997
ingestion
estimates
across
acceptable
tracers
provides
the
due
to
fluctuation
in
gastrointestinal
transit
time)
in
the
soil
most
reliable
ingestion
estimates.
The
validity
of
this
ingestion
calculation.
Because
the
recoverability
of
tracers
assumption
depends
on
the
particular
set
of
tracers
used
in
can
vary
within
any
group
of
individuals,
the
BTM
uses
a
the
study,
and
is
not
fully
assessed.
ranking
scheme
of
F/
S
ratios
to
determine
the
best
tracers
Table
4­
10.
Estimated
Distribution
of
Individual
Mean
Daily
Soil
Ingestion
Based
on
Data
for
64
Subjects
Projected
Over
365
Daysa
Range
1
­
2,268
mg/
d
50th
Percentile
(
median)
75
mg/
d
90th
Percentile
1,190
mg/
d
95th
Percentile
1,751
mg/
d
b
Based
on
fitting
a
log­
normal
distribution
to
model
daily
soil
a
ingestion
values.
Subject
with
pica
excluded.
b
Source:
Stanek
and
Calabrese,
1995a.

In
developing
the
365
day
soil
ingestion
estimates,
data
that
were
obtained
over
a
short
period
of
time
(
as
is
the
case
with
all
available
soil
ingestion
studies)
were
extrapolated
over
a
year.
The
2­
week
study
period
may
not
reflect
variability
in
tracer
element
ingestion
over
a
year.
While
Stanek
and
Calabrese
(
1995a)
attempt
to
address
this
through
lognormal
modeling
of
the
long
term
intake,
new
uncertainties
are
introduced
through
the
parametric
modeling
of
the
limited
subject
day
data.
Also,
the
sample
population
size
of
the
original
study
was
small
and
site
limited,
and,
therefore,
is
not
representative
of
the
U.
S.
population.
Study
mean
estimates
of
soil
ingestion,
such
as
the
study
mean
estimates
presented
in
Table
4­
9,
are
substantially
more
reliable
than
any
available
distributional
estimates.
Stanek
and
Calabrese
(
1995b)
­
Soil
Ingestion
Estimates
for
Use
in
Site
Evaluations
Based
on
the
Best
Tracer
Method
­
Stanek
and
Calabrese
(
1995b)
recalculated
ingestion
rates
that
were
estimated
in
three
previous
mass­
balance
studies
(
Calabrese
et
al.,
1989
and
Davis
et
al.,
1990
for
children's
soil
ingestion,
and
Calabrese
et
al.,
1990
for
adult
soil
ingestion)
using
the
Best
Tracer
Method
(
BTM).
This
method
allows
for
the
selection
of
the
most
recoverable
tracer
for
a
particular
subject
or
group
of
subjects.
The
selection
process
involves
ordering
trace
elements
for
each
subject
based
on
food/
soil
(
F/
S)
ratios.
These
ratios
are
estimated
by
dividing
the
total
amount
of
the
tracer
in
food
by
the
tracer
concentration
in
soil.
The
F/
S
ratio
is
small
when
the
tracer
concentration
in
food
is
almost
zero
when
compared
to
the
tracer
concentration
in
soil.
A
small
F/
S
ratio
is
desirable
because
it
lessens
the
impact
of
transit
time
error
(
the
error
that
occurs
when
fecal
output
does
not
reflect
food
ingestion,
for
use
in
the
ingestion
rate
calculation.
To
reduce
biases
that
may
occur
as
a
result
of
sources
of
fecal
tracers
other
than
food
or
soil,
the
median
of
soil
ingestion
estimates
based
on
the
four
lowest
F/
S
ratios
was
used
to
represent
soil
ingestion
among
individuals.
For
adults,
Stanek
and
Calabrese
(
1995b)
used
data
for
8
tracers
from
the
Calabrese
et
al.
(
1990)
study
to
estimate
soil
ingestion
by
the
BTM.
The
lowest
F/
S
ratios
were
Zr
and
Al
and
the
element
with
the
highest
F/
S
ratio
was
Mn.
For
soil
ingestion
estimates
based
on
the
median
of
the
lowest
four
F/
S
ratios,
the
tracers
contributing
most
often
to
the
soil
ingestion
estimates
were
Al,
Si,
Ti,
Y,
V,
and
Zr.
Using
the
median
of
the
soil
ingestion
rates
based
on
the
best
four
tracer
elements,
the
average
adult
soil
ingestion
rate
was
estimated
to
be
64
mg/
day
with
a
median
of
87
mg/
day.
The
90th
percentile
soil
ingestion
estimate
was
142
mg/
day.
These
estimates
are
based
on
18
subject
weeks
for
the
six
adult
volunteers
described
in
Calabrese
et
al.
(
1990).
For
children,
Stanek
and
Calabrese
(
1995b)
used
data
on
8
tracers
from
Calabrese
et
al.,
1989
and
data
on
3
tracers
from
Davis
et
al.
(
1990)
to
estimate
soil
ingestion
rates.
The
median
of
the
soil
ingestion
estimates
from
the
lowest
four
F/
S
ratios
from
the
Calabrese
et
al.
(
1989)
study
most
often
included
Al,
Si,
Ti,
Y,
and
Zr.
Based
on
the
median
of
soil
ingestion
estimates
from
the
best
four
tracers,
the
mean
soil
ingestion
rate
was
132
mg/
day
and
the
median
was
33
mg/
day.
The
95th
percentile
value
was
154
mg/
day.
These
estimates
are
based
on
data
for
128
subject
weeks
for
the
64
children
in
the
Calabrese
et
al.
(
1989)
study.
For
the
101
children
in
the
Davis
et
al.
(
1990)
study,
the
mean
soil
ingestion
rate
was
69
mg/
day
and
the
median
soil
ingestion
rate
was
44
mg/
day.
The
95th
percentile
estimate
was
246
mg/
day.
These
data
are
based
on
the
three
tracers
(
i.
e.,
Al,
Si,
and
Ti)
from
the
Davis
et
al.
(
1990)
study.
When
the
Calabrese
et
al.
(
1989)
and
Davis
et
al.
(
1990)
studies
were
combined,
soil
ingestion
was
estimated
to
be
113
mg/
day
(
mean);
37
mg/
day
(
median);
and
217
mg/
day
(
95th
percentile),
using
the
BTM.
This
study
provides
a
reevaluation
of
previous
studies.
Its
advantages
are
that
it
combines
data
from
2
studies
for
children,
one
from
California
and
one
from
Massachusetts,
which
increases
the
number
of
observations.
It
also
corrects
for
biases
associated
with
the
differences
in
tracer
metabolism.
The
limitations
associated
with
the
data
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Exposure
Factors
Handbook
Page
August
1997
4­
11
used
in
this
study
are
the
same
as
the
limitations
described
from
areas
in
and
around
London
for
lead,
and
estimating
in
the
summaries
of
the
Calabrese
et
al.
(
1989),
Davis
et
al.
the
amount
of
hand
dirt
that
a
child
might
ingest.
Duggan
(
1990)
and
Calabrese
et
al.
(
1990)
studies.
and
Williams
(
1977)
estimated
the
amount
of
dust
that
4.3.
RELEVANT
STUDIES
ON
SOIL
INTAKE
AMONG
CHILDREN
Lepow
et
al.
(
1975)
­
Investigations
Into
Sources
of
Lead
in
the
Environment
of
Urban
Children
­
Lepow
et
al.
(
1975)
used
data
from
a
previous
study
(
Lepow
et
al.,
1974)
to
estimate
daily
soil
ingestion
rates
of
children.
Lepow
et
al.
(
1974)
estimated
ingestion
of
airborne
lead
fallout
among
urban
children
by:
(
1)
analyzing
surface
dirt
and
dust
samples
from
locations
where
children
played;
(
2)
measuring
hand
dirt
by
applying
preweighed
adhesive
labels
to
the
hands
and
weighing
the
amount
of
dirt
that
was
removed;
and
(
3)
observing
"
mouthing"
behavior
over
3
to
6
hours
of
normal
play.
Twenty­
two
children
from
an
urban
area
of
Connecticut
were
included
in
the
study.
Lepow
et
from
Exposure
to
Contaminated
Soil
­
Using
existing
al.
(
1975)
used
data
from
the
1974
study
and
found
that
the
literature,
Hawley
(
1985)
developed
scenarios
for
mean
weight
of
soil/
dust
on
the
hands
was
11
mg.
estimating
exposure
of
young
children,
older
children,
and
Assuming
that
a
child
would
put
fingers
or
other
"
dirty"
adults
to
contaminated
soil.
Annual
soil
ingestion
rates
objects
into
his
mouth
about
10
times
a
day
ingesting
11
mg
were
estimated
based
on
assumed
intake
rates
of
soil
and
of
dirt
each
time,
Lepow
et
al.
(
1975)
estimated
that
the
housedust
for
indoor
and
outdoor
activities
and
assumptions
daily
soil
ingestion
rate
would
be
about
100
mg/
day.
about
the
duration
and
frequency
of
the
activities.
These
According
to
Lepow
et
al.
(
1975),
the
amount
of
hand
dirt
soil
ingestion
rates
were
based
on
the
assumption
that
the
measured
with
this
technique
is
probably
an
underestimate
contaminated
area
is
in
a
region
having
a
winter
season.
because
dirt
trapped
in
skin
folds
and
creases
was
probably
Housedust
was
assumed
to
be
comprised
of
80
percent
soil.
not
removed
by
the
adhesive
label.
Consequently,
mean
Outdoor
exposure
to
contaminated
soil
among
young
soil
ingestion
rates
may
be
somewhat
higher
than
the
values
children
(
i.
e.,
2.5
years
old)
was
assumed
to
occur
5
days
estimated
in
this
study.
per
week
during
only
6
months
of
the
year
(
i.
e.,
mid­
April
Day
et
al.
(
1975)
­
Lead
in
Urban
Street
Dust
­
Day
et
al.
(
1975)
evaluated
the
contribution
of
incidental
ingestion
of
lead­
contaminated
street
dust
and
soil
to
children's
total
daily
intake
of
lead
by
measuring
the
amount
of
lead
in
street
dust
and
soil
and
estimating
the
amount
of
dirt
ingested
by
children.
The
amount
of
soil
that
might
be
ingested
was
estimated
by
measuring
the
amount
of
dirt
that
was
transferred
to
a
"
sticky
sweet"
during
30
minutes
of
play
and
assuming
that
a
child
might
eat
from
2
to
20
such
sweets
per
day.
Based
on
"
a
small
number
of
direct
measurements,"
Day
et
al.
(
1975)
found
that
5
to
50
mg
of
dirt
from
a
child's
hands
may
be
transferred
to
a
"
sticky
sweet"
during
30
minutes
of
"
normal
playground
activity.
Assuming
that
all
of
the
dirt
is
ingested
with
the
2
to
20
"
sticky
sweets,"
Day
et
al.
(
1975)
estimated
that
intake
of
soil
among
children
could
range
from
10
to
1000
mg/
day.
Duggan
and
Williams
(
1977)
­
Lead
in
Dust
in
City
Streets
­
Duggan
and
Williams
(
1977)
assessed
the
risks
associated
with
lead
in
street
dust
by
analyzing
street
dust
would
be
retained
on
the
forefinger
and
thumb
by
removing
a
small
amount
of
dust
from
a
weighed
amount,
rubbing
the
forefinger
and
thumb
together,
and
reweighing
to
determine
the
amount
retained
on
the
finger
and
thumb.
The
results
of
"
a
number
of
tests
with
several
different
people"
indicated
that
the
mean
amount
of
dust
retained
on
the
finger
and
thumb
was
approximately
4
mg
with
a
range
of
2
to
7
mg
(
Duggan
and
Williams,
1977).
Assuming
that
a
child
would
suck
his/
her
finger
or
thumb
10
times
a
day
and
that
all
of
the
dirt
is
removed
each
time
and
replaced
with
new
dirt
prior
to
subsequent
mouthing
behavior,
Duggan
and
Williams
(
1977)
estimated
that
20
mg
of
dust
would
be
ingested
per
day.
Hawley
et
al.
(
1985)
­
Assessment
of
Health
Risk
through
mid­
October).
Children
were
assumed
to
ingest
250
mg
soil/
day
while
playing
outdoors
based
on
data
presented
in
Lepow
et
al.
(
1974;
1975)
and
Roels
et
al.
(
1980).
Indoor
exposures
among
this
population
were
based
on
the
assumption
that
young
children
ingest
100
mg
of
housedust
per
day
while
spending
all
of
their
time
indoors
during
the
winter
months,
and
50
mg
of
housedust
per
day
during
the
warmer
months
when
only
a
portion
of
their
time
is
spent
indoors.
Based
on
these
assumptions,
Hawley
(
1985)
estimated
that
the
annual
average
soil
intake
rate
for
young
children
is
150
mg/
day
(
Table
4­
11).
Older
children
(
i.
e.,
6
year
olds)
were
assumed
to
ingest
50
mg
of
soil
per
day
from
an
area
equal
to
the
area
of
the
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Page
Exposure
Factors
Handbook
4­
12
August
1997
Table
4­
11.
Estimates
of
Soil
Ingestion
for
Children
Scenarios
Media
(
mg/
day)
Activity
Content
(
mg/
day)
Exposure
Days/
Year
Fraction
Soil
Intake
Annual
Average
Soil
Young
Child
(
2.5
Years
Old)
Outdoor
Activities
(
Summer)
Soil
250
130
1
90
Indoor
Activities
(
Summer)
Dust
50
182
0.8
20
Indoor
Activities
(
Winter
Dust
100
182
0.8
40
TOTAL
SOIL
INTAKE
150
Older
Child
(
6
Years
Old)
Outdoor
Activities
(
Summer)
Soil
50
152
1
21
Indoor
Activities
(
Year­
Round)
Dust
3
365
0.8
2.4
TOTAL
SOIL
INTAKE
23.4
Source:
Hawley,
1985.

