4 Summary Report for Personal
J Chemical Exposure Informatics:
» VISUALIZATION AND EXPLORATORY RESEARCH
IN SIMULATIONS AND SYSTEMS (PerCEIVERS)
innovation
participants
'> O monitoring identify | patterns
too|s Theme gn categories
•esented-"3prajec'ur
, I social lJ~ activity [rr«i°p
i i «*>j&
EtnoasOR1) M^B™. , , ,,,,,
fmmunity . , development WOrKSnOD •^T2°nt
- ^ 1
arch and Developme
^Dnc sum man
Data
approaches 3 u-Sbie technology Breakout
d S fartnr'; ingredient = £hem,cal ,
sources |ractors „:,,,,„ r .9 one using
lames assessmen
-------
SEPA
EPA/600/R13/041 | June 26-27, 2012 | Research Triangle Park, NC
U ited States
En ironmental Protec ion
Agency
Summary Report for Personal
Chemical Exposure Informatics:
VISUALIZATION AND EXPLORATORY RESEARCH
SIMULATIONS AND SYSTEMS (PerCEIVERS)
Michael-Rock Goldsmith, Cecilia Tan, Daniel Chang, Christopher M. Grulke,
Rogelio Tornero-Velez, Daniel Vallero, Curtis C. Dary, Jeffre Johnson,
Peter Egeghy, Jade Mitchell-Blackwood, Kathleen Holm, Madeline Reich**,
Ryan Edwards** and Linda Phillips*
US-EPA/ORD/NERL
* US-EPA/ORD/NCEA
** Shaw Research Apprentice Program participants 2011 and 2012
Office of Research and Development
US Environmental Protection Agency, Washington DC, 20460
-------
Disclaimer
The information in this document has been funded in part by the U.S. Environmental Protection Agency. It has been subjected
to the Agency's peer and administrative review and has been approved for publication as an EPA document. Mention of trade
names or commercial products does not constitute endorsement or recommendation for use.
-------
Table of Contents
1.0 Introduction and Summary of Opening Remarks 1
Background, Purpose, Themes, and Topics of the Workshop 1
2.0 Summary of Workshop Presentations 3
3.0 Summary of Workshop Discussions and Outcomes 5
3.1 Breakout Discussions 5
3.2 Workshop Outcomes 6
Appendix A - Agenda A-l
Appendix B - Workshop Attendees, Presenters Bios, and participant contact details B-l
Appendix C - PerCEIVERS 2012 Break-Out Sessions C-l
Appendix D - Workshop Presentation Material D-l
Appendix E - WWW resources and URLs of interest E-l
-------
List of Figures
Figure 1: Presentation Roster of title slides (overview) for PerCEIVERS 2012 of all presentations 2
Figure 2: Phrasenet map of all text information from all presentations presented at PerCEIVERS 2012
(visualized in wwwjriany^yeSiCom., color for conceptual domains manually added.) 3
-------
1.0
Introduction and Summary of Opening Remarks
Background, Purpose, Themes, and Topics of
the Workshop
EPA Research Pathfinder Innovation Projects (PIPs), an
internal competition for Agency scientists, was launched
in 2010 to solicit innovative research proposals that would
help the Agency to advance science for sustainability.
In 2011, of the 117 proposals received from almost 300
scientists, 12 winners were awarded with seed funding to
pursue their creative solutions to environmental and human
health challenges. One of these projects was "Systems
Reality Modeling Project, Step 1: Chemical Inventory."
A team of nine scientists from the National Exposure
Research Laboratory (NERL) and the National Center for
Environmental Assessment (NCEA) (eight from NERL,
one from NCEA) proposed to develop novel informatics
and data curation approaches that both exposure assessment
communities and proactive members of the public may
use to become more aware of the chemicals present in our
living space and lifestyle. This awareness is a component of
"personal chemical exposure informatics."
A two-day workshop on Personal Chemical Exposure
Informatics was held on June 26 and 27, 2012, at the US
Environmental Protection Agency campus in Research
Triangle Park, North Carolina. This report details the
presentations and breakout group discussions to further
advance this particular research field and identify gaps for
additional efforts.
The three major topics in this workshop were:
1. Consumer product ingredient (chemical) informatics;
2. Real-time methods for monitoring human behavior/
activity; and
3. Increasing chemical exposure awareness through
motivational communication, curricula and
participatory outreach development.
There was a diverse set of themes at this workshop that
relates to personal chemical exposures informatics, including
but not limited to:
• Exposure concepts: near-field and far-field sources
• Current consumer/personal exposure assessment models
• Household/consumer product ingredient (chemical) data-
sources and open-access chemical information
• Human factors (activity and location pattern) data sources
and survey methods
• Social media analysis and reporting methods for
informing personal chemical exposure such as survey
methods, data-mining, passive inquiry, and privacy/ethical
considerations
• Data mining in support of Personal Chemical Exposure
Informatics
• Gamification (application of game design thinking to
non-game purposes) of exposure concepts and consumer
product exposure modeling with emphasis on curriculum
development for modeling processes and variables
associated with personal exposure
• Personal informatics tools development for chemical
exposures or "chemically-related" decision processes
• DIY community for tools/methods/apps development for
monitoring personal chemical exposures - crowdsourcing
initiatives and open-data exchange
• Novel data visualization/analytic representations
(communications/outreach and rendering of information)
for communities/consumers and educators/DIYers for
personal chemical exposure awareness
• Data/effort sharing Personal Chemical Exposure and
Personal Informatics (life-logging and life-streaming
concepts)
Workshop Organization
A total of 27 presenters and 43 workshop participants
attended (both physically and remotely) the PerCEIVERS
workshop (Agenda see Appendix A). The presenters came
from a variety of research disciplines/interests (i.e., exposure
and dosimetry modeling) and types of organizations - i.e.,
private industry, non-profit organizations (creative arts,
science advancement [e.g., SHODOR and NC Life &
Science Museum], and academia), and government agencies
(e.g., US EPA, USD A, NLM, and CPSC) - to discuss and
explore strategies for filling gaps in data, modeling, and
communication in the greater realm of personal chemical
exposure informatics.
This workshop was held to stimulate development and
promote collaboration on the use of ubiquitous computing
devices (smartphones, tablets, notebooks) by individual
modern consumers as a means to identify chemical exposures
-------
25
Figure 1: Presentation Roster of title slides (overview) for PerCEIVERS 2012 of all presentations.
arising from their personal actions (activities and location)
and the consumer products they use, as inspired by the
Systems Reality pathfinder innovation project.'
Another objective of the workshop was to investigate
personal chemical exposure informatics in the broader
context of exposure types and sources. This context includes
strategies for communicating personal chemical exposure
awareness through a variety of venues: from product bar-
code scanning, to activity/location (behavioral) recording via
smart-journaling or passive network monitoring, to scenario
exploration. Presenters and participants delved into the
challenges of describing human-product-chemical systems
and modeling their interactions in our everyday life.
There were seven sessions with a total of 27 presentations,
Additional information:
(a) http://www.epa.gov/ord/sciencematters/iune2011/
pathfinder.htm
(b) http://www.epa.gov/ord/sciencematters/iune2011/
innovation.htm
(c) http://www.epa.gov/sciencematters/december2011/
executivemessage.htm
(d) http://www.epa.gov/heasd/research/srm.html
of which 8 were performed remotely over multiple time-
zones (Berlin Germany, North Carolina, and San Francisco).
Streaming live-video was available, and Adobe Connect
was used for remote slide-casting. All presentations were
annotated on the fly and video-recorded for later analysis.
A full list of attendees and their biographical information,
as well as the biographies of the presenters, can be found in
Appendix B.
-------
2.0
Summary of Workshop Presentations
A list of presentation titles, presenter affiliations, and
session chairs is available in Appendix A. In addition, the
presentations can be viewed in their entirety in Appendix D.
The text from all 27 presentations was placed into many-eyes
(www.many-eyes.com) to construct a visual synopsis of the
PerCEIVERs workshop, providing a visual representation of
the major themes of the workshop (see Figure below). The
major themes have arbitrarily been colored:
• Red - Supporting exposure assessment (modeling and
informatics/knowledge-based approaches)
• Blue- Understanding personal chemical exposure from
the use of consumer products
• Green- Identifying information gaps pertaining to (a)
chemical ingredients of products; (b) human time-location
activity data; and
• Yellow: (c) product use information.
There were seven sessions in the workshop, and each session
had three to five presentations.
1. Overview and innovation
In this session, Dr. Peter Preuss (EPA) presented the EPA
Office of Research and Development's (ORD) innovation
strategy and gave an overview of the Pathfinder Innovation
Projects (PIPs). Dr. Michael-Rock Goldsmith (EPA) and
Mr. Ryan Edwards (Shaw University Summer Research
Internship 2011) gave an overview of the Systems Reality
Modeling (SRM) project, which was one of the 12
awarded PIPs. The SRM project is a multi-part project that
characterizes an individual's consumer product inventory,
links consumer products to chemical ingredients, collects
individual time/location activity patterns, and simulates
personal chemical exposures. Ms. Madeline Reich (Shaw
University Summer Research Internship 2012) used several
case studies to demonstrate, as part of the SRM project,
how social network analysis can be used to obtain human
time-location activity patterns passively.
hazard
\ _. location
information
ingredients
environment key-
studies
^—• health
consumer
* sourdes
^*~-/< \|
> f*^X- household ^chemicals
people based
human search
\r
assessment
modeling
factors
Figure 2: Phrasenet map of all text information from all presentations presented at PerCEIVERS 2012 (visualized in www.manv-eves.com, color for
conceptual domains manually added.)
-------
2. Exposure-based chemical prioritization
In this session, Dr. Elaine Cohen-Hubal (EPA) presented the
basic concepts of exposure science, and the framework of
ExpoCast™, which is a tool for chemical prioritization based
on exposure potentials. Dr. John Wambaugh (EPA) presented
an approach, in ExpoCastTM, which uses production volume
to predict exposure potentials. Dr. Jade Mitchell-Blackwood
(USDA) introduced a "Multi-Criteria Decision Modeling"
framework for prioritizing chemicals for more targeted
testing based on chemical properties and life cycle
considerations.
3. Exposure factors and informatics
In this session, Dr. Linda Phillips (EPA) introduced the EPA's
Exposure Factor Handbook and its Consumer Products Use
Data. Dr. Kristin Isaacs (EPA) described the critical attributes
of human activity patterns and their uses in exposure
assessment and exposure modeling. She also presented
current databases, tools, and projects on collecting and
analyzing human activity pattern data. Dr. Deborah Bennett
(University of California, Davis) presented her study on
using bar code scanner and motion sensors to evaluate the
use of personal and household care products with minimal
burden to the study participants. Mr. Michael Keating
described how smart phones can be used as a flexible
research tool to conduct surveys, obtain micro-level location
data, or collect data on air quality, physical activity, etc.
Dr. Curry Guinn (The University of North Carolina at
Wilmington) presented their autocoding program that maps
the text of voice diaries to the EPA's Consolidated Human
Activity Database (CHAD).
4. Consumer exposure models
In this session, Ms. Cathy Fehrenbacher gave an overview
of the current exposure tools and models used by the
EPA's Office of Pollution Prevention and Toxics (OPPT).
Dr. Christina Cowan-Ellsberry (The Lifeline Group)
presented the three types of information needed for exposure
assessment: chemical-specific, product-specific, and people-
specific information, as well as example sources for these
data. She also introduced probabilistic models to estimate
aggregate exposures. Selection of distributions for model
parameters was also discussed.
5. Chemical information for consumer products
In this session, Dr. Antony Williams (Chemspider at
Royal Society of Chemistry) gave an overview on
the free chemical database, ChemSpider. Dr. Henry
DeLima (householdproducts.nrm.nih.gov) introduced
the Household Products Database which contains over
12,000 consumer products in nine product categories.
Dr. Rogelio Tornero-Velez (EPA) presented the idea that
human activities and chemical use patterns are not random
events. Dr. Treye Thomas (US Consumer Product Safety
Commission) gave an overview of the mission and the type
of research conducted by the US Consumer Product Safety
Commission.
6. Participatory methods and personal informatics
In this session, Dr. Bill Pease (GoodGuide.com) introduced a
web-based platform (GoodGuide.com) that tracks the health,
environment, and social aspects of products, brands and
companies, for helping buyers to make purchase decisions.
Mr. Michael Nagle (OuantifiedSelf.com and the Sprouts.
org) introduced the Quantified Serf which is a collaboration
of users and tool makers who are interested in self-tracking,
such as exercise, diet, or sleep, using computers, phones,
and other methods. Dr. Michael Breen (EPA) presented
his research on using Microenvironment Tracker to record
time and duration people spent in microenvironments for
supporting exposure modeling. Ms. Shannon O'Shea (EPA)
introduced the EPA's Community-Focused Exposure and
Risk Screening Tool (C-FERST) which is a community
mapping and assessment tool to support decision making.
7. Engaging the community for personal chemical
exposure informatics: gamification, visual and computer
models, electronic and live-action role play
In this session, George Scheer (elsewhere) presented on the
artists in residence community and living museum known
as 'elsewhere' (http://www.goelsewhere.org/). where the
public and artists mingle (live action role play) and explore
our relationships with everyday things. Dr. Benjamin Balak
(Rollins College) discussed the current stasis of pedagogy
in our school systems, and presented on his use of games to
teach economics and game theory, how games enable so-
called ADD kids to focus and learn for hours at a time, and
the motivation for gamifying disciplines such as exposure
science. Dr. Robert Panoff (Shodor) presented on the benefits
of visual dynamic learning and agent based simulation
approaches developed at Shodor (http://www.shodor.org/).
and why computational chemistry needs to be within the
reach of high school students. Bech Tench (Museum of
Life and Science) spoke of the Experimonth sessions at the
Durham museum; one monthly session involved exploration
of prisoner's dilemma, another involved rhetorical analysis of
texts among an online group to identify markers of learning
(evidenced by changes in text sentiment).
-------
3.0
Summary of Workshop Discussions and Outcomes
3.1 Breakout Discussions
The breakout sessions for PerCeivers mirrored the overall
workshop themes (see Appendix C). A total of five themes
(Themes A-E) with some intentionally open-ended questions
were provided (see below) to the workshop participants to
facilitate free discussion on these topics. A summary of the
questions and discussions in each of the five Breakout Group
Discussion Themes (A through E) are provided below.
Additional web sites mentioned throughout the course of this
exercise are provided in Appendix E.
THEMEA.
Exposure factors informatics, e.g., real-time human factors
Discussion summary:
• How to make data available?
o First, identify the information needed. For example,
identify the important parameters needed to properly
describe an exposure scenario.
o A lot of data are available already. We may get those
data from structured querying or asking around (e.g.,
manufacturer or marketing group)
oOne needs to be cautious about the accuracy of the
acquired data. Also, these data often lack an estimate
of variability.
o Should we reward companies for being more open by
providing data?
• How to conduct inventories and estimate exposures?
o Linking products to exposure scenarios for estimating
exposures.
o Frequency and duration information needs to be
collected, but it is difficult to collect such information
for certain products.
oHow representative is a generic product that is linked
to an exposure scenario?
THEME B.
Consumer/personal exposure modeling
Discussion summary:
• How to protect private information?
oNot all "unprotected" information is public
information. How do we identify which information
we can use without consent?
o There are many restrictions for government agencies
to collect or use private information (e.g., Privacy Act
of 1974), but those restrictions may not apply to other
non-government agencies. Can government agencies
still use the results of these findings?
oHow reliable are the data collected from social
networks?
o There are many methods to aggregate and anonymize
data collected from individuals (e.g., Silent Spring
study).
oDo "observations" count as "surveys" (e.g., counting
numbers of joggers in 10 min in different seasons)?
• What are the innovative methods to discover trends in
human activity?
o One use for aggregated and anonymized data is to
discover trends of human activity over time, across
populations for comparison.
oThe North Carolina Museum of Life and Science
collects data from attendees to generate hypotheses
for more targeted research.
o Some example tools for discovering trends include
Google, alpha engine, RADIANG, SPSS Text
analytic, wikigroup, and concept maps.
is a data-driven social
networking site that collects disease, symptom, and
treatment information from members.
o Can Google popularity assessment be used to rank
product use in a population? Currently, there are
databases on products, but no analysis on ranking the
uses of different products (e.g., by weight purchased
per year?).
o Challenges with collecting, curating, storing,
searching, sharing, analyzing, and visualizing
Big Data.
THEME C.
Chemical informatics for consumer products, i.e., consumer
product chemical ingredient data sources, data mining,
analytics, and visualization
Discussion summary:
• What categories of products are important in terms of data
needs?
o Articles, such as furniture, appliances, and
non-household products (e.g., building materials)
oToys
• Are there any chemicals/products of emerging concern
or lifestyle changes for which consumer use information
does not exist?
oBrominated flame retardants
o Electronics (e.g., cellphones, tablets)
oNanomaterials
-------
o Lifestyle change: the advancement of technology
makes people stay indoor longer.
• What can be done to make product/chemical information
more easily attainable?
o Voluntary vs. regulatory (State or Federal)? Purchase
information from industries?
o May get access to data through 3rd party certification
groups (e.g., Green Guard)
THEME D.
Participatory methods and personal informatics tools, e.g.,
social media mining/analysis and reporting, DIY community
Discussion summary:
• When using personal monitoring or social media to gather
consumer use data, what are the data biases (e.g., groups
that have no access to/interest in social media)
o The types of biases of interest depend on whether we
are analyzing existing data or designing new studies
o Targeting vs. self-selection
o Lower socioeconomic classes are typically under-
represented. For these classes, we may provide them
access to technology (e.g., cellphones, internet access)
o Older generations are typically under-represented
when analyzing social media data.
o App use varies widely by age ("App use" covers
the number of apps, the identity of apps, and the
categories of apps that are used by different age
groups)
o Based on the methods used, can we identify the
population represented?
o Can social media tell us what is really happening?
o Is knowledge gained from old-fashioned surveys still
applicable?
o May be difficult to collect information from one
individual for a long period of time. Sometimes, the
best approach may be combining results from several
people to represent one scenario.
• How to engage DIYers to develop an app or game to
collect product/chemical use data and/or product/chemical
inventories?
o Government "challenges"
o Look for those who are already at stake
o Examples of PatientsLikeMe.com. OuantifiedSelf.
com and curetogether.com. People like to compare
themselves to others in a community.
o Examples of shopper cards or loyalty programs.
• What are the elements to make a tool, game, or learning
module ideal for combining consumer use data with
household product inventories?
o Pitch the tool, game or learning module as something
fun or useful to the users, not as a way to get
something from the users.
o Fun, low burden, visual.
• Are there privacy concerns? What is the public health
message that can be sent?
o If the purpose is behavior modification, then it is
"Big Brother".
• "Opt in" vs. "Opt out"
o Must consider biases
• Aggregating data may be needed to protect privacy,
but the utility of these data may be limited.
THEME E.
Data visualization and analytics for engagement and
communication, e.g., gamification and computational
simulations, novel data visualization/analytic representations,
sharing personal chemical exposure informatics
Discussion summary:
• Need to provide incentives to motivate participation.
• A video of walking through the house may be a better
survey than answering questions online.
• What is the best point of data collection (e.g., checkout
line at the supermarket)?
• Forming groups to build the tool together.
• Government monitoring data, such as activity mapping
or exposure factors, can be useful to parameterize the
models.
• People generally want to know more about themselves,
so setting goals or competing in a game will get them
interested.
• One idea: scan the barcode of a product and the product
becomes a character in a game. Players can fight against
each other with their "products" and the most "toxic"
one wins.
3.2 Workshop Outcomes
Introductions The opening session of the workshop
introduced some of the newer areas of innovation at the
US EPA related to the pathfinder innovation projects in OPJ).
The natural segue was to bring up the concept of systems
reality modeling and how to model the various facets of
personal chemical exposure using chemical ingredient
profiles of everyday household products, to understanding or
developing passive interrogation methods in order to capture
human-activity patterns with unprecedented temporal-spatial
resolution using geo-coded and seasonal big-data streams
and clever data mining techniques. This introductory section
brought forth the PIP innovation projects (Preuss), the
SRM PIP1 (Goldsmith) and provided two excellent examples
of sub-domains of this research performed by exceptionally
gifted highschool students (Edwards & Reich).
Presentation Session 2 focused on different prioritization
approaches on chemicals based on their exposure potentials.
It was suggested that the most critical outcome would be a
fundamental transformation in exposure science to realize the
NRC vision for toxicity testing, in which human exposure
information is critical for guiding the development and use
of toxicity information. Two examples were provided to
-------
demonstrate the beginning of such a transformation. One
example showed that despite very large uncertainties in the
exposure estimates, it is still possible to isolate a group of
chemicals with the highest predicted exposure potentials
and target those chemicals for further assessment. The
other example showed that a variety of exposure metrics
ranging from physical-chemical properties to socioeconomic
measures can be efficiently and intelligently combined to
inform screening level risk estimates.
Presentation Session 3 considered various means of
quantifying and characterizing the factors that drive exposure,
particularly information technologies that may be used to
support computational exposure sciences. The session began
with a discussion of recent updates to EPA's Exposure Factor
Handbook (EPA, 2011) with an emphasis on the handbook's
Consumer Products Use Data. The critical attributes of
human activity patterns and their uses in exposure assessment
and exposure modeling were discussed, along with currently
deployed tools and databases. Ongoing EPA projects aimed
at improving the quality of EPA's data on human activity
patterns were discussed, including better means of collecting
and analyzing human activity information. Other informatics
topics were covered, including a recently completed study
that employed bar code scanners and motion sensors to
estimate how personal and household care products are
actually used. The study also endeavored to find methods
that can reduce burdens on participants. Recruitment and
retention of subjects and participants is crucial in light of the
paucity of reliable use, habit, and practice data for products.
Smart phones were discussed as promising research tools that
may provide flexible means of conducting surveys, gathering
finely-textured location data, and collecting co-incidental
environmental and physiological information, e.g. air quality
and physical activity, respectively. The session wrapped up
with a discussion of operational program for autocoding
i.e. mapping text of voice diaries to the EPA's Consolidated
Human Activity Database (CHAD). Reference: U.S. EPA,
2011: Exposure Factors Handbook, National Center
for Environmental Assessment, Office of Research and
Development, EPA/600/R-09/052F
Presentation Session 4 outlined the current standard methods
used for exposure assessments. The ideas of generic
product categories and exposure scenarios were outline
while commonly needed inputs for assessments such as
production volume were catalogued. The need for extensive
parameterization of exposure models was highlighted
leading to the identification of the need for novel methods
of data collection. Several possible solutions were detailed
in the discussion afterward including the identification of
trade organizations as potential data providers and greater
public/private communication to enhance the collection of
relevant consumer product use data.
Presentation Session 5 highlighted the use of extant and
emerging technologies in chemical-based exposure risk
assessments related to consumer products. It was widely
recognized that chemical databases are important assets
that can be readily utilized to ascertain chemical specific
data. However, there is an important continued effort
to reduce ambiguity in chemical structure-related data
(i.e., incorrect stereoisomers). One method proposed was
by adopting more open and definitive standards such as
InChi codes - i.e., lUPAC's textual identifier for chemical
substances. Furthermore, product categories and chemical
composition data should be leveraged in chemical risk
assessment to reduce the amount of chemical information
required. However, existing databases are limited where
information regarding articles such as furniture, appliances,
and non-household products (e.g., building materials) are
incomplete or nonexistent. It was proposed that industry or
third party (i.e., NGO) collaboration would be a necessary
means to elicit data on certain products and this topic
was explored further within Breakout Theme C. Others
recognized that human factors and specific use patterns can
drive the need for identifying a subset of chemical substances
(i.e., non-random, co-occurrence of chemicals) and
thereby restrict the chemical information/testing/modeling
requirement. Integrated testing of consumer products was
highlighted as well demonstrating the clear need for chemical
information in narrowing existing data gaps.
Presentation Session 6 explored participatory methods
and personal informatics with the goal of developing an
app for engaging the public to examine their personal
chemical exposure to consumer products, and the option
of giving exposure scientists access to their personal data.
