USER'S GUIDE FOR lEUBKwiN
United States Office of Superfund Remediation EPA 9285.7-42
Environmental Protection and Technology Innovation May 2007
Agency 540-K-01-005
User's Guide for the Integrated
Exposure Uptake Biokinetic
Model for Lead in Children
(IEUBK) Windows®
Prepared for
The Technical Review Workgroup for Metals and Asbestos (TRW)
Prepared by
Syracuse Research Corporation
6225 Running Ridge Road
North Syracuse, NY 13212
-------
USER'S GUIDE FOR lEUBKwiN
Disclaimer
This document has been reviewed in accordance with U.S. EPA policy and is approved
for publication. Mention of trade names or commercial products does not constitute
endorsement or recommendation.
-------
USER'S GUIDE FOR lEUBKwiN
U.S. Environmental Protection Agency
Technical Review Workgroup for Metals and Asbestos
Members
The members of the TRW Lead Committee are technical staff from EPA Regions, Office of Solid Waste
and Emergency Response (OSWER) Headquarters, and Office of Research and Development National
Center for Environmental Assessment (ORD/NCEA). Lead Committee members generally have an active
interest and recognized scientific expertise in metals or asbestos risk assessment. For more information
see: http://www.epa.sov/superfund/lead/trw.htm.
-------
USER'S GUIDE FOR lEUBKwiN
This page intentionally left blank.
-------
USER'S GUIDE FOR lEUBKwiN v
Table of Contents
Disclaimer ii
Members iii
Table of Contents v
List of Tables vii
List of Figures ix
Acronyms and Abbreviations xi
1.0 Introduction 1
1.1 Purpose of the User's Guide 1
1.1.1 Overview of the lEUBKwin Model 1
1.2 Loading and Starting the Model 5
2.0 Inputs 7
2.1 Input Windows 9
2.2 Detailed Descriptions of Input Options 10
2.2.1 Air Data 14
2.2.2 Dietary Data 14
2.2.3 Drinking Water Data 15
2.2.4 Soil/Dust Data 15
2.2.4.1 Lead in Soil 15
2.2.4.2 Lead in Dust 16
2.2.5 Maternal Data 17
2.2.6 Alternate Source Data 17
2.2.7 GI Values/Bioavailability 18
2.2.8 Geometric Standard Deviation (GSD) 18
2.3 Save and Load Input Parameters 19
3.0 Model Analysis 20
3.1 Options for model analyses 20
3.2 Overview of Model Output 20
3.3 Selecting Output Alternatives 21
3.3.1 What Graphical Options are Included in lEUBKwin? 21
3.4 Computation Options 22
3.4.1 Run a Single Simulation 22
3.4.1.1 Single-Run Options 23
3.4.2 Calculating a Blood Lead Concentration Associated with Media-Specific
Concentration 23
3.4.3 Run Multiple Simulations 24
3.4.3.1 Multiple-Run Options 24
3.4.4 Run Batch Mode Simulation 25
4.0 Interpretation of Results 27
4.1 Understanding the Output 27
-------
USER'S GUIDE FOR lEUBKwiN
5.0 Model Application and Documentation 29
5.1 Model Application 29
5.2 Model Documentation (minimum requirements) 30
5.3 User Responsibilities 30
6.0 Example Scenarios 31
6.1 Simple Case Single Runs 31
6.2 "Sum of Individual Risks" Approach 31
6.3 Risk Assessment Using Grouped Data for a Neighborhood 33
6.4 Risk Assessment Using Neighborhood or Neighborhood-Scale Input 35
7.0 References 38
APPENDICES
A: Using Spreadsheets with lEUBKwin 40
A.I Importing lEUBKwin data to spreadsheet packages 40
A.2 Creating a batch mode data input file using a spreadsheet 43
B: Probability Distribution Function 44
C: Short Sheets Currently Available 46
-------
USER'S GUIDE FOR lEUBKwiN vii
List of Tables
2-1. Default Values for lEUBKwin Model Parameters 11
6-1. Simple Case Runs 31
6-2. Interpreted Results for Figure 6-1 33
6-3. Neighborhood Risk Estimation with Grouped Data 36
6-4. Neighborhood Risk Estimation with Coarsely Grouped Data 37
-------
USER'S GUIDE FOR lEUBKwiN viii
This page intentionally left blank.
-------
USER'S GUIDE FOR lEUBKwiN ix
List of Figures
1-1. Biological Structure of the lEUBKwin Model 4
2-1. Categories of Application of lEUBKwin 8
3-1. The Impact of GSD on Plausible Geometric Mean PbB Distributions 22
6-1. Probability Density of Blood Lead in Houses 1 to 4 32
A-l. Export lEUBKwin Data Output to Microsoft Excel 40
A-2. Export lEUBKwin Data Output to Lotus 1-2-3 41
A-3. Creating a Batch Data file with Microsoft Excel 42
-------
USER'S GUIDE FOR lEUBKwiN
This page intentionally left blank.
-------
USER'S GUIDE FOR lEUBKwiN
Acronyms and Abbreviations
CERCLA Comprehensive Emergency Response Compensation Liability Act
dL deciliter
ggram
GI gastrointestinal
GM geometric mean
GSD Geometric Standard Deviation
lEUBKwin Integrated Exposure Uptake Biokinetic Model for Lead in Children,
Windows version
FDA Food and Drug Administration
L liter
LOC level of concern
MSA Multiple Source Analysis
MSD Mass fraction of soil to dust
m3 cubic meters
OSWER Office of Solid Waste and Emergency Response
Pb lead
PbB blood lead
PbD dust lead
PbS soil lead
PDF Probability Density Function
PRG Preli minary Remediation Goal
RCRA Resource Conservation Recovery Act
TRW Technical Review Workgroup for Metals and Asbestos
(ig microgram
-------
USER'S GUIDE FOR lEUBKwiN xii
This page intentionally left blank.
-------
USER'S GUIDE FOR lEUBKwiN xiii
This page intentionally left blank.
-------
USER'S GUIDE FOR lEUBKwiN
1.0 Introduction
The Integrated Exposure Uptake Biokinetic Model for Lead (Pb) in Children, Windows (lEUBKwin)
is a stand-alone personal computer (PC)-compatible software package. The model utilizes four
interrelated modules (exposure, uptake, biokinetic, and probability distribution) to estimate blood lead
(PbB) levels in children exposed to lead (Pb)-contaminated media. lEUBKwin allows the user to
estimate, for a hypothetical child or population of children, a plausible distribution of PbB concent-
rations centered on a geometric mean PbB concentration. The geometric mean PbB is predicted from
available information about the child's or children's exposure to lead. From this distribution, the
model estimates the risk (i.e., probability) that a child's or a population of children's PbB concentration
will exceed a certain level of concern (typically 10 |ig/dL).
lEUBKwin should be viewed as a tool for making rapid calculations and recalculations of an
extremely complex set of equations which includes many exposure, uptake, and biokinetic
parameters. It has been recommended as a risk assessment tool to support the implementation of
the July 14, 1994, Office of Solid Waste and Emergency Response (OSWER) Directive entitled
Revised Interim Soil Lead Guidance for Comprehensive Environmental Response Compensation
Liability Act (CERCLA) Sites and Resource Conservation and Recovery Act (RCRA) Corrective
Action Facilities., as well as the subsequent August 1998 OSWER Directive entitled Clarification
to the 1994 Revised Interim Soil Lead Guidance for CERCLA Sites and RCRA Corrective Action
Facilities. The development of the lEUBKwin model included Independent Validation and
Verification and a limited peer review of the software and documentation.
1.1 PURPOSE OF THE USER' s GUIDE
This User's Guide is intended to serve as a comprehensive reference for use of lEUBKwin. It
provides the basics needed for accurate application and interpretation of the model, including
how to install and run the model, model inputs, model analysis, model output, interpretation of
model results, minimum requirements for risk assessment application, and example scenarios.
Users of the model can refer to the IEUBK Guidance Manual (U.S. EPA, 1994) or the IEUBK
Structure Paper (White etal., 1998) for supplemental information concerning the parameters and
correct application of lEUBKwin.
1.1.1 Overview of the lEUBKwin Model
The focus of the integrated exposure uptake biokinetic (IEUBK) model for lead in children is the
prediction of blood lead concentrations in young children exposed to lead from several sources
and by several routes. The model is a four-step process that mathematically and statistically
links environmental lead exposure to blood lead concentrations for a population of children (0-84
months of age). Figure 1-1 provides a schematic showing the sources of exposure to
-------
USER'S GUIDE FOR lEUBKwiN
environmental lead and the absorption and processing of lead by the human body (IEUBK
Guidance Manual, U.S. EPA, 1994).
The four model components each reflect a different aspect of the overall biologic process:
Exposure Component. Exposure can be thought of as the contact of a chemical, or other agent,
with the absorption or exchange boundaries of an organism, such as the gut, lungs, and
skin. Quantitation of a child's exposure to lead (jig/day) requires estimation of the
concentration of lead in the environmental media that the child contacts (usually |ig/g,
|ig/m3, or |ig/L, multiplied by a term to describe the amount of contact the child has with
the medium (usually g/day, m3/day, or liters/day), and a term for the duration of that
contact (usually days). The results from the exposure component of the IEUBK model
are estimated intake rates for the quantities of lead inhaled or ingested from
environmental media. The media addressed by the IEUBK model include soil, house
dust, drinking water, air, and food. Paint is usually addressed in terms of its contribution
to the measured concentration of lead in soil or house dust.
Uptake Component. The uptake component models the process by which lead intake (lead that
has entered the child's body through ingestion or inhalation) is transferred to the blood
plasma. Uptake (jig/day) is the quantity of lead absorbed per unit time from portals of
entry (gut, lung) into the systemic circulation of blood. Only a fraction of the lead
entering the body through the respiratory or gastrointestinal (GI) tracts is actually
absorbed into the systemic circulation. This absorption fraction is, by convention, termed
bioavailability and provides the most convenient parameterization of the uptake process.
The IEUBK model addresses the different bioavailabilities of lead from different
environmental media and provides for a partial saturation of absorption at high levels of
lead intake.
Biokinetic Component. The biokinetic component of the IEUBK model is a mathematic
expression of the movement of absorbed lead throughout the body over time by
physiologic or biochemical processes. The biokinetic component converts the total lead
uptake rate from the uptake component into an input to the central plasma-extracellular
fluid (ECF) compartment. Transfer coefficients are used to model movement of lead
between the internal compartments and to the excretion pathways. The quantities are
combined with the total lead uptake rate to continuously recalculate the lead masses in
each of the body compartments and especially the changing concentration of lead in
blood.
Variability. An important goal of the IEUBK model is to address variability in blood lead
concentrations among exposed children. Children having contact with the same
concentrations of environmental lead can develop very different blood lead
concentrations due to differences in behavior, household characteristics, and individual
patterns of lead uptake and biokinetics. The IEUBK model uses a log-normal probability
distribution to characterize this variability. The biokinetic component output provides a
central estimate of blood lead concentration, which is used to provide the geometric
standard deviation (GSD). The recommended default value for this parameter (1.6) was
-------
USER'S GUIDE FOR lEUBKwiN
derived from empirical studies with young children where both blood and environmental
lead concentrations were measured.