fingers
on
one
hand
while
playing
outdoors.
This
each
child,
mean
soil
intake
was
estimated
to
be
91
mg/
day
assumption
was
based
on
data
from
Lepow
et
al.
(
1975).
and
90th
percentile
intake
was
estimated
to
be
143
mg/
day.
Outdoor
activities
were
assumed
to
occur
each
day
over
5
Thompson
and
Burmaster
(
1991)
tested
the
months
of
the
year
(
i.
e.,
during
May
through
October).
hypothesis
that
soil
ingestion
rates
based
on
the
adjusted
These
children
were
also
assumed
to
ingest
3
mg/
day
of
Binder
et
al.
(
1986)
data
for
aluminum,
silicon
and
the
housedust
from
the
indoor
surfaces
of
the
hands
during
average
of
these
two
tracers
were
lognormally
distributed.
indoor
activities
occurring
over
the
entire
year.
Using
these
The
distribution
of
soil
intake
based
on
titanium
was
not
data,
Hawley
(
1985)
estimated
the
annual
average
soil
tested
for
lognormality
because
titanium
may
be
present
in
intake
rate
for
older
children
to
be
23.4
mg/
day
(
Table
4­
food
in
high
concentrations
and
the
Binder
et
al.
(
1986)
11).
study
did
not
correct
for
food
sources
of
titanium
Thompson
and
Burmaster
(
1991)
­
Parametric
Distributions
for
Soil
Ingestion
by
Children
­
Thompson
and
Burmaster
(
1991)
developed
parameterized
distributions
of
soil
ingestion
rates
for
children
based
on
a
reanalysis
of
the
data
collected
by
Binder
et
al.
(
1986).
In
the
original
Binder
et
al.
(
1986)
study,
an
assumed
fecal
weight
of
15
g/
day
was
used.
Thompson
and
Burmaster
reestimated
the
soil
ingestion
rates
from
the
Binder
et
al.
(
1986)
study
using
the
actual
stool
weights
of
the
study
participants
instead
of
the
assumed
stool
weights.
Because
the
actual
stool
weights
averaged
only
7.5
g/
day,
the
soil
ingestion
estimates
presented
by
Thompson
and
Burmaster
(
1991)
are
approximately
one­
half
of
those
reported
by
Binder
et
al.
(
1986).
Table
4­
12
presents
the
distribution
of
estimated
soil
ingestion
rates
calculated
by
Thompson
and
Burmaster
(
1991)
based
on
the
three
tracers
elements
(
i.
e.,
aluminum,
silicon,
and
titanium),
and
on
the
arithmetic
average
of
soil
ingestion
based
on
aluminum
and
silicon.
The
mean
soil
intake
rates
were
97
mg/
day
for
aluminum,
85
mg/
day
for
silicon,
and
1,004
mg/
day
for
titanium.
The
90th
percentile
estimates
were
197
mg/
day
for
aluminum,
166
mg/
day
for
silicon,
and
2,105
mg/
day
for
titanium.
Based
on
the
arithmetic
average
of
aluminum
and
silicon
for
(
Thompson
and
Burmaster,
1991).
Although
visual
inspection
of
the
distributions
for
aluminum,
silicon,
and
the
average
of
these
tracers
all
indicated
that
they
may
be
lognormally
distributed,
statistical
tests
indicated
that
only
silicon
and
the
average
of
the
silicon
and
aluminum
tracers
were
lognormally
distributed.
Soil
intake
rates
based
on
aluminum
were
not
lognormally
distributed.
Table
4­
12
also
presents
the
lognormal
distribution
parameters
and
underlying
normal
distribution
parameters
(
i.
e.,
the
natural
logarithms
of
the
data)
for
aluminum,
silicon,
and
the
average
of
these
two
tracers.
According
to
the
authors,
"
the
parameters
estimated
from
the
underlying
normal
distribution
are
much
more
reliable
and
robust"
(
Thompson
and
Burmaster,
1991).
The
advantages
of
this
study
are
that
it
provides
percentile
data
and
defines
the
shape
of
soil
intake
distributions.
However,
the
number
of
data
points
used
to
fit
the
distribution
was
limited.
In
addition,
the
study
did
not
generate
"
new"
data.
Instead,
it
provided
a
reanalysis
of
previously­
reported
data
using
actual
fecal
weights.
No
corrections
were
made
for
tracer
intake
from
food
or
Y
i
'
x
e
(
&
0.112
(
yr)
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Exposure
Factors
Handbook
Page
August
1997
4­
13
(
Eqn.
4­
3)

where:
Y
=
adjusted
mean
soil
ingestion
(
mg/
day)
i
x
=
a
constant
yr
=
average
age
(
2
years)
Table
4­
12.
Estimated
Soil
Ingestion
Rate
Summary
Statistics
and
Parameters
for
Distributions
Using
Binder
et
al.
(
1986)
Data
with
Actual
Fecal
Weights
Trace
Element
Basis
Soil
Intake
(
mg/
day)

A1
Si
Ti
MEANa
Mean
97
85
1,004
91
Min
11
10
1
13
10th
21
19
3
22
20th
33
23
22
34
30th
39
36
47
43
40th
43
52
172
49
Med
45
60
293
59
60th
55
65
475
69
70th
73
79
724
92
80th
104
106
1,071
100
90th
197
166
2,105
143
Max
1,201
642
14,061
921
Lognormal
Distribution
Parameters
Median
45
60
­­
59
Standard
Deviation
169
95
­­
126
Arithmetic
Mean
97
85
­­
91
Underlying
Normal
Distribution
Parameters
Mean
4.06
4.07
­­
4.13
Standard
Deviation
0.88
0.85
­­
0.80
MEAN
=
arithmetic
average
of
soil
ingestion
based
on
aluminum
and
silicon.
a
Source:
Thompson
and
Burmaster,
1991.