Good Guide is an example of a similar type of app, with
years of point-of-purchase data that is possibly available
to the Systems Realty research team. Another presentation
demonstrated how GPS technology, which is often built into
smart phones, can be used to characterize activity/location
information. Another outcome from this session was a
discussion of how to engage interest in an individual's own
personal informatics, and ways to facilitate communication
between the EPA and communities to effectively
communicate accurate information (e.g. c-FERST).
Presentation Session 7 investigated applications in
'gamifying' problems, with the expectation that this approach
may be useful in addressing the problems of engaging the
community to learn about chemical exposures. Gamification
is the application of game elements and game design
techniques to help solve real-world problems. The main
outcome was the discussion and suggestion of different
gamification strategies, including role playing, computer
simulation, and online social games. Formulating problems
as games also has the benefit of engaging participants
from diverse backgrounds who would otherwise not have
participated.
Breakout Theme A discussed the use of innovative, social
media-based methods for collecting exposure factor
information. The outcome was an increased recognition
by the attending EPA and academic scientists regarding
the potential for the development of rapid, deployable,
inexpensive, reproducible, and validated methods to collect,
mine, analyze, and disseminate exposure factor data from
social media data streams. Some of these methods have
-------
subsequently been demonstrated through the development of
multiple EPA PIPS proposals, as well as other proposed CSS
research.
Breakout Theme B acknowledged the challenges, such as
privacy concerns or issues with gathering data from a large
population, in obtaining relevant chemical use and human
behavior data needed for more detailed exposure modeling.
Besides new ideas to tackle these challenges, another
outcome was the awareness of multiple sources of databases
that currently exist and may be used to refine our knowledge
regarding chemical uses and human behaviors.
Breakout Theme C was primarily concerned with the
accessibility and usability of data relating to consumer
product ingredients and chemical informatics (i.e., from
both sources of curated as well as raw data). Questions were
raised about public sources of easily attainable ingredient
information, portals for public/regulatory/industrial exchange
about emerging concerns for emerging products, and
product ingredient prioritization (i.e. prioritization based
on actual consumption scaling). Data on articles such as
furniture, appliances, non-household products (e.g., building
materials) and toys (early life-stage) are not currently well
documented, captured or consolidated. It was also mentioned
that material additives, such as flame retardants or stain
repellants (i.e. brominated flame retardants), components
of commonplace objects (i.e. cellphones, tablets, and other
consumer electronics), and nanomaterials have not received
enough documentation, despite their ubiquitous use.
There are several ways that lists of ingredients present in
consumer products may be obtained. Participants discussed
the differences in public access to data that is voluntarily
disclosed vs. that which is mandatorily disclosed as a result
of government regulation. They also discussed what types
of data may be purchased directly from industry or via third
parties (e.g. Green Guard).
Breakout Theme D discussed the use of participatory methods
and personal informatics tools, rather than traditional
monitoring methods, to obtain exposure information. The
outcome was a greater awareness among participants of
potential biases when relying on "ubiquitous" technology
(e.g., cell phones, social media) to generate exposure
information. For example, older segments of population
and those within lower socioeconomic strata are likely to
be underrepresented. With this knowledge, efforts should
be made to increase representation of such individuals by
providing access to technology, along with appropriate
training. Another outcome from this discussion was a
possible strategy for developing a platform for individuals
to voluntarily provide information on their own habits as
consumers. Such a platform should allow people to compare
themselves to others in a community in a fun, low-burden, yet
highly visual way. Moreover, the purpose should be behavior
observation rather than modification, and privacy protection
(through data aggregation) should be explicitly guaranteed.
Breakout Theme E discussed the variety of ways data
visualization/analytics, aggregation, and sharing for the
purpose of engagement and communication with the public
could be accomplished (e.g., gamification and computational
simulations, novel data visualization/analytic representations,
and sharing personal chemical exposure informatics).
Questions were discussed regarding how to engage the public
on the issues of chemical exposure, motivate them to share
or add to data efforts, encourage them to participate in games
developed to emphasize the need for exposure research,
and communicate to them the fundamental relationship
between exposure and risk in a way that is understandable
by the lay individual. One of the many discussion points
on how to improve in these areas was the need to provide
incentives to motivate participation (although specifics
on how this would be implemented were not discussed).
The capabilities of current technology can be harnessed in
creative ways to provide information that is insufficiently
captured by traditional approaches. Instead of performing
a question-based survey to catalog consumer products in
a home, a participant could take a video with their phone
while walking through their house instead. The groups also
discussed alternative points of data collection (e.g., the
checkout line at the supermarket), or forming DIY and serf-
help interest groups to build tools that emphasize personal
chemical exposure together. The group found that the concept
of passive monitoring data, such as activity mapping or
exposure factors, can be useful to parameterize the models.
A concept of developing a game for informing oneself of
household product dangers came to light - One idea: scan the
barcode of a product and the product becomes a character
in a game. Players can fight against each other with their
"products", and the most "toxic" one wins. If young people
play the game they might engage older family members to
participate, raising awareness and expanding the domain of
research.
Conclusions and Next Steps The workshop explored and
highlighted a variety of extant technologies, methods and
innovations from a variety of scientific and non-scientific
disciplines that could be used to inform personal chemical
exposure. The diversity of interests and expertise provided
an unprecedented potential for innovative and collaborative
opportunities for scientific and technological advancement,
as well as outreach and community based involvement.
Both within and outside the Agency, current efforts are
underway to capitalize on the outcomes by developing and
strengthening many of the innovative initiatives identified.
Examples include:
• Data gap identification: Article and household
furnishing and materials as sources of exposure, as
well as key additives added to such products (such as
flame, stain, and water resistant additives added to many
commercial articles and household goods).
• Information accessibility related to consumer product
ingredients and constituents was a key area of interest,
as the public and regulatory agencies alike would
-------
like streamlined portals and dashboards that can give
them ingredient and material composition through
consolidated web-accessible databases.
Methods to capture "big data" related to exposure
factors and exposure modeling as a whole.
Approaches to more holistically integrate disparate
datasets related to exposure assessment calculations and
models.
Methods to gamify (developing a game that would
inform the player about personal chemical exposures)
and develop platforms that engage communities and
individuals to compare exposure sources and "share" (for
instance via social networks) their findings
Several of the highlighted outcomes are being developed
further among workshop participants as complete research
proposals that are currently being considered for funding
(i.e., EPA Pathfinder Innovation Projects). The ideation
process exploring the variety of topics within the workshop
is anticipated to give rise to several manuscripts and other
future collaborative initiatives for years to come.
-------
Appendix A - Agenda
Personal chemical exposure informatics: visualization, user Experience, Research in Systems modeling and
Simulations (PerCEIVERS)
ROOM C111C, US Environmental Protection Agency
Research Triangle Park, NC
Remote Call-in: 1-866-299-3188
Call-in Code: 919-541-1021
June 26-27, 2012
Day 1 - Tuesday, June 26,2012 (8:30 am - 5:00 pm)
WebConference Link, Day 1: https://epa.connectsolutions.com/perceiversdav1/
Session 1: Overview and Innovation - Chair: Linda Sheldon
8:30 - 8:45 Welcome and Opening Remarks - Peter Preuss, EPA Office of Research and Development (ORD) Chief
Innovation Officer
8:45 - 9:05 Personal Chemical Exposure Informatics - Rocky Goldsmith, EPA National Exposure Research Laboratory
(NERL)
9:05 - 9:20 SRMstepping closer to the vision!- Ryan Edwards, NCSU undergraduate/SHAW SRP
9:20-9:35 Search Terms + Tweets = Exposure Informatics 2.0? - Madeline Reich, SHAW SRP
Session 2: Exposure-Based Chemical Prioritization - Chair: Peter Egeghy, NERL US EPA
9:35-9:55 Does exposure imitate art? Recent impressions - Elaine Cohen-Hubal, EPA National Center for
Computational Toxicology
9:55 - 10:10 ExpoCast High Throughput Exposure Potential Prioritization - John Wambaugh, EPA National Center for
Computational Toxicology
10:10- 10:25 From Decision Analytics for Exposure Prioritization to dietary residue exposures - Jade Mite he 11-
Blackwood, US Department of Agriculture, Food Safety and Inspection Service
10:25-10:45 break
Session 3: Exposure Factors and informatics - Chair: Dan Vallero, NERL US EPA
10:45 - 11:00 Exposure Factors Handbook Consumer Products Data- Linda Phillips, US EPA, NCEA
11:00 - 11:15 Human Activity Data in Exposure Assessment - Kristin Isaacs, US EPA, NERL
11:15-11:30 SUPERB and Passive sampling methods - Deborah Bennett, UC-Davis
11:30-11:45 Smartphones as a Flexible Research Tool: Lessons from Early Implementations and the Consumer
Marketplace- Michael Keating, Research Triangle Institute
11:45 - 12:00 Natural Language Processing and human activity patterns - Curry Guinn, UNC Wilmington
12:00-1:15 lunch (on your own, Lake Side Cafe recommended)
Session 4: Consumer Exposure Models - - Chair: Chris Grulke, NERL US EPA
1:15 - 1:30 Consumer Exposure Assessment in aRegulatory Context under TSCA- Cathy Fehrenbacher, EPA Office of
Chemical Safety and Pollution Prevention
-------
1:30- 1:45 Chemical Use: The Key to Near-Field Chemical Exposure Estimation- Michael Jayjock, The Lifeline
Group
1:45 - 2:00 Probabilistic Exposure Assessments for Consumer Products -Christina Cowan-Ellsberry, The Lifeline
Group
Breakout Discussion Session I
Attendees will be assigned to breakout groups based on their interests and adequate coverage of themes
2:00 - 3:30 Breakout Group Discussion
Themes:
1. Exposure factors informatics, e.g., real-time human factors (Room C400A)
2. Consumer/personal exposure modeling (Room C111C)
3:30-3:45 Break (rapporteurs and chairs to prepare reports)
3:45 - 4:10 Presentation by rapporteurs (10 minutes each)
4:10-4:45 Joint Discussion
4:45 - 5:00 Preview of next day's meeting
5:00 Meeting adjourns (shuttle to return to hotel)
6:15 pm Option to meet for Dinner at "Mez Contemporary Mexican Restaurant" (5410 Page Road, Durham: (919)
941-1630)* Please note that this is serf-purchase dinner. We have made reservations to accommodate up to
25 people.
Day 2 - Wednesday, June 27, 2012 (8:30 am - 3:30 pm)
WebConference Link, Day 2: https://epa.connectsolutions.com/perceiversdav2/
Session 5: Chemical Information for Consumer Products - Chair: Danny Chang, NERL US EPA
8:30-8:45 ChemSpider—A crowdsourced community environment for hosting and validating chemistry data-
Antony Williams, Chemspider at Royal Society of Chemistry
8:45-9:00 Household Products Database and tools for consumers- Henry DeLima, DeLima Associates (with
contributions from Pertti Hakkinen from the National Library of Medicine)
9:00-9:15 Birds are Cool, Ecologists got it going on! -Mike Tornero, EPA National Exposure Research Laboratory
9:15-9:30 Human Exposure Assessment Strategies for Consumer Pro ducts - Treye Thomas, US Consumer Product
Safety Commission
Session 6: Participatory Methods and Personal Informatics - - Chair: Kathleen Holm, NERL US EPA
9:30-9:45 Consumer Empowerment atpoint-of-purchase - Bill Pease, GoodGuide.com
9:45-10:00 Producing and Promoting Personal Informatics - Michael Nagle, OuantifiedSelf.com and theSprouts.com
10:00-10:15 GPS and Exposure Assessment for Individuals- Michael Breen, EPA National Exposure Research
Laboratory
10:15-10:30 Getting Chemical Ideas and Information to Communities- Shannon O'Shea and Brad Schultz, EPA
National Exposure Research Laboratory
10:30-10:45 break
Session 7: Engaging the community for personal chemical exposure informatics: Gamification, visual and computer
models, electronic and live-action role play - Chair: Mike Tornero, NERL US EPA
10:45-11:00 Teaching and learning about oneself through gaming: Education 2.0 -Benjamin Balak, Rollins College
11:00-11:15 Computational Thinking in Chemistry: Dynamic Variations for Visual Learning -Robert Panoff, SHODOR
11:15-11:30 Evidence of learning in online social environments - Beck Tench, NC Life & Science Museum
11:30-11:45 Live-Action Role Play (LARP) with every-day stuff - George Scheer, Elsewhere Collaborative
11:45-1:00 lunch (on your own, Lake Side Cafe recommended)
Breakout Discussion Session II
Attendees will be assigned to breakout groups based on their interests and adequate coverage of themes
-------
1:00 - 2:30 Breakout Group Discussion
Themes:
1. Chemical informatics for consumer products, i.e., consumer product chemical ingredient data sources,
data mining, analytics, and visualization (Room C300C)
2. Participatory methods and personal informatics tools, e.g., social media mining/analysis and reporting,
DIY community (Room C111C)
3. Data visualization and analytics for engagement and communication, e.g., gamification and computational
simulations, novel data visualization/analytic representations, sharing personal chemical exposure
informatics (Room C111C)
2:30- 2:45 Break (rapporteurs and chairs to prepare reports)
2:45 - 3:00 Presentation by rapporteurs (5 minutes each)
3:00 - 3:30 Joint Discussion
Session 8: Summary and Closing
3:30 - 4:00 Summary and Future Work
4:00 - 4:15 Closing Comments - meeting adjourns
-------
Appendix B
Workshop Attendees, Presenters Bios, and
participant contact details
Presenter Biographies
Dr. Peter Preuss is the Chief Innovation Officer in the Office
of Research and Development (ORD), US EPA. Dr. Preuss
leads an interdisciplinary team charged with building an
innovation infrastructure for science that will move US
EPA forward on the path to sustainability. In their first year,
Dr. Preuss and his team have already introduced several
innovative ideas and approaches to ORD, including the use of
collaborative platforms for research planning and competitive
internal awards to promote high-risk, high-reward research.
The team has established a cross agency innovation
workgroup to help US EPA make effective use of open
innovation challenges, prizes and awards delegated under the
America Competes Act. Additionally, the team has launched
an environmental pavilion on IraroCeittive.CQm; a company
that specializes in open source innovation for scientific and
technical challenges. Currently Dr. Preuss and team are
working closely with ORD's National Program Directors
on high profile signature projects oriented around topics
such as sustainable alternatives to toxic chemicals and net
zero structures and communities. As his team endeavors
to promote new air monitoring sensors and applications to
enhance citizen science and citizen empowerment, Dr. Preuss
continues to work to bring innovative science and technology
research to the forefront of ORD's activities.
Dr. Michael-Rock ("Rocky") Goldsmith is a principal
investigator and Physical Research Scientist in the
Exposure Dose Research Branch (EDRB) of the National
Exposure Research Laboratory (NERL) at the US EPA.
Prior to joining the Agency in 2006, Rocky had worked
in pharmaceutical, tobacco and explosives industries in
R&D and product development, and completed his Ph.D.
in theoretical chemistry at Duke University in 2005. Since
2006 he has been active in six (6) major themed areas
of research that span molecular (1-3) and macroscale
(4-6) modeling in support of modern risk assessment of
environmental chemicals: (1) In silico / computational
modeling: (a) approaches for parameter estimation to support
the development of pharmacokinetic models, (b) screening-
level and provisional forward and reverse dosimetry models,
(c) model development for linking exposure to internal
dose (tissue dose, or biomarkers of exposure), specifically
for characterization of absorption, distribution metabolism
and elimination (ADME). (2) Ongoing research on in silico
chemical genomics methods to complement the IVIVE (in
vitro to in vivo extrapolation) paradigm for systems modeling
and toxicogenomics efforts in modern risk assessment (3)
Bringing stereochemistry and its implications on quantitative
risk assessment of racemates and effects research to
the forefront of critical scientific awareness: misguided
mixtures research (4) Development of low-burden, personal
household chemical exposure informatics "apps" to better
understand chemical exposure arising from the "things we
expose ourselves to", coupled to comparative analytics
to social streams of one's nearest neighbor for exposure
scenario simulation using life simulation strategy engines.
(5) The development of novel robust and rugged low-cost
remote sensing technologies using in vivo assays, and (6)
Quantitative mapping, visualization and modeling using
uncommon medium and unconventional technology from
3D anaglyph images, PS3 desktop supercomputing, poster
with a digital screen add-on, and tablet device (iPod Touch)
household product scanning and actigraphy integration.
Mr. Ryan Edwards recently graduated Southeast Raleigh
High School. He is in the honor roll, the national honors
society, the Technology Student Association, and the
FIRST robotics club. This past summer, he worked at the
Environmental Protection Agency where he helped research
the possible ways to monitor daily human exposure to
various products and chemicals, and aided in the debugging
and development of an MSDS database creation tool. He
is an eagle scout, and a black belt, having participated in
both activities for over ten years. After earning his black
belt, he now acts as an instructor to both children and
adults, and he sometimes helps out in the special needs
classes. Ryan hopes to go into a career in robotics, and
plans to go to North Carolina State University this coming
fall. Besides building several robots with his robotics club,
and participating in several engineering classes, Ryan has
also participated in an independent study course where he
researched humanoid robotics and constructed a robotic hand.
Ms. Madeline Reich is a rising senior at Fuquay Varina
High School and is an Apprentice with the Shaw University
and US Environmental Protection Agency's Research
Apprenticeship Program. She is spending six weeks at
the RTF Environmental Protection Agency to work on
"Social Network Analysis for Personal Chemical Exposure
Informatics" with her mentors Michael Goldsmith, Daniel
Chang, and Chris Grulke.
Dr. Elaine Cohen Hubal is currently a senior scientist
in US EPA's National Center for Computational
Toxicology (NCCT). The NCCT has a mission to integrate
modern computing and information technology with
molecular biology to improve Agency prioritization of
data requirements and risk assessment of chemicals.
Dr. Cohen Hubal leads ExpoCast, the EPA research program
in exposure science to support chemical prioritization
-------
and toxicity testing. Her primary research interests are in
characterizing human exposure and developing approaches
for using human exposure metrics to inform health studies
and public health policy. The current focus of her research
is on applying a systems approach to characterize complex
relationships between environmental factors and health
outcomes with an emphasis on vulnerable populations.
Previously, she was Acting Associate Director for Human
Exposure Modeling in the Human Exposure and Atmospheric
Sciences Division of the US EPA's National Exposure
Research Laboratory (NERL) where she worked to develop
and direct NERL's human exposure modeling research
program. Dr. Cohen Hubal has published in the areas
of children's exposure and human health risk modeling.
Dr. Cohen Hubal has served as an expert on a variety of
scientific panels and committees including the Voluntary
Children's Chemical Evaluation Program (VCCEP) Peer
Consultation Panel and the Study Design Working Group
for the National Children's Study. Currently, she serves
as chair of the WHO IPCS working group on "Identifying
Important Life Stages for Monitoring and Assessing Risks
from Exposures to Environmental Contaminants." Dr.
Cohen Hubal also serves as an associate editor for reviews
for the Journal of Exposure Science and Environmental
Epidemiology. Dr. Cohen Hubal received herPh.D. and
M.S. in Chemical Engineering from North Carolina State
University and a B.S. in Chemical Engineering from
Massachusetts Institute of Technology.
Dr. John Wambaugh is a Physical Scientist with the United
States Environmental Protection Agency's National Center
for Computational Toxicology (NCCT). His areas of active
research include virtual tissues, high throughput exposure
modeling, and biostatistics. He co-leads the EPA ExpoCast
project team and is a member of the Virtual Liver and
ToxCast teams. John's research on these projects focuses
on predicting chemical effects in, and exposures to, humans
using in vitro laboratory measurements and computer
simulations. John Wambaugh received his Ph.D. in 2006 from
Duke University (physics) for work in experimental non-
equilibrium statistical mechanics; in particular how large-
scale behaviors can depend on small-scale differences. John
worked with Woodrow Setzer (EPA/NCCT) and Hugh Barton
(Pfizer, formerly EPA/NCCT) as a post-doctoral researcher
at the NCCT; studying the statistical analysis of biological
models, with an emphasis on Bayesian methods and
integrating multiple data types. He received his B.S. (physics)
from the University of Michigan, Ann Arbor, obtained a M.S.
(physics) from Georgia Institute of Technology, and a M.S.
(computer science) from Duke University.
Dr. Jade Mitchell-Blackwood is currently a Risk Analyst
with the USDA Food Safety Inspection Service. She works
on hazard identification and prioritization of chemicals in
food products. Earlier this year, Jade completed a post-
doctoral fellowship in the National Exposure Research
Laboratory (NERL) at the US EPA in Research Triangle Park,
NC. She worked in exposure modeling research to develop
innovative approaches to exposure-based prioritization of
chemicals which fall under the Toxic Substance Control Act
(TSCA) for rapid risk screening. Her interests in the project
included informing the type, quantity and quality of data or
information needed to prioritize chemicals based on exposure
potential using statistical, mechanistic and decision models.
Jade has a Ph.D. is in Environmental Engineering from
Drexel University in Philadelphia, PA where she focused on
decision making under high uncertainty for managing risks
associated with bioaerosols of pathogenic agents.
Dr. Linda Phillips is an environmental biologist in EPA's
National Center for Environmental Assessment, Exposure
and Risk Characterization Group where she provides
technical support to program offices and regions on topics
related to exposure assessment and risk analysis. A primary
focus of her work has been in support of NCEA's Exposure
Factors Program which has produced documents such as
the Child-specific Exposure Factors Handbook and the
Exposure Factors Handbook. She is currently managing
the development of a web-based toolbox to make exposure
assessment tools more accessible to the user community.
Dr. Kristin Isaacs is a Research Physical Scientist in EPA's
NERL. Her current research focuses on the development
and evaluation of human exposure and dosimetry models
and associated algorithms for use in risk assessment of air
pollutants and chemicals. Her specific interests include
development of physiology-based energy expenditure
prediction methods and associated inhalation and dietary dose
algorithms, monitoring and assessment of human activity
patterns, development and evaluation of indoor chemical
source-to-concentration models, and development and
application of sensitivity analyses for exposure/dose models.
She has 10 years' experience supporting EPA modeling
research. She received her Ph.D. in Biomedical Engineering
from Vanderbilt University in 2002, where her doctoral work
involved development of visualization-based sensitivity
analysis methods for physiological and pharmacokinetic
models. She subsequently completed postdoctoral training in
EPA's National Health and Environmental Effects Research
Laboratory (NHEERL), focusing on the development
and application of lung dosimetry models for paniculate
matter. From 2004-2010 she was an Environmental Project
Scientist with Alion Science and Technology Inc., providing
research and analysis support for a number of EPA models,
including the Air Pollutant Exposure (APEX) model of the
Office of Air Quality Planning and Standards (OAQPS) and
NERL's Stochastic Human Exposure and Dose Simulation
(SHEDS) models. She joined NERL in 2010. She has over
a dozen publications in the Journal of Exposure Science and
Environmental Epidemiology, Environmental Modelling
and Software, Journal of Aerosol Medicine, Cellular and
Biochemical Biophysics, and Journal of Pharmacokinetics
and Pharmacology. She has also co-authored two chapters in
the CRC Press Aerosols Handbook (2011).
Dr. Deborah Bennett is an associate professor in
Environmental and Occupational Health in the Department
of Public Health Sciences at the University of California,
-------
Davis. Dr. Bennett's research focuses on the fate, transport,
and exposure of chemicals in both the indoor and multimedia
environments within the context of both environmental risk
assessment and environmental epidemiology. Her work
utilizes both modeling and measurement techniques, bridging
the gap between these two lines of inquiry. Current research
interests include exposure to pesticides from indoor uses,
relating environmental measures to biological measures
for flame retardants, exposures and resulting risks from
hazardous air pollutants, supporting exposure assessments
in Autism studies, quantifying intake fraction values
and measurement of exposures to agricultural workers.