-------
USER'S GUIDE FOR lEUBKwiN
-------
USER'S GUIDE FOR lEUBKwiN
1.2 LOADING AND STARTING THE MODEL
IEUBKWIN.EXE; AIR.INP; DIET.INP; GIBIO.INP;
MATERNAL.INP; MULTISRC.INP; OTHER.INP; SOIL.INP;
WATER.INP; BATCHRUN.INP; BATCHRUN.DAT; UNINST.ISU
The "IEUBKWIN.EXE" file is the main program file; the "*.INP" files contain the default values
for the Parameter Input pull-down options; the "BATCHRUN.DAT" file is the example input file
for batch mode runs; and the "UNINST.ISU" file is used by Windows to completely uninstall the
program.
Following extraction, the user must agree to the software license agreement, register the software
with the U.S. EPA, and specify a location for installation. lEUBKwin is intended to be run as a
desktop application rather than from a network installation. It is recommended that a local copy
of the application be saved onto the PC from which it will be run. As part of the model
installation, a desktop icon labeled lEUBKwin will be installed. To run the model, double-click
the lEUBKwin icon from the Programs menu. A Disclaimer will appear on the screen. Read the
Disclaimer then close the Disclaimer window by clicking on the «Close» button.
The most current information pertaining to lEUBKwin model and Pb risk assessment is available
on the Technical Review Workgroup (TRW) website. The TRW home page may be found at
http://www.epa.gov/superfund/lead.
Questions concerning the use and interpretation of lEUBKwin model may be submitted to the
IEUBK technical support center via the TRW web page.
lEUBKwin is menu driven, with on-line, context-sensitive help available in almost every menu
or window. In addition to the buttons that provide quick access to commonly-used model
features, the main pull-down menus and the associated options remain. The contents of each
these menus are listed below:
1. File Menu
• Select Run Mode
• Print
• Print Setup
• Close
• Exit
Under Select Run Mode the user may choose to run the model in Research or Site Risk
Assessment mode, depending upon individual needs. If the user chooses to run the model in the
Site Risk Assessment mode, the model will prompt the user for a user name, date, site name, and
operable unit. In addition, the model will require the user to input a comment into a comment
screen whenever the user modifies the default value of an input parameter.
-------
USER'S GUIDE FOR lEUBKwiN
2. Parameter Input Menu Options
• Air Data
• Dietary Data
• Drinking Water Data
• Soil/Dust Data
• Maternal Data
• Alternate Source Data
• GI Values/Bioavailability
• Reset All Parameters
• Load Input From File
• Save Input To File
3. Computation Menu Options
• Run the Model (Single Run)
• Multiple Runs
• Blood Pb vs. Media Pb
• Find Blood Pb Concentration
• Batch Mode Model Runs
4. Output Menu Options
• View Text Result File
• Rename Text Result File
• Rename Blood Media Graph File
• Rename Single Run Graph File
• Rename Multiple Run Graph File
5. Graph Menu Options
• Blood Pb Concentration vs. Media Pb Concentration
• Distribution Probability Percent
• Distribution Probability Density Histogram
• Distribution Probability for Multiple Runs
• Probability Density for Multiple Runs
6. Help Options
• Contact Information
• General Help
• Menu Information
• File Extensions in lEUBKwin
• Significant Figures in lEUBKwin
• About lEUBKwin Lead Model
-------
USER'S GUIDE FOR lEUBKwiN
2.0 Inputs
lEUBKwin contains more than 100 input parameters that are initially set to default values. Of
these 100 input parameters, there are 46 external input parameters that may be changed by the
user; the remaining internal model parameters are set to fixed default values. The default values
represent national averages or plausible central values that were developed based on many years
of research. However, to most effectively use the model, it is necessary to use valid site-specific
information as much as is feasible. This section provides a brief description of how a user can
change lEUBKwin parameters to better match the conditions at a specific site.
The array of valid applications for which lEUBKwin can be used is as follows:
Category A: One 1 ocation.
Al: One living unit, one child.1
A2: One living unit, more than one child.
A3: More than one living unit, more than one child, homogeneous
media concentrations.
Category B: Multiple locations, one neighborhood, homogeneous media
concentrations.
Category C: Multiple locations, one neighborhood, heterogeneous media
concentrations.
Category D: Multiple locations, more than one neighborhood, potentially
heterogeneous neighborhood ingestion/absorption parameters.
A single run of lEUBKwin is sufficient for categories A and B. Categories C and D require
classification or disaggregation of the neighborhood into distinct exposure subgroups, with the
possibility of different ingestion or absorption parameters for different neighborhoods in
Category D. Neighborhood-scale and community-scale risk estimation requires aggregating the
risk estimates for individuals or subgroups. The differences between these levels are sketched in
Figure 2-1. Category A requires calculating only a single PbB distribution. Category B requires
calculating a PbB distribution for each child, but since each child of the same age has the same
exposure scenario, a single run of the model is sufficient to characterize risk for this subgroup.
In Category C, there are different media concentrations for different individual. Risk estimates
must be calculated for each individual or homogeneous subgroup, then added up across sites and
children. Category D allows for the case where not only exposure concentrations, but other basic
exposure and bioavailability parameters differ among the exposed groups.
IEUBK model is not intended to be used as a substitute for blood lead measurement
and medical evaluation of a specific child at risk. Medical evaluation can take into account
specific information on behavior and risk factors, such as nutritional status, that are beyond the
scope of a broadly applicable environmental risk model. (See FAQ on PbB studies at
(http://www.epa.gov/superfund/lead)).
-------
USER'S GUIDE FOR lEUBKwiN
A. Single child
Single site of exposure
Probability
Blood Lead
B. Multiple children or population
Single location or locations with homogeneous
concentrations
Child 1 Child 2 Child 3
C. Multiple children or population in neighborhood
Multiple sites with exposures to different media concentrations
'
s/' ^\ '^tz^i o\ '1V"~~~"') |\\ /^ \,
s^ --.. i^— •} \V A^ \nJ /^^ ^\
Cl
IK\
fe?-
T ax/ o^T
«e^A
/vV^
W
k A
r\
\ \
/ -~~^ k.
k A
1 \
/ \
/ V k.
k A
r\
/ \
/ V k.
k
/
/ 1
/ V k.
Site 1 Site 2 Site 2 Site 3
Child 1 Child 2 Child 3 Child 4
D. Multiple children or population in heterogeneous community
or region, different neighbohood exposure scenarios
Figure 2-1. Categories of Application of lEUBKwin.
-------
USER'S GUIDE FOR lEUBKwiN
2.1 INPUT WINDOWS
There are seven parameter input windows where parameters can be modified to reflect the site-
specific conditions. lEUBKwin input options are button at the top of the screen and are as
follows:
• Air Data
• Dietary Data
• Drinking Water Data
• Soil/Dust Data
• Maternal Data
• Alternative Source Data
• GI Values/Bioavailability
The parameter values shown in the input windows are set to EPA-specified default values.
Where site-specific data are available, the user is encouraged to alter inputs for environmental
media Pb concentrations. However, the user should change parameters such as the mass fraction
of soil to dust (MSD), bioavailability, geometric standard deviation (GSD), soil ingestion rates,
and dietary data only if the user has a thorough understanding of the methodology behind those
parameters and valid site-specific monitoring data that are specifically applicable to the
parameters. Generally, the TRW considers that "valid" site-specific information will reflect data
quality indicators such as completeness, comparability, representativeness, precision, and
accuracy.
The user can reset any modified parameters to their default values by using the Reset All
Parameters command on the Parameter Input pull-down menu. If further information about an
input parameter or default value is required, consult the IEUBK Guidance Manual (U.S. EPA,
1994). The input windows were designed to allow users to tab between fields to facilitate the
data entry process. Note that a value is not saved for use in a model run until the user clicks the
«OK» button on the input screen. If «Cancel» is selected (or if the window is closed using
the "X" in the upper right corner of the window), the model retains the initial default values. The
model does not substitute default values for soil, water, or air prior to performing calculations
when any input values are missing.
Additionally, the user has the option to load a complete set of input data from a previously saved
file (see Section 2.3). The Load Input From File option is located on the Parameter Input pull-
down menu. The loaded site-specific data should provide the model with the necessary
information to predict PbB levels in children who live on or near the site in question.
When changing any of the input parameters, the user may enter values with as many as six
significant figures. In lEUBKwin, three significant figures are reported to the right of the
decimal point for all output data except blood lead concentration which is reported to one digit to
the right of the decimal point. However, it should be noted that the true precision of any
calculated output can be strongly influenced by the least precise input value. An explanation of
how significant figures are used in lEUBKwin can be found in the Help pull-down menu.
-------
10
USER'S GUIDE FOR lEUBKwiN
Each Parameter Input menu option includes context-sensitive help windows to aid the user.
These windows are available for most input parameters and are accessed by clicking on the
«Help ?» button. This feature opens a new window which shows the available guidance for
the subject. In addition, the user has online access to the complete documentation for the model
under the Help pull down menu.
Finally, lEUBKwin has been developed to warn the user when she or he has entered
unacceptable input values. Currently, lEUBKwin contains three classes of error messages that
will appear when the user enters inappropriate input values. These classes of error messages
address:
• Input values that result in greater than 100 percent. In cases where the user enters an
input value in the form of a percent, the value may not exceed 100 percent. Similarly,
when percent values are summed, the total may not exceed 100 percent.
• Negative values. Negative input values are not allowed in lEUBKwin.
• Valid model inputs. Although the user is permitted to enter any values into the
lEUBKwin model (given the limits above), in certain instances, the input is limited
based on scientific evidence (for example, the plausible range of GSD values).
2.2 DETAILED DESCRIPTIONS OF INPUT OPTIONS
The following is a discussion of the Parameter Input pull-down menu options for lEUBKwin.
Help menus and the IEUBK Guidance Manual (U.S. EPA, 1994) provide further assistance and
discussion on data entry. A detailed list of default values for lEUBKwin model parameters are
presented in Table 2-1, and the user should refer to the IEUBK Guidance Manual (U.S. EPA,
1994) for additional information.
Aside from the mathematical limits described in Section 2.1, there is no limit to the range of
values permitted in the lEUBKwin model per se. However, empirical validation of the model
did not address situations where the predicted blood lead concentration exceeds 30 |ig/dL;
therefore, such results must be interpreted with caution (Zaragoza and Hogan, 1998). In
addition, values for some input parameters are limited by physiology or scientific data as
discussed in the IEUBK Guidance Manual (U.S. EPA, 1994). Additional information on
plausible limits for certain parameters is also available from the Input Parameters section of the
FAQs of the TRW website (http://www.epa.gov/superfund/lead).