medicine
and
the
results
may
not
be
representative
of
longterm
intake
rates
because
the
data
were
derived
from
a
short­
term
study.
Sedman
and
Mahmood
(
1994)
­
Soil
Ingestion
by
Children
and
Adults
Reconsidered
Using
the
Results
of
Recent
Tracer
Studies
­
Sedman
and
Mahmood
(
1994)
used
the
results
of
two
recent
children's
(
Calabrese
et
al.
1989;
Davis
et
al.
1990)
tracer
studies
to
determine
estimates
of
average
daily
soil
ingestion
in
young
children
and
for
over
a
lifetime.
In
the
two
studies,
the
intake
and
excretion
of
a
variety
of
tracers
were
monitored,
and
concentrations
of
tracers
in
soil
adjacent
to
the
children's
dwellings
were
determined
(
Sedman
and
Mahmood,
1994).
From
a
mass
balance
approach,
estimates
of
soil
ingestion
in
these
children
were
determined
by
dividing
the
excess
tracer
intake
(
i.
e.,
quantity
of
tracer
recovered
in
the
feces
in
excess
of
the
measured
intake)
by
the
average
concentration
of
tracer
in
soil
samples
from
each
child's
dwelling.
Sedman
and
Mahmood
(
1994)
adjusted
the
mean
estimates
of
soil
ingestion
in
children
for
each
tracer
(
Y)
from
both
studies
to
reflect
that
of
a
2­
year
old
child
using
the
following
equation:
In
addition
to
the
study
in
young
children,
a
study
(
Calabrese
et
al.,
1989)
in
adults
was
conducted
to
evaluate
the
tracer
methodology.
In
the
adult
studies,
percent
recoveries
of
tracers
were
determined
in
six
adults
who
ingested
known
quantities
of
tracers
in
1.5
or
0.3
grams
of
soil.
The
distribution
of
tracer
recoveries
from
adults
was
evaluated
using
data
analysis
techniques
involving
visualization
and
exploratory
data
analysis
(
Sedman
and
Mahmood,
1994).
From
the
results
obtained
in
these
studies,
the
distribution
of
tracer
recoveries
from
adults
were
determined.
In
addition,
an
analysis
of
variance
(
ANOVA)
and
Tukey's
multiple
comparison
methodologies
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Page
Exposure
Factors
Handbook
4­
14
August
1997
were
employed
to
identify
differences
in
the
recoveries
of
in
a
2­
year
old
child,
an
average
daily
soil
ingestion
over
a
the
various
tracers
(
Sedman
and
Mahmood,
1994).
lifetime
was
estimated
to
be
70
mg/
day.
The
lifetime
From
the
adult
studies,
the
ANOVA
of
the
natural
estimates
were
derived
using
the
equation
presented
above
logarithm
of
the
recoveries
of
tracers
from
0.3
or
1.5
g
of
that
describes
changes
in
soil
ingestion
with
age
(
Sedman
ingested
soil
showed
a
significant
difference
(
%
=
0.05)
and
Mahmood,
1994).
among
the
estimates
of
recovery
of
the
tracers
regardless
of
whether
the
recoveries
were
combined
or
analyzed
separately
(
Sedman
and
Mahmood,
1994).
Sedman
and
Mahmood
(
1994)
also
reported
that
barium,
manganese,
and
zirconium
yielded
significantly
different
estimates
of
soil
ingestion
than
the
other
tracers
(
aluminum,
silicon,
yttrium,
titanium,
and
vanadium).
Table
4­
13
presents
the
Tukey's
multiple
comparison
of
mean
log
tracer
recovery
in
adults
ingesting
known
quantities
of
soil.
The
average
ages
of
children
in
the
two
recent
studies
were
2.4
years
in
Calabrese,
et
al.
(
1989)
and
4.7
years
in
Davis
et
al.
(
1990).
The
mean
of
the
adjusted
levels
of
soil
ingestion
for
a
two
year
old
child
was
220
mg/
kg
for
the
Calabrese
et
al.
(
1989)
study
and
170
mg/
kg
for
the
Davis
et
al.
(
1990)
study
(
Sedman
and
Mahmood,
1994).
From
the
adjusted
soil
ingestion
estimates,
based
on
a
normal
distribution
of
means,
the
mean
estimate
for
a
2­
year
old
child
was
195
mg/
day
and
the
overall
mean
of
soil
ingestion
and
the
standard
error
of
the
mean
was
53
mg/
day
(
Sedman
and
Mahmood,
1994).
Based
on
uncertainties
associated
with
the
method
employed,
Sedman
and
Mahmood
(
1994)
recommended
a
conservative
estimate
of
soil
ingestion
in
young
children
of
250
mg/
day.
Based
on
the
250
mg/
day
ingestion
rate
AIHC
Exposure
Factors
Sourcebook
(
1994)
­
The
Exposure
Factors
Sourcebook
(
AIHC,
1994)
uses
data
from
the
Calabrese
et
al.
(
1990)
study
to
derive
soil
ingestion
rates
using
zirconium
as
the
tracer.
More
recent
papers
indicate
that
zirconium
is
not
a
good
tracer.
Therefore,
the
values
recommended
in
the
AIHC
Sourcebook
are
not
appropriate.
Furthermore,
because
individuals
were
only
studied
for
a
short
period
of
time,
deriving
a
distribution
of
usual
intake
is
not
possible
and
is
inappropriate.
Calabrese
and
Stanek
(
1995)
­
Resolving
Intertracer
Inconsistencies
in
Soil
Ingestion
Estimation
­
Calabrese
and
Stanek
(
1995)
explored
sources
and
magnitude
of
positive
and
negative
errors
in
soil
ingestion
estimates
for
children
on
a
subject­
week
and
trace
element
basis.
Calabrese
and
Stanek
(
1995)
identified
possible
sources
of
positive
errors
to
be
the
following:

C
Ingestion
of
high
levels
of
tracers
before
the
study
starts
and
low
ingestion
during
study
period
may
result
in
over
estimation
of
soil
ingestion;
and
C
Ingestion
of
element
tracers
from
a
non­
food
or
non­
soil
source
during
the
study
period.

Table
4­
13.
Tukey's
Multiple
Comparison
of
Mean
Log
Tracer
Recovery
in
Adults
Ingesting
Known
Quantities
of
Soil
Tracer
Reported
Mean
Age
Adjusted
Mean
(
mg/
day)
(
mg/
day)

Calabrese
et
al.,
1989
Study
Aluminum
153
160
Silicon
154
161
Titanium
218
228
Vanadium
459
480
Yttrium
85
89
Davis
et
al.,
1990
Study
Aluminum
39
53
Silicon
81
111
Titanium
246
333
Age
adjusted
mean
estimates
of
soil
ingestion
in
young
children.
Mean
estimates
of
soil
ingestion
for
each
tracer
in
each
study
were
a
adjusted
using
the
following
equation:
Y
=
x
e
,
where
Y
=
adjusted
mean
soil
ingestion
(
mg/
day),
x
=
a
constant,
and
yr
=
age
in
years.
(­
0.112
*
yr)