Dr. Bennett received her doctoral degree in mechanical
engineering from UC Berkeley, worked as a scientist at the
Lawrence Berkeley National Laboratory, and was a member
of the faculty at the Harvard School of Public Health. She
has served on both the EPA Science Advisory Board and
the Science Advisory Panel and other EPA committees and
was a US representative to OECD/UNEP Workshop on the
use of Multimedia models. Dr. Bennett received the Early
Career Award from the International Society of Exposure
Assessment and was an EPA STAR Fellow.
Mr. Michael Keating, who joined RTI in 2008, is a survey
manager in the Program on Digital Technology and Society.
Mr. Keating has a broad range of survey research experience,
including leadership of field, Web, and virtual data collection
efforts; management of data collection field staff; data quality
control; instrument development, programming, and testing;
and creation of survey materials. His research interests
focus on the use of new technologies in survey research and
data collection, including crowdsourcing methodologies,
smartphone panels, virtual world data collection methods,
and cloud computing to improve field study efficiencies. His
academic training is in political science.
Dr. Curry I. Guinn is an Associate Professor at the
University of North Carolina Wilmington and formerly a
Research Engineer at RTI International. Using pioneering
spoken human-computer dialogue algorithms, Dr. Guinn
has integrated advanced spoken dialogue capabilities in a
variety of virtual reality-based applications. Dr. Guinn was
co-Principal Investigator on a National Science Foundation
grant leading the development of interactive virtual humans
with emotions that affect their body and facial gestures,
decision-making, and language generation. Dr. Guinn has
led research in using both symbolic and statistical techniques
in natural language parsing and understanding. Funded
primarily by STRICOM's ACT II program, Dr. Guinn led the
development of a system that allows military maintenance
personnel to talk to a computerized assistant in a virtual
reality environment during the diagnosis and repair of
equipment. His work has been supported by research grants
and contracts from U.S. Department of Defense, the National
Science Foundation, National Institute of Justice, National
Institute of Health, the Environmental Protection Agency,
and commercial businesses such as IBM, Michelin, Lexxle,
and John Deere. Dr. Guinn received his B.S. from Virginia
Polytechnic Institute and State University, his M.S. and Ph.D.
degrees from Duke University and is the author or co-author
of 40 peer reviewed articles.
Ms. Cathy Fehrenbacher is the Chief of the Exposure
Assessment Branch in the Office of Pollution Prevention
and Toxics at the US Environmental Protection Agency in
Washington, D.C. Cathy has a B.S. in Environmental Science
from Lamar University in Texas, and a M.S. in Industrial
Hygiene from Texas A&M University. She is a Certified
Industrial Hygienist and has over 25 years of experience in
various aspects of industrial hygiene, exposure assessment,
and environmental fate and transport. She has authored or
co-authored several publications and chapters on the use of
modeling approaches for predicting inhalation and dermal
exposure, and has given numerous presentations and lectures
on EPA's programs, methods, and policies for assessing and
managing chemical risks. She currently serves as co-chair of
the OECD Exposure Assessment Task Force.
Dr. Michael A. Jayjock is a senior member of LifeLine
Group's management team. Prior to joining LLG, he was
the Senior Research and EHS Fellow and Manager, Risk
Assessment, in the Toxicology Department of Rohm and
Haas Company, where he has served in various technical
positions since 1969. He has a Ph.D. in Environmental
Engineering from Drexel University, Philadelphia,
Pennsylvania, where he also received his Master of Science
degree in Environmental Science and Occupational
Health. Dr. Jayjock is a Fellow of the American Industrial
Hygiene Association and Diplomat of the American
Board of Industrial Hygiene (CIH). He has served on
various committees of the American Industrial Hygiene
Association: Committee on Exposure Assessment Strategies,
Exposure Strategies Modeling Subcommittee, Exposure
Strategies Expert System Subcommittee, Committee on
Risk Assessment, and Low-Dose Estimation Task Group.
Dr. Jayjock's principal research interest includes the
development of better-estimating and more cost-efficient
exposure models. He has expertise in such areas as
exposure modeling and human exposure to environmental
pollutants, human health risk assessment, and uncertainty
analysis. He has published extensively in peer-reviewed
publications and served from 1998-2003 as an Editorial
Board Member for the American Industrial Hygiene
Journal. He has made numerous technical presentations,
including at the American Industrial Hygiene Conference,
International Society of Exposure Assessment Conference,
and the Air Toxics Monitoring Workshop to Support EPA's
Integrated Urban Air Toxics Strategy. His wide service on
advisory committees includes: US EPA - Office of Pollution
Prevention & Toxics - Voluntary Children's Chemical
Evaluation Program (VCCEP); Peer Consultation Panel
on Flame Retardants, 2003; US EPA Science Advisory
Board, Executive Committee, Human Health Research
Strategy Panel, November 2002; US EPA Science Advisory
Board Consultant 2001-2003 - Integrated Human Exposure
Committee; US EPA Science Advisory Board Member
1998-2001 - Integrated Human Exposure Committee (IHEC);
and National Research Council - National Academy of
-------
Sciences, as a Member of the Committee to Review Risk
Management in the DOE's Environmental Remediation
Program, the Committee on Advances in Assessing Human
Exposure to Airborne Pollutants, and the Committee
on Toxicology - Subcommittee on Risk Assessment of
Flame-Retardant Chemicals.
Dr. Christina E. Cowan-Ellsberry recently retired as a
Principal Scientist in the Environmental Sciences and Human
Safety Departments of The Procter & Gamble Company in
Cincinnati, OH. Dr. Cowan-Ellsberry has worked in the area
of environmental fate and risk assessment for over 30 years.
While working in the environmental field, she conducted
fate studies and developed models for predicting the fate of
both inorganic and organic chemicals in the environment. Dr.
Cowan-Ellsberry has also served as a technical representative
for industry to the US EPA's Endocrine Disrupter's Priority
Setting workshop, Environment Canada's "Categorization
and Screening of the DSL" project, and numerous
international panels including the OECD's Environmental
Exposure Task Force, the OECD working group for
developing an internationally harmonized classification
scheme for hazardous to the Aquatic environment, and both
the NAFTA Commission for Environmental Cooperation
and the UNEP Criteria Expert Groups for developing the
criteria and process for identifying candidate Persistent,
Bioaccumulative and Toxic substances for international
management which is now incorporated into the recently
adopted Persistent Organic Pollutants protocol. For several
years, she has been applying her exposure modeling expertise
to improve human exposure assessment approaches and
models for consumer products. She has been especially
active in advancing the field of probabilistic and aggregate
exposure assessment for humans by conducting research
to understand and provide recommendations on how to
conduct probabilistic and aggregate exposure assessments
for consumer products. In addition, she was a key participant
in several international activities such as the recent SDA
compilation of habits and practices data for consumer
product exposure which has greatly improved public access
to these important data thereby improving consumer product
exposure assessments. She has been a guest lecturer at several
international universities and recently led a training session
under UN sponsorship in Africa, gave testimony before
the US Congress on TSCA reform in the area of Persistent,
Bioaccumulative and Toxic substances and was a technical
expert in the successful D5 Board of Review in Canada. She
has also authored or co-authored over 60 scientific papers,
5 book chapters and 4 books and holds one US patent. She
has been a member of SETAC for over 20 years and of
ISES for over 10 years and served as ISES counselor from
2007 to 2009.
Dr. Antony Williams is the VP of Strategic Development at
the Royal Society of Chemistry. He obtained his PhD from
the University of London focused on Nuclear Magnetic
Resonance and postdoc'ed at the National Research Council
in Ottawa, Canada and ran the NMR Facility at Ottawa
University. He was the NMR Technology Leader at the
Eastman Kodak Company in Rochester then joined Advanced
Chemistry Development (ACD/Labs) as their Chief Science
Officer working on structure representation, nomenclature
and analytical data handling. With a small team of passionate
individuals interested in sharing chemistry data with the
community he oversaw the development of the ChemSpider
database as a hobby project. ChemSpider quickly developed
into one of the community's primary online chemistry
resources and was acquired by the Royal Society of
Chemistry. He is the ChemConnector in the chemistry social
network environment."
Mr. Henry DeLima is a mechanical engineer and owner of
DeLima Associates, a management consulting firm founded
in the San Francisco Bay area over 26 years ago. Henry
has provided consulting services in the areas of energy
and environmental health to commercial, institutional and
federal government clients. Services in the environmental
health area include authoring materials to aid primary care
physicians in diagnosing and treating patients exposed
to specific environmental toxins, developing medical
management guidelines for first responders and emergency
room physicians in treating victims of acute chemical
incidents and developing databases of environmentally
preferable construction products for EPA and DOD. Some
of the products developed by DeLima Associates include the
Household Products Database and the Dietary Supplements
Database currently hosted by the National Library of
Medicine.
Dr. Pertii (Bert) Hakkinen is the Senior Toxicologist and
Toxicology and Environmental Health Science Advisor in
the Division of Specialized Information Services, National
Library of Medicine (NLM), (US) National Institutes of
Health (NIH). He provides leadership on the development
of new resources in toxicology, exposure science, risk
assessment and enhancements to existing NLM resources
in these fields. Dr. Hakkinen is the project leader for the
Wireless Information System for Emergency Responders
(WISER) and Chemical Hazards Emergency Medical
Management (CHEMM) tools, represents NLM on various
committees, and provides leadership for NLM's participation
in national and international efforts in toxicology-, exposure-,
and risk assessment-related information. He is an Adjunct
Associate Professor in Biomedical Informatics and the
co-director of a public health informatics course offered
since 2009 at the Uniformed Services University of the
Health Sciences (USUHS) in Bethesda, Maryland. Further,
he is the Vice-chair of the Scientific Advisory Panel for
the Mickey Leland National Urban Air Toxics Research
Center (NUATRC) in Houston, Texas. During his career
Dr. Hakkinen has held numerous leadership positions in
the field of toxicology and risk assessment. Before joining
the NIH in 2008, Dr. Hakkinen served for several years
on the auxiliary staff of the European Commission (EC)
at the EC's Institute for Health and Consumer Protection,
Joint Research Centre, in Italy. He has also held positions
with Toxicology Excellence for Risk Assessment (TERA)
and Gradient Corporation in the US, and at the Procter and
Gamble Company in the US and Japan. Dr. Hakkinen earned
a B. A. in Biochemistry and Molecular Biology from the
-------
University of California, Santa Barbara, and received his
Ph.D. in Comparative Pharmacology and Toxicology from
the University of California, San Francisco. Dr. Hakkinen is
a member of the Society of Toxicology (SOT) and a charter
member of the Society for Risk Analysis (SRA) and the
International Society of Exposure Science (ISES). He is a co-
editor and co-author of the latest edition of the Encyclopedia
of Toxicology, and of the last two editions of the Information
Resources in Toxicology book.
Dr. Rogelio (Mike) Tornero-Velez is a scientist with the
National Exposure Research Laboratory of the US EPA.
Dr. Tornero has led efforts within the Agency to couple
probabilistic exposure models with physiologically-based
pharmacokinetic models to investigate cumulative exposures
to pyrethroids pesticides. He has adapted methods from the
field of community ecology to investigate the co-occurrence
of chemicals in the environment with anthropogenic
origin. He received a Ph.D. in Environmental Sciences
and Engineering from the University of North Carolina at
Chapel Hill in 2001.
Dr. Treye A. Thomas is a toxicologist and leader of the
Chemical Hazards Program team in the US Consumer
Product Safety Commission's (CPSC) Office of Hazard
Identification and Reduction. His duties include establishing
priorities and projects to identify and mitigate potential
health risks to consumers resulting from chemical
exposures during product use. Dr. Thomas has conducted
comprehensive exposure assessment studies of chemicals in
consumer products and quantified the potential health risks
to consumers exposed to these chemicals. Specific activities
have included conducting exposure and/or health hazard
assessments of flame retardant (FR) chemicals, combustion
by-products, indoor air pollutants, and compounds used
to pressure-treat wood. Dr. Thomas is the leader of the
CPSC nanotechnology team, and is responsible for
developing agency activities and policy for nanotechnology.
Dr. Thomas has served as a CPSC representative on a
number of nanotechnology committees including the ILSI/
HESI Nanomaterial Environmental, Health, and Safety
Subcommittee, the Federal NSET and NEHI sub-committees,
and the International Council on Nanotechnology (ICON). Dr.
Thomas received a Bachelor's degree in Chemistry from the
University of California, Riverside, an MS in Environmental
Health Sciences from UCLA, and a PhD in Environmental
Sciences at the University of Texas, Health Science Center,
Houston. He completed a post-doctoral fellowship in
Industrial Toxicology at the Warner-Lambert Corporation
(now Pfizer Pharmaceutical).
Dr. Bill Pease is Chief Scientist of GoodGuide, where
he is responsible for the systems used to rate products
and companies by their health, environmental and social
impacts. Bill is an environmental scientist and has served
on the faculty at the School of Public Health, University of
California at Berkeley. Bill was also Director of Internet
Projects at Environmental Defense Fund, where he created
scorecardLorg, the web's top-ranked site for localized
environmental information. He has worked for Cal-EPA,
the California Department of Health Services, and the
Massachusetts Department of Environmental Management.
His academic areas of interest include informatics and
quantitative risk assessment. Bill holds a B A from Yale
University, and an MS in Energy and Resources and a PhD
in Environmental Health Sciences from the University
of California, Berkeley. He was also a Rhodes Scholar at
Oxford University in England.
Dr. Michael Nagle (@nagle5000) runs the Boston Quantified
Self Meetup and is a co-founder of sprout, a non-profit in
Somerville working to make science a cultural activity. He is
fascinated by how people learn. While studying theoretical
math at MIT, he began working to build environments that
support sustainable inquiry for kids, and later, adults. If he
didn't think education was in such vital need of renewal,
he would probably be a hardcore mathematician and
a softcore DJ.
Dr. Michael S. Breen is a Research Physical Scientist
in the National Exposure Research Laboratory at the US
Environmental Protection Agency (EPA). His research
focuses on development of air pollution exposure models,
integrated with novel personal sensor technologies, to
improve exposure assessments for individuals in health
studies. Dr. Breen is co-investigator for three near-road health
studies assessing exposure to traffic-related air pollutants,
one study with asthmatic children and two studies with a
coronary artery disease cohort. Currently, he is developing
three exposure models: Exposure Model for Individuals
(EMI), GPS-based Microenvironment Tracker (MicroTrac)
model, and Personal Exposure Index (PEI) model. He serves
on various science steering committees and conference
technical organizing committees, member of the EPA Ozone
Integrated Science Assessment Review Team, and peer-
reviewer for three journals. Dr. Breen received his Doctorate
in Biomedical Engineering from Case Western Reserve
University, Cleveland, Ohio, and joined EPA in 2005. He
authored over 20 publications, holds two US patents, and
received various scientific awards including the Sally Liu
Outstanding New Researcher Award from the International
Society of Exposure Science, the Biological Modeling
Specialty Section Award from the Society of Toxicology, and
the EPA Pathfinder Innovation Project Award.
Ms. Shannon O'Shea is a contractor in the EPA Office
of Research and Development's Sustainable and Healthy
Communities Research Program. Her primary focus is
development of EPA's Community-Focused Exposure and
Risk Screening Tool (C-FERST), including populating
the tool with information and tracking and incorporating
feedback from stakeholders and pilot end-users. Shannon
received her undergraduate degree in Biological Sciences
from NC State University and worked at Duke University in
Molecular Biology prior to pursuing her M.S.P.H degree in
Environmental Sciences and Engineering from UNC School
of Public Health. At UNC Shannon studied Environmental
Health and Toxicology and became interested in community-
based research.
-------
Professor Benjamin Balak is an Associate Professor of
Economics at the Department of Economics at Rollins
College, and is currently the Chairperson of the department.
Professor Balak holds a PhD in economics from the
University of North Carolina at Chapel Hill and, prior to
that, studied abroad at the American University of Paris
(FR) where he earned a BA in economics. He also holds
a postgraduate diploma from the University of Kent at
Canterbury (UK). Professor Balak's areas of interest
include the history of economic thought, economic history,
philosophy and ethics, and comparative economic systems
and cultures.
Dr. Robert M. Panoff is founder and Executive Director
of the Shodor Education Foundation (www.shodor.
org), a non-profit education and research corporation in
Durham, NC, dedicated to reform and improvement of
mathematics and science education through computational
and communication technologies. As PI on several National
Science Foundation (NSF) and US Department of Education
grants that explore interactions between technology and
education, he develops interactive simulation modules that
combine standards, curriculum, supercomputing resources
and desktop computers. In recognition of Dr. Panoff's efforts
in college faculty enhancement and curriculum development,
the Shodor Foundation was named as a NSF Foundation
Partner for the revitalization of undergraduate education. In
1998, Shodor established the Shodor Computational Science
Institute, which was expanded with NSF funding in 2001 to
become the National Computational Science Institute (www.
computationalscience.org'). Shodor's Computational Science
Education Reference Desk (www. shodororg/refdesk) serves
as a Pathway portal of the National Science Digital Library.
Dr. Panoff consults at several national laboratories and is
a frequent presenter at NSF workshops on visualization,
supercomputing, and networking. Dr. Panoff has served
on the NSF advisory panel for Applications of Advanced
Technology program, and is a founding partner of NSF-
affiliated Corporate and Foundation Alliance. Dr. Panoff
received his M.A. and Ph.D. in theoretical physics from
Washington University in St. Louis, with both pre- and
postdoctoral work at the Courant Institute of Mathematical
Sciences at New York University. Wofford College
awarded Dr. Panoff an honorary Doctor of Science degree
in recognition of his leadership in computational science
education.
Ms. Beck Tench is a simplifier, illustrator, story teller and
technologist. Formally trained as a graphics designer at the
University of North Carolina's School of Journalism and
Mass Communication, she has spent her career elbow deep
in web work of all sorts - from the knowledge work of
information architecture and design to the hands dirty work
of writing code and testing user experiences. Currently, she
serves as Director for Innovation and Digital Engagement
at the Museum of Life and Science in Durham, NC where
she studies and experiments with how visitors and staff use
technology to experience risk-taking, community-making and
science in their everyday lives. Specialties: Visual Thinking,
Informal Science Education, Data Visualization, Human
Computer Interaction
Mr. George Scheer is co-founder and Collaborative Director
of Elsewhere, a living museum and international residency
program set within a former thrift store in Greensboro, NC.
He is also a co-curator of Kulturpark, a project to animate an
abandoned amusement park in East Berlin. His theoretical
and artistic projects take place at the intersection of aesthetics
and social change. George holds an MA in Critical Theory
and Visual Culture from Duke University and a BA from the
University of Pennsylvania in Political Communications.
He is currently pursuing a PhD in Communication Studies
and Performance at UNC-Chapel Hill. George's currents
projects can be found online at elsewhereelsewhere.org and
kulturpark.org
-------
PerCEIVERS 2012 Workshop Attendees
NAME
Rocky Goldsmith
Antony Williams
Beck Tench
Benjamin Balak
Cathy Fehrenbacher
Cecilia Tan
Chris Grulke
Christina Cowan-Ellsberry
Curry Guinn
Curtis C. Dary
Dan Vallero
Daniel Chang
Daniel M. Stout II
Deborah (Debbie) Bennett
Elaine Cohen Hubal
Haluk Ozkaynak
Henry DeLima
TITLE
Research Physical Scientist
VP of Strategic Development
Director for Innovation and
Digital Engagement
Associate Professor
Chief, Exposure Assessment
Branch
Physical Scientist
Research Physical Scientist
Expert Consultant
Associate Professor
Senior Scientist
Environmental Researcher
Research Physical Scientist
Biological Scientist
Associate Professor
Senior Scientist
Senior Scientist
Owner
INSTITUTION
U.S. Environmental Protection
Agency
Royal Society of Chemistry
Museum of Life and Science
Rollins College
USEPA/OCSPPOPPT
U.S. Environmental Protection
Agency
U.S. Environmental Protection
Agency
The Lifeline Group
UNC Wilmington
USEPA/NERL/HEASD
USEPA/NERL
U.S. Environmental Protection
Agency
USEPA/NERL/HEASD
University of California, Davis
U.S. EPA, National Center for
Computational Toxicology
U.S. EPA, National Exposure
Research Laboratory
DeLima Associates
EMAIL ADDRESS
aoldsmith.rockv(3)epa. aov
williamsa(3)rsc.ora
beck.tench@ncmls.org
bbalak(3)rollins.edu
fehrenbacher.cathy(5)epa.
gov
tan.cecilia@epa.aov
arulke.chris@epa.aov
cellsberrv@amail.com
guinnc@uncw.edu
darv.cutris@epa.aov
vallero.daniel@epa.aov
chana.daniel@epa.aov
stout.dan@epa.aov
dhbennett(5)phs.ucdavis.
edu
hubal.elaine@epa.aov
ozkaynak.haluk@epa.aov
HDL603@AOL.COM
PHONE
NUMBER
919-541-0497
919-201-1516
919-475-3421
321-948-9803
202-564-8551
919-541-2542
919-541-4198
513-240-6689
910-962-7937
702-798-2286
919-541-3306
919-541-0875
919-541-5767
530-754-8282
919-541-4077
919-541-5172
703-448-9653
-------
PerCEIVERS 2012 Workshop Attendees
NAME
Jade Mitchell-Blackwood
James Rabinowitz
Jeffre Johnson
Jennifer Houchins
Jennifer Orme-Zavaleta
John Wambaugh
Kathleen Holm
Kristin Isaacs
Linda Phillips
Linda Sheldon
Lynn Delpire
Madeline Reich
Michael Breen
Michael Jayjock
Michael Keating
Michael Nagle
Myriam Medina-Vera
TITLE
Risk Analyst
Senior Scientist
Research Physical Scientist
Computational Mathematics
Mentor
Director NERL
Research Physical Scientist
Post Doc Statistician
Research Physical Scientist
Biologist
NERL Associate Director for
Health
Chemist
EPA Intern
Research Physical Scientist
Senior Analyst
Survey Manager
Co-Founder
Acting Deputy HEASD
INSTITUTION
USDA, Food Safety Inspection
Service
U.S. EPA, National Center for
Computational Toxicology
USEPA/NERL/HEASD
Shodor
U.S. EPA, National Exposure
Research Laboratory
U.S. EPA, National Center for
Computational Toxicology
U.S. Environmental Protection
Agency
U.S. Environmental Protection
Agency
USEPA/ORD/NCEA
U.S. EPA, National Exposure
Research Laboratory
U.S. Environmental Protection
Agency
Shaw University/EPA Reseach
Apprenticeship Program
U.S. Environmental Protection
Agency
The LifeLine Group
RTI International
Sprout & Co.
U.S. Environmental Protection
Agency
EMAIL ADDRESS
jade.mitchell-blackwood(5)fsis.
usdMfiv
rabinowitz.james(S>epa.aov
johnson.jeffre@epa.aov
jhouchins(5)shodor.ora
orme-zavaleta.jennifer©
epa.gov
wambauqh.john@epa. aov
holm.kathleen(3)epa.aov
isaacs.kristin@epa.aov
philliDS.linda@epa.aov
sheldon. linda@epa.aov
delpire.lynn@epa.aov
madeline.reich@amail.com
breen.michael@epa.aov
mjavjock@amail.com
mkeatina@rti.ora
naale@thesprouts.ora
medina-vera.mvriam@epa.aov
PHONE
NUMBER
919-208-2929
919-541-5714
702-798-2177
919-530-1911
919-541-7641
919-541-0859
919-541-2785
703-347-0366
919-541-2205
202-564-8531
919-802-8485
919-541-9409
215-968-3102
703-216-3350
617-571-1369
919-541-5016
-------
PerCEIVERS 2012 Workshop Attendees
NAME
Nicolle Tulve
Peter Egeghy
Robert M. Panoff
Rogelio Tornero-Velez
Roy Fortmann
Ryan Edwards
Shannon O'Shea
Tina Bahadori
Treye Thomas
TITLE
Research Physical Scientist
Environmental Health Scientist
President and Executive
Director
Research Scientist
Acting Director HEASD
Former Shaw Research Intern
NC State Engineering student
Contractor
National Program Director
Chemical Safety for Sustain-
ability
Toxicologist
INSTITUTION
U.S. Environmental Protection
Agency
U.S. Environmental Protection
Agency
Shodor Education Foundation
U.S. Environmental Protection
Agency
U.S. EPA-ORD/NERL/HEASD
N.C. State University
U.S. Environmental Protection
Agency
U.S. EPA, Office of Research and
Development
Consumer Product Safety Com-
mission
EMAIL ADDRESS
tulve.nicollefiiepa.aov
eaeahv.peter@epa.aov
rpanoff@shodor.org
tornero-velez.roaelio@epa.aov
fortmann.rov@epa.aov
edwardsr75@amail.com
O'Shea. Shannon@epa.aov
bahadori.tina@epa.aov
tthomas@cpsc.aov
PHONE
NUMBER
919-541-1077
919-541-4103
919-530-1911
919-541-4997
919-541-2454
919-303-8698
919-541-2036
202-564-0428
301-504-7738
-------
Appendix C
PerCEIVERS 2012 Break-Out Sessions
Group Assignments
We ask that you consider or think about some of these open
breakout charge questions and can even jot down ideas for
multiple sessions (even one's you don't think you will attend)
that could potentially contribute to discussion.
