Jump to main content.


Model Report

Stochastic Human Exposure and Dose Simulation Model for Wood Preservatives

Last Revision Date: 08/31/2009 View as PDF
General Information Back to Top
Model Abbreviated Name:

Model Extended Name:

Stochastic Human Exposure and Dose Simulation Model for Wood Preservatives
Model Overview/Abstract:
SHEDS-Wood (Stochastic Human Exposure and Dose Simulation Model for Wood Preservatives) is a physically-based stochastic model that was developed to quantify exposure and dose of children to wood preservatives on treated playsets and residential decks. Probabilistic inputs are combined in physical/mechanistic algorithms to estimate exposure and dose using a two dimensional Monte-Carlo methodology that quantifies variability and uncertainty in model inputs and outputs. To date the model has focused on simulating aggregate exposures of children to chromated copper arsenate (CCA)-treated wood.
Model Technical Contact Information:
Dr. Valerie Zartarian

Dr. Haluk Ozkaynak

Model Homepage: http://www.epa.gov/scipoly/sap/tools/atozindex/sheds.htm

User Information Back to Top
Technical Requirements
Computer Hardware
PC Computer, at least 128 MB RAM, at least 500 MHz, at least 10GB hard drive, CD-ROM drive; faster processor and higher RAM recommended
Compatible Operating Systems
Windows NT, Windows 2000, Windows XP
Download Information
Documentation of model assumptions, data inputs, model outputs and analyses can be viewed and downloaded from the OPP FIFRA SAP December 3-5, 2003 web sites:


A Probabilistic Exposure Assessment for Children Who Contact CCA-Treated Playsets and Decks, Draft Preliminary Report, September 25, 2003.

Using the Model
Basic Model Inputs
CHAD time-location-activity diaries, other activity pattern information (e.g., days per year child plays on/around playsets/decks, fraction outdoor time child contacts treated wood), soil concentrations and wood surface residues, exposure factors (e.g., surface-to-skin residue transfer efficiency, frequency of hand-to-mouth contact, fraction skin surface area contacted, soil ingestion rates), dermal and GI absorption rates
Basic Model Outputs
Output options include individual daily exposure and dose time profiles, population cdfs, summary statistics tables, contributions by route and pathway, sensitivity analyses, and uncertainty analyses for predicted absorbed dose via the dermal contact and non-dietary ingestion routes for wood residues and soil concentrations.

The model offers user the options for generating correlations, multiple regressions and uncertainty analysis using multiple stepwise regression analysis techniques to identify the variables and parameters that are most influential and those that contribute to greatest uncertainty in the predicted estimates.

User Support
User's Guide Available?
The User's Guide is not yet available for the current version.

Model Science Back to Top
Problem Identification
SHEDS-Wood explicitly characterizes both the variability and uncertainty in the predicted human exposures and doses resulting from personal exposures to wood preservatives on playsets and decks via the dermal and non-dietary ingestion exposure routes.
Summary of Model Structure and Methods
Time-location-activity diaries are sampled to generate a population of simulated individuals. For each sequential location-activity combination, model inputs are sampled from probability distributions for macro- and micro-level activity data, medium-specific concentrations and residues, exposure factors, and dose factors. These are combined in physically-based equations to simulate individuals? daily exposure and dose time profiles for dermal and non-dietary ingestion exposure routes. The daily absorbed dose time profiles are aggregated across all routes and pathways. Two-stage Monte-Carlo simulation is used to estimate inter-individual variability in the population and uncertainty in estimated exposure and dose distributions.

Real-time exposure and dose time profiles based on EPA CHAD diaries are computed for the various routes considered. An algorithm, based on actual longitudinal activity data for children, using 8 CHAD diaries is applied to simulate a 1-year diary for a simulated individual. User-specified information is applied to simulate exposure days and wood/soil contact events within a day.

Dermal exposure and dose from surface residues and soil are computed for each macroactivity indoor and outdoor event in the CHAD diaries by combining surface residues and soil concentrations with factors including dermal transfer coefficients, skin surface area contacted, soil-skin adherence, and dermal absorption rates; removal from the skin is accounted for via hand to mouth transfer of residues, hand washing, and bathing using information on hand-to-mouth contact frequencies, hand washing and bathing frequency, and washing removal efficiencies. Non-dietary ingestion via hand-to-mouth contact activities are simulated using videography data for contact frequencies and available saliva removal efficiency information.

Model Evaluation
No biomonitoring studies are currently available to compare against SHEDS-Wood model predictions. Initial model evaluation has been conducted by comparing SHEDS-Wood dose predictions against other deterministic model results for CCA (e.g., CPSC, Exponent, Gradient). Although model-to-model comparison revealed that the model produces reasonable results compared to other models, model evaluation will be ongoing as new human exposure field measurements become available.

Conceptual, mathematical and chemical/physical verification has been performed by model developers. The SHEDS-Wood methodology was reviewed by the OPP FIFRA SAP in August 2002, and the refined model applied to a CCA case study. The SHEDS-Wood CCA assessment was presented to and favorably reviewed by the December 2003 OPP FIFRA SAP. The technical report for this assessment is posted on the SAP web site for that meeting. This report describes algorithms and model inputs in detail, as well model predictions for dose and results of sensitivity and uncertainty analyses to identify key inputs and parameters influencing results.

Local Navigation

Jump to main content.