Source:
Sedman
and
Mahmood,
1994.
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Exposure
Factors
Handbook
Page
August
1997
4­
15
Possible
sources
of
negative
bias
identified
by
Calabrese
considered
the
best
estimate
for
that
particular
day.
The
and
Stanek
(
1995)
are
the
following:
magnitude
of
positive
or
negative
error
for
a
specific
tracer
C
Ingestion
of
tracers
in
food,
but
the
tracers
are
the
value
for
the
tracer
and
the
median
value;
(
4)
negative
not
captured
in
the
fecal
sample
either
due
to
errors
due
to
missing
fecal
samples
at
the
end
of
the
study
slow
lag
time
or
not
having
a
fecal
sample
period
were
also
determined
(
Calabrese
and
Stanek,
1995).
available
on
the
final
study
day;
and
Table
4­
14
presents
the
estimated
magnitude
of
C
Sample
measurement
errors
which
result
in
children's
study
(
i.
e.,
conducted
by
Calabrese
et
al.,
1989).
diminished
detection
of
fecal
tracers,
but
not
in
The
original
mean
soil
ingestion
rates
ranged
from
a
low
of
soil
tracer
levels.
21
mg/
day
based
on
zirconium
to
a
high
of
459
mg/
day
The
authors
developed
an
approach
which
attempted
to
ingestion
rate
after
correcting
for
negative
and
positive
reduce
the
magnitude
of
error
in
the
individual
trace
errors
ranged
from
97
mg/
day
based
on
yttrium
to
208
element
ingestion
estimates.
Results
from
a
previous
study
mg/
day
based
on
titanium
(
Table
4­
14).
Calabrese
and
conducted
by
Calabrese
et
al.
(
1989)
were
used
to
quantify
Stanek
(
1995)
concluded
that
correcting
for
errors
at
the
these
errors
based
on
the
following
criteria:
(
1)
a
lag
period
individual
level
for
each
tracer
element
provides
more
of
28
hours
was
assumed
for
the
passage
of
tracers
ingested
reliable
estimates
of
soil
ingestion.
in
food
to
the
feces
(
this
value
was
applied
to
all
subject­
day
This
report
is
valuable
in
providing
additional
estimates);
(
2)
daily
soil
ingestion
rate
was
estimated
for
understanding
of
the
nature
of
potential
errors
in
trace
each
tracer
for
each
24­
hr
day
a
fecal
sample
was
obtained;
element
specific
estimates
of
soil
ingestion.
However,
the
(
3)
the
median
tracer­
based
soil
ingestion
rate
for
each
operational
definition
used
for
estimating
the
error
in
a
trace
subject­
day
was
determined.
Also,
upper
and
lower
bound
element
estimate
was
the
observed
difference
of
that
tracer
estimates
were
determined
based
on
criteria
formed
using
from
a
median
tracer
value.
Specific
identification
of
an
assumption
of
the
magnitude
of
the
relative
standard
sources
of
error,
or
direct
evidence
that
individual
tracers
deviation
(
RSD)
presented
in
another
study
conducted
by
were
indeed
in
error
was
not
developed.
Corrections
to
Stanek
and
Calabrese
(
1995a).
Daily
soil
ingestion
rates
individual
tracer
means
were
then
made
according
to
how
for
tracers
that
fell
beyond
the
upper
and
lower
ranges
were
different
values
for
that
tracer
were
from
the
median
values.
excluded
from
subsequent
calculations,
and
the
median
soil
This
approach
is
based
on
the
hypothesis
that
the
median
ingestion
rates
of
the
remaining
tracer
elements
were
tracer
value
is
the
most
per
day
was
derived
by
determining
the
difference
between
positive
and
negative
error
for
six
tracer
elements
in
the
based
on
titanium
(
Table
4­
14).
The
adjusted
mean
soil
Table
4­
14.
Positive/
Negative
Error
(
bias)
in
Soil
Ingestion
Estimates
in
the
Calabrese
et
al.
(
1989)
Mass­
balance
Study:
Effect
on
Mean
Soil
Ingestion
Estimate
(
mg/
day)
a
Negative
Error
Lack
of
Fecal
Sample
on
Final
Other
Causes
Total
Negative
Total
Positive
Original
Mean
Adjusted
Study
Day
Error
Error
Net
Error
Mean
b
Aluminum
14
11
25
43
+
18
153
136
Silicon
15
6
21
41
+
20
154
133
Titanium
82
187
269
282
+
13
218
208
Vanadium
66
55
121
432
+
311
459
148
Yttrium
8
26
34
22
­
12
85
97
Zirconium
6
91
97
5
­
92
21
113
How
to
read
table:
for
example,
aluminum
as
a
soil
tracer
displayed
both
negative
and
positive
error.
The
cumulative
total
negative
error
is
a
estimated
to
bias
the
mean
estimate
by
25
mg/
day
downward.
However,
aluminum
has
positive
error
biasing
the
original
mean
upward
by
43
mg/
day.
The
net
bias
in
the
original
mean
was
18
mg/
day
positive
bias.
Thus,
the
original
156
mg/
day
mean
for
aluminum
should
be
corrected
downward
to
136
mg/
day.
Values
indicate
impact
on
mean
of
128­
subject­
weeks
in
milligrams
of
soil
ingested
per
day.
b
Source:
Calabrese
and
Stanek,
1995.
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Page
Exposure
Factors
Handbook
4­
16
August
1997
accurate
estimate
of
soil
ingestion,
and
the
validity
of
this
Sheppard's
(
1995)
estimates,
based
on
activity
and
assumption
depends
on
the
specific
set
of
tracers
used
in
the
exposure
duration,
are
quite
similar
to
the
mean
values
from
study
and
need
not
be
correct.
The
approach
used
for
the
intake
rate
estimates
described
in
previous
sections.
The
estimation
of
daily
tracer
intake
is
the
same
as
in
Stanek
and
advantages
of
this
study
are
that
the
model
can
be
used
to
Calabrese
(
1995a),
and
some
limitations
of
that
approach
calculate
the
ingestion
rate
from
non­
food
sources
with
are
mentioned
in
the
review
of
that
study.
variability
in
exposure
ingestion
rates
and
exposure
Sheppard
(
1995)
­
Parameter
Values
to
Model
the
Soil
Ingestion
Pathway
­
Sheppard
(
1995)
summarized
the
available
literature
on
soil
ingestion
to
estimate
the
amount
of
soil
ingestion
in
humans
for
the
purposes
of
risk
assessment.
Sheppard
(
1995)
categorized
the
available
soil
ingestion
studies
into
two
general
approaches:
(
1)
those
that
measured
the
soil
intake
rate
with
the
use
of
tracers
in
the
soil,
and
(
2)
those
that
estimated
soil
ingestion
based
on
Exposure
to
Contaminated
Soil
­
Information
on
soil
activity
(
e.
g.,
hand­
to­
mouth)
and
exposure
duration.
ingestion
among
adults
is
very
limited.
Hawley
(
1985)
Sheppard
(
1995)
provided
estimates
of
soil
intake
based
on
estimated
soil
ingestion
among
adults
based
on
assumptions
previously
published
tracer
studies.
The
data
from
these
regarding
activity
patterns
and
corresponding
ingestion
studies
were
assumed
to
be
lognormally
distributed
due
to
amounts.
Hawley
(
1985)
assumed
that
adults
ingest
the
broad
range,
the
concept
that
soil
ingestion
is
never
outdoor
soil
at
a
rate
of
480
mg/
day
while
engaged
in
zero,
and
the
possibility
of
very
high
values.
In
order
to
yardwork
or
other
physical
activity.
These
outdoor
account
for
skewness
in
the
data,
geometric
means
rather
exposures
were
assumed
to
occur
2
days/
week
during
5
than
arithmetic
means,
were
calculated
by
age,
excluding
months
of
the
year
(
i.
e.,
May
through
October).
The
pica
and
geophagy
values.
The
geometric
mean
for
soil
ingestion
estimate
was
based
on
the
assumption
that
a
50
ingestion
rate
for
children
under
six
was
estimated
to
be
F
m/
thick
layer
of
soil
is
ingested
from
the
inside
surfaces
of
100
mg/
day.
For
children
over
six
and
adults,
the
geometric
the
thumb
and
fingers
of
one
hand.
Ingestion
of
indoor
mean
intake
rate
was
estimated
to
be
20
mg/
day.
Sheppard
housedust
was
assumed
to
occur
from
typical
living
space
(
1995)
also
provided
soil
ingestion
estimates
for
indoor
and
activities
such
as
eating
and
smoking,
and
work
in
attics
or
outdoor
activities
based
on
data
from
Hawley
(
1985)
and
other
uncleaned
areas
of
the
house.
Hawley
(
1985)
assumptions
regarding
duration
of
exposure
(
Table
4­
15).
assumed
that
adults
ingest
an
average
of
0.56
mg
durations.
The
limitation
of
this
study
is
that
it
does
not
introduce
new
data;
previous
data
are
re­
evaluated.
In
addition,
because
the
model
is
based
on
previous
data,
the
same
advantages
and
limitations
of
those
studies
apply.

4.4.
SOIL
INTAKE
AMONG
ADULTS
Hawley
1985
­
Assessment
of
Health
Risk
from
housedust/
day
during
typical
living
space
activities
and
110
mg
housedust/
day
while
working
in
attics.
Attic
work
Table
4­
15.
Soil
Ingestion
Rates
for
Assessment
Purposes
Receptor
Age
Setting
Hands
Rate
Durations
Ingestion
Soil
Load
on
Soil
Exposure
Ingestion
Suggested
Exposure
Average
Daily
Soil
(
mg/
cm
)
(
mg/
hr)
(
hr/
yr)
(
mg/
day)
2
Pica
Child
­­­
1,000
200
500
2.5
yrs
Outdoor
0.5
20
1,000
50
Indoor
0.4
3
Remaining
60
a
6
yrs
Outdoor
0.5
10
700
20
Indoor
0.04
0.15
5,000
2
Adult
Gardening
1.0
20
300
20
Indoor
0.04
0.03
5,000
0.4
Hawley
(
1985)
assumed
the
child
spent
all
the
time
at
home,
so
that
the
indoor
time
was
8,760
hours/
year
minus
the
outdoor
time.
a
Source:
Sheppard,
1995
was
assumed
to
occur
12
days/
year.
Hawley
(
1985)
also
Based
on
these
assumptions
about
soil
intake
and
the
assumed
that
soil
comprises
80
percent
of
household
dust.
frequency
of
indoor
and
outdoor
activities,
Hawley
(
1985)
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Exposure
Factors
Handbook
Page
August
1997
4­
17
Table
4­
16.
Estimates
of
Soil
Ingestion
for
Adults
Scenarios
Media
Exposure
(
mg/
day)
Days/
Year
Activity
Fraction
Soil
Content
Annual
Average
Soil
Intake
(
mg/
day)

Adult
Work
in
attic
(
year­
round)
Dust
110
12
0.8
3
Living
Space
(
year­
round)
Dust
0.56
365
0.8
0.5
Outdoor
Work
(
summer)
Soil
480
43
1
57
TOTAL
SOIL
INTAKE
60.5
Source:
Hawley,
1985.
estimated
the
annual
average
soil
intake
rate
for
adults
to
be
during
each
of
the
3
weeks.
In
addition,
all
medications
and
60.5
mg/
day
(
Table
4­
16).
vitamins
ingested
by
the
adults
were
collected.
Total
The
soil
intake
value
estimated
by
Hawley
(
1985)
is
excretory
output
were
collected
from
Monday
noon
through
consistent
with
adult
soil
intake
rates
suggested
by
other
Friday
midnight
over
3
consecutive
weeks.
Table
4­
17
researchers.
Calabrese
et
al.
(
1987)
suggested
that
soil
provides
the
mean
and
median
values
of
soil
ingestion
for
intake
among
adults
ranges
from
1
to
100
mg/
day.
each
element
by
week.
Data
obtained
from
the
first
week,

According
to
Calabrese
et
al.
(
1987),
these
values
"
are
when
empty
gelatin
capsules
were
ingested,
may
be
used
to
conjectural
and
based
on
fractional
estimates"
of
earlier
derive
an
estimate
of
soil
intake
by
adults.
The
mean
intake
Center
for
Disease
Control
(
CDC)
estimates.
In
an
rates
for
the
eight
tracers
are:
Al,
110
mg;
Ba,
­
232
mg;
evaluation
of
the
scientific
literature
concerning
soil
Mn,
330
mg;
Si,
30
mg;
Ti,
71
mg;
V,
1,288
mg;
Y,
63
mg;
ingestion
rates
for
children
and
adults
(
Krablin,
1989),
Arco
and
Zr,
134
mg.
Coal
Company
suggested
that
10
mg/
day
may
be
an
The
advantage
of
this
study
is
that
it
provides
appropriate
value
for
adult
soil
ingestion.
This
value
is
quantitative
estimates
of
soil
ingestion
for
adults.
The
study
based
on
"
extrapolation
from
urine
arsenic
epidemiological
also
corrected
for
tracer
concentrations
in
foods
and
studies
and
information
on
mouthing
behavior
and
time
medicines.
However,
a
limitation
of
this
study
is
that
a
activity
patterns"
(
Krablin,
1989).
limited
number
of
subjects
were
studied.
In
addition,
the
Calabrese
et
al.
(
1990)
­
Preliminary
Adult
Soil
Ingestion
Estimates:
Results
of
a
Pilot
Study­
Calabrese
et
al.
(
1990)
studied
six
adults
to
evaluate
the
extent
to
which
they
ingest
soil.
This
adult
study
was
originally
part
of
the
children
soil
ingestion
study
conducted
by
Calabrese
and
was
used
to
validate
part
of
the
analytical
methodology
used
in
the
children
study.
The
participants
were
six
healthy
adults,
three
males
and
three
females,
25­
41
years
old.
Each
volunteer
ingested
one
empty
gelatin
capsule
at
breakfast
and
one
at
dinner
Monday,
Tuesday,
and
Wednesday
during
the
first
week
of
the
study.
During
the
second
week,
they
ingested
50
mg
of
sterilized
soil
within
a
gelatin
capsule
at
breakfast
and
at
dinner
(
a
total
of
100
mg
of
sterilized
soil
per
day)
for
3
days.
For
the
third
week,
the
participants
ingested
250
mg
of
sterilized
soil
in
a
gelatin
capsule
at
breakfast
and
at
dinner
(
a
total
of
500
mg
of
soil
per
day)
during
the
three
days.
Duplicate
meal
samples
(
food
and
beverage)
were
collected
from
the
six
adults.
The
sample
included
all
foods
ingested
from
breakfast
Monday,
through
the
evening
meal
Wednesday
subjects
were
only
studied
for
one
week
before
soil
capsules
were
ingested.