-------
11
USER'S GUIDE FOR lEUBKwiN
Table 2-1. Default Values for lEUBKwin Model Parameters
Parameter
Indoor air lead concentration (% of outdoor)
Default Value
30
Units
%
AIR (by year)
Air concentration
Age = 0-1 year (0-1 1 mo)
1-2 years (12-23 mo)
2-3 years (24-35 mo)
3-4 years (36-47 mo)
4-5 years (48-59 mo)
5-6 years (60-71 mo)
6-7 years (72-84 mo)
Time outdoors
Age = 0-1 year (0-1 1 mo)
1-2 years (12-23 mo)
2-3 years (24-35 mo)
3-7 years (36-84 mo)
Ventilation rate
Age = 0-1 year (0-1 1 mo)
1-2 years (12-23 mo)
2-3 years (24-35 mo)
3-4 years (36-47 mo)
4-5 years (48-59 mo)
5-6 years (60-71 mo)
6-7 years (72-84 mo)
Lung absorption
0.10
0.10
0.10
0.10
0.10
0.10
0.10
1
2
o
J
4
2
o
J
5
5
5
7
7
32
ug/m3
ug/m3
ug/m3
ug/m3
ug/m3
ug/m3
ug/m3
h/day
h/day
h/day
h/day
m3/day
m3/day
m3/day
m3/day
m3/day
m3/day
m3/day
%
DATA ENTRY FOR DIET (by year)
Dietary lead intake
Age = 0-1 year (0-1 1 mo)
1-2 years (12-23 mo)
2-3 years (24-35 mo)
3-4 years (36-47 mo)
4-5 years (48-59 mo)
5-6 years (60-71 mo)
6-7 years (72-84 mo)
2.26
1.96
2.13
2.04
1.95
2.05
2.22
ug Pb /day
ug Pb /day
ug Pb /day
ug Pb /day
ug Pb /day
ug Pb /day
ug Pb /day
DATA ENTRY FOR ALTERNATE DIET SOURCES (by food class)
Concentration:
home-grown fruits
home-grown vegetables
fish from fishing
game animals from hunting
Percent of food class:
home-grown fruits
home-grown vegetables
fish from fishing
game animals from hunting
0
0
0
0
0
0
0
0
ugPb/g
ugPb/g
ugPb/g
ugPb/g
%
%
%
%
-------
12
USER'S GUIDE FOR lEUBKwiN
Table 2-1. Default Values for lEUBKwin Model Parameters
Parameter
Default Value
Units
DATA ENTRY FOR DRINKING WATER
Lead concentration in drinking water
Ingestion rate
Age = 0-1 year (0-1 1 mo)
1-2 years (12-23 mo)
2-3 years (24-35 mo)
3-4 years (36-47 mo)
4-5 years (48-59 mo)
5-6 years (60-71 mo)
6-7 years (72-84 mo)
4
0.20
0.50
0.52
0.53
0.55
0.58
0.59
ug/L
L/day
L/day
L/day
L/day
L/day
L/day
L/day
DATA ENTRY FOR ALTERNATE DRINKING WATER SOURCES
Concentration
first-draw water
flushed water
fountain water
Percentage of total intake
first-draw water
flushed water
fountain water
4
1
10
50
100 minus first draw
and fountain
15
ug/L
ug/L
ug/L
%
%
DATA ENTRY FOR SOIL/DUST (constant over time)
Concentration (starting values to be modified using
appropriate site data)
soil
dust
Soil/dust ingestion weighting factor (percent soil)
200
200
45
^g/g
^g/g
%
DATA ENTRY FOR SOIL/DUST INGESTION (by year)
Soil/Dust ingestion
Age = 0-1 year (0-1 1 mo)
1-2 years (12-23 mo)
2-3 years (24-35 mo)
3-4 years (36-47 mo)
4-5 years (48-59 mo)
5-6 years (60-71 mo)
6-7 years (72-84 mo)
0.085
0.135
0.135
0.135
0.100
0.090
0.085
g/day
g/day
g/day
g/day
g/day
g/day
g/day
DATA ENTRY FOR SOIL/DUST MULTIPLE SOURCE ANALYSIS (constant over time)
Fraction of indoor dust lead attributable to soil
(MSD)
Ratio of dust lead concentration to outdoor air lead
concentration
0.70
100
unitless
ug Pb/g dust per ug Pb/m3 air
-------
13
USER'S GUIDE FOR lEUBKwiN
Table 2-1. Default Values for lEUBKwin Model Parameters
Parameter
Default Value
Units
DATA ENTRY FOR SOIL/DUST MULTIPLE SOURCE ANALYSIS WITH
ALTERNATIVE HOUSEHOLD DUST LEAD SOURCES (constant over time)
Concentration (starting values to be modified using
appropriate site data)
household dust (calculated value)
secondary occupational dust
school dust
daycare center dust
second home
interior lead-based paint
Percentage
household dust (calculated value)
secondary occupational dust
school dust
daycare center dust
second home
interior lead-based paint
150
1,200
200
200
200
1,200
100 minus all other
0
0
0
0
0
^g/g
^g/g
^g/g
Hg/g
Hg/g
Hg/g
%
%
%
%
%
%
BIO AVAIL ABILITY DATA ENTRY FOR ALL GUT ABSORPTION PATHWAYS
Total lead absorption (at low intake)
diet
drinking water
soil
dust
alternate source
Fraction of total net absorption at low intake rate that
is attributable to non saturable (passive) processes.
50
50
30
30
0
0.2 u
%
%
%
%
%
nitless
DATA ENTRY FOR ALTERNATE SOURCES (by year)
Total lead intake
Age = 0-1 year (0-1 1 mo)
1-2 years (12-23 mo)
2-3 years (24-35 mo)
3-4 years (36-47 mo)
4-5 years (48-59 mo)
5-6 years (60-71 mo)
6-7 years (72-84 mo)
0
0
0
0
0
0
0
ug/day
ug/day
ug/day
ug/day
ug/day
ug/day
ug/day
DATA ENTRY MENU FOR MATERNAL-TO-NEWBORN LEAD EXPOSURE
Mothers blood lead concentration at childbirth
1.0
ug/dL
DATA ENTRY MENU FOR PLOTTING AND RISK ESTIMATION
Geometric standard deviation (GSD) for blood lead
Blood lead level of concern, or cutoff
1.6
10
unitless
ug/dL
COMPUTATION OPTIONS
Iteration time step for numerical integration
4
h
-------
14
USER'S GUIDE FOR lEUBKwiN
2.2.1 Air Data
The Air Data option on the Parameter Input pull-down menu opens the Air Data input window.
Model default values can be replaced with site-specific data for the following parameters:
• Indoor air Pb concentration as a percentage of outdoor air Pb concentration (|ig/m3)
• Outdoor air Pb concentration (|ig/m3)
• Time spent outdoors (hr/day)
• Ventilation rate (m3/day)
• Lung absorption (0/-
The default outdoor air Pb concentration is 0.1 |ig/m3. As a default, the model calculates indoor
air Pb concentration as 30 percent of the outdoor air Pb concentration. Because the indoor air Pb
concentration is treated as a fraction of outdoor concentration, the model does not allow a value
greater than 100 percent to be entered in this field. If the user is working on a site where indoor
air Pb concentration is in fact greater than the outdoor air Pb concentration, contact the TRW
(see Section 1.2 for contact information). With the exception of the indoor air Pb concentration,
the model allows the user to vary the value of each parameter by age.
2.2.2 Dietary Data
The Dietary Data input window includes information on that portion of total Pb intake which
enters the body through the consumption of food. The default daily dietary Pb intake values for
each age apply to a typical child in the United States. These estimates are derived based upon
Food and Drug Administration (FDA) food monitoring data collected 1995-2003 (FDA, 2006).
The model also allows the user to directly alter the dietary lead intake amounts by age. If site-
specific data suggest that default dietary intake rates differ due to the consumption of local
produce, fish, or game animals, that information may be entered by answering «YES» to Use
Alternate Dietary Values? in the Dietary Data input window. The value in the first column of
the Dietary Data input window is the observed concentration of Pb in the local food item. The
second column is the amount of food of a given type obtained from local sources, as a percent of
total consumption of that food type. Alternate dietary data should not be used in combination
with changes to default age-specific lead intake data. Local fish and game animal meat are part
of the same "meat" category, and the percentage values for this food category may not be greater
than 100. Home grown fruits and home grown vegetables are part of the "fruit and vegetable"
category. Limitations on inputs for the "meat" or the "fruit and vegetable" category are
explained in the «?» button for each input parameter. Additional guidance is available from
the FAQs of the TRW website (http://www.epa.gov/superfund/lead).
-------
15
USER'S GUIDE FOR lEUBKwiN
2.2.3 Drinking Water Data
The Drinking Water Data input window is divided into two sections: water consumption rates
and environmental concentrations. The default consumption rates are age-dependent and based
on national averages. These consumption rates should be changed only when valid site-specific
monitoring data is available. When entering a water Pb concentration, the user has two options:
either use one concentration for all water sources (Option 1), or use source-dependent values
(Option 2). Option 1 is selected by answering «NO» to Use Alternate Water Values? If the
user elects Option 1, the user can either select the default value (4 |ig/L) or change the water Pb
default concentration to reflect conditions at the site. Option 2 is selected by answering
«YES» to the Use Alternate Water Values? By selecting Option 2, the user can enter site-
specific values for the following parameters:
• Percentage of total consumed as first draw (%) [default=50%]
• Concentration of lead in first draw (|ig/L) [default=4 |ig/L]
• Concentration of lead in flushed (|ig/L) [default=l |ig/L]
• Percentage of total consumed from fountains (%) [default=15%]
• Concentration of lead in fountain water (|ig/L) [default=10 |ig/L]
Note that total water consumed is the sum of first draw, flushed, and fountain water.
2.2.4 Soil/Dust Data
Because soil and dust are often the primary sources of Pb exposure at a site, the Soil/Dust Data
menu option is the most detailed and complex of the input windows. The first parameter that can
be modified by the user is the Soil/Dust Ingestion Weighting Factor. This value represents the
percentage of total soil/dust ingestion that is derived from soil and the percent that is derived
from dust. The value in the parameter input window of the model is the percent from soil. The
combined soil/dust ingestion weighting factor equals 100 percent.
• The Soil/Dust Data input window allows the user to adjust the rate at which a child
ingests soil and dust. These values should only be adjusted if site-specific, technically-
supportable data that are applicable to this parameter are available. The user should
consult the Ingestion Rate short sheet that is available on the TRW website for further
information concerning the use of site-specific soil ingestion values. (See Appendix C or
[Short Sheet: IEUBK Model Soil/Dust Ingestion Rates, EPA #540-F-00-007, OSWER
#9285.7-33, December 1999] http://www.epa.gov/superfund/lead)
2.2.4.1 Lead in Soil
When site-specific soil values are used, the TRW recommends replacing the default constant
dust value (200 |ig/g) with valid site-specific data. At a minimum, soil concentration data is
needed to obtain modeling results that are predictive of site risks. If no valid site-specific dust
data are available, it is strongly recommended that user use the Multiple Source Analysis (MSA)
-------
16
USER'S GUIDE FOR lEUBKwiN
option to estimate dust lead concentrations as a function of soil lead concentrations. This
approach presumes, however, that there are not significant indoor sources of lead dust, i.e.,
deteriorating lead-based paint.