For Day 1, we will have about 20 people in each group.
Participants can choose their two preferences before the
meeting, and we assign them to different groups based on
their choices and # of people in each group.
For Day 2, we will have about 15 people in each group.
Participants can choose their preferences before the meeting,
and we assign them to different groups based on their choices
and # of people in each group.
Breakout session charge questions For Day 1 and 2 and
specific theme areas.
Same questions are asked in different groups, so that we can
see different perspectives on the same issue.
Day 1
THEME A. (Room C400A)
Exposure factors informatics, e.g., real-time human factors
1. What is the best way to define activities related to
consumer product use?
2. Are there any chemicals/products of emerging
concern or lifestyle changes for which consumer use
information does not exist?
3. What categories or products are most important in
terms of data needs?
THEME B. (Room C111C)
Consumer/personal exposure modeling
1. What can be done to make product/chemical
information more easily attainable?
2. What is the best way to define activities related to
consumer product use?
3. How can household inventories best be used with
activity information to estimate potential exposure?
Day 2
THEME C. (Room C300C)
Chemical informatics for consumer products, i.e., consumer
product chemical ingredient data sources, data mining,
analytics, and visualization
1. What can be done to make product/chemical
information more easily attainable?
2. Are there any chemicals/products of emerging
concern or lifestyle changes for which consumer use
information does not exist?
3. What categories or products are most important in
terms of data needs?
THEME D. (Room C111C)
Participatory methods and personal informatics tools, e.g.,
social media mining/analysis and reporting, DIY community
1. When using personal monitoring or social media to
gather consumer use data, what are the data biases (e.g.,
groups that have no access/interest to social media)
2.2. How to engage DIYers to develop an app or game
to collect product/chemical use data and/or product/
chemical inventories?
3. What are the elements to make a tool, game, or learning
module ideal for combining consumer use data with
household product inventories?
4. Are there privacy concerns? What is the public health
message that can be sent?
Day 1
A) Exposure factors
B) Exposure modeling
Day 2
C) Chemical informatics
D) Personal informatics tools
E) Engagement/Communication
Chair
Kristin Isaacs
Cecilia Tan
Jade Mitchell-Blackwood
Mike Breen
Rocky Goldsmith
Rapporteur
Dan Vallero
Chris Grulke
Dan Chang
Peter Egeghy
Kathleen Holm
Room location
C400A
C111C
-------
THEME E. (Room C111C)
Data visualization and analytics for engagement and
communication, e.g., gamification and computational
simulations, noval data visualization/analytic representations,
sharing personal chemical exposure informatics
1. What are the motivation factors for people to participate
in product/chemical inventories or use survey/study?
2. How could the act of learning about one's exposures be
"gamified"
3. Can LARP / technology assisted games co-exist with
computer simulated environments such as Sims /
Second life / Civilization or with agent-based models to
engage the participant?
4. How could one visualize or make the visual imagery
motivational and informative with regards to personal
chemical exposures?
-------
Appendix D
Workshop Presentation Material
-------
PerCEIVERS
~! US-EPA in RTP-NC
Day I Presentations
-June 26, 2012
ORD Innovation
June 26th, 2012
Peter Preuss, PhD
Chief Innovation Officer
EPA Office of Research and Development
US-Environmental Protection Agency
Office of Research and Development
-------
What we can do - what America does better than anyone
else -- is spark the creativity and imagination
of our people...
...In America, innovation doesn't just change our lives. It is
n
how we make our living.
- President Barack Obama
Innovation at the EPA
"BY coming together to advance sustainability
and innovation, we will in turn enhance our security
for decades to come.
- Administrator Jackson, EPA
"ORD must help drive that innovation, because in
d."
- PaulAnastas, EPA
its absence, our mission cannot be achieved.
US-Environmental Protection Agency
Office of Research and Development
-------
ORD Innovation Strategy
Shifting from individual projects to a systematic approach
What?
How?
Support innovation at the bench in ORD laboratories
By fostering a dynamic work environment that rewards and
recognizes creative problem solving
• Pathfinder Innovation Projects
• Innovation in Research Plans
• PeerOvation Awards
• Apps and Sensors for Air Pollution (ASAP)
Demonstrate the power of transdisciplinary research
Learning how to connect scientists in new ways and engage
practitioners and users
• Within ORD (IdeaScale)
• Federal Environmental Research Network
• Design Labs
• Apps and Sensors for Air Pollution (ASAP)
Use open innovation to broaden network of
environmental problem solvers
Bring in new ideas and creative solutions from external scientists
and others
• InnoCentive and TopCoder challenges
• Challenge.gov
• Partnering with OSTP, NASA, HHS, DOD
•Apps and Sensors for Air Pollution (ASAP)
Showcase research that exemplifies the principles of
Path Forward
Expand understanding of innovative research and sustainability
• Signature projects
• Pathfinder Innovation Projects
• Apps and Sensors for Air Pollution (ASAP)
Pathfinder Innovation Projects (PIP):
From Libraries to Pilot Facilities
US-Environmental Protection Agency
Office of Research and Development
-------
Unna SiliH
Open Innovation
v>EPA Real-time Air Quality for Human Health
Air Sensor Technology
Citizen
Science
Environmental Health Data
US-Environmental Protection Agency
Office of Research and Development
-------
, ,:.- .,
r •,. •• • ,:-•
Workshops: Bringing Innovators Together
" " M
Creating an Innovative Research Culture
PeerOvations - peer-driven recognition for research
and administrative innovation
Federal Environmental Research Network (FERN)
connecting ideas and experts across agencies
.^ Interactive Educational Tools - Community-driven
mobile tools for stormwater management decisions
US-Environmental Protection Agency
Office of Research and Development
-------
Innovation Moving Forward
We believe that innovation in ORD will prosper with:
•Visible leadership and commitment. Consistent support from lab and center directors and the
new National Program Directors, demonstrated by aligned resources and incentives, is a critical
determinant of innovation success.
• Design-thinking and experimentation. An organization willing to experiment, test, and learn
will be able to produce and sustain innovation over the long haul.
•Smart risk taking. Not all innovation activities will succeed, but we can still learn from, and
benefit from, creative projects that don't achieve the anticipated results.
•Creative empowerment. Because innovation can come from anyone, it is essential that we
empower people across ORD and be open to external ideas and processes.
•Teams and partners. ORD needs to become better at working across disciplines and with users
and practitioners both in ORD and EPA, and with a variety of external partners.
• Measurement and accountability. We must strive to understand which ORD investments in
innovation are yielding the greatest results, and continuously look for ways to improve ORD's
innovation processes and infrastructure.
Social networking and
c Rtt
Exposure Informatics: • m ',
Systems Reality Modeling
Michael-Rock "Rocky" Goldsmith
US-Environmental Protection Agency
Office of Research and Development
-------
Big Exposure Questions
& Information Needs
*
Three big questions in personal chemical exposure informatics
i
1. What do we expose ourselves to everyday?
2. What chemicals are really in our products?
3. What product changes, or lifestyle changes could be modified to
reduce exposures?
Two major data requirements essential for consumer product chemical
exposure modeling:
1. Consumer product ingredients data
2. Human behavioral/action data (time/location/activity journals)
Burden, passive interrogation, and
personal informatics
Can we evaluate our product inventories in the context of others
around us?
• require a means of reporting that is implemented everywhere (ubiquitous),
fast and dirty
• Need a mechanism of acquiring product data that is ubiquitous and low
burden
Can we evaluate our activities and behaviors in the context of others?
• require a means of rapidly assessing personal activities (i.e. actigraphy)
• Requires a means of identifying and categorizing local environmental
human behavioral patterns in an unsupervised fashion. (NLP mining of
social network feeds)
US-Environmental Protection Agency
Office of Research and Development
-------
Bigger Picture Questions:
How can we get real-time chemical ingredients
or activity/location social/behavioral data?
How can we capture "Real-life" like a "fly-on-the-wall"
— How does one obtain data to understand real systems in real-time
— Dynamic updating of activity/location information at a specified
geographic location
— Twitter relationships
• Big data that captures activity and location information and relateds to exposure
assessment
How can we update and invigorate current data streams for
activity/location information and how will this impact exposure
models?
— directly update and renew CHAD-like activity/location data for studies
that need to be geographically, life-stage, culturally, or gender
segmented.
— will translate on-line conversation and instances into relevant data for
Exposure Modeling efforts such as CHAD or SHEDS?
How do we get required data
elements for Modern Exposure
Assessment?
•Require a vision of current data-streams and data needs
•Information model
•Chemical ingredients in products database
•Means to personally assess activities and to document them
— Creation of the Systems Reality Modeling (SRM) project
• www.systemsreality.org
US-Environmental Protection Agency
Office of Research and Development „
-------
Mash-up of a variety of environmental, human,
and chemical factors to perform personal chemical ' •£ •
exposure informatics
Information Model
Required for Systems Reality
^—!*-»*
/
X^X "^
Or* mrm, atnrul
7/7e Systems Reality Modeling workflow and
Human Matter Interactions
US-Environmental Protection Agency
Office of Research and Development
-------
Our Consumer Product Chemical
Ingredient inventory Interface
•>70% curation of 22K product High-
market share consumer product MSDS
inventory
• N-sampled, crowdsourced over 10
experts
• We enter chemical ingredient name,
CAStt, link out to ACToR, and percent
composition in a given product when
available.
CHAD: provides context driven human
behavior patterns
•Priors analysis on time-slot of
activity relationships to locations
• Will provide first step for an
intelligent exposure related
journaling system.
US-Environmental Protection Agency
Office of Research and Development
10
-------
How does one relate to many? Converting
spoken word into CHAD activity / location
entries
Extrapolation
From singleton
To plurality of
Exposure related
events
SocialStream
lifeStream
NLP:(Natural
Language
Processing)
Text-mining and
search queries
on micro-
bloggingto
obtain "map" of
exposure
activities and
locations
Supporting NERL, program offices and
CSS 2.3.1
Provides mechanism to wrap multiple required data-streams fore personal
chemical exposure informatics into a ubiquitous, low-burden personal computing
device; the smart-phone.
— timely approaches based on devices that most people already own, know
how to use, lower-burden, and could reduce study attrition.
— This data can feed into ExpoCast, CHAD, and SHEDs while also
— filling a community based effort for personal chemical exposure informatics
and supporting CSS!
r
Link to product page
Or www.systemreality.org
PREZI interactive
Presentation:
http://goo.gl/wPPc
US-Environmental Protection Agency
Office of Research and Development
11
-------
/stem s Reality
Modeling
Mark-up
Mining
An EPA Pathfinder Innovation Project
Created by Ryan Edwards
Mentored by: Rocky Goldsmith
Project Outline
Chemical Inventory (Mark up)
• MSDS Database Design and Acquisition
• Using Smartphone devices to scan product inventories and
provide user with augmented reality of chemical exposures
from personal products
Data Mining
• Move towards Integrating CHAD with social network
streams
• Capturing human activity patterns
Computational Modeling
• Simulating peoples daily activities used to explore personal
chemical exposure scenarios
US-Environmental Protection Agency
Office of Research and Development
12
-------
SDS Manual duration
US-Environmental Protection Agency
Office of Research and Development
13
-------
yzing Human Activity Data
CHAD (Consolidated Human Activity
Database)
Data taken from
8 surveys —
j 123,542 entries
• 21,723 subjects
• Data taken by the top 25% by location
Activity and Location Codes
Natural Language Processor
US-Environmental Protection Agency
Office of Research and Development
14
-------
Example Diary Entry
'In the kitchen about to make some eggs'
Location CHAD code: 30121 Kitchen
Activity CHAD code: 11100 Prepare and clean
up food
US-Environmental Protection Agency
Office of Research and Development
15
-------
SocialStream
1
ll
ill
1111
lllll
n-* ',
•«..,<.«. \
US-Environmental Protection Agency
Office of Research and Development
16
-------
Madeline Reich
SHAW University Summer Research Internship 2012
JEMMB ^TWEETS -
INFORMATICS
2.0?
US-Environmental Protection Agency
Office of Research and Development
17
-------
A bit about Maddie
® Rising Senior at FVHS
® Shaw University and the Environmental
Protection Agency Research
Apprenticeship Program
® Social Network Analysis for Personal
Chemical Exposure Informatics
Identifying susceptible populations:
SNA and search queries?
® Variables of interest
• Geographic areas
• Instances of disease related terms, weather,
or human activities
• Timelines
® Tools:
• Google insights, Google maps
• Twitter search, Twitter Maps
• The Archivist
CDC, NOAA
US-Environmental Protection Agency
Office of Research and Development
18
-------
Disorders of interest
® Migraines, ~12B$ / ~75M people annually
® Obesity, ~190B$ / ~111M people annually
® Asthma, ~18B$/~25M people annually
Uin T«mp«rotur« (F)
06/17/2012
US-Environmental Protection Agency
Office of Research and Development
19
-------
Mississippi
::
-
....
Asthma
13 17 18 25 26 34 35 44 45 54 55+
Current Asthma Prevalence Percents by
Age, Sex, and Race, United States, 2010
Age Sex Race/Ethnicity
Sowce. ftfeonalHeakK interview Survey, NattaiulCcnttr for Health SUtBtlci.
Center* for bbe»e Control and MrvvtntKM
CW>H!
White Black
Middle
sian I Hispanic Eastern
14
18 14
US-Environmental Protection Agency
Office of Research and Development
20
-------
Does Exposure Imitate Art?
Recent Impressions
Personal Chemical Exposure Informatics: visualization, user
experience, research in systems modeling and simulations.
June 26-27, 2012
Research Triangle Park, NC
Elaine Cohen Hubal
National Center for Computational Toxicology
Renoir, On the Terrace, 1881
I Office of Research and Development
I National Center for Computational Toxicology
Disclaimer.
Although this work was reviewed by EPA and approved for
presentation, it may not necessarily reflect official Agency policy.
June 26, 2012
US-Environmental Protection Agency
Office of Research and Development
21
-------
X-/EPA
Exposure is the contact between a stressor and a receptor.
^dian
To assess exposure to a particular stressor we need to know
• Properties of the stressor
• Sources, pathways, routes
• Pattern of exposure (magnitude, frequency, duration, location)
• Characteristics of receptor
SOURCE / STRESSOR
FORMATION
K
TRANSPORT,
TRANSFORMATION, and FATE
PROCESS MODELS
Office c' Research and Development
National Ct. '"r for Computational Toxicology
Context: Chemical Evaluation and Risk Assessment
Mandate: Assess Thousands of Chemicals
Need to develop methods to evaluate a large number of
environmental chemicals for potential human-health risks
Data Collection
I Office of Research and Development
| National Center for Computational Toxicology
Richard Judson
US-Environmental Protection Agency
Office of Research and Development
22
-------
High-Throughput Screening Assays
batch testing of chemicals for pharmacological/lexicological
endpoints using automated liquid handling, detectors, and data
acquisition
LTS
MTS
HTS
uHTS
10s-100s/day
10,000s-
100,000s/day
Human Relevance/
Cost/Complexity
I Office of Research and Development
| National Center for Computational Toxicology
Toxicity Testing in the Twenty-first Century:
A Vision and a Strategy
Key aspect of the NRC vision is that new tools are available to examine
toxicity pathways in a depth and breadth that has not been possible
An explosion of high-throughput-screening (HTS) data for in vitro toxicity
assays will become available over the next few years — Data are
available now!
How will this new toxicity
information be integrated
with exposure information
to assess potential for real-
world human health risk?
I Office of Research and Development
| National Center for Computational Toxicology
MAS, June 2007.
US-Environmental Protection Agency
Office of Research and Development
23
-------
Transformation in Exposure
is required to realize potential of
NRC Vision for Toxicity Testing
Claude Monet, Impression, soleil levant, 1872
Does Exposure Imitate Art?
• System
- Moved from studio out into modern world
-Open compositions, realistic scenes
• Resolution
- Exquisite detail (smoothly blended) of surrogate representation
-Abstraction (distillation) of key determinants to address mechanism
- Free brush strokes of pure color to emphasize vivid overall effects
rather than details
• Determinants
- Light (changing qualities)
- Color (bright and varied)
- Form (loose brush strokes)
I Office of Research and Development
| National Center for Computational Toxicology
48
US-Environmental Protection Agency
Office of Research and Development
24
-------
Open System, Relevant Resolution
Jacques-Louis David, The Comtesse
Vilain Xllll and Her Daughter (1816)
Pierre-Auguste Renoir, Le Moulin de la Galette, 1876
Key Determinants
Fraqonard, The Sv^\r\^,
Renoir, The Swing (La Balangoire), 1876
US-Environmental Protection Agency
Office of Research and Development
25
-------
Variability, Vulnerability and Life-stage Aspects Integral
" • •
y^
*•
Monet, Grainstacks 1890-1890
ExpoCast™: Exposure Science for
Prioritization and Toxicity Testing
Recognizes critical need for exposure information to inform
-Chemical design and evaluation
- Health risk management
Goal
- Advance characterization of exposure required to translate findings in
computational toxicology to support exposure and risk assessment
- Together with ToxCast™ help EPA determine priority chemicals
Approach
- Mine and apply scientific advances and tools in a broad range of fields
- Develop novel approaches for evaluating chemicals based on potential for
biologically-relevant human exposure
I Office of Research and Development
| National Center for Computational Toxicology
US-Environmental Protection Agency
Office of Research and Development
26
-------
Environment
Human
nd&
Population
Exposure
tun'
Biotransforma tion
Biomonitoring
IHosf
Susceptibility
Rapid
Prioritization
Relate real-world exposures with
toxicity pathway perturbations
Select doses for c, Translate in vitro results
toxicity testing AN Toxicity for risk assessment
^L JL *L — onrfnn/nfc
,f
» —- i Irt i stun
HTS assays
/.'
bioassays'
Data
Repositories
Mechanistic
Models
Informatics
Approaches
Knowledge
Systems
Network
Models
Exposome
ExpoCast: Recent Activities
• Chemical Prioritization
- Incorporating and Linking Exposure Information into ACToR
- ExpoCastDB
- Integrated Chemical Prioritization Scheme
- Partnering to Develop Exposure Indices for Rapid Prioritization of Chemicals in
Consumer Products
- High Throughput Exposure Estimates
- Rapid modeling ofSVOC exposure in indoor environment
- Intake Production Ratio
' Informing Design of Toxicity Testing
- Selecting Doses for ToxCast In Vitro Testing - Nanomaterials
- Identify Priorities for Mixture Research - Modelinh Chemical Co-Occurrence
• Translate in vitro Results for Risk Assessment
- Combining ToxCast, Dosimetry and Exposure
- ExpoDat2012: Exposure determinants for high throughput risk assessment
• Relate Real-World Exposures with Tox Pathway Perturbations
- ExO: An Exposure Ontology
I Office of Research and Development
| National Center for Computational Toxicology
US-Environmental Protection Agency
Office of Research and Development
27
-------
X-/EPA
Prioritization:
Using Hazard and Exposure Information
High exposure potential
HE
ToxCast Low
Hazard
Prediction
r
Lower Priority for
Testing and Monitoring
I Office of Research and Development
| National Center for Computational Toxicology
ToxCast Hazard Prediction
High Priority
Low Priority
Low exposure potential
HE
Low Priority for
Bioactivity Profiling
Intelligent, Targeted Testing
Human Biomonitoring
Richard Judson
Knowledge Management and Decision Support Tools for CSS
• Data Management Warehouse (e.g., ACToR)
- federate raw data generated by CSS/EPA and available in the public domain on:
chemical structure, production, environmental fate, human use, ecological and
health effects, exposure, etc.
• Ontologies for Interoperability
- publicly available ontologies will be used to specify the semantics to integrate
experimental data from multiple sources, as well as the inputs and outputs of
diverse predictive tools (e.g. empirical models, pathway analysis, systemsmodels,
etc.).
• Knowledge-based management system (KB):
- Develop KB systems that use the above ontologies to acquire, organize, store and
share the complex information flows across diverse CSS activities on chemical
inherency, production, exposure, hazard, pathways and sustainability metrics.
I Office of Research and Development
| National Center for Computational Toxicology
US-Environmental Protection Agency
Office of Research and Development
28
-------
X-/EPA
Exposure Ontology, ExO: Definitions of Central Concepts
• Exposure Stressor-An agent, stimulus, activity, or event that causes
stress or tension on an organism and interacts with an exposure receptor
during an exposure event.
• Exposure Receptor-An entity (e.g., a human, human population, or a
human organ) that interacts with an exposure stressor during an
exposure event.
• Exposure Event - An interaction between an exposure stressor and an
exposure receptor.
• Exposure Outcome - Entity that results from the interaction between an
exposure receptor and an exposure stressor during an exposure event.
I Office of Research and Development
| National Center for Computational Toxicology
Mattinglyetal, EST, 2012
/ Biolog.
' Agent
•Location >'__Biomecl
•Process Ss^ Agent
•Transport Path \xs
\ x_ Phys.
\ Agent
Individ ua
N£sychosoc.
Agent
Exposure
.Human Anthro Receptor
Pop. sphere
•Location
•Genetic Background
•Lifestage
•Health Status
•Socioeconomic Status
I Offlce-QAHHMttond Development
| National Center for Computational Toxicology
Relational View of Selected ExO Domains
Public
Policy
/ inter-
Exposure /--'vention
Outcome N.
\ ^ Disease
Biolog. s^
ResponsiS.
\\_
\ Symptom
Molecular
Response
Exposure
Stressor
•Location
•Temporal Pattern
•Intensity
•Route
•Assay
•Medium
•Method
•Location
Mattingly et al, 2012, EST
US-Environmental Protection Agency
Office of Research and Development
29
-------
High-level schematic of Exposure Ontology (ExO) integration
within a broader biological context.
Encode
Annotated
>\f
Biological F
Molecular Function
Cellular Component
e.g., Gene Ontology
I Office of Research and Development
| National Center for Computational Toxicology
Mattinglyetal, EST, 2012
Art is born of the observation and
investigation of nature.
- Cicero (106-43 BCE)
I am among those who think that
science has great beauty.
-Marie Curie (1867-1934)
Acknowlgements
Carolyn Mattingly, NCSU
Tom McKone, LBNL
US-Environmental Protection Agency
Office of Research and Development
30
-------
ExpoCast:
High Throughput Exposure
Potential Prediction
John Wambaugh
Introduction
Environmental Fate and Transport
Goal: There are thousands of chemicals, many without enough data for evaluation -
working to provide a high-throughput exposure approach to use with the ToxCast chemical
hazard identification.