4.5.
PREVALENCE
OF
PICA
The
scientific
literature
define
pica
as
"
the
repeated
eating
of
non­
nutritive
substances"
(
Feldman,
1986).
For
the
purposes
of
this
handbook,
pica
is
defined
as
an
deliberately
high
soil
ingestion
rate.
Numerous
articles
have
been
published
that
report
on
the
incidence
of
pica
among
various
populations.
However,
most
of
these
papers
describe
pica
for
substances
other
than
soil
including
sand,
clay,
paint,
plaster,
hair,
string,
cloth,
glass,
matches,
paper,
feces,
and
various
other
items.
These
papers
indicate
that
the
pica
occurs
in
approximately
half
of
all
children
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Page
Exposure
Factors
Handbook
4­
18
August
1997
Table
4­
17.
Adult
Daily
Soil
Ingestion
Estimates
by
Week
and
Tracer
Element
After
Subtracting
Food
and
Capsule
Ingestion,
Based
on
Median
Amherst
Soil
Concentrations:
Means
and
Medians
Over
Subjects
(
mg)
a
Week
Al
Ba
Mn
Si
Ti
V
Y
Zr
Means
1
2
3
110
98
28
­
232
12,265
201
330
1,306
790
30
14
­
23
71
25
896
1,288
43
532
63
21
67
134
58
­
74
Medians
1
2
3
60
85
66
­
71
597
386
388
1,368
831
31
15
­
27
102
112
156
1,192
150
047
44
35
60
124
65
­
144
Data
were
converted
to
milligrams
a
Negative
values
occur
because
of
correction
for
food
and
capsule
ingestion.
b
Source:
Calabrese
et
al.,
1990
between
the
ages
of
1
and
3
years
(
Sayetta,
1986).
The
nonpregnant
women.
However,
"
dirt"
was
not
clearly
incidence
of
deliberate
ingestion
behavior
in
children
has
defined.
The
Bruhn
and
Pangborn
(
1971)
study
was
been
shown
to
differ
for
different
subpopulations.
The
conducted
among
91
non­
black,
low
income
families
of
incidence
rate
appears
to
be
higher
for
black
children
than
migrant
agricultural
workers
in
California.
Based
on
the
for
white
children.
Approximately
30
percent
of
black
data
from
the
five
key
tracer
studies
(
Binder
et
al.,
1986;
children
aged
1
to
6
years
are
reported
to
have
deliberate
Clausing
et
al.,
1987;
Van
Wïjnen
et
al.,
1990;
Davis
et
al.,
ingestion
behavior,
compared
with
10
to
18
percent
of
1990;
and
Calabrese
et
al.,
1989)
only
one
child
out
of
the
white
children
in
the
same
age
group
(
Danford,
1982).
more
than
600
children
involved
in
all
of
these
studies
There
does
not
appear
to
be
any
sex
differences
in
the
ingested
an
amount
of
soil
significantly
greater
than
the
incidence
rates
for
males
or
females
(
Kaplan
and
Sadock,
range
for
other
children.
Although
these
studies
did
not
1985).
Lourie
et
al.
(
1963)
states
that
the
incidence
of
pica
include
data
for
all
populations
and
were
representative
of
is
higher
among
children
in
lower
socioeconomic
groups
short­
term
ingestions
only,
it
can
be
assumed
that
the
(
i.
e.,
50
to
60
percent)
than
in
higher
income
families
(
i.
e.,
incidence
rate
of
deliberate
soil
ingestion
behavior
in
the
about
30
percent).
Deliberate
soil
ingestion
behavior
general
population
is
low.
However,
it
is
incumbent
upon
appears
to
be
more
common
in
rural
areas
(
Vermeer
and
the
user
to
use
the
appropriate
value
for
their
specific
study
Frate,
1979).
A
higher
rate
of
pica
has
also
been
reported
population.
for
pregnant
women
and
individuals
with
poor
nutritional
status
(
Danford,
1982).
In
general,
deliberate
ingestion
behavior
is
more
frequent
and
more
severe
in
mentally
retarded
children
than
in
children
in
the
general
population
Information
on
the
amount
of
soil
ingested
by
(
Behrman
and
Vaughan
1983,
Danford
1982,
Forfar
and
children
with
abnormal
soil
ingestion
behavior
is
limited.
Arneil
1984,
Illingworth
1983,
Sayetta
1986).
However,
some
evidence
suggests
that
a
rate
on
the
order
of
It
should
be
noted
that
the
pica
statistics
cited
above
10
g/
day
may
not
be
unreasonable.
apply
to
the
incidence
of
general
pica
and
not
soil
pica.
Information
on
the
incidence
of
soil
pica
is
limited,
but
it
appears
that
soil
pica
is
less
common.
A
study
by
Vermeer
and
Frate
(
1979)
showed
that
the
incidence
of
geophagia
(
i.
e.,
earth­
eating)
was
about
16
percent
among
children
from
a
rural
black
community
in
Mississippi.
However,
geophagia
was
described
as
a
cultural
practice
among
the
community
surveyed
and
may
not
be
representative
of
the
general
population.
Average
daily
consumption
of
soil
was
estimated
to
be
50
g/
day.
Bruhn
and
Pangborn
(
1971)
reported
the
incidence
of
pica
for
"
dirt"
to
be
19
percent
in
children,
14
percent
in
pregnant
women,
and
3
percent
in
4.6.
DELIBERATE
SOIL
INGESTION
AMONG
CHILDREN
Calabrese
et
al.
(
1991)
­
Evidence
of
Soil
Pica
Behavior
and
Quantification
of
Soil
Ingestion
­
Calabrese
et
al.
(
1991)
estimated
that
upper
range
soil
ingestion
values
may
range
from
approximately
5­
7
grams/
day.
This
estimate
was
based
on
observations
of
one
pica
child
among
the
64
children
who
participated
in
the
study.
In
the
study,
a
3.5­
year
old
female
exhibited
extremely
high
soil
ingestion
behavior
during
one
of
the
two
weeks
of
observation.
Intake
ranged
from
74
mg/
day
to
2.2
g/
day
during
the
first
week
of
observation
and
10.1
to
13.6
g/
day
during
the
second
week
of
observation
(
Table
4­
18).
These
results
are
based
on
mass­
balance
analyses
for
seven
(
i.
e.,
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Exposure
Factors
Handbook
Page
August
1997
4­
19
aluminum,
barium,
manganese,
silicon,
titanium,
vanadium,
ingested
10­
13
grams
of
soil
per
day
over
the
second
week
and
yttrium)
of
the
eight
tracer
elements
used.
Intake
rates
of
a
2­
week
soil
ingestion
study.
Also,
the
previous
study
based
on
zirconium
was
significantly
lower
but
Calabrese
utilized
a
soil
tracer
methodology
with
eight
different
tracers
et
al.
(
1991)
indicated
that
this
may
have
"
resulted
from
a
(
Al,
Ba,
Mn,
Si,
Ti,
V,
Y,
Zr).
The
reader
is
referred
to
limitation
in
the
analytical
protocol."
Calabrese
et
al.
(
1989)
for
a
detailed
description
and
results
Table
4­
18.
Daily
Soil
Ingestion
Estimation
in
a
Soil­
Pica
Child
by
Tracer
and
by
Week
(
mg/
day)

Tracer
Estimated
Soil
Ingestion
Estimated
Soil
Ingestion
Week
1
Week
2
Al
74
13,600
Ba
458
12,088
Mn
2,221
12,341
Si
142
10,955
Ti
1,543
11,870
V
1,269
10,071
Y
147
13,325
Zr
86
2,695
Source:
Calabrese
et
al.,
1991
Calabrese
and
Stanek
(
1992)
­
Distinguishing
Outdoor
Soil
Ingestion
from
Indoor
Dust
Ingestion
in
a
Soil
Pica
Child
­
Calabrese
and
Stanek
(
1992)
quantitatively
distinguished
the
amount
of
outdoor
soil
ingestion
from
indoor
dust
ingestion
in
a
soil
pica
child.
This
study
was
based
on
a
previous
mass­
balance
study
(
conducted
in
1991)
in
which
a
3­
1/
2
year
old
child
of
the
soil
ingestion
study.
Calabrese
and
Stanek
(
1992)
distinguished
indoor
dust
from
outdoor
soil
in
ingested
soil
based
on
a
methodology
which
compared
differential
element
ratios.
Table
4­
19
presents
tracer
ratios
of
soil,
dust,
and
residual
fecal
samples
in
the
soil
pica
child.
Calabrese
and
Stanek
(
1992)
reported
that
there
was
a
maximum
total
of
28
pairs
of
tracer
ratios
based
on
eight
tracers.
However,
only
19
pairs
of
tracer
ratios
were
available
for
quantitative
evaluation
as
shown
in
Table
4­
19.
Of
these
19
pairs,
9
fecal
tracer
ratios
fell
within
the
boundaries
for
soil
and
dust
(
Table
4­
19).
For
these
9
tracer
soils,
an
interpolation
was
performed
to
estimate
the
relative
contribution
of
soil
and
dust
to
the
residual
fecal
tracer
ratio.
The
other
10
fecal
tracer
ratios
that
fell
outside
the
soil
and
dust
boundaries
were
concluded
to
be
100
percent
of
the
fecal
tracer
ratios
from
soil
origin
(
Calabrese
and
Stanek,
1992).
Also,
the
9
residual
fecal
samples
within
the
boundaries
revealed
that
a
high
percentage
(
71­
99
percent)
of
the
residual
fecal
tracers
were
estimated
to
be
of
soil
origin.
Therefore,
Calabrese
and
Stanek
(
1992)
concluded
that
the
predominant
proportion
of
the
fecal
tracers
was
from
outdoor
soil
and
not
from
indoor
dust
origin.