When specifying a concentration of Pb in soil, the user has the following two options:
• Use an alternate site-specific constant value in place of the default value (200 |ig/g).
When the user inputs a constant value for soil (or dust), the value is applied to all age
ranges. The 200 |ig/g value, while serving as a plausible value for many urban
settings, is not intended to be a substitute for site data.
• Use the Variable Values option to change the concentration of soil to which children
are exposed by age. If the user chooses to enter a value instead of using the model
default or a site-specific constant value across all age ranges, the user must select the
Variable Values option and enter the new value into each age field.
2.2.4.2 Lead in Dust
By default, the IEUBK model defaults to Multiple Source Analysis because it assumes that site-
specific dust concentrations are not measured and therefore, Multiple Source Analysis should be
used to determine the appropriate dust concentration. If site-specific dust measurements are
available, this information can be entered as a Constant Value or Variable Values (by age) before
running the model by selecting the appropriate mode in the soil dust input window.
When specifying a concentration of Pb in dust, the user has the following three options:
• Use an alternate site-specific constant value in place of the default value (200 |ig/g).
• Use the Variable Values option to change the concentration of dust that children are
exposed to by age. If the user chooses to enter variable values instead of using the
model default or a site-specific constant value, the user must select the Variable
Values option and enter the new value into each age field.
• Use Multiple Source Analysis for dust concentrations. Multiple Source Analysis
allows the use of information about the contribution of soil lead (PbS), air Pb, and
other sources to derive a value for indoor dust lead (PbD). The data entry line in
Multiple Source Analysis represents the contribution of Pb in soil to the concentration
of Pb in indoor dust. If there were no other sources to contribute to indoor PbD (such
as lead-based paint), this would be the ratio of the PbD concentration to the PbS
concentration. As indicated in the IEUBK Guidance Manual (U.S. EPA, 1994), the
default value of 0.70 is appropriate to neighborhoods or residences in which soil
materials are available for contact and track-in into the house. In the absence of
indoor sources of lead of established significance, the TRW recommends that the soil
contribution to dust lead be evaluated by comparing the average or arithmetic mean
of soil lead concentrations from a representative area in the child's yard, and an
average of dust lead concentrations from representative areas frequented by children
-------
17
USER'S GUIDE FOR lEUBKwiN
inside the house. There are complicated statistical and practical matters involved in
evaluating this parameter. See the TRW short sheet on MSD, available from the
TRW website. (See Appendix C or [Short Sheet: IEUBK Model Mass Fraction of
Soil in Indoor Dust (MSD) Variable, EPA #540-F-00-008, OSWER #9285.7-34, June
1998] http://www.epa.gov/superfund/lead).
The second data entry line is the contribution to indoor dust from the deposition of airborne Pb,
over and above the PbS contribution. The current default value is an additive increment of
100 |ig/g Pb in house dust for each jig Pb/m3 air.
The final set of inputs allows the user to consider additional sources of indoor dust by entering
data into the Indoor Dust Lead Sources portion of the MSA screen. The user has the option to
enter a concentration and the fraction of dust intake to baseline indoor dust; secondary
occupational dust; dust at school, daycare or second home; and the exposure to Pb in dust from
household paint measured as a percentage of total dust ingestion and concentration.
The sum of all these alternate source percentages may not be greater than 100. The model will
automatically adjust the indoor dust percentage so that the sum of all sources is 100. When all
the alternate source data have been entered, the model will calculate the Multiple Source Average
concentration and display it in the Soil/Dust Data window.
2.2.5 Maternal Data
The Maternal Data input window, on the Parameter Input pull-down menu, allows the user to
consider the impact of Pb transferred from the mother to the fetus in utero. The Pb that is stored
in the tissues of the newborn child is calculated by entering the maternal PbB value at the time of
delivery (default =1.0 |ig/dL). The empirical validation effort for the IEUBK model did not
include data specific to the disposition of maternally supplied lead in a young child, and model
predictions in this regard should be interpreted with caution.
2.2.6 Alternate Source Data
The Alternate Source Data input window, on the Parameter Input pull-down menu, allows the
user to input Pb concentrations from sources that are not covered under other menus in the
model. Examples include the direct ingestion of lead-based paint and the use of cosmetics or
home remedies. In cases where alternate sources exist, the intake of Pb (jig/day) must be
calculated from the site-specific data available. If the alternate source is lead-based paint, this
exposure is in addition to exposure to lead-based paint in house dust, which is covered in the
Multiple Source Analysis input window on the Soil/Dust Data option. If values are entered in the
Alternate Source Data input window, then the Total Percent Accessible for Alternate must be
changed in the GI/Bioavailability input window on the Parameter Input pull-down menu (see
Section 2.2.7).
-------
18
USER'S GUIDE FOR lEUBKwiN
Building an exposure scenario using the Alternate Source Data should be done with care. The
model assumes all entries represent chronic exposure. Remember that the model's output
represents only those children defined by the exposure scenario.
2.2.7 GI Values/Bioavailability
The GI Values/Bioavailability option on the Parameter Input pull-down menu allows the user to
make adjustments to the gastrointestinal absorption coefficient to account for site-specific
information on bioavailability. If this option is selected, a disclaimer window appears which
indicates that the user should only change default values if alternative bioavailability information
is available. Relative bioavailability is indexed by measuring the bioavailability of a particular
substance relative to the bioavailability of a standardized reference material, such as soluble lead
acetate. For the IEUBK model, soluble lead in water and food is estimated to have 50 percent
absolute bioavailability based on the bioavailability of soluble lead acetate.
Relative bioavailability is indexed by measuring the bioavailability of a particular substance
relative to the bioavailability of a standardized reference material, such as soluble lead acetate.
For the IEUBK model, soluble lead in water and food is estimated to have 50 percent absolute
bioavailability based on the bioavailability of soluble lead acetate. Furthermore, qualitative
estimates of relative bioavailability can be made in the uncertainty section of a risk assessment.
After selecting «OK», the GI Values/Bioavailability Information input window appears. The
TRW recommends not changing the bioavailability default without a thorough understanding of
the absorption methodology. Additional guidance on bioavailability is available from the TRW
website. (See Appendix C or [Short Sheet: IEUBK Model Bioavailability Variable, EPA #540-
F-00-006, OSWER #9285.7-32] http://www.epa.gov/superfund/lead).
2.2.8 Geometric Standard Deviation (GSD)
Geometric Standard Deviation (GSD) is a measure of the relative variability in PbB of a child of
a specified age, or children from a hypothetical population whose Pb exposures are known. The
GSD in lEUBKwin encompasses biological and behavioral differences, measurement variability
from repeat sampling, variability as a result of sample locations, and analytical variability. The
default value of 1.6 is recommended. This GSD value is based on specific analyses of data from
several neighborhoods with paired data for environmental Pb concentration and measured PbB
levels. The TRW recommends that site-specific estimates of GSD not be substituted for the
default value without detailed, scientifically defensible studies documenting site-specific
differences in child behavior or Pb kinetics. Should the user decide to modify the GSD,
lEUBKwin currently accepts values within the range of 1.3 to 1.8 for the GSD based on results
of site-specific data. If the user enters values outside of this range, a warning screen will appear
telling the user to select a different value. A GSD short sheet is currently under development and
will be available at the TRW website upon completion. Additional guidance is available through
the FAQ How can I derive a site-specific geometric standard deviation (GSD) ? on the TRW
website (http://www.epa.gov/superfund/lead).
-------
19
USER'S GUIDE FOR lEUBKwiN
2.3 SAVE AND LOAD INPUT PARAMETERS
If the user wishes to retain the current set of model input parameters (media concentrations,
intake rates, and bioavailability values) as the starting point for another analysis, the data can be
saved using the «Save Input to File» command on the Parameter Input pull-down menu. The
user may create a filename (eight characters or shorter), which will be stored in the form
[NAMEJ.svd. It should be noted that, the lEUBKwin model has a 16-bit dynamic library, and is
not a full 32-bit application, and therefore does not support long filenames. The model also uses
a directory structure such that the user may specify the location of the saved input file; by default
the model will save the input file in the lEUBKwin subfolder under the Programs directory
which is where all file types (i.e., input files, results files) are stored unless otherwise specified
by the user. Saved parameter data can be loaded using the «Load Input from a File» command
found on the Parameter Input pull-down menu. It should be noted that, if the user loads a saved
text input file, the same checks will be performed by the model to determine the appropriateness
of the input values as when the user enters input values from the model input modules. The
model does not read tab or comma-delimited batch mode files.
-------
20
USER'S GUIDE FOR lEUBKwiN
3.0 Model Analysis
This section describes the types of analyses that can be run with lEUBKwin. Further discussion
can be found in the IEUBK Guidance Manual (U.S. EPA, 1994).
3.1 OPTIONS FOR MODEL ANALYSES
Effective use of lEUBKwin requires entering data that are appropriate to the site(s),
parameter(s), and subject(s). The most convenient way to accomplish this is to construct a multi-
media, site-specific exposure scenario. The user should be aware that a site-specific risk
assessment requires site-specific soil concentrations which may be augmented by site-specific
data for dust, air, food, and other adjustable defaults.
The use of site-specific data is recommended to tailor the exposure to the conditions at the site.
Site-specific data are most commonly used in place of the model default values for Pb
concentrations in soil, dust, air, and water. Site-specific data for locally caught fish or home-
grown vegetables are sometimes used to augment dietary default values. However, such data
should be evaluated for validity prior to use. Model default values should only be replaced when
site-specific data specific to the parameters are more representative. lEUBKwin default values
should never be adjusted simply to attain a better match between modeled and empirical PbB
concentrations.
Prior to using lEUBKwin to calculate PbB levels, the user should decide what type of analysis
should be run: single simulation run, multiple simulation run, or batch mode. Each of these is
appropriate for certain scenarios, and inappropriate for others. A single simulation allows the
user to run scenarios for a single location, or multiple locations with homogenous media
concentrations (Categories A and B of Figure 2-1, Section 2.0). By contrast, multiple runs allow
the user to specify a range Pb concentrations for a specific environmental media and obtain
results of model runs at multiple points within that range. Batch mode allows the user to run
input data for many locations having different media concentrations in a single run of the model.
Additionally, batch mode allows the use of different ages or different values for Pb concentration
in soil, dust, water, air, and alternate Pb sources for each exposure scenario. The media intake
and absorption parameters are the same for every exposure scenario in a batch mode run and
must be specified before using the Run Batch Mode Model option, unless default values are used.
Output options for results from Single, Multiple, and Batch Mode runs are presented below.