TSCA21: Prioritization of ~500 Toxic Substances
Control Act (TSCA) chemicals
EDSP21: Prioritization of ~2000 Endocrine
Disrupters Screening Program (EDSP) chemicals
OW21: Development of next chemical
contaminants list (CCL)
Using fate and transport models to predict the
contribution from manufacture and industrial use to
overall exposure rapidly and efficiently
Applying and developing new high throughput
models of consumer use and indoor exposure
lififl Office of Research and Development
Consumer Use and Indoor Exposure
US-Environmental Protection Agency
Office of Research and Development
31
-------
Exposure-related Models Amenable to
High Throughput Operation
-
Exposure-based prioritization challenge identified two models capable of HT
operation (RAIDAR and USEtox)
Harmonized chemical descriptors (EPI Suite)
USEtox
Olivier Jolliet
RAIDAR
Jon Arnot
Default release profiles needed (two variations used, either pesticidal or water)
ural Air
Freshwater
Sea water
I Office of Research and Development
Predictions for 1678 Chemicals
Models predict
partitioning of >1600
chemicals into
environmental media,
and describe human
interaction with that
media
Use models like related
high-throughput assays
How do we ground-truth
these predictions?
I Office of Research and Development
US-Environmental Protection Agency
Office of Research and Development
32
-------
I Office of Research and Development
Ground—truth
with CDC NHANES
urine data
Focusing on U.S.
median initially
Capable of adding
population
variability, but will
need consumer
use models
Data Availability for Model
Predictions and Ground-truthing
Chemicals of
Interest (2127)
Chemicals
Current
Models can
Handle
(1678)
Production /Release
Data
IUR (6759 compounds
with production of
>25,000 Ibs a year)
CPRI (242 pesticides with
total Ibs applied)
Data Availability for Model
Predictions and Ground-truthing
"Ground-truthing"
Chemicals
I Office of Research and Development
Chemicals of
Interest (2127)
Chemicals
Current
Models can
Handle
(1678)
Production / Release
Data
IUR (6759 compounds
with production of
>25,000 Ibs a year)
CPRI (242 pesticides with
total Ibs applied)
NHANES
volatile,
insoluble
US-Environmental Protection Agency
Office of Research and Development
33
-------
&EPA
Linking NHANES Urine Data and
Exposure
(mg/kg/day ),. =
Steady-state assumption
1 me*
* 111&i
Parent
Lafe'nd and Naiman (2008)
\ Office of Research and Development
Stoichiometry of NHANES
Parents and Metabolites
•-«
r • .•—••: ~
~/. x...
, .4. -
% •
c^«'\-^-°x V
& "«0 ^.0
I Office of Research and Development
One to one mappings of
parent to urinary
products (metabolites)
are the exception, not the
rule!
US-Environmental Protection Agency
Office of Research and Development
34
-------
&EPA
JtllJOll EtlfAM
=M •o-'numiri
Calibrate ExpoCast Predictions
to CDC NHANES Data
Y~b1+b2*N + m2 log( VM) + m3 log( vr)
Comparison between model
predictions and biomonitoring
data indicates positive correlation
Consumer use is critical:
Compounds with near-field
applications on average llOOOx
greater
Rigorous statistical analysis gives
calibration of predictions
Same analysis gives uncertainty
(confidence) in those predictions
I Office of Research and Development
ie-10 le-oa le-oe ie-c-4 ie-o:
Conwn*u» Pr«dlct«
-------
Conclusions
Production volume (a multiplicative factor in USEtox/RAIDAR predictions) is
a primary determinant of predicted exposure
Indoor/consumer use is a primary determinant
Next steps:
• HT models for exposure from consumer use and indoor environment
• Use and evaluate these models as additional HT exposure assays
I Office of Research and Development
U.S. E.P.A. Office of Research and Development ExpoCast Team
NCCT
Elaine Cohen Hubal
David Dix
Alicia Frame
Sumit Gangwal
Richard Judson
Robert Kavlock
Thomas Knudsen
Stephen Little
Shad Mosher
James Rabinowitz
NERL
Peter Eghehy
Kathie Dionisio
Dan Vallero
University of Toronto at Scarborough
Jon Arnot
University of Michigan
Olivier Jolliet
USFDAFSIS
Jade Mitchell Blackwood
pressed in this presentation are those of the author and do not
Bflect the views or policies of the U.S. EPA
US-Environmental Protection Agency
Office of Research and Development
36
-------
USDA
From Decision Analytics for
Exposure Prioritization to dietary
residue exposures
Jade Mitchell-Blackwood
Risk Analyst
U.S. Department of Agriculture
Food Safety and Inspection Service
jade.mitchell-blackwood@fsis.usda.gov
PerCEIVERS
Personal Chemical Exposure Informatics: visualization, user Experience, Research in Systems
modeling and Simulations
U.S. EPA, Research Triangle Park, NC
June 26-27, 2012
\
Overview of Research/Work
Modeling approaches for multi-media, multi-pathway
exposure screening for prioritization of chemicals
Sampling ot
meat, poultry,
nd egg products
Veterinary drugs
Pesticides
Environmental
ontaminants
US-Environmental Protection Agency
Office of Research and Development
37
-------
Hazard
High throughput in vitro
experiments measuring bioactivity
Receptor
Evaluating ADME
(Absorption,
Distribution,
Metabolism and
Elimination) parameters
to prioritize exposure
based on biological
relevance
RISK Exposure
Evaluating exposure models
(like mechanistic fate and
transport models) to assess
exposure potential from
indirect, diffuse sources
(i.e., concentrations in food,
air, and water)
Using ADME for
Prioritization
Exposure to target
Exposure Model
Challenge
Source to
concentration
Concentration to
exposure
Screening Level
Uncertainty
Analysis
Multi-Criteria
Decision
Analytic
Framework
US-Environmental Protection Agency
Office of Research and Development
38
-------
Multi-criteria Decision Modeling
Source
Transport
Environmental
Concentration
Exposure
'
Testing decision analytic approaches to:
Integrate disparate data types (criteria and attributes)
Overcome limitations of statistical or mechanistic models
Provide a framework for value of information analysis
Communicate results that are transparently and scientifically
defensible
Exposure
Potential
Chemical
Properties
Life Cycle
Properties
Multi-criteria Decision Modeling
Chemical
Properties
Life Cycle
Properties
Physical
Hazard
Potential
Disposal
US-Environmental Protection Agency
Office of Research and Development
39
-------
Future FSIS Application
Exposure and toxicity
information for certain
classes of chemicals, like
veterinary drugs and
pesticides are available
Other classes, like many
environmental
contaminants, lack
sufficient data
Potency
—> n
Usage data
Exposure Factors Handbook:
Consumer Products Data
Linda Phillips
U.S. EPA, ORD, NCEA
Personal Chemical Exposure Informatics: visualization, user
Experience, Research in Systems modeling and Simulations
(PerCEIVERS) Meeting
June 26/27, 2012
US-Environmental Protection Agency
Office of Research and Development
40
-------
SEPA History of the
f-...-,.-;.". r- * ;. .
Exposure Factors Handbook (EFH)
* Precursor: Development of Statistical Distributions or Ranges of
Standard Factors Used in Exposure Assessment 1985
* EFH first published 1989 |
* EFH updated 1997
* Child specific EFH 2008
EFH updated again 2011
http://cfpub.epa.gov/ncea/risk/recordisplay.cfm?deid=236252
Highlights of the EFH 2011
http://cfpub.epa.gov/ncea/risk/recordisplay.cfm?deid=221023
Toolbox (web based) edition currently under development
Related documents
F -,•..-.! .. r ,•- :.;
Topics Covered in the 2011 EFH
Executive Summary
Introduction
Variability and Uncertainty
Food and Water Intake
Mouthing Behavior ._
Soil Ingestion
Inhalation Rates
Dermal Factors
Body Weight
Consumer Products (added in 1997)
Activity Patterns
Life Expectancy
Building Characteristics
US-Environmental Protection Agency
Office of Research and Development
41
-------
UMtolSuM*
r -. ..-,.-_-. . r- ; ;. .
*a«nn
2011 Revision of the EFH
* Incorporated children's data from 2008 Child Specific EFH
* Included data for other special populations (e.g., pregnant women)
* Improved organization and consistency on data presentation
* Expanded discussions on data limitations
* Enhanced selection criteria approach
* Added new data and analyses
- e.g., new consumer products data added
* Developed new chapters/sections to address additional factors
* Revised recommendations
EPA-Expo-Box
(EPA Exposure Assessment Tool Box)
Exposure Factors Module
* Currently under
development
* Highlights of
each factor
* Full detail for
each factor
* Bookmarks for
easy navigation
* Links to source
references via HERO
* Spreadsheets in
downloadable form
* Links to
related resources
* Search capabilities
(i.e., keywords/topics)
.
US-Environmental Protection Agency
Office of Research and Development
42
-------
r -. ..-,.-_-. . r- ; ;. .
EFH Consumer Products Data
Primary Data Sources/Surveys
* Cosmetic, Toiletry, and Fragrance Assoc. (1983)
* Westat (1987)
Household solvents
Household cleaning products
Interior painters survey
•« Abt (1992) Methylene Chloride Survey
«« EPA National Human Activity Patterns (1996)
«« Bass et al. (2001) Household Pesticides
«« Weegels and van Veen (2001) Household Products
«« Loretz et al. (2005, 2008) Cosmetics
«« Hall et al. (2007) Cosmetics
«« Sathyanarayana et al. (2008) Baby Care Products
SEFA
F -. ..:,,- r •- :. H
EFH Consumer Products Data
EFH Consumer Products Information Summaries
* Survey descriptions
study elements and scope
parameters (products, populations, and scale)
* Data tables
frequency
duration of use
amount of product used per event
* Limitations and Uncertainties
limited data (manufacturers data generally proprietary)
age of data (changes in uses overtime)
* Recommended values not provided due to diversity of
product types
US-Environmental Protection Agency
Office of Research and Development
43
-------
Example of EFH
Consumer Products Data
Table 17-40. Frequency of Use of Personal Care Products
Product Type
Hairspray (aerosol)
Hairspray (pump)
Liquid Foundation
Spray Perfume
Body Wash
Shampoo
Solid Antiperspirant
165b
162
326
326
340
340
340
Average Number of Applications per Use Daya
Mean
1.49
1.51
1.24
1.67
1.37
1.11
1.30
SD
0.63
0.64
0.32
1.10
0.58
0.24
0.40
Min
1.00
1.00
1.00
1.00
1.00
1.00
1.00
Max
5.36
4.22
2.00
11.64
6.36
2.14
4.00
Derived as the ratio of the number of applications to the number of use days.
Subjects who completed the study but did not report their number of applications were excluded.
= Number of subjects (women, ages 18 to 65 years).
1 SD = Standard deviation.
I Source: Loretzetal., 2006.
US-Environmental Protection Agency
Office of Research and Development
44
-------
r -. ..-,.-_-. . r- ; ;. .
Mapping of EFH Data to
HPD Product Categories
RY
NCE
S
SUB-
CATEGORY
PAINT
LAUNDRY
CLEANER
CLEANER
EYE
CARE/MAKEUP
EYE
CARE/MAKEUP
EYE
CARE/MAKEUP
NSECTICIDE
PRODUCT
NTERIOR
LATEX
GLOSS
SPOT
REMOVER
TOILET
BOWL
TUBfTILE
CONTACT
LENS
CLEANER
EYE
MAKEUP
REMOVER
EYE
SHADOW
NSECT
REPELLENT
IN
EFH?
(yes=1;
no=0|
1
1
1
1
0
1
1
1
REFERENCE
TABLE
amount
used
Table 17-6
Table 17-15
Table 17-6
Table 17-37
Table 17-52
Table 17-53
frequency
Table 17-4
Table 17-1 4
Table 17-4
Table 17-37
Table 17-10
Table 17-3
Table 17-3
Table 17-51
Table 17-34
Table 17-35
Table 17-36
duration
Table 17-5
Table 17-1 3
Table 17-1 4
Table 17-23
Table 17-5
Table 17-28
Table 17-37
Table 17-11
Table 17-30
total time
exposed
Table 17-8
Table 17-9
Table 17-1 2
time
exposed
after use
Table 17-7
Table 17-7
REFERENC
SOURCE
Weslat1987
Weslat, 198
Weslat, 198
US EPA, 19
Weslat 1987
US EPA, 19
Weegels an
VanVeen, 20
Westat, 198
Westat, 198
Westat, 198
CTFA, 198
CTFA, 198
Loretz etal., 2
Loretzetal.,2
US EPA, 19
US EPA, 19
Bassetal.,2C
INSIDE THE
IHOME
(INSIDE THE
IHOME
(INSIDE THE
HOME
PERSONAL
CARE
PERSONAL
CARE
PERSONAL
CARE
F -. ..:,,- r •- :. R
Examples of New
Consumer Product Use
Data Not in 2011 EFH
Just, et al. (2010) Urinary and air phthalate concentrations and self
reported use of personal care products among minority pregnant
women in New York City. J Expo Sci Environ Epidem 20(7) 625 633.
Wu, X. et al. (2010) Usage pattern of personal care products in
California households. Food and Chemical Toxicology 48:3109 3119.
Portland State University (PSU), Survey Research Lab (2011).
Personal Care Products Survey. Available online at:
httD://www.oeconline.ora/our-work/healthier-lives/whats-in-mv-
makeuD-baa/Dersona -care-Droduct-survev-reDort/
Moran, R.E., et al. (2012) Frequency and longitudinal trends of
household care product use. Atmospheric Environment 55: 417 424.
US-Environmental Protection Agency
Office of Research and Development
45
-------
&EFK
.'•'..; -.In
Human Activity Patterns in Exposure
Assessment and Exposure Modeling
Movement and activities of humans
(receptor) in time and space
Location-* microenvironment ->
pollutant concentrations
Activity -> energy expenditure -»
ventilation and/or caloric intake ->
intake dose
Exposure-related activities or
microactivities
Pollutant-generating activities ->
microenvironmental sources
I Office of Research and Development
I National Exposure Research Laboratory
Microactivity:
Hand-to-mouth
behavior
US-Environmental Protection Agency
Office of Research and Development
46
-------
Critical Attributes of Human Activity Data
Time Spent Outdoors
Longitudinal Information
Quantification of mean behavior of individuals
• Characterization of intra- and inter-individual variability
• Temporal patterns in individuals (season, daytype)
• Trends over time within individuals with growth and development or aging
(change in lifestage)
Representative Information
• Age/Gender
• Race
• SES
• Culture
• Geography
Timely Information
• Population shifts in behavior
due to societal changes
Children, age 5-12
Minutes/day
2000-2009
I Office of Research and Development
I National Exposure Research Laboratory
&EFK
.'•'..; -.In
Current Databases and Tools
NERL's Consolidated Human
Activity Database (CHAD)
- 41,600 real 24-hour human
activity diaries
-19 studies 1980-2007
- New data being added
EPA's Exposure Models using
CHAD
- Based upon building a simulated
population of people
- Stochastic Human Exposure
Simulations (SHEDS) Models
(Multimedia chemicals, PM, Air
Toxics)
- Air Pollutants Exposure Model
(APEX) - OAPQS
- Screening-tier exposure models
(SHEDS-Lite)
I Office of Research and Development
I National Exposure Research Laboratory
* Census
• Human Activity
• Ambient Cone.
• Food Residues
• Recipe/Food Diary
Exposure Factor
Distributions
US-Environmental Protection Agency
Office of Research and Development
47
-------
'"'••'-' Human Activity Data and Personal Chemical Informatics
•Natural linkages between human activities and PCI
•Locations/activities can drive contact with chemicals: Framework for
modeling Usage and Exposure
~UAril i- Chemical or Consumer
CHAD Locations _, ,,„..,
Product Category
•Development of technology for tracking activity can track chemical use as wel
I Office of Research and Development
I National Exposure Research Laboratory
(,£v"'"'" Current Projects to Address Human Activity and Potentially
Chemical Exposures
• Collection of detailed human activity data is
burdensome
• Use of new technologies will be
essential
- GPS (location) -MicroTrack
- Accelerometry (activity level for intake
dose)
- Active collection
• Smartphone methods for collection
of data
• Innovative data streams
- Social media
• Natural language processing (NLP)
of Twitter feed archives
• Geographic component
I Office of Research and Development
I National Exposure Research Laboratory
US-Environmental Protection Agency
Office of Research and Development
48
-------
&EFK
Future Work
Continue to rapidly update of CHAD to include detailed available human activity
information on a minute-by-minute resolution for higher tier assessments
Continue research into building longitudinal activity patterns from cross-sectional
data
Leveraging other public data and engaging the public - "participatory sensing"
Energy expenditure (EE) research
- Better characterization of EE for activities, ventilation, intake dose rates across
life stage
- Currently collaborating with exercise and obesity researchers CDC, NIH, NCI
to build and curate a new database of individual EE measurements from
academic and government labs across the U.S.
- Linkage of average EE rates with dietary intake of food/chemicals
Start to consider modeling paradigms for linking CHAD-type HA information with
chemical use data in a meaningful way
lsaacs.kristin@epa.gov
I Office of Research and Development
I National Exposure Research Laboratory
Passive Sampling Methods to
Determine Personal
and Household Care Product Use
Deborah Bennett1, Xiangmei (May) Wu1, Candice league1
Kiyoung Lee3, Beate Ritz2, Diana Cassady1, Irva Hertz-Picciotto1
University of California, Davis, Davis, CA, U.S.A.
2 University of California, Los Angeles, Los Angeles, CA, U.S.A.
3 Seoul National University, Seoul, Korea
US-Environmental Protection Agency
Office of Research and Development
49
-------
STUDY GOALS
Study of Use of Products and Exposure Related
Behaviors: SUPERB Study
Pesticides, personal care products, household care
products, food intake, time activity
Multiple Tiers
Tier 1: Largest Tier collected 3 years of phone interviews
Tier 2: Internet based questionnaire every month
Tier 3: Home based passive methods
Tier 4: Environmental and biological samples
Longitudinal changes
Background
Traditionally, data has been collected through
questionnaires, which is very time consuming for
participants.
Determining use of personal and household care
products is of interest, both for use in epidemiology
studies as well as in determining exposure for risk
assessments.
New sampling method is desired to minimize
participants' effort.
Identify products used and amount used.
US-Environmental Protection Agency
Office of Research and Development
50
-------
Study Population
453 families, children <6 yrs
Birth Certificate records
22 Northern CA Counties
30 families in Tier 3
152 households, adult >55
yrs
Probability sampling
by number of housing units
3 Central California counties,
with high agricultural
productivity
17 individuals in Tier 3
CALIFORNIA s
58COUNTIFS
Methods
• Bar codes readily found on consumer products quickly and
reliably determine what products people used in their homes.
• Scan and weigh products at beginning and end of week. Mark
scanned products.
• Determined the change in mass of the product over a one
week period to assess the potential magnitude of exposure.
• Visit each home 4 times periods to capture longitudinal
variability
US-Environmental Protection Agency
Office of Research and Development
51
-------
( Scan barcode to identify product )
dentifiable bat codes • S tKvcocJei entered nfo • % no barcod
US-Environmental Protection Agency
Office of Research and Development
52
-------
Parents of Young Children • Older adults
US-Environmental Protection Agency
Office of Research and Development
53
-------
Product Use Scenarios
Used - Product found both times, and mass decreased
Not Used - Product found both times, and mass the same
Increased Mass - Product found both times, and mass
increased
Removed - Product found at beginning of week but not at
end of week
New - Product found only at end of week
Rediscovered - Product found only at the end of the
week, but had been seen before
Replaced - Product found at the beginning and end of the
week, but there was a new container at end of week
US-Environmental Protection Agency
Office of Research and Development
54
-------
US-Environmental Protection Agency
Office of Research and Development
55
-------
What about over 4 months?
Significant increases in the percent of products used for
• oven cleaning products (13.2 vs 4.5% )
• metal cleaning products (22.7 vs 10.7%)
Slight increases for
• bathroom products (22% from 17.8%)
disinfectant sprays (23% from 20.6%)
pesticides (15.5% from 13.9%)
However, the sum of the percent of products removed,
new, and replaced increased to over 50%.
Household Use Scenarios
For each category of products, scenarios with useable
information:
• All products found, Used
1 All products found, Not Used
j Not Owned
For those in the remaining groups, mass used was not
quantifiable:
• Removed Only
1 New Only
j Replaced Only
• Rediscovered Only
^ Removed Plus
1 Multiple Difficulties
US-Environmental Protection Agency
Office of Research and Development
56
-------
Household Use - Personal Care
KM
60%
in
20%
0%
Not Clearly Interpretable
Multiply
Ollfn Hllto
• Rwfcu nvvi * cl
• N.- . Only
M
• ftemoveclptus
• ffPTItir • !
-------
Mass (g) used in one week
Product Type
Antibacterial Soap
Baby Bath
Babv Lotion
^aby Shampoo^
Body Wash
Bubble Bath
Facial A/fnjftfnriyer
Foundation J
Fragrance Men
Fragrance Women
Flair Styling
Fland Sanitizer
Liquid Soap
Lotion Hand and
4 Naif Polish ""}
Shampoo/Conditioner
Sun Block
Households with
young children
Mean
22.3
18.4
Oi
O6-6,
47.1
42.1
8.8
5.8
—
1.1
25.9
6.5
43.9
30.1
1.3
64.5
8.4
Median
12.9
11.5
4.9
) 11.0
25.8
30.1
^L
C 2.7
"~" - ~~
0.9
12.4
4.0
25.5
15.8
( 1.3
^rrf
5.7
Households with
older adults
Mean
28.2
—
—
—
21.2
83.8
6.2
) 2.8
1.6
1.2
11.9
1.9
24.2
20.5
) 3.9
49.4
9.3
Median
18.1
—
14.4
83.8
3.6
2.9
1.5
1.2
10.6
1.6
12.7
10.9
3.9
28.3
1.4
O
US-Environmental Protection Agency
Office of Research and Development
58
-------
Mass (g) used in one week
Product Type
Air Freshener products
All Purpose products
Ammonia products
Bathroom products
Disinfectant Sprays
Glass Cleaner products
Hobby products
Metal Polish
Oven Cleaners
Pesticide products
Households with young children
Mean
17.8
126.4
66.2
88.6
45.6
33.3
4.5
9.9
17.3
35.9
Median
6.3
75.9
5.3
21.7
8.9
19.2
3.3
2.0
3.0
21.8
Households with older adults
Mean
8.7
59.0
101.7
35.4
1.4
36.9
1.0
12.0
148.6
18.7
Median
4.5
24.4
109.4
18.7
1.4
7.2
1.0
5.0
148.6
16.1
Single Use Mass
For shampoo and hand soap, we asked the
participant to dispense the amount of each
product they typically used onto a plastic sheet
that they placed on their hand.
Individual use amount (g) and estimated number of uses per week
Product Type
Liquid Soap
Shampoo/
Conditioner
location
PYC
OA
PYC
OA
Single Use Amount (g)
Mean
10.3
4.5
40.8
7.6
Median
3.6
1.9
6.8
3.8
75th%
4.6
2.4
57.5
13.3
Uses per week based on Single Use Amount
Mean
34.0
15.8
11.2
10.0
25th%
2.7
1.4
1.7
2.2
50th%
6.6
9.6
6.0
8.2
75th%
25.6
24.0
16.8
14.9
95th%
224.1
57.5
34.5
24.2
US-Environmental Protection Agency
Office of Research and Development
59
-------
Motion Sensor to Record Use
Actical Accelerometers were strapped on two most frequently
used products to record how often the product was moved.
• The use frequencies during
different sampling weeks for
each household were
moderately consistent.
• The majority of the usage of
cleaning products (92%)
happened between 7am and
9pm, with peaks at mealtime,
i.e., Sam, 12pm, 3pm, and 7pm.
(a) Use frequency during the
sampling week
11.7%
25.0%
• 1-2
• 3-6
• 7-13
• 14 and up
(b) Duration of each use (minute)
1.3%
4.2%l
Conclusions and Future Directions
The bar code scanner obtains actual products used the
majority of the time relatively quickly.
Participants store, but don't use, a large number of
products. Participants could be asked to show staff which
products are used.
Instead of determining mass used, participants could be
asked frequency of use.
Overall, the use of bar code data and motion sensors are
promising methods for evaluating use of personal and
household care products with minimal burden to the
participants.