Table
4­
19.
Ratios
of
Soil,
Dust,
and
Residual
Fecal
Samples
in
the
Soil
Pica
Child
Tracer
Ratio
Pairs
Soil
Fecal
Dust
Origin
as
Predicted
by
Specific
Tracer
Ratios
Estimated
%
of
Residual
Fecal
Tracers
of
Soil
1.
Mn/
Ti
208.368
215.241
260.126
87
2.
Ba/
Ti
187.448
206.191
115.837
100
3.
Si/
Ti
148.117
136.662
7.490
92
4.
V/
Ti
14.603
10.261
17.887
100
5.
Ai/
Ti
18.410
21.087
13.326
100
6.
Y/
Ti
8.577
9.621
5.669
100
7.
Mn/
Y
24.293
22.373
45.882
100
8.
Ba/
Y
21.854
21.432
20.432
71
9.
Si/
Y
17.268
14.205
1.321
81
10.
V/
Y
1.702
1.067
3.155
100
11.
Al/
Y
2.146
2.192
2.351
88
12.
Mn/
Al
11.318
10.207
19.520
100
13.
Ba/
Al
10.182
9.778
8.692
73
14.
Si/
Al
8.045
6.481
0.562
81
15.
V/
Al
0.793
0.487
1.342
100
16.
Si/
V
10.143
13.318
0.419
100
17.
Mn/
Si
1.407
1.575
34.732
99
18.
Ba/
Si
1.266
1.509
15.466
83
19.
Mn/
Ba
1.112
1.044
2.246
100
Source:
Calabrese
and
Stanek,
1992.
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Page
Exposure
Factors
Handbook
4­
20
August
1997
In
conducting
a
risk
assessment
for
TCDD,
U.
S.
EPA
summer,
exposure
during
the
winter
months
when
the
(
1984)
used
5
g/
day
to
represent
the
soil
intake
rate
for
pica
ground
is
frozen
or
snow
covered
should
not
be
considered
children.
The
Centers
for
Disease
Control
(
CDC)
also
as
zero.
Exposure
during
these
months,
although
lower
investigated
the
potential
for
exposure
to
TCDD
through
the
than
in
the
summer
months,
would
not
be
zero
because
soil
ingestion
route.
CDC
used
a
value
of
10
g/
day
to
some
portion
of
the
house
dust
comes
from
outdoor
soil.
represent
the
amount
of
soil
that
a
child
with
deliberate
soil
ingestion
behavior
might
ingest
(
Kimbrough
et
al.,
1984).
These
values
are
consistent
with
those
observed
by
Calabrese
et
al.
(
1991).

4.7.
RECOMMENDATIONS
Results
obtained
using
titanium
as
a
tracer
in
the
Binder
et
The
key
studies
described
in
this
section
were
used
to
recommend
values
for
soil
intake
among
children.
The
key
and
relevant
studies
used
different
survey
designs
and
study
populations.
These
studies
are
summarized
in
Table
4­
20.
For
example,
some
of
the
studies
considered
food
and
nonfood
sources
of
trace
elements,
while
others
did
not.
In
other
studies,
soil
ingestion
estimates
were
adjusted
to
account
for
the
contribution
of
house
dust
to
this
estimate.
Despite
these
differences,
the
mean
and
upper­
percentile
estimates
reported
for
these
studies
are
relatively
consistent.
The
confidence
rating
for
soil
intake
recommendations
is
presented
in
Table
4­
21.
It
is
important,
however,
to
understand
the
various
uncertainties
associated
with
these
values.
First,
individuals
were
not
studied
for
sufficient
periods
of
time
to
get
a
good
estimate
of
the
usual
intake.
Therefore,
the
values
presented
in
this
section
may
not
be
representative
of
long
term
exposures.
Second,
the
experimental
error
in
measuring
soil
ingestion
values
for
individual
children
is
also
a
source
of
uncertainty.
For
example,
incomplete
sample
collection
of
both
input
(
i.
e.,
food
and
nonfood
sources)
and
output
(
i.
e.,
urine
and
feces)
is
a
limitation
for
some
of
the
studies
conducted.
In
addition,
an
individual's
soil
ingestion
value
may
be
artificially
high
or
low
depending
on
the
extent
to
which
a
mismatch
between
input
and
output
occurs
due
to
individual
variation
in
the
gastrointestinal
transit
time.
Third,
the
degree
to
which
the
tracer
elements
used
in
these
studies
are
absorbed
in
the
human
body
is
uncertain.
Accuracy
of
the
soil
ingestion
estimates
depends
on
how
good
this
assumption
is.
Fourth,
there
is
uncertainty
with
regard
to
the
homogeneity
of
soil
samples
and
the
accuracy
of
parent's
knowledge
about
their
child's
playing
areas.
Fifth,
all
the
soil
ingestion
studies
presented
in
this
section
with
the
exception
of
Calabrese
et
al.
(
1989)
were
conducted
during
the
summer
when
soil
contact
is
more
likely.
Although
the
recommendations
presented
below
are
derived
from
studies
which
were
mostly
conducted
in
the
Soil
Ingestion
Among
Children
­
Estimates
of
the
amount
of
soil
ingested
by
children
are
summarized
in
Table
4­
22.
The
mean
values
ranged
from
39
mg/
day
to
271
mg/
day
with
an
average
of
146
mg/
day
for
soil
ingestion
and
191
mg/
day
for
soil
and
dust
ingestion.

al.
(
1986)
and
Clausing
et
al.
(
1987)
studies
were
not
considered
in
the
derivation
of
this
recommendation
because
these
studies
did
not
take
into
consideration
other
sources
of
the
element
in
the
diet
which
for
titanium
seems
to
be
significant.
Therefore,
these
values
may
overestimate
the
soil
intake.
One
can
note
that
this
group
of
mean
values
is
consistent
with
the
200
mg/
day
value
that
EPA
programs
have
used
as
a
conservative
mean
estimate.
Taking
into
consideration
that
the
highest
values
were
seen
with
titanium,
which
may
exhibit
greater
variability
than
the
other
tracers,
and
the
fact
that
the
Calabrese
et
al.
(
1989)
study
included
a
pica
child,
100
mg/
day
is
the
best
estimate
of
the
mean
for
children
under
6
years
of
age.
However,
since
the
children
were
studied
for
short
periods
of
time
and
the
prevalence
of
pica
behavior
is
not
known,
excluding
the
pica
child
from
the
calculations
may
underestimate
soil
intake
rates.
It
is
plausible
that
many
children
may
exhibit
some
pica
behavior
if
studied
for
longer
periods
of
time.
Over
the
period
of
study,
upper
percentile
values
ranged
from
106
mg/
day
to
1,432
mg/
day
with
an
average
of
383
mg/
day
for
soil
ingestion
and
587
mg/
day
for
soil
and
dust
ingestion.
Rounding
to
one
significant
figure,
the
recommended
upper
percentile
soil
ingestion
rate
for
children
is
400
mg/
day.
However,
since
the
period
of
study
was
short,
these
values
are
not
estimates
of
usual
intake.
The
recommended
values
for
soil
ingestion
among
children
and
adults
are
summarized
in
Table
4­
23.
Data
on
soil
ingestion
rates
for
children
who
deliberately
ingest
soil
are
also
limited.
An
ingestion
rate
of
10
g/
day
is
a
reasonable
value
for
use
in
acute
exposure
assessments,
based
on
the
available
information.
It
should
be
noted,
however,
that
this
value
is
based
on
only
one
pica
child
observed
in
the
Calabrese
et
al.
(
1989)
study.
Soil
Ingestion
Among
Adults
­
Only
three
studies
have
attempted
to
estimate
adult
soil
ingestion.
Hawley
(
1985)
suggested
a
value
of
480
mg/
day
for
adults
engaged
in
outdoor
activities
and
a
range
of
0.56
to
110
mg/
day
of
house
dust
during
indoor
activities.
These
estimates
were
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Exposure
Factors
Handbook
Page
August
1997
4­
21
derived
from
assumptions
about
soil/
dust
levels
on
hands
representativeness
of
the
general
population
is
unknown
and
mouthing
behavior;
no
supporting
measurements
were
due
to
the
small
study
size
(
n=
6);
and
(
2)
representativeness
made.
Making
further
assumptions
about
frequencies
of
of
long­
term
behavior
is
unknown
since
the
study
was
indoor
and
outdoor
activities,
Hawley
(
1985)
derived
an
conducted
over
only
2
weeks.
In
the
past,
many
EPA
risk
annual
average
of
60.5
mg/
day.
Given
the
lack
of
assessments
have
assumed
an
adult
soil
ingestion
rate
of
50
supporting
measurements,
these
estimates
must
be
mg/
day
for
industrial
settings
and
100
mg/
day
for
residential
considered
conjectural.
Krablin
(
1989)
used
arsenic
levels
and
agricultural
scenarios.
These
values
are
within
the
in
urine
(
n=
26)
combined
with
information
on
mouthing
range
of
estimates
from
the
studies
discussed
above.
Thus,
behavior
and
activity
patterns
to
suggest
an
estimate
for
50
mg/
day
still
represents
a
reasonable
central
estimate
of
adult
soil
ingestion
of
10
mg/
day.
The
study
protocols
are
adult
soil
ingestion
and
is
the
recommended
value
in
this
not
well
described
and
has
not
been
formally
published.
handbook.
This
recommendation
is
clearly
highly
Finally,
Calabrese
et
al.
(
1990)
conducted
a
tracer
study
on
uncertain;
however,
and
as
indicated
in
Table
4­
21,
is
given
6
adults
and
found
a
range
of
30
to
100
mg/
day.
This
study
a
low
confidence
rating.
Considering
the
uncertainties
in
is
probably
the
most
reliable
of
the
three,
but
still
has
two
the
central
estimate,
a
recommendation
for
an
upper
significant
uncertainties:
(
1)
percentile
value
would
be
inappropriate.
Table
4­
23
summarizes
soil
ingestion
recommendations
for
adults.
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Page
Exposure
Factors
Handbook
4­
22
August
1997
Table
4­
20.
Soil
Intake
Studies
Study
Study
Type
Number
of
Observations
Age
Population
Studied
Comments
CHILDREN
KEY
STUDIES:

Binder
et
al.,
1986
Tracer
study
using
aluminum,
silicon,
and
titanium
59
children
1­
3
years
Children
living
near
lead
smelter
in
Montana
Did
not
account
for
tracer
in
food
and
medicine;
used
assumed
fecal
weight
of
15
g/
day;
short­
term
study
conducted
over
3
days
Calabrese
et
al.,
1989
Tracer
­
mass
balance
study
using
aluminum,

barium,
manganese,
silicon,
titanium,

vanadium,
ytrium,
and
zirconium
64
Children
1­
4
years
Children
from
greater
Amherst
area
of
Massachusetts;

highlyeducated
parents
Corrected
for
tracer
in
food
and
medicine;
study
conducted
over
twoweek
period;
used
adults
to
validate
methods;
one
pica
child
in
study
group.