3.2 OVERVIEW OF MODEL OUTPUT
lEUBKwin can calculate four primary outputs:
Geometric mean PbB levels, by age
• Percent of children with PbB in excess of a user-specified level of concern (typically
10 |ig/dL)
• Average media-specific daily Pb uptake rates
• Media-specific remediation goals
-------
21
USER'S GUIDE FOR lEUBKwiN
The output data can be saved and presented in tables, graphs, or text files. By default, the model
will generate a text file of the output unless otherwise specified by the user prior to running the
model. Screen captures of graphs (for use in technical reports) can be performed by applying the
Print Screen feature available on most PC keyboards, re-sizing the image to the appropriate
dimensions, and exporting the image to the desired application (e.g., Corel WordPerfect or
Microsoft Word). While printed tables and graphs allow convenient documentation of
lEUBKwin results in risk assessments, text files can also be exported to most statistical software
packages for additional analysis. The appropriate procedures to follow when exporting output
data to Lotus 1-2-3 or Microsoft Excel are described in Appendix A.
3.3 SELECTING OUTPUT ALTERNATIVES
Results of lEUBKwin simulations may be saved in several forms. The user should select the
most useful forms in advance, because the results of some interactive simulations cannot be
recovered without repeating the analysis once the user has bypassed the opportunity to save the
results.
3.3.1 What Graphical Options are Included in lEUBKwin?
lEUBKwin incorporates some output options that facilitate presentation risk assessment results.
The graphical output choices in the lEUBKwin Graph pull-down menu include the following
choices:
• PbB concentration vs. media Pb concentration
• Distribution probability percent
• Distribution probability density
• Distribution probability for multiple runs
• Probability density for multiple runs
Some plots may be used for single runs, while others are applicable for multiple runs. These
plots present estimates of a plausible distribution of PbB levels that are centered on the
calculated geometric mean blood lead. To derive the distribution around the geometric mean,
lEUBKwin uses a geometric standard deviation-a value describing the spread around the
geometric mean PbB. From this distribution, lEUBKwin calculates the probability that a child's
PbB will exceed a user-selected PbB level of concern (typically 10 |ig/dL) according to the
equation z = (ln(10) - ln(GM)/ln(GSD). For more information see Appendix B.
As discussed above, the GSD is a measure of relative variability in PbB levels of children of a
specific age or a hypothetical population of children whose Pb exposure is known. The GSD in
lEUBKwin encompasses biological and behavioral differences, measurement variability from
repeat sampling, variability as a result of sample locations, and analytical variability. The
default value of 1.6 is recommended. Figure 3-1 illustrates the impact of the relative positions of
the level of concern (LOG) and the geometric mean (GM) on the proportion of children "at risk"
for two populations with different GSDs. If LOC>GM, then the area for children at risk (shaded
plus solid) for GSD=1.7 is greater than the area (solid) for GSD=1.4. In situations where the
-------
22
USER'S GUIDE FOR lEUBKwiN
typical GM blood lead concentrations are predicted to be higher than the LOG, the graph would
look different.
GSD= 1.4
Children
not at i
risk i
i
LOG GM LOG
Figure 3-1. The Impact of GSD on Plausible Geometric Mean PbB Distributions.
If LOCXGSD, then the area for children not at risk (shaded plus solid in lower tail) for GSD=1.7
is larger than the area for children not at risk (solid in lower tail) for GSD=1.4. While this might
at first appear to suggest that risks are lower in this case with a higher GSD, the user should note
that the graph with the high GSD also indicates an increased risk of even higher blood lead levels
at values in excess of the LOG. Both distributions presented in the figure are based on the same
data set.
3.4 COMPUTATION OPTIONS
3.4.1 Run a Single Simulation
This option uses only the currently loaded parameter set (the default values and site-specific data
specified by the user). After the user selects Run the Model (Single Run) from the Computation
pull-down menu, the model prompts the user to choose a time step for numerical iteration. This
is a technical matter which does not usually need to be changed from the default. The default
time step is 4 hours. The time step does not determine the averaging time used to define average
daily intakes of lead. The time step is used in the biokinetic component of the model in
-------
23
USER'S GUIDE FOR lEUBKwiN
combination with compartmental transfer times to calculate the distribution of lead among bodily
tissues.
If you want to save the results for reference or further analysis, the user can assign a unique file
name, otherwise the default file name will be assigned. If you want to analyze the data using a
spreadsheet, checking this box will save the data in a tab-delimited format for import into a
statistical software package or spreadsheet program (see details in Appendix A).
3.4.1.1 Single-Run Options
Single-run plotting options in lEUBKwin include the following:
• The average of the geometric mean PbB concentrations for children in sequential 1-
year age intervals, the input concentrations for several media, and the media-specific
daily Pb uptake for each year are saved internally. The output file that is automatically
generated is RunModel.txt.
• Plotting the cumulative probability distribution for exceeding any user-specified PbB
level of concern for a single exposure scenario (Option 2—Distribution Probability
Percent). This is sometimes called the exceedance probability distribution.
• Plotting the log-normal probability density function of PbB levels for a single
exposure scenario predicting geometric mean PbB (Option 3—Distribution
Probability Density).
The probability density function gives users a better idea of the spread of PbB levels for children
exposed to a single set of environmental Pb concentrations. The plot of the exceedance
probability distribution also shows the fraction of children above a PbB level of concern for the
single-exposure case (e.g., what fractions of children are above 10 |ig/dL). The user may print
the probability plots after viewing. The single-run media plots are saved as [name].GRF. If
saved, these graphs may be viewed at a later date using the Graph pull down menu.
3.4.2 Calculating a Blood Lead Concentration Associated with Media-Specific
Concentration
The user may select Find Blood Pb Concentration from the Computation pull-down menu and
enter a known media lead concentration to determine the predicted concentration of lead in blood
associated with that media concentration.
This module generates a plot of the relationship between blood lead concentration and the
exposure to lead in a specific, user-selected, environmental medium (e.g., soil, dust, drinking
water, air, or diet). The module can be used to explore the sensitivity of the predicted blood lead
concentration to assumptions about exposures to lead in specific media. The module can also be
used to explore the sensitivity to other exposure assumptions. For example, the effect of
-------
24
USER'S GUIDE FOR lEUBKwiN
alternative assumptions regarding the soil lead absorption fraction can be explored by comparing
plots of blood lead concentration vs soil lead concentration, when different values for the lead
absorption fraction are assumed.
3.4.3 Run Multiple Simulations
This option uses the currently loaded parameter set (default values and site-specific data
specified by the user) but repeats the run for either a range of concentrations for a user-specified
medium. To conduct this simulation run, select the medium, the range of values for that
medium, and the number of time steps for numerical iteration. With Multiple Runs the user can
determine PbB concentration by age group based on a range of media Pb concentrations and can
search for the medium Pb concentration associated with an age-specified PbB concentration.
3.4.3.1 Multiple-Run Options
lEUBKwin has additional features that allow the user to combine output from several runs onto
single plots. These multiple-run options include both blood lead concentration vs. media lead
concentration plots and a sequence of graphics-overlay simulation runs (multiple runs) as
described below. For both of these options, the media range option saves input data for
geometric mean PbB vs. media Pb plots for a range of evenly spaced media Pb concentrations.
For example, the user may generate plots of PbB vs. PbS concentrations smoothly interpolated
from calculated values at 250, 500, 750, and 1,000 |ig/g for children aged 12 to 24 months.
Results are saved as filename [name] .LIN. The data in the [name] .LAY overlay file include the
geometric mean PbB for children in the age range, and the Pb concentration in soil and other
media. The actual plots of probability density or cumulative distribution functions depend on the
GSD value selected for use in risk estimates, and these plots can be printed. While this option is
useful in recognizing the pattern between increasing media concentrations and increasing blood
lead levels, the results are not intended to directly support risk-based decisions. These graphs
show the predicted geometric mean blood lead levels, and not the probability of exceeding
10 |ig/dL as a risk criterion in EPA's risk management guidance for lead.
Geometric mean blood lead concentration vs. media lead concentration graph, where
each run increases the Pb concentration in a specified medium by a user-defined
amount across a range (Option 1—Distribution Probability for Multiple Runs).
Overlaid plots of the cumulative exceedance probability distributions of PbB levels
for multiple media concentrations, where each run increases the Pb concentration in a
specified medium by a user-defined amount (Option 4—Distribution Probability for
Multiple Runs).
Overlaid plots of the log-normal probability density functions of PbB levels for
multiple exposure scenarios, where each run increases the Pb concentration in a
specified medium by a user-defined amount (Option 5—Probability Density for
Multiple Runs).
-------
25
USER'S GUIDE FOR lEUBKwiN
Overlaid plots cannot be used unless the user has previously created an output file from the
Computation menu, with the file extension *.LAY. The overlaid probability density functions
give most users a better idea of how the probability of exceeding a PbB level of concern
increases with each incremental increase in environmental lead. The exceedance probability
distributions may be used to estimate the increases in the fraction of elevated PbB levels or to
visually estimate the environmental Pb levels corresponding to a specified fraction of
unprotected children above the level of concern.
3.4.4 Run Batch Mode Simulation
The input data file for a batch run must be created outside of lEUBKwin using any text editor
(e.g., Notepad). The following conventions must be observed in creating the batch file:
• The input data file must be an ASCII file with no special characters. The model does
not read tab or comma-delimited batch mode files.
• The data set must have a DAT extension (i.e., [name].dat).
The first three lines of the input data file can be any identifiers that the user may
require; usually the first line is the run name, the second line is the modeling options
used in the run, and the third line consists of the headers for variables in the data set.
The data fields are entered format-free, although the use of regular spacing and
alignment of decimal points are recommended to improve readability.
• The maximum width is 8 spaces per column.
The variable values should be separated by spaces.
• For quicker runs, the recommended limit is 500 rows of data.
• The missing values must be shown by a single decimal point.
• Each line in the input data file must contain data or blank values for missing data for
the following 10 variables:
1. Child identifier code 6. PbD concentration in |ig/g
2. Family or residence unit identifier code 7. Drinking water Pb concentration in |ig/L*
3. Area or neighborhood identifier code 8. Air Pb concentration in |ig/m3*
4. Child's age in months 9. Daily intake of Pb from "alternate"
5. PbS concentration in |ig/g sources in jig Pb/g
10. Observed child PbB level in jig/dL*
Any of these (*) may be left blank by just placing a period in the middle of the data column.
There is no need for PbB values, since the model can calculate risk for a group of locations or
-------
26
USER'S GUIDE FOR lEUBKwiN
residences. For many users the easiest way to create a batch mode input file is to use a
spreadsheet (see Appendix A). As noted above, the child's age (in months) must be entered.
The media concentrations for water and air from the parameter input windows will be entered if
values for these media are missing. Either a PbS concentration or a PbD value is needed for the
simulation, along with a stand-in value (imputation rule) if one is missing (e.g., if the user does
not fill in PbD values, the current default is to replace a missing PbD concentration with the PbS
concentration). The user may prefer to create an input data file with missing PbD concentrations
replaced by some fraction of the PbS concentration.
Note that lEUBKwin considers only two significant digits (to the right of the decimal point)
from the data input file for running batch mode; however, the model uses double precision (eight
digits) in all its calculations. In the current stand alone version, lEUBKwin is not capable of
executing consecutive batch mode files from a list of files.