US-Environmental Protection Agency
Office of Research and Development
60
-------
Questionnaire Household Cleaners
• Products: All purpose cleaner, Car cleaner,
carpet cleaner, floor cleaner, glass cleaner,
oven cleaner, polish, tub/shower cleaner,
various types of air cleaners
• Cleaning Habits: Dry mop, Wet mop, sweep,
and vacuum hard floors, vacuum carpets
• Frequency, correlations between product use,
demographic differences
• Frequency and Longitudinal Trends of Household Care Products Use,
Atmospheric Environment, in press
Longitudinal Consistency
Car Cleaner
Polish
Oven Cleaner
Carpet Cleaner
0 1-3 4-23 24+ 0 1-3 4-23 24+ 0 1-3 4-23 24+
Mean use per month
0 1-3 4-23 24+
US-Environmental Protection Agency
Office of Research and Development
61
-------
Personal Care Product Use
Shampoo, Bath gel, body lotion, hand lotion,
deodorant, liquid soap, waterless hand sanitizer,
facial cleanser, facial moisturizer, mask, anti aging
cream, lip balm/lipstick, sunscreen hot/cool,
Hair: dye, perm, spray, mousse
Makeup: foundation, mascara, nail polish, fragrance
Frequency, correlations, demographic differences,
scented/unscented
Wu X, Bennett D.H., Ritz B., Frost J., Cassady D., Lee K. Hertz Picciotto I. Usage Pattern of Personal Care
Products in California Households. Food and Chemical Toxicology, 48(11): 3109 3119, 2010.
Brand
Loyalty
(a) parents of young children in
northern California
US-Environmental Protection Agency
Office of Research and Development
62
-------
Acknowledgement
The project was supported by United States
Environmental Protection Agency (grant# RD
83154001 0).
Many thanks to all the SUPERB participants.
Special thanks to our field staff, Jessica Riley,
Laura Gonzalez, Brianna Diaz and Karen
Wagner.
HRTI
!•? «(••* IQII*I
Smart phones as a Flexible Research Tool:
Lessons from Early Implementations and the
Consumer Marketplace
Michael D. Keating
RTI International
PerCEIVERS 2012
mkeating(S)rti org
www rti org
US-Environmental Protection Agency
Office of Research and Development
63
-------
How fast is smartphone growth?
U.S, Smartphorw Perforation
-..-x
50%
.
It's fast. Really fast.
• .>
fcim
Samsung Galaxy Nexus
• 4 65 inch HD screen
• 5 megapixel camera
• Global positioning
systems
Accelerometer
Bluetooth
Access to fast data
connections
US-Environmental Protection Agency
Office of Research and Development
64
-------
Questionnaire Administration
Self-administration
survey successes!
High response rates.
Promising usability.
Journal entry is possible.
- Text entry
- Photo entry
Be mindful of screen
sizes..
••I..J- H 'i-1
Kill I
Global Positioning Systems
Capture micro-
level location
data
Passive data
collection
opportunities
Match location
data with other
data
,
GPS sensors should enhance data quality!
fclll I
US-Environmental Protection Agency
Office of Research and Development
65
-------
Sensory Technology
Bluetooth connection to
sensory technology
offers the possibility for
collection of a rich set of
data.
- Indoor location
- Air quality
Accelerometers can
track physical activity.
In-aje
Kill I
Unanswered Questions
Human subjects
literature is still a work in
progress.
The privacy implications
of personal data
collection.
Data security is a
concern
fclll 1
US-Environmental Protection Agency
Office of Research and Development
66
-------
Natural Language Processing and
Human Activity Patterns:
USING A SPOKEN DIARY AND HEART RATE
MONITOR IN MODELING HUMAN EXPOSURE FOR
EPA'S CONSOLIDATED HUMAN ACTIVITY
DATABASE
Curry I. Guinn, UNC Wilmington
Daniel J. Rayburn Reeves, UNC Chapel Hill
Collecting Human Activity Data
Purpose
• To develop a method of generating an
activity/location/time/energy expenditure database of sufficient
detail to accurately predict human exposures and dose.
To Evaluate
• the use of digital voice recordings
• the use of the ambulatory heart rate monitor
• participant/instrumentation interactions
To Develop
• a protocol for automating the processing of voice recordings
• an autocoding program that will be able to map the text of the
diary entries to CHAD
US-Environmental Protection Agency
Office of Research and Development
67
-------
Problems with Collecting Human
Activity Data
Recall Data
• Failure to recollect many daily activities
• Lack of detail
Real Time Paper Diaries
• Increased number of reports/better detail
Burdensom
Direct Observation
Greatest number of reports/most de
• Inefficient and expensive
Platform for Solution
Audio diary using a digital voice recorder
. Ambulatory Monitoring System that monitors heart rate and
prompts subjects to provide diary entries when heart rate increases
by a specified criterion level.
Database Sample
Recorded Utterance
in the bedroom starting housework
carrying clothes to the laundry room
the bedroom getting more clothes
loading the washing machine in the
laundry room
sitting down going to watch twenty
minutes of Regis
I'm going to be brushing the dog in
the family room
CHAD
Location
30125
Bedroom
30128 Utility
room / Laundry
room
30125
Bedroom
30128 Utility
room / Laundry
room
30122 Living
room / family
room
30122 Living
room / family
CHAD Activity
11200 Indoor
chores
11410 Wash
clothes
11410 Wash
clothes
11410 Wash
clothes
17223 Watch
TV
11800 Care for
pets/animals
US-Environmental Protection Agency
Office of Research and Development
68
-------
Voice Diaries
Average: 29 entries/ day
With average monitoring time of
8.56 hours, 3.39 recordings/hour
First 3 days of trial: 34.44/ day
Last 2 days of trial: 20.65/ d.
1 out of 63 reporting periods dat
lost (1.6%)
345
Day of Study
US-Environmental Protection Agency
Office of Research and Development
69
-------
Quality of Diary Entries
Entry Length
• 9.39 words average
Some entries invalid because of length (subject failed to
turn off recording)
• 1/30 recordings (3%)
Heart rate change and diary entry
• Avg. of 28.8 beeps per day; 36.8% compliance
Computer classification
66% accuracy in activity; 76% in location
Significant improvements with less granular CHAD
encodings
US-Environmental Protection Agency
Office of Research and Development
70
-------
&EPA
United States
Environmental Protection
Agency
Consumer Exposure Assessment
for New Chemicals
Cathy Fehrenbacher, Chief
Exposure Assessment Branch
June 26, 2012
Office of Pollution Prevention and Toxics
US-Environmental Protection Agency
Office of Research and Development
71
-------
Overview EPAs New Chemicals (PMN) Review Process
Day 8-12
Chemistry
Meeting
Day 9-13
Hazard
Meeting
Assessment/
Management
Meetin
10%
_
55%
Drop
Day 10-15
EPA reviews about 1,500 PMNs per year!
Submitters of PMNs are not required to conduct any new testing
Office of Pollution Prevention and Toxics |
OPPT's Exposure Tools and Models Include:
Methods used by OPPT for exposure assessment, in the
absence of, or to supplement data
Computerized models and accompanying databases
Default assumptions which can be modified by the user
Online help and transparent guidance in using the
models and databases
Capability to address adults, children, and infant
populations
Some population and demographics data, and
information on endangered species
Some geospatial and graphing capabilities
Office of Pollution Prevention and Toxics |
US-Environmental Protection Agency
Office of Research and Development
72
-------
Application of OPPT's Models: workplace, home, community,
environment (http://www.epa.gov/oppt/exposure/)
Office of Pollution Prevention and Toxics |
OPPT's Models for Consumer Exposure Assessment
AMEM-chemical
migration through
polymers (under
development)
E-FAST — screening
level modeling suite,
includes Consumer
Exposure Module
MCCEM-higher
tier consumer exposure
• WPEM-wall paints
• FIAM — formaldehyde in
pressed wood products
(new)
Office of Pollution Prevention and Toxics | 5
US-Environmental Protection Agency
Office of Research and Development
73
-------
Consumer Exposure Data and Tools
Consumer Product Scenarios
— Developed in 1986; multi-volume set
— Few were programmed into E-FAST; with user-defined
scenario which is commonly used with the scenarios
— Contain formulation data, weight fractions of functional
components, exposure factors, use pattern, use conditions,
frequency and duration, etc.
Consumer Products Database
— Confidential Business Information
— Based on formulation-related data for consumer and
commercial products available
— Undergoing internal review
Office of Pollution Prevention and Toxics |
MHiffli OO(F
-------
Chemical Use: The Key to Near-
Field Chemical Exposure
Estimation
Christina Cowan-Ellsberry
The Lifeline Group
Cincinnati, OH
onten
Public Availability of Data
How to fill Chemical presence and Use
gaps
Product use profile data
6/2012 © The LifeLine Group 150
US-Environmental Protection Agency
Office of Research and Development -,,-
-------
ecessary Informatio
Three types of information needed for exposure
assessment
Chemical Specific Information
' Chemical properties
Chemical presence
Chemical fraction in product
Product Specific Information
Use profile
People Specific Information
demographics
Information = data, surrogate data, derivations,
default values, assumptions
6/2012
© The LifeLine Group
Product Specific Information
c Use scenarios—how products are used by different
people during different seasons and conditions
Product co-uses and competitive uses
f~ Information applicable to many chemicals, so when
product use profiles are constructed, much of the
information is reusable across many chemicals
r Much of the information is publicly available
6/2012
© The LifeLine Group
US-Environmental Protection Agency
Office of Research and Development
76
-------
ources of Informal!
Publically available BUT
Not necessarily in final form required
This chemical is used in specific product at x%
Everyone uses 2 times per day etc.
• Instead may need some interpretation and bringing
together different parts of data
Chemical is used as fragrance in product, as
emulsifier etc.
Type of product used in are face cream, body wash,
etc.
Publicly available information usually provides good initial
listing of uses and potential products for consideration
6/2012
© The LifeLine Group
153
hemical Soecific Information
How are Chemicals Used in Products? Possible health concerns and tox reviews
•wikipedia.org
• product safety assessments (online) by manufacturer (i.e. Dow Chemical)
• government chemical and product reviews such as
http://www.nicnas.govau/publications/information_sheets/existing_chemical
_information_sheets
Characteristics of the chemical, sources, related information
• US Nat'l. Ctrfor Biotechnology
• Forms, sources, links to other info sites
Concentrations in products, possible substitutes, functions in products and use
scenarios/profiles
• (example: Household Products Database )
• European evaluations under REACh. 14,000 dossiers expected to be
publicly available relatively soon
•The Cosmetic, Toiletry, and Fragrance Association (2006) International
Cosmetic Ingredient Dictionary and Handbook, Washington DC, Ed
Wenninger JA McEwan GN
6/2012
© The LifeLine Group
154
US-Environmental Protection Agency
Office of Research and Development
77
-------
Necessary Information: Chemical Specific
c
Chemical Specific Information
Hazard/toxicology information
Exposure related information
Physical/Chemical properties - Lipophilicity, size, vapor
pressure, reactivity, etc.
Functionality because of its chemical properties -
Surfactants, solvents, colorants, stabilizer, etc.
Considering competing chemicals and additional
uses/products containing the chemical
" Ranges of concentrations for functional ingredients
Usually readily publicly available - manufacturers discuss
how chemical can be used and example products
6/2012 © The LifeLine Group IE
sample: Sodium Laurvl Sulfate
Detergent surfactant function as a wetting agent, dispersing
agent, emulsifying and/or foaming agent.
Appears in a wide range of products because it is highly
effective, relatively inexpensive, and especially useful for
opaque, pearlescent or cream products, can be used in
powdered or tablet product forms, provides high foam with good
viscosity and is readily solubilized in cool water.
Functionality determines both the product list and the likely
concentration range in each product type within these two
product categories. Even within a particular product type, the
concentration range can vary considerably depending on the
functionality required and the presence of other ingredients
which serve the same or similar functionality
6/2012
© The LifeLine Group
156
US-Environmental Protection Agency
Office of Research and Development
78
-------
.xample: Sodium Lauryl Suit
Heavy duty cleaning products where very effective removal of
heavy grease and stains (e.g., heavy duty degreaser washes) is
required the concentration of SLS can reach 50 %
In less heavy duty cleaning products, such as carwashes, carpet
cleaners, dish washing liquids, pet shampoos and upholstery
cleaners; SLS is typically in the 3 to 30% range. In most
household cleaners, SLS concentrations range from <1 to 5
percent.
Product form can also determine the concentration range of
SLS. For example, liquid laundry detergents and fabric
softeners contain SLS at about 30% whereas the powder and
tablet forms of these products may contain up to 93% SLS,
which is then diluted in use.
6/2012
© The LifeLine Group
157
xample: Sodium Laurvl Sulf
Depending on desired amount of lathering, the concentration in
personal care products can also vary considerably. For
example, a few hand soaps, body washes and shampoos can
contain SLS in a wide range of concentrations up to 30 percent
whereas most hand soaps contain 1 to 5% and face soaps 1 to
2.5% SLS. Notably, some children's toothpastes can contain up
to 5 % SLS (Barkvol, 1989) because these products require
more bubbles/foaming to make the tooth cleaning activity more
fun. Shaving creams' concentrations range from 1 to 10% with
the higher concentrations occurring in products that produce
heavier and thicker foams or gels.
6/2012
© The LifeLine Group
158
US-Environmental Protection Agency
Office of Research and Development
79
-------
xample from One Websi
Typical Face Cream Ingredient with Typical Concentration Ranges
Emulsifiers 2-6%
Emollients 10-35%
Thickener 0.1-1%
Deionized Water Q.S.
Preservatives 0.2-1 %
Humectants 1-8%
Consistency factors 1-6%
Antioxidants 0.01-0.05%
UV filters 0.01-0.5%
Chelating Agents 0-0.02 %
Fragrance 0.1-1 %
Active agents 0.1-2%
Coloring agents Q.S.
Aesthetic enhancers 0.1-5%
6/2012 © The LifeLine Group 159
lemical Informal!'
r Physical/chemical properties define
functionality
Functionality defines how used in
products and likely product forms
Functionality defines concentration
range in typical products
Bring together the best information and
make best initial judgment
6/2012
© The LifeLine Group
160
US-Environmental Protection Agency
Office of Research and Development
80
-------
Sources of Habits and Practices Data-US Data
Personal Care Product Council Studies:
• Exposure data for cosmetic products: Facial cleanser, hair
conditioner, and eye shadow. Food Chemical Tox Loretz et al.
2008vol:46pg:1516
Exposure data for personal care products: Hairspray, spray
perfume, liquid foundation, shampoo, body wash, and solid
antiperspirant. Food and Chem Tox Loretz et al., 2006 vol:44
pg:2008
• Exposure data for cosmetic products: lipstick, body lotion, and
face cream. Food Chem Tox Loretz et al. 2005 vol:43 pg:279.
American Cleaning Institute-
• Global Exposure and Risk Screening Methods for Consumer
Product Ingredients 2005 and updates
EPA Exposure Factors Handbooks - 1997 and updates
6/2012
© The LifeLine Group
161
r COLIPA (European Cosmetics Association)
Studies
SCCNFP/0321/02 and referenced in THE SCCP'S
NOTES OF GUIDANCE FOR THE TESTING OF
COSMETIC INGREDIENTS AND THEIR SAFETY
EVALUATION
r Company marketing product will have data on
' Other populations, sub-populations
Special product forms/use scenarios
6/2012
© The LifeLine Group
162
US-Environmental Protection Agency
Office of Research and Development
81
-------
er Product Form
Example, Azo and Benzidine dyes in temporary
tattoos and skin paint
Concentrations can range from 0.4 to 30 % and the
area covered can range from quite small, for example
children's hand stickers, to full torso body art (Louis
Vuitton 2011, Chanel 2011).
Concentration of the colorant in the product depends
on the product formulation as well as the desired
durability of the tattoo.
6/2012
© The LifeLine Group
163
People Specific Information
Morphometrics and physiological parameters
(height/weight, breathing rates, etc)
" Age dependent activity profiles
I Demographic, econometric and ethnic activity-related
influences (special subgroups)
Once information is developed, useful for all subsequent
analyses
<" Independent of chemical but dependent on the product
6/2012
© The LifeLine Group
US-Environmental Protection Agency
Office of Research and Development
82
-------
y Learnin
1. Information gathering, and initial use profiling
requires 4 hours or less per chemical for most
chemicals.
1. Must have structured approach with stop points
2. Exposure scenarios for products are "reusable"
across many chemicals, so efficiency increases
as the process continues.
3. Needed information is increasingly available,
spurred by other regulatory and private
transparency initiatives, in organized public and
commercial databases
6/2012 © The LifeLine Group 165
.eferen
Using publicly available information
to create exposure and risk-based
ranking of chemicals used in the
workplace and consumer products
MICHAEL A. JAYJOCK, CHRISTINE F. CHAISSON, CLAIRE A.
FRANKLIN, SUSAN ARNOLD, AND PAUL S. PRICE
Journal of Exposure Science and Environmental
Epidemiology (2009) 19, 515-524
6/2012 © The LifeLine Group 166
US-Environmental Protection Agency
Office of Research and Development oo
-------
Probabilistic Exposure
Assessments for Consumer
Products
Christina Cowan-Ellsberry
The Lifeline Group
Cincinnati, OH
Define different types of Probabilistic
Exposure for consumer products
Describe the sources of the Habits and
Practices/Use data for products
r Strengths and limitations
Discuss some of the important things to
consider when doing probabilistic
exposure assessments
r Key Learnings
6/2012 © The LifeLine Group 168
US-Environmental Protection Agency
Office of Research and Development 04
-------
Habits and Practices or Use Data
•Amount /use
• Frequency of use/day
Duration (continuous/intermittent; fraction of a
lifetime)
• Method of application (e.g., rinse-off)
Etc. - Modifying factors
• Body site(s) applied to
Demographic Data
Size of area of application or Body Weight
6/2012
© The LifeLine Group
169
obabilistic Exposure Assessmen
Individual Product Exposure
' Variability in Use Patterns by age, gender, ethnicity
10% use 1 time per day, 30% use 2 times per day,
etc.
Aggregate Exposure
f Include variability in non- and co-use patterns of
individual products
Population Exposure
Ensure that population or sub-population is correctly
represented by Use Pattern data
6/2012
© The LifeLine Group
170
US-Environmental Protection Agency
Office of Research and Development
85
-------
P/Use .
termine
Several sources
' Focus Group Studies
r Product Placement Studies
Market Research Surveys
In all cases there is a lot of data collected,
although not necessarily in form required
Often need to combine data from several
studies to get a complete picture of how the
product is used.
Need to recognize strengths and limitations of
each source
6/2012 © The LifeLine Group 171
oup Studie
• Observe product handling under normal use conditions
(e.g., for fragrance Products: Typical spray distances, Sites
of exposure)
• Bottles are weighed before & after application (i.e., Amount
applied per application (g))
• Deposition area determined by measurement of spread onto
a paper collar applied to the preferred body target (i.e.,
Surface area (cm2) covered by product)
• Questionnaire to probe further (e.g., Number of sprays per
application, Number of applications/site per day)
Limitations: Very costly, one time focus on product use, may
not reflect actual use
Advantages: Personal observation, ask additional questions
6/2012 © The LifeLine Group 172
US-Environmental Protection Agency
Office of Research and Development
86
-------
roduct Placement Studies
Participant is given product, asked to use for set period of
time, record when use, etc
• For fragrance products: Number of sprays per
application, number of applications per day, site of
application
• Bottles are weighed before & after time period to
determine amount used over entire period
Often coupled with information on how they like the
product, demographics, etc.
Limitations: inaccurate/incomplete recording, other people
may use product or participant may retain remaining product
Advantages: direct measurement of product amount used
and frequency over multiple uses
6/2012
© The LifeLine Group
173
Questionnaire - mail/internet/street/phone
• Target large numbers of consumers (> 1000)
• Include multiple countries, ethnic, economic groups to
understand differences in use.
User Preferences and Use of Products
• Product Form: sprays, splash, parfum, etc. for fragrance
products
• Frequency of use - daily, occasional, 1x, 2x + per day...
• Use with other products
Limitations: Recall is not perfect, bias in recording based on social
acceptability of behavior, amount used data is missing
Advantages: large number of people can participate, multiple
product use (co-use and non-use patterns)
6/2012
© The LifeLine Group
174
US-Environmental Protection Agency
Office of Research and Development
87
-------
Representing Habits and Practices data
distribution
Correlation between amount and
frequency of use
Choice of percentile of exposure
Use patterns different for age, sex,
ethnic group, etc.
6/2012
© The LifeLine Group
175
Represented Data Distributio
Appropriate statistical distributions for the habits and practices is critical.
Use raw data as much as possible rather than force fit a statistical
distribution
1) fitted distributions may over-estimate high end users because they are
continuous or user must determine where to truncate distribution; 2) raw
data is often bi- or tri-modal; and 3) there is usually no "best fit" across all
sub-sets of the data (e.g., Normal or Log-Normal is seldom the "best fit").
Comparison of Input Distribution anc
Gamma(2.23,4.O8)
6/2012
Comparison of Input Distribution and
Gamma(1.90,4.52)
Values in 10A1
© The LifeLine Group
176
US-Environmental Protection Agency
Office of Research and Development
88
-------
orrelation of Use Da
For most products, data from several surveys will need to be
combined since one survey may not contain data for all parts
of the target population (e.g., non-users, age groups) nor data
on both frequency of product use and amount used per
application.
If the frequency and amount used are positively correlated
then the resulting exposure will be under estimated. However,
data analysis for representative Beauty Care and Oral Care
product studies where both frequency and amount have been
measured shows that frequency of use and amount used per
application are either negatively correlated or independent
and none are positively correlated; therefore, exposure will be
over-estimated for these products if independence is
assumed when combining data from independent studies.
Really important when looking at high-end exposure
6/2012
© The LifeLine Group
177
xamole for Sham
10 20 30 40 50 60 70 80 90 100
Percent of Exposure
6/2012
© The LifeLine Group
178
US-Environmental Protection Agency
Office of Research and Development
89
-------
Forecast: Systemic load
Frequency Chart
95th Percentile Deterministic Value was 1.93 gm/day
6/2012 © The LifeLine Group 179
reaate Exoosur
Include non-use and co-use information
r Extent of ingredient use in that type of
product
6/2012
© The LifeLine Group
180
US-Environmental Protection Agency
Office of Research and Development
90
-------
ggregate Exposure Considerations
r Typical aggregate exposure assessment involves
adding the exposures to the individual products -
very conservative
Basic assumption of the additive approach is
that a consumer uses all the product types
frequently
r Co-use and non-use patterns of different product
types are important determinants of exposure
If a person does not use a product or uses it
very infrequently then this product will not
contribute to aggregate exposure for that
person
Extent that ingredient is used in a particular
product type will vary
The extent of use will determine likelihood
that person will come in contact with that
ingredient when used that product type
Paraben
Methyl
Propyl
Ethyl
Butyl
Total
Added
Exposure
(mg/kg/d)
1.25
1.25
0.93
0.47
3.9
6/2012
© The LifeLine Group
181
robabilistic Aggregate Exposr
For many products the percent of the population that
are non-users can be very large.
Consumer Product
Toothpaste
Conditioner
APDO
Mouthwash
Shampoo
Approximate Percent or
Population tliut arc Nun
Concumer*
2%
30 ID 75% depending on M v* F
4%
29 to T31*
^ 5 So 5O% depending an type of
ihampoo
6/2012
© The LifeLine Group
182
US-Environmental Protection Agency
Office of Research and Development
91
-------
Product co-use combination
BC + BL + HL + FM + FC
BL + HL + FM + FC
BC + BL + HL + FC
BL + HL
BC + BL + HL + FM
BC + BL + HL
BL + HL + FM
Non-Use
Percent of participants
27%
16.9 %
5.5 %
6.6 %
4.7 %
4.1%
4.5 %
31.3%
BC - body cream, BL-body lotion, HL- hand lotion, FM - facial
moisturizer, and FC - facial cleanser.