Clausing
et
al.,
1987
Tracer
study
using
aluminum,
acid
insoluble
residue,
and
titanium
18
nursery
school
children;
6
hospitalized
children
2­
4
years
Dutch
children
Did
not
account
for
tracer
in
food
and
medicines;
used
tracer­
based
intake
rates
for
hospitalized
children
as
background
values;
short­
term
study
conducted
over
5
days
Davis
et
al.,
1990
Tracer
­
mass
balance
study
using
aluminum
silicon
and
titanium
104
children
2­
7
years
Children
from
3­
city
area
in
Washington
State
Corrected
for
tracer
in
food
and
medicine;
short­
term
study
conducted
over
seven­
day
period;
collected
information
on
demographic
characteristics
affecting
soil
intake.

Stanek
and
Calabrese,
1995a
Adjusted
soil
intake
estimates
64
children
1­
4
years
Same
children
as
in
Calabrese
et
al.,
1989
Based
on
data
from
Calabrese
et
al.,

1989
Stanek
and
Calabrese,
1995b
Recalculated
intake
rates
based
on
three
previous
mass­
balance
studies
using
the
Best
Tracer
Method
164
children
6
adults
1­
7
years
25­
41
years
Children
from
three
massbalance
studies
Based
on
studies
of
Calabrese
et
al.,

1989;
Davis
et
al.,
1990;
and
Calabrese
et
al.,
1990.

Van
Wïjnen
et
al.,
1990
Tracer
study
using
aluminum,
acid
insoluble
residue,
and
titanium
292
daycare
children;
78
campers;
15
hospitalized
children
1­
5
years
Dutch
children
Did
not
account
for
tracer
in
food
and
medicines;
used
tracer­
based
intake
for
hospitalized
children
as
background
values;
evaluated
population
(
campers)

with
greater
access
to
soil;
evaluated
differences
in
soil
intake
due
to
weather
conditions.

CHILDREN
RELEVANT
STUDIES:

AIHC,
1994
Reanalysis
of
data
from
Calabrese
et
al.,
1990
6
adults
21­
41
years
Health
adults
Used
data
from
Calabrese
et
al.
(
1990)

study
to
derive
soil
ingestion
rates
using
zirconium
as
a
tracer;
recent
studies
indicate
that
zirconium
is
not
a
good
tracer
Calabrese
and
Stanek,
1995
Evaluated
errors
in
soil
ingestion
estimates
64
children
1­
4
years
Study
population
of
Calabrese
et
al.,
1989
Based
on
Calabrese
et
al.,
1989
data.
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Exposure
Factors
Handbook
Page
August
1997
4­
23
Table
4­
20.
Soil
Intake
Studies
(
continued)

Study
Study
Type
Number
of
Observations
Age
Population
Studied
Comments
CHILDREN
RELEVANT
STUDIES
(
continued):

Day
et
al.,
1977
Measured
dirt
on
sticky
sweets
and
assumed
number
of
sweets
eaten
per
day
Not
specified
Not
specified
Not
specified
Based
on
observations
and
crude
measurements.

Duggan
and
Williams,
1977
Measured
soil
on
fingers
and
observed
mouthing
behavior
Not
specified
Not
specified
Areas
around
London
Based
on
observations
and
crude
measurements.

Hawley,
1985
Assumed
soil
intake
rates
based
on
nature
and
duration
of
activities
Not
specified
Young
children,

older
children,

adults
Not
specified
No
data
on
soil
intake
collected;
estimates
based
on
assumptions
regarding
data
from
previous
studies.

Lepow
et
al.,
1974;
1975
Measured
soil
on
hands
and
observed
mouthing
behavior
22
children
2­
6
years
Urban
children
from
Connecticut
Based
on
observations
over
3­
6
hours
of
play
and
crude
measurement
techniques.

Sedman
and
Mahmood,
1994
Adjusted
data
from
earlier
tracer­
mass
balance
studies
to
generate
mean
soil
intake
rates
for
a
2­
year
old
child
64
children
from
Calabrese
et
al.,
1989
study
and
104
children
from
Davis
et
al.,
1990
study
Adjusted
to
2­
year
old
child
Same
children
as
in
Calabrese
et
al.,
1989
and
Davis
et
al.,
1990
study
Based
on
data
from
Calabrese
et
al.,
1989
and
Davis
et
al.,
1990.

Sheppard,
1995
Provides
estimates
based
on
the
current
literature
on
soil
ingestion
from
tracer
methods
and
recommends
values
for
use
in
assessments
Not
specified
1
year­
adults
(
age
not
specified)
Various
Presents
mean
estimates
for
children
and
adults;
provides
ingestion
estimates
for
indoor
and
outdoor
activities
based
on
Hawley,
1985.

Thompson
and
Burmaster,
1991
Re­
evaluation
of
Binder
et
al.,
1986
data
59
children
1­
3
years
Children
living
near
lead
smelter
in
Montana
Re­
calculated
soil
intake
rates
from
Binder
et
al.,
1986
data
using
actual
fecal
weights
instead
of
assumed
weights.

ADULT
SOIL
INTAKE
STUDIES:

Hawley,
1985
Assumed
soil
intake
rates
based
on
nature
and
duration
of
activities
Not
specified
Young
children,

older
children,

adults
Not
specified
No
data
on
soil
intake
collected;
estimates
based
on
assumptions
regarding
data
from
previous
studies.

Calabrese
et
al.,
1990
Measured
excretory
output
after
ingestion
of
capsules
with
sterilized
soil
6
adults
21­
41
years
Healthy
adult
volunteers
Data
used
to
validate
the
analytical
methodology
used
in
the
children's
study
(
Calabrese,
1989).

PICA
STUDIES:

Calabrese
et
al.,
1991
Tracer
­
mass
balance
1
pica
child
3.5
years
1
pica
child
from
greater
Amherst
area
of
Massachusetts
Child
was
observed
as
part
of
the
Calabrese
et
al.,
1989
study.

Calabrese
and
Stanek,
1992
Reanalysis
of
data
from
Calabrese
et
al.,
1991
1
pica
child
3.5
years
1
pica
child
from
greater
Amherst
area
of
Massachusetts
Distinguished
between
outdoor
soil
ingestion
and
indoor
dust
ingestion
in
a
soil
pica
child.
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Page
Exposure
Factors
Handbook
4­
24
August
1997
Table
4­
21.
Confidence
in
Soil
Intake
Recommendation
Considerations
Rationale
Rating
Study
Elements
C
Level
of
peer
review
All
key
studies
are
from
peer
review
literature.
High
C
Accessibility
Papers
are
widely
available
from
peer
review
journals.
High
C
Reproducibility
Methodology
used
was
presented,
but
results
are
difficult
to
Medium
reproduce.

C
Focus
on
factor
of
interest
The
focus
of
the
studies
was
on
estimating
soil
intake
rate
by
High
(
for
children)
children;
studies
did
not
focus
on
intake
rate
by
adults.
Low
(
for
adults)

C
Data
pertinent
to
U.
S.
Two
of
the
key
studies
focused
on
Dutch
children;
other
studies
used
Medium
children
from
specific
areas
of
the
U.
S.

C
Primary
data
All
the
studies
were
based
on
primary
data.
High
C
Currency
Studies
were
conducted
after
1980.
High
C
Adequacy
of
data
collection
period
Children
were
not
studied
long
enough
to
fully
characterize
day
to
Medium
day
variability.

C
Validity
of
approach
The
basic
approach
is
the
only
practical
way
to
study
soil
intake,
but
Medium
refinements
are
needed
in
tracer
selection
and
matching
input
with
outputs.
The
more
recent
studies
corrected
the
data
for
sources
of
the
tracers
in
food.
There
are,
however,
some
concerns
about
absorption
of
the
tracers
into
the
body
and
lag
time
between
input
and
output.

C
Study
size
The
sample
sizes
used
in
the
key
studies
were
adequate
for
children.
Medium
(
for
children)
However,
only
few
adults
have
been
studied.
Low
(
for
adults)

C
Representativeness
of
the
population
The
study
population
may
not
be
representative
of
the
U.
S.
in
terms
Low
of
race,
socio­
economics,
and
geographical
location;
Studies
focused
on
specific
areas;
two
of
the
studies
used
Dutch
children.

C
Characterization
of
variability
Day­
to­
day
variability
was
not
very
well
characterized.
Low
C
Lack
of
bias
in
study
design
(
high
The
selection
of
the
population
studied
may
introduce
some
bias
in
Medium
rating
is
desirable)
the
results
(
i.
e.,
children
near
a
smelter
site,
volunteers
in
nursery
school,
Dutch
children).

C
Measurement
error
Errors
may
result
due
to
problems
with
absorption
of
the
tracers
in
Medium
the
body
and
mismatching
inputs
and
outputs.

Other
Elements
C
Number
of
studies
There
are
7
key
studies.
High
C
Agreement
between
researchers
Despite
the
variability,
there
is
general
agreement
among
researchers
Medium
on
central
estimates
of
daily
intake
for
children.