The batch mode option can be used to develop estimates for statistical analyses of simulated
community PbB distributions with or without observed PbB levels. When PbB data are
available, batch mode also provides a tool for developing empirical comparisons of the model
predictions and measured blood lead concentrations.
Output consists of files [name].TXT that contain predicted PbB concentrations for each case or
record (child) in the input data file. The output files also document the missing value
imputations when some of the input data on residential media Pb concentrations are missing.
The files may be used as input for statistical analysis in other software programs (see
Appendix A).
A feature that is new to the 32-bit version allows for processing of multiple batch mode input
files. The Batch Run Dialog includes a list box, which will list the batch input data files
available for processing. The user may add or remove files from the queue. The results are
written to a single file BATCHRUN.TXT (unless another file name is specified). Each input file
is processed sequentially, and the results appended to the results file below the previous run.
Results are identified by repeating the header, which identifies the input file name.
Additional guidance on Batch Mode inputs and use is available in the FAQs on the TRW
website, (http://www.epa.gov/superfund/lead)
-------
27
USER'S GUIDE FOR lEUBKwiN
4.0 Interpretation of Results
4.1 UNDERSTANDING THE OUTPUT
Each of the output options described above allows the user to examine a different aspect of
lEUBKwin simulation. The output from a single simulation run may be displayed in several
forms. Most users wish to see the variability associated with a predicted PbB concentration. This
range can be demonstrated graphically by selecting the intrinsic variability GSD, and then
plotting a cumulative probability distribution. The range of plausible PbB values may be
determined graphically as defined by upper and lower percentiles of the distribution. For
example, the 5th and 95th percentiles of the distribution will include 90 percent of the children
with the given site-specific or household-specific exposure scenario. Note that the plausible
range reflects predicted variability among individuals with the same exposure scenario, and
should not be confused with a statistical confidence limit (which is a measure of statistical
uncertainty in a predicted value such as a geometric mean).
The other output characteristic that many users wish to see is the estimated probability of
exceeding the specified PbB level of concern, corresponding to the given exposure scenario or
scenarios (for multiple runs in a given medium). According to the IEUBK Guidance Manual
(U.S. EPA, 1994), there are two valid interpretations for this output:
1. The output of the model may be considered to be the best estimate of a plausible
range of PbB concentrations for a hypothetical child under a specific Pb exposure
scenario. The range of values is centered on the geometric mean PbB concentration
expected for a typical child with this exposure scenario. The portion of the upper tail
of the probability distribution exceeding some chosen PbB level of concern provides
an estimate of the risk of exceeding that level for a typical child of that age residing in
the same household and with the same exposure history.
2. The output of the model may also be considered to be the predicted geometric mean
PbB of a population of children under the same Pb exposure scenario. That portion
of the upper tail that exceeds some chosen PbB level of concern indicates the fraction
of the population exceeding that level when all of these children have the same
exposure history.
A common misinterpretation of lEUBKwin is that it predicts community geometric mean PbB
and the fraction of the population of children at risk when the input is the mean or geometric
mean of household-specific environmental Pb concentrations. That misinterpretation arises,
particularly when the environmental variables have a wide distribution among the neighborhoods
of the community. A correct approach requires applying the model to each individual home (or
area with homogeneous lead concentrations) and combining these results as an aggregate to form
an estimate of neighborhood or community risk.
It should be noted that the user may experience some discrepancies in the appearance of the
graphs produced with lEUBKwin as a result of the limitations of Visual Basic C++, the
-------
28
USER'S GUIDE FOR lEUBKwiN
programming tool used to develop the model. Specifically, scientific notation is used for values
less than 0.01 (e.g., 10"2) on graphical outputs; however, improper formatting of the scientific
notation occurs as a result of the Visual Basic C++ programming functions and cannot be
corrected. This discrepancy should be eliminated in future versions of the model, and currently
does not functionally impede upon the user's ability to read and interpret graphs.
The National Contingency Plan2 states that the Superfund process for establishing cleanup
numbers for contaminated sites consider nine criteria concerning risks associated with remedial
alternatives analysis that apply to. The nine criteria are as follows:
Prim ary Criteria
1. Overall protection of human health and the environment
2. Compliance with applicable or relevant and appropriate standards, limitations,
criteria, and requirements (ARARs)
Balancing Criteria
3. Long-term effectiveness and permanence
4. Reduction of toxicity, mobility and volume
5. Short-term effectiveness
6. Implementability
7. Cost
Modifying Criteria
8. State acceptance
9. Community acceptance
2For more information, see 40CFR §300.430.
-------
29
USER'S GUIDE FOR lEUBKwiN
5.0 Model Application and Documentation
The output of lEUBKwin helps users develop an understanding of the potential risk for elevated
PbB levels in children. Through different model run scenarios, these potential risks can be
calculated for a number of different hypothetical situations, including risks for a single child, a
group of neighborhood children, or an entire community. With proper application and
documentation, lEUBKwin can effectively and accurately predict the risks at lead-contaminated
sites.
5.1 MODEL APPLICATION
lEUBKwin is designed to facilitate two processes: rapid delineation of the relationship between
environmental Pb and PbB in children and calculation of their risk of elevated PbB (i.e., the
probability that a given child or group of children will have PbB concentrations exceeding a
specified level of concern). As such, lEUBKwin provides a tool for site-specific risk assessment
for young children exposed to Pb from different media and through different pathways in their
environment. The intended applications of lEUBKwin are to:
• Provide a summary of children's long-term, primary residential exposure to lead.
• Provide a best estimate of the geometric mean PbB concentration for a typical child,
aged 6 months to 84 months, assumed to reside at a given residence.
• Provide a basis for estimating the risk of elevated PbB (i.e., for exceeding a
designated PbB concentration of concern) for a hypothetical child of specified age
with a given site-specific residential Pb exposure.
• Provide a basis for estimating the risk of elevated PbB concentrations among
populations of children under age 7 in a given neighborhood by aggregating the
individual residential risk estimates.
• Predict likely changes in risk of elevated PbB concentrations from exposure to soil,
dust, water, or air Pb following abatement actions designed to reduce exposure levels
from one or more environmental media.
• Provide assistance in determining appropriate soil or dust Pb target cleanup levels at
specific residential sites.
• Provide assistance in estimating PbB concentrations associated with soil or dust Pb
concentrations at undeveloped residential sites that may be developed in the future.
EPA has established default values for all of the user-specified parameters in lEUBKwin. Use of
the default parameters is recommended unless valid site-specific monitoring data exist to define
values that are higher or lower. Site-specific data are commonly used in place of the model
default values for Pb concentrations in soil, dust, air, and water. However, for intake rates,
-------
30
USER'S GUIDE FOR lEUBKwiN
bioavailability and PbB distribution parameters, the TRW recommends that the default values be
used, unless accompanied by strong evidence supporting the use of site-specific values. Model
default values should never be adjusted simply to attain a better match between modeled and
empirical blood lead. Information on situations where it might not be advisable to use the
IEUBK model can be found in the IEUBK Guidance Manual (U.S. EPA, 1994) and Section 2.0
of this manual.
5.2 MODEL DOCUMENTATION (MINIMUM REQUIREMENTS)
By reviewing every adjustable parameter in the model and noting which ones have been
modified in a particular run, the user has a permanent record of the input. The data used in the
model runs should be saved so the results can be reproduced. Different levels of documentation
are required for specific applications of the model. The text of risk assessments should contain a
full discussion describing the inputs, use, and outputs of the model. Any model output
referenced should be included in appendices of the risk assessment, including distribution and
density curves, text outputs, and batch mode files. A listing of model inputs should be included
in the text or as an appendix to the risk assessment. lEUBKwin captures data and comments in
the risk assessment mode in a tabular format. This information can be exported to a spreadsheet
application (e.g., Lotus 1-2-3, Microsoft Excel) for submission to EPA (see Appendix A).
5.3 USER RESPONSIBILITIES
The user is responsible for using model input parameters that are appropriate to the site.
Generally the minimum site-specific data collection and analyses include, but are not necessarily
limited to, collection of an adequate number of representative soil and dust samples, and
determination of the Pb concentrations and physical or chemical properties that affect transport
and bioavailability. The typical input data include a multimedia household environmental Pb
study that includes soil, dust, paint, water, and air; information on Pb exposures outside a child's
home; information on family demographics and child behavior patterns in the community that
may affect access to Pb sources; and characterization of physical and chemical properties that
affect bioaccessibility and bioavailability.
-------
31
USER'S GUIDE FOR lEUBKwiN
6.0 Example Scenarios
This section presents examples and results obtained from reasonably straight-forward
applications of lEUBKwin. The first part of this section presents results obtained for single runs
of the model, when only soil, dust, and water lead concentrations are entered into the model.
Additional examples cover the sum of the individual risks approach, and the assessment of risk
using grouped data for a neighborhood. These examples are relatively simple and are intended to
cover the most common applications of lEUBKwin. Several additional examples, some more
complex in nature, are presented for the user in the IEUBK Guidance Manual (U.S. EPA, 1994).
6.1 SIMPLE CASE SINGLE RUNS
If the model is run using the following inputs, the user should calculate the following outputs for
children aged 0 to 84 months. These data can be used as a quick test of the model.
Table 6-1. Simple Case Runs
Runl
Run 2
Run 3
Run 4
Run 5
Input Parameters
Soil
(mg/kg)
Default D<
500
500
1,000
1,000
Dust
(mg/kg)
:fa ult
500
500
1,000
1,000
Water
(Hg/L)
Default
Default
50
Default
50
Age
(months)
0-84
0-84
0-84
0-84
0-84
All other
inputs
Default 2.
Default
Default
Default
Default
Output Parameters
Geometric
mean ((ig/dL)
7
6.0
8.8
10.2
12.6
% Above
10 (ig/dL
0.287
13.899
39.572
51.496
68.553
6.2 "SUM OF INDIVIDUAL RISKS" APPROACH
Suppose that there are data on four households with children in a neighborhood. Residents of
each household are exposed to lead-contaminated soil. The first house has 250 jig/g Pb in soil,
the second house has 250 jig/g, the third house has 1,000 jig/g, and the fourth house has
1,000 |ig/g. Dust lead (PbD) concentrations are assumed to be 70 percent of the PbS
concentration in houses 2 and 4, and 15 percent of the PbS concentration in houses 1 and 3. As
an estimate of air Pb contribution to PbD at 0.1 jig Pb/m3, 10 jig/g was added to PbD. The
respective PbD concentrations are 47.5 ^g/g, 185 ^g/g, 160 ^g/g, and 710 ^g/g. All other
parameters are set to default values.
The samples are assumed to be representative of the exposure distribution in the neighborhood.
The selection of four houses is for illustration purposes only. The risk estimates are intended to
be unbiased estimates of potential risk for other years in which different children, not in the
current sample, may occupy the same or other houses in the neighborhood. Obviously, a reliable
-------
32
USER'S GUIDE FOR lEUBKwiN
estimate of neighborhood risk will require more than four houses. The probability density of
PbB for the four houses is shown in Figure 6-1 and the interpreted results in Table 6-2.