6/2012
© The LifeLine Group
183
Extent of Use Data: Methyl vs Ethyl Paraben
Percent of Product Formulations Containing Paraben
\
Factor 71 Factor 15.1
*From Elder 1984. J. Amer. College of Toxic. 35:147-209.
6/2012 © The LifeLine Group
Products or Product Categories
Eye make-up
Make-up
Skin Cleansing
Face, body and hand skin case
Moisturizing skin care \
Night skin care / \
Across all formulations
Methyl
35
20
62
67
71
62
28
Propyl
34
25
51
56
64
51
24
Butyl
3
2
8
12
12
15
3
Ethyl
0.2
0.2
2
4
1
1
0.8
184
US-Environmental Protection Agency
Office of Research and Development
92
-------
ggregate Exposure Summary
Paraben
Methyl
Propyl
Ethyl
Butyl
Total
Summed Aggregate
Exposure (mg/kg-d)
1.25
1.25
0.93
0.47
3.9
Refined Estimate
(mg/kg-d)
0.56
0.28
0.012 to 0.007
0.003
0.855
» Reduction in aggregate exposure by factor of 3 to 5 by applying
refinements
» Potentially larger if more products included
Cowan-ElIsberry and Robison. Regulatory Toxicology and Pharmacology 55 (2009) 321-329
6/2012 © The LifeLine Group 185
Population or Sub-Population Exposure
Differences in use patterns by age, sex, ethnicity etc are very important
and can be significant.
Population exposure determined by estimating exposure sub-
populations and then combine to calculate population exposure using
census data
Typically consumer surveys under represent young and old, and ethnic
groups (e.g., African Americans, Hispanics) and focus on product users.
If the survey population does not have the same demographics as the
target population or data from several surveys are combined, exposure
will need to be calculated as the weighted exposure by demographic
group.
Weighting can significantly modify (often reduce) exposure.
6/2012
© The LifeLine Group
186
US-Environmental Protection Agency
Office of Research and Development
93
-------
xampl
High-end aggregate exposure from
multiple products is reduced by a factor
>4x by proper weighting for gender and
age. More detail more accuracy.
Miles
50
c
! 30
10
D Cere is Data
i: - :•;
n
hrilrflr
1 Hi rfl m
'
.
<1S 10-24 £34 £44 4554 5564 65+ Total
M
AgtGraups
Females
10
m riTrn m rfl rn
<1B &24 2534 £-44 4554 5564 65+ Total
F
ilge Groups
a 200'
m ™ .
150
t 50^
0
-•-Weighted Male and Female
Weidhted Ade Grduos alsd
D 20% 40% 60% 80
Percentile of Exposure
,
1
/
/ .
/ ,
o
r-^/
% 100%
6/2012
© The LifeLine Group
187
ev Learnin
Consumer panels may not accurately reflect the general population or
sub-population of interest
Data on high-end frequency and consumption patterns are often less
substantiated compared to average frequency and consumption. For
example, when > x times per day use of product represents a significant
percent of the responses.
Co-use or non-use of products as well as extent of use of ingredient has
an important impact on aggregate and population exposure estimates
The higher the complexity of the model (i.e. the better exposure variability
is reflected in the model) the more accurate the exposure estimate
Probabilistic approaches to aggregation can result in a factor of 2 to 3
decrease in exposure compared to conventional additive approach
6/2012 © The LifeLine Group
188
US-Environmental Protection Agency
Office of Research and Development
94
-------
PerCEIVERS
~! US-EPA in RTP-NC
Day II Presentations
-June 27, 2012
RSC ChemSpider - A crowdsourced
community environment for hosting
and validating chemistry data
PERCEIVERS Meeting, June 2012
US-Environmental Protection Agency
Office of Research and Development
-------
ChemSpider
• The Free Chemical Database
• A central hub for chemists to source information
• >26 million unique chemical records
• Aggregated from >400 data sources
• Chemicals, spectra, GIF files, movies, images,
podcasts, links to patents, publications,
predictions
• A central hub for chemists to deposit & curate data
RSCI SSS&. sKj ChemSpider
We Want to Answer Questions
• Questions a chemist might ask...
• What is the melting point of n-heptanol?
• What is the chemical structure of Xanax?
• Chemically, what is phenolphthalein?
• What are the stereocenters of cholesterol?
• Where can I find publications about xylene?
• What are the different trade names for Ketoconazole?
• What is the NMR spectrum of Aspirin?
• What are the safety handling issues for Thymol Blue?
RSCISSSSSU SO! ChemSpider
US-Environmental Protection Agency
Office of Research and Development
-------
I want to know about "Vincristine"
• 2O 30 S*ve Zoom
•V
Vincristine
CflCTiSpKJer ID 97M
Mc4ecuiar Formula
MoooaotopK mass 824 399644 Da
(2o 2T& 3o 4a,sp.i9p).22<»ovir>ealeul
» SMILES and mew*
Wttbox
EmdcO
Deprecal*
wacn ma recort
; I,., |K ,l S| • . .
f=y ChemSpider
Vincristine: Identifiers and Properties
» Names and Ideiitifiere
MO Jit I iTV.F
W -M I
MHlMH
^illlMIMlllllllll ^
vlN
* Properties
il.l.
O[ ChemSpider
US-Environmental Protection Agency
Office of Research and Development
-------
Vincristine: Vendors and Sources
Chemical Vendors
m • »
*
S-PTNJ1W
1011 10118
• Data Sources
MM1D
•
; I,., |K ,1 S| • . .
f=y ChemSpider
Vincristine: Patents
' Patents
OM«kPM>mi lAPtpOlMM USPI04«jik«oM tlinwwi
itat fit
-------
Vincristine: Articles
OSCf**. ».*)«]• VMMUWM. <..^fl-..
*» h«n b _4 W W iltan* (HUM
• *»NB» ven-cvAO*
Jtomto 2011*1
*W ^ In :a l«Ju-. AMI
O |«««.*
«» tfwir
Vduw 147 ttuuj
< F«« Jte feMMnto «l NM«
i fkUii tU MllMKl uw5i *im me same *taripion
' Uve mn molecule in a ttruouie searcfi
S*arch •«t*rn]l sit*s tor this structur*:
4 Seaich Google Scnctai (by synonym)
•?! S<;wch Google to« exact siiu
*1 Search Google (o> siiuctucrj with same sneielon
O[ ChemSpider
US-Environmental Protection Agency
Office of Research and Development
-------
Validated Names for Searching...
scholar ~v«c"**w"|wvTO'*«*""Ofi"*7a-ro|l
Scholar Mr.\n ttuuw HUHI ' m,limt - mtlalt cumin - [~TjO.M. .m^ J^»MII»J 1 -1« rf jteu HUM HMtl
flirt , u: nr.ir •ymcnssne" OR
OR "lybpi
OR "V-g-T" OR ^g-OiiCTnrealpiiKobla^inB'
22-onp. I
rq o resistance m P388 )a*emu m vivo and •m>ot»on[*iafh«pcBd
t tunic '•' VCD-nMUffl
t sianrticanlt) iimfaos t» ou
l V
I UnMnc> rf Munch
rf H«m»l UMtcmt» Umtnty H««M ix»-i»^-Hc<*»ti f j
HMDM St Oo"J hintwg Dt|)»llllK«l id MftnH kVr»rn. II I. :n>nm
In mimil Knaniatmn alvtnaMatmHladnfmKin indiuaJMislanl P388 tono eels (olowinQ
IIK^MftQTI llftf frUIT*
Care** O0*Tvt*>rwy !">*•« Jvpimt** FouvMipn ** C*tr*< RC*C«K* TwNmUnj fifty*
T70 JBFW -Spr^c t»c»j"i antigcmifs vttf c«n»o(M«> irffMcgrs cntw
lt»*l» u* vMcmwk» ano Ad^-artr/ttn in -*ncin tv* •"! Aftmnr^cn-fmiiiPt P3M .
GO
And InChls.
1
OGWKCGZFUXNPOA-XQKSVPLVSA^N
. . • -
v .. ; •• : c
-
CNHSMMOIO nCMKfy nCNK^-OOWKffjrUWTO*
i(CHEBI2S445}
thtmSpdwC l.'M (totootti tamu, CMHWUO10
UVOMIOIK mwj >?< )9WUOi jyrtii JIL iwit -
ixA»i«;(iff IWHPIM xQKivm.ru N
-.--
ViitJiiiline CAS 57-22 -7
US-Environmental Protection Agency
Office of Research and Development
-------
Chemistry Databases on the Internet
• Public databases are "trusted" as primary sources
Trust is granted without investigation of the
content ~
Online data vary dramatically in quality!
ChemSpider
With Great Fanfare.
ifi)rtjn»r 3inioi
NIH News
National Inililutet of Health
'_• . . •- • '< Contact
Wednesday Ann 27. 2011 G*an Soe*.* NHGRI
2pm EOT 30MD2-OT11
NIH researchers create comprehensive collection of approved drug* to Identify new therapies for
rare and neglected diseases
Reseatcnere haw begun screening the first oeflnnve cMecton of irxxjsarws of appravea drugs T« ciirucal use
against tare and neglected diseases Tney are hunting tar additional uses of the drugs hopeig to flnd off-taoel
thefapies. lot some o( irte 6.000 rare diseates that affki 75 milttxi Amercans The effort is coordinated t>> ine
National Insututes of HeaW s Chemical Genomics Center {NCGC t
US-Environmental Protection Agency
Office of Research and Development
-------
NPC Browser
Manginese ptot«
-------
NPC Browser
I.! DvcumcMiff) • Crodurt Ljfcrt (0) F>»9menoWl
••,..!-. • . .
Oj ChemSpider
PHYSPROP Database
The freely downloadable
database under the EPI
Suite prediction software
Very Basic filters suggest
data quality issues
I
.
£V ChemSpider
US-Environmental Protection Agency
Office of Research and Development
9
-------
The Stereochemistry challenge.
12500 chemicals with "missed" stereo
; I,., |K ,l S| • . .
rV ChemSpider
MIST Webbook
('holesterul
• l«n.UC H..O
..^kl 1M.6'.'
i-t
.i.i:-i:.;t«. !-;.;;-:•«,l-j«*ttW.ai-,;::.;:• :.^..' K
Ik.«lllulll»ukBl>f>!Illll£l> I I
(r.... ,-CkolrtTOl. Oaln S«> SMI Cbolnmr Owlim.ol
H. S
ChoUnnt
ol CV>lmn» Own Focal CH L»)U
-------
FDA's DailyMed
e»»ny1 estradwl
•-C-CH
i=CH
; I,., |K ,1 S| • . .
jrV ChemSpider
FDA's DailyMed
O[ ChemSpider
US-Environmental Protection Agency
Office of Research and Development
11
-------
FDA's DailyMed
-p
r
•:, ,'.'
i. ij
; I,., |K ,l S| • . .
f=y ChemSpider
PubChem
PubChem
Compound
Roort* $
-•
1 TAXOL '"'Am
nw lUtiituitna w t.«,Jio,
i
TAXOC
US-Environmental Protection Agency
Office of Research and Development
12
-------
Crowdsourced "Annotations
• Users can add
• Descriptions/Syntheses/Commentaries
• Links to PubMed articles
• Links to articles via DOIs
• Add spectral data
• Add Crystallographic Information Files
• Add photos
• Add MP3 files
• Add Videos
; I,., |K ,1 S| • . .
jrV ChemSpider
~ Description
i i«
•in Oowrmaft 9mM C
* ..
o! ChemSpider
US-Environmental Protection Agency
Office of Research and Development
13
-------
Community Contribution to ChemSpider
»tpr •-•'«•=
.
MBH1
I •
i
•
itutu
•.tax
tax
-ma
225 »> i':. 110 i» i» w M n o
; I,., |K ,1 S| • . .
f=y ChemSpider
Structure Database Lookup
jry ChemSpider
US-Environmental Protection Agency
Office of Research and Development
14
-------
SciMobileApps.com
Scientific
Mobile
Applications
ChemSplder Mobile
o ii i aui »••<»»« "mips
«* t**ty
4 MVP
, -nw'i >»»«»•»'••
•* tfnidh**. w t«*» *^ • M«»q^rt <*lmi t»t «pt • **•»
«» !*»*»• »i « *K •»V«^ It ^tlrt
»ON ••>» w H«rdi top truat^ff h
tal WMK rflM •»(»• NtlMwp
- —. r' »—.TI(. , 5
«*t*M.«ta • • = -.; •«•»#'.£«
Future of Chemistry on the Web?
• Public compound databases federate & build
a linked environment of validated data!
Data validation needs are not ignored
Public-Private databases can be linked
Open Data proliferate
jry ChemSpider
US-Environmental Protection Agency
Office of Research and Development
15
-------
Conclusions
ChemSpider is a FREE resource for the community
Grows daily with new data
Concerned about data quality!
Crowdsourced and algorithmic curation is working
API is available to access data
Anybody interested in depositing data?
Any interest in accessing API?
; I,., |K ,1 S| • . .
jrV ChemSpider
Thank you
Email: williamsa@rsc.org
Twitter: ChemConnector
Personal Blog: www.chemconnector.com
SLIDES: www.slideshare.net/AntonvWilliams
jrV ChemSpider
Building community f
-------
Household Products Database
and tools for consumers
Presentation at:
Personal chemical exposure informatics: visualization, user Experience,
Research in Systems modeling and Simulations (PerCEIVERS)
Research Triangle Park, NC
June 26-27, 2012
Henry DeLima
DeLima Associates • 1227 Providence Terrace • McLean • Virginia • 22101 • 703-448-9653
Pertti Hakkinen, PHD
Acting Head, Office of Clinical Toxicology • National Library of Medicine, National Institutes of
Health • 6707 Democracy Blvd. Suite 510 • Bethesda • Maryland • 20892 • 301-827-4222
and
Adjunct Associate Professor in Biomedical Informatics • Uniformed Services University of the Health
Sciences • F. Edward Hebert School of Medicine
Household Products Database
Overview
* Background
* What's in it?
* How do we select Brands for inclusion?
* Sources for Brands-Specific Data
* Target Audience
* Site Statistics
* Proposed Enhancements
Household Products Database- 062612
34
US-Environmental Protection Agency
Office of Research and Development
17
-------
Household Products Database
Background
* Sponsor: Centers for Disease Control & Prevention
* Objective: Develop and maintain a brands-specific
consumer product database with:
o List of ingredients
o Acute and chronic health effects
o First aid information
o Safe handling and disposal procedures
* Launched: In 2001 by National Library of Medicine
Consumers' 24/7 global online gateway into NLM's databases: 30,000
page views/day
Household Products Database- 062612
35
Household Products Database
What's in it?
•:• For each of over 12,000 Consumer Products in 9 Product
Categories: m^^^^^^^^^^^^^^m
Product Image & Description
Manufacturer Information
Ingredients from Labels
and Safety Data Sheets
Properties and Data for
Individual Ingredients from
NLM's Suite of Databases
Health Effects for Product
First Aid Guidance
*T
«»
Household Products Database- 062612
36
US-Environmental Protection Agency
Office of Research and Development
18
-------
Household Products Database
How do we Select Brands?
* Selection of Brands
o Major Manufacturers and Market Share for Each
Subcategory
o Shelf Presence in Stores
o Consumer Requests through National Library of Medicine
o Unsolicited Requests from Manufacturers
o Brand Highlighted by Media
37
Household Products Database- 062612
Household Products Database
Where do we get Brand Information?
* Data Sources (Established, On-going Process)
o Labels of Products
o Manufacturer's Web Sites
o Safety Data Sheets obtained directly from Manufacturers
o Manufacturers' Health & Safety/ Regulatory Affairs Offices
o Manufacturers' Customer Service Agents
Household Products Database- 062612 „
US-Environmental Protection Agency
Office of Research and Development 19
-------
Household Products Database
Target Audiences
•:• Consumers
o To identify the chemicals in products
3 To determine the health effects of product ingredients
o To try to avoid brands that have some ingredients
o To contact emergency health line
o To have access to brand-specific First-Aid and risk management information
•:• Researchers
o To help with design and conduct human exposure studies and risk assessments
•:• Government Regulatory Agencies
o To identify additional chemicals to watch, and for possible regulation
o To determine compliance with occupational and environmental laws
•:• Physicians, and Hospital Emergency Departments
o To identify chemicals in products used by patients
o To determine the health effects of product ingredients
o To contact the product manufacturers for patient management information
39
Household Products Database- 062612
Household Products Database
Site Statistics
* Average Daily Page Views: >30,000
o Household Products Database (http://hpd.nlm.nih.qovj
o Consumer Product Information Database (www.whatsinproducts.com)
* Top Search Terms
o Ingredients
o Safety Data Sheets
* Rankings of Users by Country
o USA: 1
o Canada:2
o EU (Combined): 4
o UK: 6
o Germany: 8
40
Household Products Database- 062612
US-Environmental Protection Agency
Office of Research and Development 20
-------
Household Products Database
Current and Proposed Enhancements
* Global Taxonomy of Product Categories and Products
o Product Categories and Subcategories are based on U.S. Practices
o Propose Alternate Categories to be Compatible with E.U. and Other Product Nomenclature
•:• Include Products that contain Nanomaterials/nanoparticles:
o Identify specific ingredients present in nanoform
o NLM has MOU's with CPSC and EPA relevant to this
* Identify Products That Comply With Environmental Standards
o Allow Users to Search Products that Comply with Government-endorsed Environmental Standards
such as EPA's DfE and ECO Labels in Europe
•:• Provide Links for Worldwide Information Relevant to Product Subcategories
o Examples of Information Sources: RIVM, BfR, AFSSET, KTL, EC-DG/JRC
o WebSites, Reports, and Peer-reviewed publications
o Provide search strategies, e.g., for a product category, for using NLM's PubMed database
* Provide Educational Module(s) for Consumers and Others
o Alternative Products and Proper Use, Storage and Disposal of Products
Household Products Database- 062612
41
Household Products Database
In Progress: Pilot EU-Version of Database
* Contains Products Sold in 8 Countries:
DE, EE, ES, Fl, FR, IT, NL, UK
•:• Health Effects Information Provided in:
Country-Specific Languages
Chemicals Classifications Provided for Each Ingredient
o Hazard Symbols, Risk and Safety Phrases (67/548/EEC)
o Hazard and Precautionary Statements and Pictograms (GHS)
o Substances of Very High Concern (SVHCs) Identified
Chemical Property Links to HSDB, Toxnet, ECHA-Chem, Etc.
Safety Data Sheets provided for all Products (except Cosmetics)
Product Ingredients can be Compared between Countries
Household Products Database- 062612
42
US-Environmental Protection Agency
Office of Research and Development
21
-------
Entropy in Personal Chemical Informatics
:. iQ/rnt Jjit3& o/te coot, 6col-ot it ciomxx
-------
ife , contrary* to t&& f&ve^ae trndme^ dictatedt>a the, 21 caw,
decreases or maintains its entropu du, fteedinQ on negative entropy
-Erwin Schrodinger
I d tfoofc ftor an entropu redaction, since this mst de a aeneratf
characteristic oft cifte
-James Lovelock on a theoretical life detection system for NASA
Limits of Predictability in
Human Mobility
0(e>a.tfa.re>tf OK wtrotoM
vo. x" tomes
fffj /l/imUr oft &&Mplume* tow&F-s tn'a>a>&t^o[ i/vajx^er oft noct&s u-isitedj
+ 7~i'tn& sp&nt @ tow&r- (+ P^oliaJufit^ oft w'sitintfi a. nod&J
+ &au.e-fi&e oft towers f+ Qro(e>t<- oft ins/tinj> nodesJ
dad£3 jt'0&$ 93% &t*edi&ta&fi
US-Environmental Protection Agency
Office of Research and Development 23
-------
Entropy is a measure of predictability in activity
Activity is essential to exposure model
ft (A&tivjfy, Cke
-------
Tulve et al. 2006. Environ. Sci. Technol., 40, 6269-6274
• National Survey of pesticide residues in child care centers
• 168 child care centers
• Surface wipe samples from indoor surfaces
Consider subset 168 x 15 pyrethroid concentration matrix
•
n C
OK Sbe&i
e&es
Sites
Species
Carduelis dominicensis
Loxia leucoptera
Volatina jacarina
Sprophilia nigrricolis
Melopyrha nigra
Loxigilla portoricensis
Loxigilla violacea
Loxigilla noxis
Melanospiza richardsoni
Tiara olivacea
Tiara bicolor
Tiara canora
Loxipasser anoxanthus
Saltatoralbicollis
Torreornis inexpectata
Ammodramus savanna rum
Zonotrichia capensis
010000
010000
000000
000000
100000
000100
011000
000011
000000
111100
1 0 1 1 1 1 1
100000
001000
000011
100000
011100
010000
475433
0
0
0
0
0
0
0
1
0
0
1
0
0
1
0
0
0
3
000000000000
000000000000
000100000000
000100000000
000000100000
000000000000
000000000000
111110011110
100000000000
000000100000
111111011111
000000000000
000000000000
100000000000
000000000000
000000000000
000000000000
422421222221
1
1
1
1
2
1
2
12
1
5
17 |
1
1
4
1
3
1
55
3-way mixture
US-Environmental Protection Agency
Office of Research and Development
25
-------
OK if
Environmental; surface loading (ng/cm2) -> presence/absence (0,1)
Community Ecology; species abundance (X) -> presence/absence (0,1)
... ... .
,|.DO. .••»'=»'>-MID
4
•
OKKu
•
.
-- .
' .
stm: , .
•*•
*»J
*•«
• IJ
- M
•
, t
•
Klit^i-ditf nplm-«l Anul\«l«. i.f < hriHK-jil < n-Ocmrrcmc UMIII
<« lcl
-------
okseriwit watr-ix
to MM nodec OK interest
U
decreasing
entropy
Table 1. Nine null models based on the observed presence/absence matrix
kColums
Equally likely
Rows Sim 1
Equally Likely P(Xij)=l/RC
Constraint: N
Rows
Proportional
Sim 7
P(Xij>Si/NC
Constraint: N
Sim 2
Sim 6
P(Xij>Tj/NR
Constraint: N
Sim 8
P(Xij>SiTj/N2
Constraint: N
Sim 4
P(Xij>Tj/N
Sim 3
P(Xij>l/R
Constraint: Tj
Sim5
P(Xij>Si/N
Constraint:!]
Sim 9
PfXijHMarkov pi
Constraint: Si,Tj
Adapted from Gotelli etal. (2000). Each entry gives the abbreviation for the null model, and a formula
for calculating the probability of occupancy for the first cell in the matrix, P(Xij); N= total occurences, R=
number of rows, C=number of columns, Si= sum of i-th row, Tj= sum of j-th column
Birdlike!
Co-Occurrence Pattern in Tulve's Child Care Center
data shows structure like West Indian Finch Matrix
I WJ. II Sul M..1 J AMVV1 .1 !k C1UU < i
UMIII .11 MJ-JH limivd WefltalUi
>'•
Ml
NWI
Hilt
•
DIM klU
)4
11 *l
14 W
> . »*mo
...Mill.
«iW
i:.i»tki
.III. Kl K
»
wr
.
KlMfKI
h>M"
Kfl
l?tfwa
nil inn
M
I» Sp-
in 44
•'
.
.M ,!
Ill
1. M--
IT l»"
1* 14"
«- -MI j\im ihrm «nJ nrrt(wwl% I Md II
niiliiMiti iiuiili i ii«l. irian<«cOn«. pcnxorx. nl pin-inn I'- i III . Kl K
B U U u. *\ H-1 .Z «».«. IT i iiMbn ul «mli(uK»).
: onrt. ••-««• '-- 'UTS
US-Environmental Protection Agency
Office of Research and Development
27
-------
e&OKoa/f&aK patterns
•7 .«.iini»iininiiii
are Etner$w
to chetn/ca.1? exboswes f
anal uo«sr Patterns
HUMAN EXPOSURE ASSESSMENT STRATEGIES FOR
CONSUMER PRODUCTS
Treye A. Thomas, Ph.D.