Overall
Rating
Studies
were
well
designed;
results
were
fairly
consistent;
sample
Medium
(
for
children
­
size
was
adequate
for
children
and
very
small
for
adults;
accuracy
of
long­
term
central
methodology
is
uncertain;
variability
cannot
be
characterized
due
to
estimate)
limitations
in
data
collection
period.
Insufficient
data
to
recommend
Low
(
for
adults)
upper
percentile
estimates
for
both
children
and
adults.
Low
(
for
upper
percentile)
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Exposure
Factors
Handbook
Page
August
1997
4­
25
Table
4­
22.
Summary
of
Estimates
of
Soil
Ingestion
by
Children
Mean
(
mg/
day)
Upper
Percentile
(
mg/
day)
References
Al
Si
AIR
Ti
Y
Al
Si
Ti
Y
a
181
184
584
578
Binder
et
al.
1986
230
129
Clausing
et
al.
1987
39
82
245.5
Davis
et
al.
1990
64.5
160
268.4
b
153
154
218
85
223
276
1,432
106
Calabrese
et
al.
1989
154
483
170
65
478
653
1,059
159
b
122
139
­­
271
165
254
224
279
144
Stanek
and
Calabrese,
1995a
133
217
Stanek
and
Calabrese,
1995b
c
69­
120
Van
Wïjnen
et
al.
1990
d
b
b
b
b
b
b
c
b
b
b
Average
=
146
mg/
day
soil
383
mg/
day
soil
191
mg/
day
soil
and
dust
587
mg/
day
soil
and
dust
combined
combined
AIR
=
Acid
Insoluble
Residue
a
Soil
and
dust
combined
b
BTM
c
LTM;
corrected
value
d
Table
4­
23.
Summary
of
Recommended
Values
for
Soil
Ingestion
Population
Mean
Upper
Percentile
Children
100
mg/
day
400
mg/
day
Adults
50
mg/
day
­­
Pica
child
10
g/
day
­­­
a
c
b
200
mg/
day
may
be
used
as
a
conservative
estimate
of
the
mean
(
see
text).
a
Study
period
was
short;
therefore,
these
values
are
not
estimates
of
usual
intake.
b
To
be
used
in
acute
exposure
assessments.
Based
on
only
one
pica
child
(
Calabrese
et
al.,
1989).
c
4.8.
REFERENCES
FOR
CHAPTER
4
American
Industrial
Health
Council
(
AIHC).
(
1994)
consideration
for
environmental
health
risk
Exposure
factors
sourcebook.
AIHC,
Washington,
assessment.
In
press
(
Comments
in
Toxicology).
DC.
Calabrese,
E.
J.;
Pastides,
H.;
Barnes,
R.;
Edwards,
C.;
Binder,
S.;
Sokal,
D.;
Maughan,
D.
(
1986)
Estimating
Kostecki,
P.
T.;
et
al.
(
1989)
How
much
soil
do
soil
ingestion:
the
use
of
tracer
elements
in
young
children
ingest:
an
epidemiologic
study.
In:
estimating
the
amount
of
soil
ingested
by
young
Petroleum
Contaminated
Soils,
Lewis
Publishers,
children.
Arch.
Environ.
Health.
41(
6):
341­
345.
Chelsea,
MI.
pp.
363­
397.
Behrman,
L.
E.;
Vaughan,
V.
C.,
III.
(
1983)
Textbook
of
Calabrese,
E.
J.;
Stanek,
E.
J.;
Gilbert,
C.
E.;
Barnes,
R.
M.
Pediatrics.
Philadelphia,
PA:
W.
B.
Saunders
(
1990)
Preliminary
adult
soil
ingestion
estimates;
Company.
results
of
a
pilot
study.
Regul.
Toxicol.
Pharmacol.
Bruhn,
C.
M.;
Pangborn,
R.
M.
(
1971)
Reported
12:
88­
95.
incidence
of
pica
among
migrant
families.
J.
of
the
Calabrese,
E.
J.;
Stanek,
E.
J.;
Gilbert,
C.
E.
(
1991)
Am.
Diet.
Assoc.
58:
417­
420.
Evidence
of
soil­
pica
behavior
and
quantification
of
Calabrese,
E.
J.;
Kostecki,
P.
T.;
Gilbert,
C.
E.
(
1987)
How
much
soil
do
children
eat?
An
emerging
soil
ingested.
Hum.
Exp.
Toxicol.
10:
245­
249.
Volume
I
­
General
Factors
Chapter
4
­
Soil
Ingestion
and
Pica
Page
Exposure
Factors
Handbook
4­
26
August
1997
Calabrese,
E.
J.;
Stanek,
E.
J.
(
1992)
Distinguishing
Lepow,
M.
L.;
Bruckman,
L.;
Robino,
R.
A.;
Markowitz,
outdoor
soil
ingestion
from
indoor
dust
ingestion
in
a
S.;
Gillette,
M.;
et
al.
(
1974)
Role
of
airborne
lead
soil
pica
child.
Regul.
Toxicol.
Pharmacol.
15:
83­
in
increased
body
burden
of
lead
in
Hartford
85.
children.
Environ.
Health
Perspect.
6:
99­
101.
Calabrese,
E.
J.;
Stanek,
E.
J.
(
1995)
Resolving
intertracer
Lepow,
M.
L.;
Buckman,
L.;
Gillette,
M.;
Markowitz,
S.;
inconsistencies
in
soil
ingestion
estimation.
Environ.
Robino,
R.;
et
al.
(
1975)
Investigations
into
sources
Health
Perspect.
103(
5):
454­
456.
of
lead
in
the
environment
of
urban
children.
Clausing,
P.;
Brunekreef,
B.;
Van
Wijnen,
J.
H.
(
1987)
Environ.
Res.
10:
415­
426.
A
method
for
estimating
soil
ingestion
by
children.
Lourie,
R.
S.;
Layman,
E.
M.;
Millican,
F.
K.
(
1963)
Why
Int.
Arch.
Occup.
Environ.
Health
(
W.
Germany)
children
eat
things
that
are
not
food.
Children
59(
1):
73­
82.
10:
143­
146.
Danford,
D.
C.
(
1982)
Pica
and
nutrition.
Annual
Roels,
H.;
Buchet,
J.
P.;
Lauwerys,
R.
R.
(
1980)
Exposure
Review
of
Nutrition.
2:
303­
322.
to
lead
by
the
oral
and
pulminary
route
of
children
Davis,
S.;
Waller,
P.;
Buschbon,
R.;
Ballou,
J.;
White,
P.
living
in
the
vicinity
of
a
primary
lead
smelter.
(
1990)
Quantitative
estimates
of
soil
ingestion
in
Environ.
Res.
22:
81­
94.
normal
children
between
the
ages
of
2
and
7
years:
Sayetta,
R.
B.
(
1986)
Pica:
An
overview.
American
population
based
estimates
using
aluminum,
silicon,
Family
Physician
33(
5):
181­
185.
and
titanium
as
soil
tracer
elements.
Arch.
Environ.
Sedman,
R.;
Mahmood,
R.
S.
(
1994)
Soil
ingestion
by
Hlth.
45:
112­
122.
children
and
adults
reconsidered
using
the
results
of
Day,
J.
P.;
Hart,
M.;
Robinson,
M.
S.
(
1975)
Lead
in
recent
tracer
studies.
Air
and
Waste,
44:
141­
144.
urban
street
dust.
Nature
253:
343­
345.
Sheppard,
S.
C.
(
1995)
Parameter
values
to
model
the
Duggan,
M.
J.;
Williams,
S.
(
1977)
Lead
in
dust
in
city
soil
ingestion
pathway.
Environmental
Monitoring
streets.
Sci.
Total
Environ.
7:
91­
97.
and
Assessment
34:
27­
44.
Feldman,
M.
D.
(
1986)
Pica:
current
perspectives.
Stanek,
E.
J.;
Calabrese,
E.
J.
(
1995a)
Daily
estimates
of
Psychosomatics
(
USA)
27(
7):
519­
523.
soil
ingestion
in
children.
Environ.
Health
Perspect.
Forfar,
J.
O.;
Arneil,
G.
C.,
eds.
(
1984)
Textbook
of
103(
3):
276­
285.
Paediatrics.
3rd
ed.
London:
Churchill
Livingstone.
Stanek,
E.
J.;
Calabrese,
E.
J.
(
1995b)
Soil
ingestion
Hawley,
J.
K.
(
1985)
Assessment
of
health
risk
from
estimates
for
use
in
site
evaluations
based
on
the
best
exposure
to
contaminated
soil.
Risk
Anal.
5:
289­
tracer
method.
Human
and
Ecological
Risk
302.
Assessment.
1:
133­
156.
Illingworth,
R.
S.
(
1983)
The
normal
child.
New
York:
Thompson,
K.
M.;
Burmaster,
D.
E.
(
1991)
Parametric
Churchill
Livingstone.
distributions
for
soil
ingestion
by
children.
Risk
Kaplan,
H.
I.;
Sadock,
B.
J.
(
1985)
Comprehensive
Analysis.
11:
339­
342.
textbook
of
psychiatry/
IV.
Baltimore,
MD:
Williams
U.
S.
EPA.
(
1984)
Risk
analysis
of
TCDD
contaminated
and
Wilkins.
soil.
Washington,
DC:
U.
S.
Environmental
Kimbrough,
R.;
Falk,
H.;
Stemr,
P.;
Fries,
G.
(
1984)
Protection
Agency,
Office
of
Health
and
Health
implications
of
2,3,7,8­
tetrachlorodibenzo­
Environmental
Assessment.
EPA
600/
8­
84­
031.
p­
dioxin
(
TCDD)
contamination
of
residential
soil.
Van
Wijnen,
J.
H.;
Clausing,
P.;
Brunekreff,
B.
(
1990)
J.
Toxicol.
Environ.
Health
14:
47­
93.
Estimated
soil
ingestion
by
children.
Environ.
Res.
Krablin,
R.
(
1989)
[
Letter
to
Jonathan
Z.
Cannon
51:
147­
162.
concerning
soil
ingestion
rates.]
Denver,
CO:
Arco
Vermeer,
D.
E.;
Frate,
D.
A.
(
1979)
Geophagia
in
rural
Coal
Co.;
October
13,
1989.
Mississippi:
environmental
and
cultural
contexts
and
nutritional
implications.
Am.
J.
Clin.
Nutr.
32:
2129­
2135.