0.4
0.3
CD
Q
= 0.2
.a
to
.a
0.1
0.0
0
House4
HouseS
House 2
House 1
10 15
Blood Lead,|ig/dl_
20
25
Figure 6-1. Hypothetical Probability Density of Blood Lead in Houses 1 to 4.
-------
33
USER'S GUIDE FOR lEUBKwiN
Table 6-2. Interpreted Results for Figure 6-1
House
number
1
2
3
4
PbS concentration
(ng/g)
250
250
1,000
1,000
PbD concentration
(ng/g)
47.5aO.
185bO.
160al
710b4
JRisk from all houses 1-4
Probability of exceeding
10 (ig/dL
106
736
6.729
0.534
58.105
a Calculated using multiple source analysis and an MSB of 0.15
b Calculated using multiple source analysis and an MSD of 0.70
The sum of the risks for these four houses 0.106% + 0.736% + 16.729% + 40.534% = 58.105%.
That is, 0.581 children are expected to have PbB concentrations exceeding 10 |ig/dL, or an
average risk for a child living in this neighborhood would be 58.1%/4=14.5%. The risk at two of
the residences (houses 3 and 4) in the neighborhood is above the OSWER specified level of
protectiveness which aims to limit exposure to PbS levels such that a typical child or a group of
similarly exposed children would have an estimated risk of no more than 5 percent exceeding the
PbB level of concern (typically 10 |ig/dL).
The use of aggregate neighborhood input data requires that the user compare the probability
density function (PDF) and elevated PbB risk calculated from aggregate parameters with the
correct PDF and elevated PbB risk functions, which are the mathematical composites of the
individual PDF and risk functions. Expressed mathematically:
true neighborhood PDF = (PDF(site 1) + PDF(site 2) +...)/N
true neighborhood risk = (risk(site 1) + risk(site 2) +...)/N
(Equation 6-1)
(Equation 6-2)
The approach outlined here does not require any mathematical assumptions about the distribution
of soil and dust Pb concentrations, nor of any other parameters or variables except for PbB. The
conditional distribution of individual PbB is assumed to be log-normal with a constant GSD
(given specified values of Pb exposure variables that determine the geometric mean PbB for
individuals in that exact environment). The method suggested is the most convenient and
flexible framework determined for neighborhood assessment concerning the effect of PbS
abatement.
6.3 RISK ASSESSMENT USING GROUPED DATA FOR A NEIGHBORHOOD
The example in the preceding section had a "neighborhood" with only four houses, so that the
amount of work required was not very burdensome. In the real world, the site manager or risk
assessor may be dealing with relatively homogeneous neighborhoods or small communities with
several hundred households. These calculations can be simplified by grouping soil and dust lead
levels into small cells with fixed ranges of values. The grouped data within each cell are all
assigned the same value, such as the midpoint of the interval.
-------
34
USER'S GUIDE FOR lEUBKwiN
Each cell is then assigned a statistical weight. The statistical weights could be one or more of the
following:
1. The number of housing units with soil and dust lead concentrations in the interval.
2. The number of children observed or expected to live in housing units with soil and
dust lead concentrations in the interval.
3. The fraction of housing in a neighborhood that is expected to have soil and dust lead
concentrations in the interval.
4. The fraction of area in as yet undeveloped neighborhoods with soil and dust lead
concentrations in the interval.
The PDF and risk of elevated PbB in children reflects the weighted sum of the cell PDF or cell
risks. If the respective weights are denoted weight (cell 1), weight (cell 2), etc., and the PDFs
are denoted PDF (cell 1), PDF (cell 2), etc., and the risks are denoted risk (cell 1), risk (cell 2),
etc., then:
neighborhood PDF = [weight (cell 1) * PDF (cell 1) + weight (cell 2) *
PDF (cell 2) + etc.] / [weight (cell 1) + weight (cell 2) + etc.] (Equation 6-3)
neighborhood risk = [weight (cell 1) * risk (cell 1) + weight call (cell 2) *
risk (cell 2) + etc.] / [weight (cell 1) + weight (cell 2) + etc.] (Equation 6-4)
The following hypothetical example may illustrate these points. Suppose that a random sample
of 250 houses and apartments has been obtained in a neighborhood. The number of houses in
each interval of 250 jig/g soil and 250 jig/g dust lead is shown in Table 6-3. This same example
is shown in Table 6-4 in intervals of 500 |ig/g in soil and 500 |ig/g in dust. There is no
requirement that there be equal interval lengths in either soil or dust.
The user may then calculate neighborhood risk in any one of three ways:
• Sum of risks for 250 housing units
• Sum of risks for 14 cells or groups of width 250 |ig/g soil and dust
• Sum of risks for seven cells or groups of width 500 |ig/g in soil and dust
The results of calculations are shown in Tables 6-3 and 6-4. The total risk for the data shown in
Table 6-3 is calculated as follows:
(30 * 0.0.065% + 50 * 1.680% + 20 * 7.360% + 10 * 1.127% + 40 * 6.005% + 30 *
14.814% + 20 * 25.597% + 10 * 13.001% + 20 * 23.582% + 10 * 34.585% + 3 *
44.829% + 4 * 52.278% + 1 * 72.123% + 2 * 65.597%)/250 = 11.74% (Equation 6-5)
The risk calculation for the data in Table 6-4 is similar.
The total risk in Table 6-4 is calculated as follows:
(130 * 1.386% + 70 * 15.745% + 10 * 12.123% + 30 * 33.604% + 7 * 53.049% + 1 *
67.238% + 2 * 65.031%)/250 = 11.92% (Equation 6-6)
If there are not too many cells, the amount of calculation can be noticeably reduced. However,
as the intervals are made wider, there is a corresponding loss of accuracy in the neighborhood
-------
35
USER'S GUIDE FOR lEUBKwiN
risk estimate. The extra effort in calculating risks with 250 |ig/g intervals (14 cells) is probably
compensated by the increased precision, with an estimate of 11.74% instead of 11.92%. The
individual risk, as shown by the calculated risk for the residences with higher soil and dust
concentrations, for many residences in the neighborhood is above the OSWER specified level of
protectiveness (no more than 5 percent exceeding typically 10 |ig/dL). This illustrates the
importance of calculating individual risk, rather than community risk, to achieve EPA's public
health goal.
6.4 RISK ASSESSMENT USING NEIGHBORHOOD OR NEIGHBORHOOD-SCALE INPUT
There are situations in which it is either inconvenient or impossible to apply the IEUBK model at
the intended household residence scale. For example, if only neighborhood mean values or
geometric mean values of input parameters such as soil and dust lead are available, the model
estimate may be far less reliable than if individual residential measurements were made.
Another possibility is that there are a substantial number of soil and dust lead measurements at a
site, but not at houses or locations within the site where blood lead and EBL risk estimates are
needed, for example, to compare with blood leads observed at residences where there are no
environmental data. There are some circumstances in which this is clearly not a valid application
of the model. As we do not clearly understand the range of conditions under which the IEUBK
model may be used with large-scale input data at this time, we must discourage use of the
IEUBK model except with single-residence or residential lot-sized input data, or with data
grouped into cells as in Section 6.3.
-------
USER'S GUIDE FOR lEUBKwiN
36
Table 6-3. Neighborhood Risk Estimation with Grouped Data
Hypothetical example of grouped data for a neighborhood with dust and soil samples
of 250 sites of house yards. Intervals are 250 (ig/g in PbS and PbD.
Soil
interval
0-249
0-249
0-249
250-499
250-499
250-499
250-499
500-749
500-749
500-749
500-749
750-999
750-999
1,000-1,249
Soil
midpoint
125
125
125
375
375
375
375
625
625
625
625
875
875
125
Dust
interval
0-249
250-499
500-749
0-249
250-499
500-749
750-999
250-499
500-749
750-999
1,000-1,249
1,000-1,249
1,750-1,999
1,250-1,499
Dust
midpoint
125
375
625
125
375
625
875
375
625
875
1,125
1,125
1,875
1,375
TOTAL
Statistical
weight
30
50
20
10
40
30
20
10
20
10
3
4
1
2
250
Geometric
mean
(Mg/dL)1
2.2
3.7
5.1
3.4
4.8
6.1
7.3
5.9
7.1
8.3
9.4
10.3
13.2
12.1
Risk2
Probability
of exceeding
10 (ig/dL
0.065
1.680
7.360
1.127
6.005
14.814
25.579
13.001
23.582
34.585
44.829
52.278
72.123
65.597
11.74
1 Calculated from IEUBK model with default parameters
2 Assuming GSD=1 .6
-------
37
USER'S GUIDE FOR lEUBKwiN
Table 6-4. Neighborhood Risk Estimation with Coarsely Grouped Data
Hypothetical example of grouped data for a neighborhood
with intervals of 500 (ig/g in soil Pb and dust Pb.
Soil
interval
0-499
0-499
500-999
500-999
500-999
500-999
1,000-1,499
Soil
midpoint
250
250
750
750
750
750
250
Dust
interval
0-499
500-999
0-499
500-999
1,000-1,499
1,500-1,999
1,000-1,499
Dust
midpoint
250
750
250
750
1,250
1,750
1,250
TOTAL
Statistical
weight
130
70
10
30
7
1
2
250
Geometric
mean
3.6
6.2
5.8
8.2
10.4
12.3
12.0
Risk2
Probability
of exceeding
10 (ig/dL
1.386
15.745
12.123
33.604
53.049
67.238
65.031
11.92
1 Calculated from IEUBK model with default parameters, ages 0 to 84 months
2 Assuming GSD=1.6
-------
38
USER'S GUIDE FOR lEUBKwiN
7.0 References
Hastings, C. Approximations for Digital Computers. Princeton University Press. 1955. In:
Handbook of Mathematical Functions with Formulas, Graphs and Mathematical Tables. M.
Abramowitz, LA. Stegun, Eds. National Bureau of Standards Applied Mathematics Series 55.
Seventh Printing: 1968, p. 932.
U.S. Environmental Protection Agency. Office of Solid Waste and Emergency Response.
Clarification to the 1994 Revised Interim Soil Lead Guidance for CERCLA Sites andRCRA
Corrective Action Facilities. OSWER Directive No. 9200.4-27P. Washington, DC: 1998.
U.S. Environmental Protection Agency. Technical Review Workgroup for Lead. Integrated
Exposure Uptake Biokinetic Model for Lead in Children (IEUBK). Version 0.99d. NTIS No.
PB94-501517. Washington, DC: 1994.
U.S. Environmental Protection Agency. Technical Review Workgroup for Lead. Guidance
Manual for the Integrated Exposure Uptake Biokinetic Model for Lead in Children. OSWER
Directive No. 9285.7-15-1, Document No. EPA/540/R-93/081. Washington, DC: 1994.
U.S. Environmental Protection Agency. Office of Solid Waste and Emergency Response.
Revised Interim Soil Lead Guidance for CERCLA Sites andRCRA Corrective Action Facilities.
OSWER Directive No. 9355.4-12. Washington, DC: 1994.