U.S. Consumer Product Safety Commission
Office of Hazard Identification and Reduction
Bethesda, MD
June 26-27, 2O12
This report was prepared by CPSC staff; it has not been reviewed or approved by,
and may not necessarily reflect the views of, the Commission.
US-Environmental Protection Agency
Office of Research and Development
28
-------
/ Pnmmiccinn
I I Q Pnnci impr PrnHi irt
Independent regulatory agency (1973)
Thousands of products in and around the home
Generally, food, drugs, cosmetics, medical
devices, pesticides, automobiles not included
Does include child-resistant packaging for
household chemicals, drugs, and cosmetics
Staff of -540; budget of $118 M
5 Commissioners appointed by President
CPSC National Product Testing & Evaluation Center
• New, modern lab-office location - S Research Place, Rockville, MD
• 63,000 sq ft (32,000 sq ft of laboratory testing space vs 13,000 at old site)
• Lease Awarded & Design Initiated May 2009
• Construction started April 2010
• Moved in and Operational May 2011
• ~75 Engineers, Scientists, and Support Staff
US-Environmental Protection Agency
Office of Research and Development
29
-------
CPSC National Product Testing & Evaluation Center
Testing Areas
Toy Test Lab Children's Products Lab Pool and Spa Products Lab Impact Lab (Bike Helmets)
General Product Test Lab Outdoor Power Sports Lab E|ectrica| Products Test lab Chemistry Lab
Combustion Products Test Lab Modern Conference Space Machine Shop Flammability/Fire Test Lab
Federal Hazardous Substances Act (FHSA)
• Risk-based
• Considers toxicity, exposure, and
bioavailability
• Human experience takes precedence over
animal data
• Includes acute and chronic effects
• Includes reasonably foreseeable misuse
• Mouthing by children
• Does not require specific testing for chronic
hazards
US-Environmental Protection Agency
Office of Research and Development
30
-------
Chronic Hazard Guidelines
• Released in 1992
• Provide guidance in assessing risks from acute and chronic
hazards
• Carcinogenicity, neurotoxicity, and
reproductive/developmental toxicity
• Exposure
• Consider all sources of information
• Bioavailability
• Acceptable daily intake (ADI)
• Route-specific exposure limits
• Inhalation
• Establishing limits for susceptible populations
• Risk assessment
• Acceptable risk
Guidelines - Exposure Assessment
• Field data preferred
• Pollutant levels in indoor air
• Laboratory data
• Emission or migration data
• Supplemented with mathematical models
• No available exposure data
• Surrogate chemical
• Theoretical model
US-Environmental Protection Agency
Office of Research and Development 31
-------
Consumers
• Best estimate of "typical consumer" (50th percentile)
• Upper (95th percentile) and lower bound (5th
percentile) screening
• Demographics
• Variation in use patterns
• Frequency of use
• Time activity patterns
• More time spent indoors
Especially among children
Data Gaps for Exposures from Consumer
Products
* Product formulations
• Product release and residue data
• Variation by chemical and product
• Frequency and duration of use of product
• The proportion of the population using product
• Scope of uses associated with products
Secondary chemical by-products of health concern
• Diversity of products
• Matrices (e.g., plastic, textile, household
chemicals)
• Variations within a product class
• Coatings and paints
US-Environmental Protection Agency
Office of Research and Development 32
-------
Considerations for Exposure Assessment
• Consumer environment
• Housing types
• Size and configuration
Ventilation
Sinks
• Carpets, furnishings
Chemical Fate in the Indoor Environment
Ambient air concentrations
Form of the compound released
• Gas
• Particulate
• Chemical transformations???
• Concentrations of reactive compounds
• Photolysis
• Effects of by-products
• Deposition and sequestration
Absorption and re-release
• Removal mechanisms
• Ventilation
• Seasonal variation
• Air cleaning devices
US-Environmental Protection Agency
Office of Research and Development 33
-------
Estimating Exposure by Route
• Inhalation
• Direct monitoring
• Available "validated" methods
• Collection efficiency
• Modeling
• Surrogate data
• Ingestion
• Extraction with simulated saliva or gastric juices
• Hand-to-mouth activity
• Concentrations on hands
• Removal efficiency
Estimating Exposure by Route
• Dermal
• Quantify material leaching from product
• Estimating amount of substance in contact with
skin
• Surface area of skin contacted
• Duration, frequency of contact, thickness of
liquid interfacial layer
• Hand wipes (e.g., CCA validation study)
US-Environmental Protection Agency
Office of Research and Development 34
-------
Surveillance Data Bases
n Injury and Potential Injury Incident
Data (IPII)
n Death Certificates (DTHS)
n In-Depth Investigations (INDP)
n National Electronic Injury Surveillance
System (NEISS)
Overview/History
n National sample of 96 hospitals from all U.S.
hospitals with at least 6 beds and 24-hour
emergency service.
n Each hospital reports information on
emergency treatments to CPSC.
n Hospital coder enters data in local PC and
transmits the data to CPSC over the internet.
n System collects ~ 400,000 product-related
injury reports each year.
- (~ 300,000 non-CPSC injury reports each
year).
n Multi-level system.
n Supports CPSC and other agencies.
US-Environmental Protection Agency
Office of Research and Development 35
-------
Consumer Product Safety
Improvement Act of 2008
Signed into law in August 2008
Sec. 212. Establishment of a Public
Consumer Product Safety Information
Database.
Searching for Incidents and Recalls
. . .
• • I * • V —
-
Search
US-Environmental Protection Agency
Office of Research and Development
36
-------
Thank vou
• CPSC Website
• www.cpsc.gov
• Incident reports and recalls
• www.saferproducts.gov
• Chronic Hazard Guidelines
• http://www.cpsc.gov/BUSINFO/chronic.pdf
What is GoodGuide?
Web-based platform that
tracks the health,
environmental and social
aspects of products,
brands and companies -
covering over 175,000
products from 5,000
companies
Tools that deliver this
information at the point
of purchase - changing
the buying decisions of
consumers, retailers and
institutional purchasers
74 Confidential
GoodGuide
US-Environmental Protection Agency
Office of Research and Development
37
-------
Scan a Bar Code -> Instant Shopping Advice
Gillette Foamy Shaving Cream.
Regular
Overall
3 Health
5.4 Environment
5.5 Socwty
Your Fitters
Detailed Information about Product
Ingredients, their Potential Health Hazards &
Regulatory Status
3.0 '•" pw*«s COTMW on* a mm nf*omm KH mm * mMun m* t*
O MMTOrt H»«tth lmp*clm
•QHMTOM •ur.i» V HrtHKIK
76 Confidential
GoodGuide
US-Environmental Protection Agency
Office of Research and Development
38
-------
Product Health Ratings - Personal Care
Categorize ingredients by level of
health concern based on:
1) Number of recognized & suspected
health effects
2) Relative toxic potency
3) Detection in US population
4) Adequacy of toxicity data set
Assign product scores by counting
ingredients by level of concern:
0 -1 High concern
2-4 Medium concern
5-8 Low concern
10 No ingredients of concern
GoodGuide
Categorizing Chemicals by Level of Concern
GoodGuide's classification
system combines data from
hazard identification, potency
estimation and exposure
assessment
- Improves on simpler
classification systems that are
hazard only
Twelve health effects are
tracked, with strength of hazard
identification reflected in
suspected vs recognized list of
hazards
GoodGuide is not making its
own expert judgment that
chemical x causes health effect
y
78 Confidential
Health effect lists are compiled
from authoritative scientific and
regulatory sources
- Scientific sources include IARC,
NTP, medical and toxicology
texts and review articles
- Regulatory sources include
CalEPA, EPA IRIS, EU CLP
Health effects are not identified
based on single or controversial
scientific reports - available
scientific data must be reviewed
by a scientific or regulatory 3rd
party that then publishes a
hazard identification
— Different approach than EWG
or other NGOs
GoodGuide
US-Environmental Protection Agency
Office of Research and Development
39
-------
Modifying Screening-level Ratings with Better
Data
Presence of "bad actor"
ingredients in a product is a
warning signal
To further evaluate whether
there is a potential health risk,
need:
— % ingredient in formulation
— Exposure potential from typical
product use
— These data are often not
available from public sources
Rating system can adjust product
scores to suppress an ingredient
when data indicate the
ingredient is unlikely to pose a
health risk because it is
— present below a regulatory
threshold, or
— present in a product category
unlikely to result in significant
exposure
Pre-empts criticism of hazard-
based scoring by challenging
manufacturer to provide the
data required for risk assessment
Example safe use determination:
— Ethyl acetate is "known to be
neurotoxic in man"
- Ethyl acetate is used as a solvent
in nail care cosmetics
- Low % formulations and product
category usage does not result
in the high doses required to
elicit neurotoxicity
| Level of Health Concern of Ingredients [-] ©
HIGH CONCERN Ethyl Aootalo B«tow T)ma*)ld
LOW CONCERN Butyl Acetate
Collodion
GoodGuide
Backing Up Our Health Ratings
Ultra Dawn Antibacterial
I Dishwashing Liquid, Orange
0
c* Wy »*v Gsmw ««»
OMCM pnpwn ww • D*rg
80 Confidential
GoodGuide
US-Environmental Protection Agency
Office of Research and Development
40
-------
Ingredient Profiles
Tnclosan In Dishwashing Guide
Scientific
& Regulatory
Basis for
Level of
Concern
fi IW*M • hlfh Hml
-------
Purchase Analyzer - Acquiring a Consumer's
Entire Product History
BooodGuKJe
B
u-
y «_
e
U :
GoodGuide
Relevance to Personal Exposure Informatics
Comprehensive catalogue of
current consumer products,
their ingredients and other
attributes relevant to
assessing health risk
Capacity to acquire personal
product purchase histories
for consumers
Distribution platform that
engages over 1 million
consumers per month who
are interested in health
impacts of consumer
products
Key data required to integrate
exposure considerations into
health ratings are unavailable:
- Limited characterization of
exposure potential associated
with specific categories of
consumer products
— Manufacturers generally
unwilling to disclose ingredient
percent formulations, so
impossible to assess
compliance with safety
benchmarks
- Essentially no data available on
actual exposures experienced
by consumers under different
use scenarios
84 Confidential
GoodGuide
US-Environmental Protection Agency
Office of Research and Development
42
-------
Increasing Consumer Ease of Use :
Augmented Reality App
GoodGuide
The Quantified Self &
Chemical Sensing
Michael Nagle
http://www.thesprouts.org
@.naale5000
US-Environmental Protection Agency
Office of Research and Development
43
-------
What is the
Quantified Self?
«»••
»• •••• ••»••••
•• •••• •••••••
•• •••• •*•*•
«••• «»•» **••
*********** ••••••••»
•*•*•••• ••*•»
Quantified Self
Example 1:
Verifying Hypotheses
US-Environmental Protection Agency
Office of Research and Development
44
-------
Example 2:
Identifying Causes
Sensing Chemicals 1
Portland Smells
Portland metis
US-Environmental Protection Agency
Office of Research and Development
45
-------
Sensing Chemicals 2
Asthamapolis
Sensing Chemicals 3:
Public Laboratory
US-Environmental Protection Agency
Office of Research and Development
46
-------
Thank you!
Drop me a line if you have questions
:: nagle@thesprouts.org
GPS and Exposure Assessment
for Individuals
Michael Breen
National Exposure Research Laboratory
Human Exposure and Atmospheric Science Division
Exposure Modeling Research Branch
US-Environmental Protection Agency
Office of Research and Development
47
-------
Exposure Model for Individuals (EMI)
•IMMW* Modd tor ImfcvtrfiMfe lIM)
I Office of Research and Development
I National Exposure Research Laboratory, Human Exposure and Atmospheric Sci
EMI Web site: www.epa.gov/heasd/emi
Exposure Model for Individuals (EMI)
I Office of Research and Development
I National Exposure Research Laboratory, Human Exposure and Atmospheric Sciences Division
EMI Web site: www.epa.gov/heasd/emi
US-Environmental Protection Agency
Office of Research and Development
48
-------
Tiered EMI Metrics
Inputs
Building
Characteristics
A Operation
MtUorotogy
C«ntrr»l-»i!* Outdoor
Concentration*
Building Outdoor
Concentration*
EMI
Indoor Air Quality
Information
Needs &
Metrics Complexity
Air Exchange
Rate Model
infiltration
Tlme-Locatton-
Actrvltw*
Time-Location
Rates
Indoor
Concentration*
Personal
Exposure*
I Office of Research and Development
I National Exposure Research Laboratory, Hu
an Exposure and Atmospheric Sciences Division
EMI Web site: www.epa.gov/heasd/emi
Tiered EMI Metrics
Inputs
EMI
Indoor Air Quality
Characteristic*
& Operation
Meteorology
Central-Kite Outdoor
ConcentiatKant
Building Outdoor
Concentrations
Air Exchange
Rate Model
Infiltration
Model
Information
Needs &
Metrics Complexity
Air Exchange
Rates
Infiltiation
Factor*
Indoor
Concentration*
Tiroe-Locatmn-
AcUvltte*
le-Locatlon Model
I Office of Research and Development
I National Exposure Research Laboratory, Human Exposure and Atmospheric Sciences Division
EMI Web site: www.epa.gov/heasd/emi
US-Environmental Protection Agency
Office of Research and Development
49
-------
Microenvironment Tracker (MicroTrac)
Qstarz GPS Travel Recorder
(BT-Q1000XT)
• Addresses critical need to develop exposure
metrics for health studies:
• Estimation of time spent in various
microenvironments
• Classification algorithm to estimate time-of-day
and duration in microenvironments (in-transit,
home, school, work) based on:
• Position and speed timelines from GPS data
loggers
• Marked boundaries of buildings using
Google Earth
• Potential to reduce challenges with diaries
(participant burden, inaccuracies, missing data)
• Ability to use smartphone GPS data
nramrtntai Ptmctton
Google Earth: Marked Boundaries
School
Home
Work
100
US-Environmental Protection Agency
Office of Research and Development
50
-------
&EBV
Classification Algorithm: Microenvironments
-
Work
-
Other (store)
•
Home
1. Home-indoors
2. Home-outdoors
3. Work/School-indoors
4. Work/School-outdoors
5. Other-indoors
6. Other-outdoors
7. In-transit
Ernramrtntai Ptmctton
.!„..„
Classification Algorithm: Home
Home
GPS positions,
Building boundaries
Yes
Dope:
Home-
No
Position
within 50m
radius of
sidence?
Continue
Done:
Home-
outdoors
102
US-Environmental Protection Agency
Office of Research and Development
51
-------
&EBV
I i cslSlHM
l*TtFsrmtrai
Vi-
-------
MicroTrac Evaluation for Participant 1
24 hr dataset (17,280 samples): processing time = 36 sec
Home-indoors
Work-indoors
In-transit
School-indoors
Work-outdoors
School-outdoors
Home-outdoors
Other-indoors
Other-outdoors
10
^^m 5.5
• 5.0
1 0.8
1 0.7
1 0.6
1 0.5
1 0.7
0.1
1 1.0
0.0
0.0
0.0
0.0
1 60.6
LI Measured (Diaries)
I I Modeled (MicroTrac)
20
30 40
Percentage of day
50
60
70
***«!
Summary of MicroTrac
• Evaluating feasibility from pilot study with concurrent GPS
and diary data
• Limitations:
• GPS spatial resolution (~3m)
• Possible spatial errors near large buildings and dense
clusters of trees (multipath errors)
• Small, lightweight, and low cost GPS devices have potential
to address challenges with diaries
• Using smartphone GPS data, MicroTrac allows for real-time
determination of microenvironment to support personal
exposure APPs
• Office of Research and Development
• National Exposure Research Laboratory, Human Exposure and Atmospheric Sciences Division
US-Environmental Protection Agency
Office of Research and Development
53
-------
Getting Chemical Information
to Communities
The EPA's Community-Focused Exposure and
Risk Screening Tool (C-FERST)
Shannon O'Shea, Andrew Geller, Brad Schultz, and Valerie Zartarian
CFERSTMail@epa.gov
PerCEIVERS workshop | RTP, NC | June 27, 2012
What is C-FERST?
C-FERST is a community mapping and assessment tool to
inform environmental public health decisions
What does C-FERST do?
Enhances access to info for
community decision-making
Provides venue for technical
assistance, science
Assists with identification and
prioritization of community
environmental health issues
Fosters sustainable and
communication, collaborations healthy communities
U.S. Environmental Protection Agency
108
US-Environmental Protection Agency
Office of Research and Development
54
-------
Conceptual Framework
^modeled local exposures
"'guidance on local measurements
^cumulative risk science
Bother info useful to communities
•;s info
nes and risks
in best practices in other communities
G Generate community reports I
5/14/2013
U.S. Environmental Protection Agency
109
Structure of Estimates
& Indicators in C-FERST
Ambient
concentrations
Human exposure
estimates
Biomarker estimates
Risks/Health impacts
- Cancer, Asthma, Early neurotoxicity effects, etc.
U.S. Environmental Protection Agency
110
US-Environmental Protection Agency
Office of Research and Development
55
-------
5/14/20
l~- Jill... •.. «»^-.,..»».-
5/14/2013 U.S. Environmental Protection Agency
US-Environmental Protection Agency
Office of Research and Development
56
-------
Using C-FERST with
Community Assessment Guidance
• CARE Roadmap
• PACE-EH
• EJ Toolkit
• HIA roadmap
• Tribal assessment roadmap
• Community-Cumulative
Assessment Tool
5/14/2013
U.S. Environmental Protection Agency
113
Future Plans
Cumulative exposure and risk science
- Community-level exposures: lead, radon, ETS
- Pathways: fish consumption, near-roadway
- Effects: lung cancer, asthma, early neurotoxicity
- Measurements: citizen science, community sensors, etc.
Integrate with other programs and tools
Sustainability and risk management aspects
Continue to refine, incorporate feedback and
broaden use of tool for community applications
Peer review for public release via website
U.S. Environmental Protection Agency
114
US-Environmental Protection Agency
Office of Research and Development
57
-------
Acknowledgements
C-FERST Development Team
- Collaborators in ORD (NERL, OSP, SHC program)
- CARE collaborators (regional project officers, community grantees)
- EPA Region 1 CIS and Technical Team
- EPA Office of Environmental Information (OEI), National Computing
Center (NCC) and contractors
Other Contributors
- EPA Office of Environmental Justice (OEJ) & EJ Coordinators
- EPA Regions & Program Offices
- National EPA-Tribal Science Council & Tribal Partners
EPA/ORD Management for C-FERST Support
5/14/2013
U.S. Environmental Protection Agency
115
George Scheer - Director
Elsewhere @ elsewhereelsewhere.org
Greensboro, NC
Calling in from Berlin, Germany
http://www.youtube.com/watch?NR=1&feature=endsc
reen&v=eTHTnbqel_ss
US-Environmental Protection Agency
Office of Research and Development
-------
Prof. Benjamin Balak
Rollins College, Florida
Balak presented aspects of his technologically-enhanced
pedagogy in a session titled, "Engaging the community for
personal chemical exposure informatics: Gamification, visual
and computer models, electronic and live-action role play,"
participated in breakout sessions, and is working on a video
game to gamify personal exposure.
Dr. Robert Panoff
SHODOR - Computational Science
Non-Profit
http://www.shodor.org/talks/ct-chem/
US-Environmental Protection Agency
Office of Research and Development
59
-------
•% Mutton
50C/AL /
ONGOING
INiep. ACTIONS
Motor IrTXjencos
ENVIRONMENT*
US-Environmental Protection Agency
Office of Research and Development
60
-------
{explanations, claims...}
~i'sci^i;"""~"*-
"me thing that has struck me hardest during thrs whole
month is that 'ever/ones a tittle bit racist' and it is how
you undertstand that racism and how you relate It back
I to the world around you that is the real lesson,'
' \ A. —\/ A
^•^1. :\A-/\ -v\
• *-*
US-Environmental Protection Agency
Office of Research and Development
61
-------
Change
turning DlKourn Facilitation
OltMt thMV)e mttvt
, . . •..
Ih4l *OMn 4 ding* in
Vlicublkn ul J ilw«l -jX
ItNMW
CiuUrubom
ImoarbonrfptKJ
• '
imoucmrf itmn
II-.«.JIH"
UtxtfrfhU
Lh» o> IMUnduti to Ktaw
Suggest Wkwr up *coom
Hntunonil—
It* IVkfMU hit
.1
US-Environmental Protection Agency
Office of Research and Development
62
-------
US-Environmental Protection Agency
Office of Research and Development
63
-------
Appendix E
WWW resources and URLs of interest
This table provides links to non-EPA web sites that provide
additional information and resources. EPA cannot attest to
the accuracy of information on that non-EPA page. Providing
links to a non-EPA Web site is not an endorsement of the
other site or the information it contains by EPA or any of its
employees. The links are accurate as of May 21, 2013.
CP
E
03
z
EXPOCAST
TOXCAST
ACTOR
RAIDAR
USETOX
EPISUITE
SYSTEMSREALITY
EXPOSURE FACTORS
HANDBOOK
CHAD: CONSOLIDATED
HUMAN ACTIVITY
DATABASE
GOOGLE INSIGHTS
CHEMSPIDER
CHEMICALIZE
GOODGUIDE
AMEM
E-FAST
MCCEM
HOUSEHOLD PRODUCT
DATABASE
WHAT'S IN PRODUCTS
SKIN DEEP
SPROUTS
_i
D±
ID
http://www.epa.gov/ncct/expocast/
http://www.epa.gov/ncct/toxcast/
htto://actor. eoa.gov
http://www.trentu.ca/academic/aminss/
envmodel/models/RAI DAR100.html
http://www.usatox.org/
http://www.epa.gov/oppt/exposure/
pubs/episuite.htm
www.svstemsrealitv.org
http://cfpub.epa.gov/ncea/risk/
record isplav.cfm?deid=236252
http:www.epa.gov/chadnetl/chad
2003.html
www.google.com/insights/
www.chemspider.com
http://www.chemicalize.org
http://www.goodguide.com
http://www.epa.gov/oppt/exposure/
http://www.epa.gov/oppt/exposure/
http://www.epa.gov/oppt/exposure/
http://hpd.nlm.nih.gov
http://www.whatinsproducts.com
http://www.ewg.org/skindeep/
http://www.thesprouts.com
consumer product data
X
X
X
X
X
X
X
03
03
-0
cp
(ft
13
-t— '
O
13
T3
O
Q.
X
X
X
X
X
X
_03
"03
T3
"c
CP
T3
CP
no
C
"o
13
T3
O
Q.
X
X
X
X
03
"03
T3
>i
'>
O
03
c
03
E
^
.c
X
X
chemical data
X
X
X
X
X
exposure models
X
X
X
X
X
X
open symptoms data
X
social archiving
X
infoveillance
X
chemical genomics data
X
X
exposure related data, models or frameworks
X
X
X
X
X
-------
CP
E
03
Z
QUANTIFIED SELF
CURETOGETHER
PATIENTS LIKE ME
TWEET ARCHIVIST
CONSEXPO
COMPARATIVE
TOXICOGENOMICS
DATABASE
D±
ID
http://www.quantifiedself.com
http://www.curetogether.com
htto://www. oat ientslikeme.com
http://www.tweetarchivist.com/
archivist-desktop/
http://www.rivm.nI/en/Topics/Topics/C/
ConsExoo
ctdbase.org
03
"03
U
O
CP
b
13
(fl
o
o
X
03
03
CP
(fl
O
O
Q.
X
"03
u
no
o
o
Q.
X
03
03
T3
'^3
03
03
E
13
-C
O
E
CP
o
(fi
CP
O
E
CP
^
(fl
o
a.
CP
-HJ
-o
Ul
£
o
Q.
b
CP
Q.
O
X
X
X
no
_c
o
03
03
'o
00
X
CP
o
03
:=
Ci
—
X
03
03
(fi
^O
O
no
o
E
CP
o
X
o
CP
F
03
CD
(/>
CP
o
E
03"
03
CP
CP
CP
^
(fl
O
a.
CP
X
------- |