U.S. Environmental Protection Agency. Technical Review Workgroup for Lead. Technical
Support Document: Parameters and Equations Used in the Integrated Exposure Uptake
Biokinetic (IEUBK) Model for Lead in Children. Version 0.99d. OSWER Directive No. 9285.7-
22, NTIS No. PB 94-963505. Washington, DC: 1994.
U.S. Environmental Protection Agency. Technical Review Workgroup for Lead. Reference
Manual for the Integrated Exposure Uptake Biokinetic Model for Lead in Children, Windows
Version. EPA/540/K-01/007. Washington, DC: 2001.
U.S. Environmental Protection Agency. Technical Review Workgroup for Lead. System
Requirements and Design for the Integrated Exposure Uptake Biokinetic Model for Lead in
Children, Windows Version. EPA/540/K-01/006. Washington, DC: 2001.
U.S. Food and Drug Administration. 2006. Total Diet Study. U. S. Food and Drug
Administration Center for Food Safety and Applied Nutrition Office of Plant and Dairy Foods
and Beverages (May 16, 2006). Available online at The U.S. Food and Drug Administration
Center for Food Safety and Applied Nutrition Web site (http://www.cfsan.fda.gov/~comm/tds-
toc.html).
White, P., P. Van Leeuwen, B. Davis, M. Maddaloni, K. Hogan, A. Marcus, and R. Elias. 1998.
U.S. Environmental Protection Agency. The Conceptual Structure of the Integrated Exposure
Uptake Biokinetic Model for Lead in Children. Environmental Health Perspectives 106
(Suppl. 6): 1513.
-------
39
USER'S GUIDE FOR lEUBKwiN
Zaragoza, L. and K. Hogan. 1998. U.S. Environmental Protection Agency. The Integrated
Exposure Uptake Biokinetic Model for Lead in Children: Independent Validation and
Verification. Environmental Health Perspectives 106 (Suppl. 6): 1555.
-------
40
USER'S GUIDE FOR lEUBKwiN
Appendix A: Using Spreadsheets with lEUBKwin
A.I IMPORTING lEUBKwiN DATA TO SPREADSHEET PACKAGES
lEUBKwin data, including batch mode output can be readily exported to spreadsheets. The
following information is provided to facilitate this process.
The lEUBKwin output data (*.txt) for the risk assessment mode can be exported to a spreadsheet
application (e.g., Excel or Lotus 1-2-3) for submission to EPA. The data is captured in a tabular
format in lEUBKwin. To export output data to Excel, the user should employ the following
procedure:
1. Start Excel
2. Click on File/Open and change Files of Type to All Files, locate your *.txt output file,
and click Open.
3. On the Text Import Wizard screen choose Delimited and click Next. Select Tab as the
Delimiter and General as Column Data Format. Select Finish.
4. Save the file in Excel format (*.xls).
nsert Format lools C;ata Window Help
a
rtn Q x r!-, irni - Tools
The Text wizard has determined that your data is Fixed Width.
IF this is correct, choose Next, or choose the data type that best describes your data,
-Original data type
Choose the rile type that best describes your data;
• [Delimited: - Characters such as commas or tabs separate each field.
f Fixed width - Fields are aligned in columns v..ii!:h spaces between each field.
This screen lets you set the delimiters your data contains. You can see
Ei how your text is affected in the preview below.
Tab f~ Semjcolon I~ Comma
P Space l~ Other:
This screen lets you select
the Data Format.
Column data format
^ Genera
values to dates, and all remaining values to text
Advanced,..
r Date:
Do not [mport colurnri
Figure A-l. Export lEUBKwin Data Output to Microsoft Excel.
-------
41
USER'S GUIDE FOR lEUBKwiN
To export output data to Lotus 1-2-3, the user should employ the following procedure:
1. Start Lotus 1-2-3
2. Click on file/Open and change Files of Type to All(*), locate your *.txt output file,
and click Open.
3. On the Text File Options screen choose Automatically Parse Based on File Layout.
4. Save file in Lotus 1-2-3 format (*.123).
To import data into Lotus 1-2-3, the user would employ the following procedure:
1. Start Lotus 1-2-3.
2. Click on File/Open and change Files of Type to All Files, locate your *.txt output file,
and click Open.
3. On the Text File Options screen choose Automatically Parse Based on File Layout.
4. Save file in Lotus 1-2-3 format (*.123).
fe AO*>»f'
|J FLAGDAVK.TXT
11 PAINTBCH.TXT
RunModel.grf
sj RunModel.txt
XT [f] WIN1_19.TXT
Files of type: | All Files (")
Description
Open as read only
\~ CuTibine with current workbook
Parking options
r Start a new column at each
I Tab
C Raise as CSV (comma separated value) tile
'•' Auh:,rn.ahccilly par:e ba:ed on tile layout
Put everything in
Character set: (Windows
Figure A-2. Export lEUBKwin Data Output to Lotus 1-2-3.
-------
42
USER'S GUIDE FOR lEUBKwiN
To import data to Microsoft Excel, the user would employ this procedure:
1. Start Excel.
2. Click on File/Open, change Files of Type to All (*), locate your *.txt output file, and
click Open.
3. On the Text Import Wizard screen choose Fixed Width and click Next.
- Excel may have left out some column breaks or inserted extra column breaks.
To insert additional column breaks, click at the desired position. To remove a
column break, double-click it, click Next, select General Format, and click
Finish.
4. Save the file in Excel format (*.xls).
Note that Excel will break the header text of the file into columns as well. Also, an lEUBKwin
batch mode output file requires little to no editing before it is imported to other statistical
programs. If the data file contains a null row, this may cause import problems with other
spreadsheet packages. Once the results are loaded into a spreadsheet or statistical program, the
user can apply a variety of graphical and statistical techniques in evaluating the output of
lEUBKwin batch mode runs. Batch mode calculates the probability of exceedance as described
in Appendix B.
[J Microsoft fucel - Bookl
HJ Fte Ed* View Insert Format Toob Data Wirrfon Help
Save as type: | Formatted Test (Space deWted) (*.pm)
Ready
Figure A-3. Creating a Batch Data file with Microsoft Excel.
-------
43
USER'S GUIDE FOR lEUBKwiN
A.2 CREATING A BATCH MODE DATA INPUT FILE USING A SPREADSHEET
As described in Chapter 3, batch mode input files require rigorous formatting. These files can be
can be quickly and easily created by using spreadsheets. As an example, the following
procedure describes creation of a batch mode data file in Microsoft Excel:
1. Select the first 10 columns in your spreadsheet.
2. Format the columns to have a width of 8.0.
3. Right align all data (including the text descriptions at the top of the worksheet).
4. Save the file with a Formatted text (space delimited) (*.prn) extension. The name you
select should have no more than 8 characters or else the IEUBK model will truncate
it.
5. Manually change the file from *.prn to *.dat in your file directory.
6. Run batch mode using the newly created file.
-------
44
USER'S GUIDE FOR lEUBKwiN
Appendix B: Probability Distribution Function
The Z score is the distance between the PbB of 10 |ig/dL and the geometric mean (GM) PbB, in
standard normal units (Z score is also known as the standard normal deviate). Furthermore, the
Z score is the inverse of the standard normal cumulative distribution function (i.e., normal
distribution with mean of 0 and standard deviation of 1). The probability of a child's blood lead
level exceeding 10 |ig/dL is calculated as 1- the area under the standard normal cumulative
distribution function to the left of the Z score. For example:
Given:
GM PbB = 5 (ig/dL (e.g., from lEUBKwin)
GSD=1.6
Determine probability of exceedance:
[In (10)-In (5)]/
Z= /ln(l.6) = l48
Probability of child's PbB exceeding 10 (ig/dL (1-p) =
1 - [Probability(Z < 1.48))] = 0.93 = 0.07
After calculating the Z score, the lEUBKwin model simulates a standard normal distribution
using the calculated GM and selected GSD. A difference approximation method is then used to
calculate the probability of exceedance (1-p). This method has simplified the calculation from
the IEUBK DOS version (TEUBK 0.99d).
The z-score can also be used to estimate a percentile value for a lognormal distribution given a
geometric mean and standard deviation:
Xp = GM-GSDZf
where,
Xp = blood lead concentration at the pth percentile of the lognormal distribution
GM = geometric mean of the distribution
GSD = geometric standard deviation of the distribution
Zp = z-score corresponding to the pth percentile of the standard normal cumulative distribution
Note that taking the logarithm of both sides of the preceding equation and solving for Zp yields
the equation for Z given above. The following is an example of using the preceding equation for
calculating the blood lead concentration corresponding to a selected probability of exceedance
(1-P):
Calculate the 95th percentile blood lead concentration (Xp)
Given:
GM PbB = 5 (ig/dL (e.g., from lEUBKwin)
GSD = 1.6
Zp =1.645 (the 95th percentile Z score)
Xp= 5-(l.6)L645 = W.^g/dL
-------
45
USER'S GUIDE FOR lEUBKwiN
The results indicate there is a 5% probability of a child's blood lead concentration exceeding 10.8 (ig/dL,
given a GM PbB concentration of 5 (ig/dL and a GSD of 1.6.
-------
46
USER'S GUIDE FOR lEUBKwiN
Appendix C: Short Sheets Currently Available
The TRW Lead Committee website has a complete listing of all Short Sheets that are related to lead risk
assessments: http://www.epa.gov/superfund/lead/guidance.htm
2000 Short Sheet: TRW Recommendations for Sampling and Analysis of Soil at Lead (Pb) Sites [EPA
#540-F-00-010, OSWER #9285.7-38] (April 2000) to be consulted when using the IEUBK or the
Adult Lead Methodology.
Provides recommendations and protocols for the collection, preparation, and analysis of lead in soil
and dust for use in lead modeling.
1999 Short Sheet: IEUBK Model Soil/Dust Ingestion Rates [EPA #540-F-00-007, OSWER #9285.7-33]
(December 1999) to be consulted when using the IEUBK model.
Provides recommendations on the substitution of default soil/dust ingestion rates in the IEUBK
model.
1999 Short Sheet: IEUBK Model Bioavailabilitv Variable [EPA #540-F-00-006, OSWER #9285.7-32] to
be consulted when using the IEUBK or the Adult Lead Methodology. Discusses issues to consider
and applicable methods for determining a site-specific bioavailability value for soil-borne lead.
1998 Short Sheet: IEUBK Model Mass Fraction of Soil in Indoor Dust (MSP) Variable [EPA #540-F-00-
008, OSWER #9285.7-34] (June 1998) to be consulted when using the IEUBK model.
Provides updated guidance on the Mass Fraction of Soil in Indoor Dust (MSD), the fraction of house
dust that is derived from the outdoor soil. The MSD is a variable in the dust lead Multiple Source
Analysis module of the IEUBK model.
1997 Short Sheet: Overview of the IEUBK Model for Lead in Children [NTIS #PB99-9635-8, OSWER
#9285.7-31] (August 1997) to be consulted when using the IEUBK model.
Provides general information on the IEUBK model
------- |