-
Title
-
Eng
Washington State Regulations Governing Endocrine Disrupting Chemicals: Current Limitations and Potential Improvements
-
Date
-
2018
-
Creator
-
Eng
Kohnen, Nicholas
-
Subject
-
Eng
Environmental Studies
-
extracted text
-
Washington State Regulations Governing Endocrine-Disrupting Chemicals:
Current Limitations and Potential Improvements
by
Nicholas Kohnen
A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
January 2018
©2018 by Nicholas Kohnen. All rights reserved.
This Thesis for the Master of Environmental Studies Degree
by
Nicholas Kohnen
has been approved for
The Evergreen State College
by
________________________
Edward A. Whitesell, Ph.D.
Member of the Faculty
________________________
Date
ABSTRACT
Washington State Regulations Governing Endocrine-Disrupting Chemicals:
Current Limitations and Potential Improvements
Nicholas Kohnen
Endocrine disruptors (EDs) are a large and pervasive group of manufactured
chemicals that are prevalent in the environment and in our bodies, and that are currently
underregulated. Though the outcomes of exposure to these chemicals vary, scientific
techniques for observing and predicting their effects are shared in common. The most
recent, substantial regulation related to EDs in Washington, the Children’s Safe Products
Act (CSPA), was passed in 2008. Since then, both scientific understanding and public
awareness of EDs has increased dramatically. Based upon what we know now, and on
novel regulatory approaches employed in the EU, this research sought to identify
limitations hindering current and proposed future policies, and suggestions for how future
policies could be improved, based upon scientific advances and areas of agreement
among stakeholders. Through qualitative policy analysis and stakeholder interviews, this
research determined that scientific uncertainty and the lack of political will, defensible
data, and funding were the main impediments to current policy improvements in
Washington State. This thesis recommends that future policy be enacted that employs
improvement in modeling technology to predict the toxicity of chemicals prior to their
use, and that Washington State follow the momentum produced by CSPA, by expanding
the products regulated by the act, basing the regulation on chemical class rather than
individual chemical, and designating the authority for chemical regulation to the
Department of Ecology.
Table of Contents
Table of Contents...........................................................................................................................iv
List of Figures and Tables..............................................................................................................vi
Acknowledgements......................................................................................................................vii
1. Introduction................................................................................................................................1
1.1 Lead as an example of the case for regulation.......................................................................2
1.2 Issues related to recognizing and regulating endocrine disruptors........................................5
1.3 Qualitative policy analysis of endocrine disruptor regulation in Washington State...............8
2. Background...............................................................................................................................10
2.1 Definition............................................................................................................................10
2.2 Human effects of endocrine disruptors characterized by life stage......................................12
2.3 Salient characteristics of known endocrine disruptors.........................................................14
2.4 Historical context of endocrine disruptor proliferation and knowledge...............................17
2.5 Justification for review of Washington State endocrine disruptor policy............................20
2.6 Established approaches to toxics policy..............................................................................22
2.6.1 Regulatory mechanisms...............................................................................................25
2.7 Broader policy considerations.............................................................................................27
2.8 Influences of agencies and extra-judicial policies on state policies.....................................29
2.8.1 State Agencies..............................................................................................................29
2.8.2 Federal and interstate policies......................................................................................30
2.8.3 International policies....................................................................................................32
2.9 Enacted Washington State policies......................................................................................33
2.10 Proposed state policies......................................................................................................36
3. Literature Review.....................................................................................................................40
3.1 State of the science of endocrine disruption........................................................................40
3.1.1 Non-monotonic and low-dose effects...........................................................................41
3.1.2Critical periods of exposure..........................................................................................44
3.1.3 Multiple modes of action..............................................................................................46
3.1.4 Mixture and interaction effects.....................................................................................47
3.2 Transitioning from toxicology to endocrinology.................................................................48
3.3 Models of science-policy interaction...................................................................................51
iv
3.4 Relevant case studies..........................................................................................................57
4. Methods and Results.................................................................................................................61
4.1 Qualitative policy analysis..................................................................................................61
4.2 Research design..................................................................................................................62
4.3 Data collection....................................................................................................................63
4.3.1 Policy documents.........................................................................................................63
4.3.2 Stakeholder interviews.................................................................................................64
4.3.3 Iterative approach.........................................................................................................65
4.4 Data analysis.......................................................................................................................65
4.5 Results................................................................................................................................66
4.5.1 Current limitations of Washington State endocrine disruptor policy............................66
4.5.2 Scientific advancements and endocrine disruptor policy..............................................71
4.5.3 Policy recommendations..............................................................................................74
5. Discussion, Conclusion and Recommendations........................................................................78
5.1 Discussion...........................................................................................................................78
5.2 Conclusion and recommendations.......................................................................................79
References Cited...........................................................................................................................81
Appendices...................................................................................................................................94
Appendix 1: List of interview questions...................................................................................94
Appendix 2: List of codes.........................................................................................................96
v
List of Figures and Tables
Figure 1: Monotonic and non-monotonic dose-response curves
Table 1: Policy mechanisms applicable to endocrine-disrupting chemicals
Table 2: Summary of Eisner’s criteria relevant to policy choice for applicable
mechanisms
Table 3: Division of departmental authority regulating endocrine-disrupting
chemicals
Table 4: Enacted Washington State policies relating to endocrine-disrupting
chemicals
Table 5: List of proposed state policies related to endocrine-disrupting chemicals
Table 6: Policy models in order from most to least reliance upon certainty
Table 7: Stakeholder assessment of current impediments
Table 8: Elements and specific applications of scientific advances
Table 9: Recommended policy changes, grouped by degree of alteration and target
audience
17
23
24
30
34
37
56
67
72
74
vi
Acknowledgements
I would like to thank my reader Ted Whitesell for his hard work, dedication, and
constant helpful suggestions. This paper would have been drastically impoverished
without his guidance. Thanks to Craig Partridge for his assistance in bridging the gap
between my vision and the sphere of policy research. I would also like to thank the MES
faculty for the broad perspectives that they shared freely, many of which found purchase
in the pages of this work. Thanks finally to my family, Alina, Dad & Gram.
This thesis is dedicated to the memory of Lorrie Brown.
vii
1. Introduction
Human beings are exposed to hundreds of endocrine-disrupting chemicals on a daily
basis, and man-made endocrine-disrupting chemicals can now be found in nearly every
environment on earth (Khetan, 2014). As a class of chemicals, they alter brain
development and decrease intelligence (Schug, Blawas, Gray, Heindel, & Lawler, 2015),
alter thyroid maintenance, promoting obesity (Heindel, Newbold & Schug, 2015), and
alter the development of reproductive organs and decrease fertility of humans and
animals worldwide (Kabir, Rahman, & Rahman, 2015; Knez, 2013).
Unfortunately, endocrine disruptors are one of many poorly understood and underregulated classes of chemicals or materials whose recent increase in prevalence threaten
the sanctity of life on earth (Bergman et al., 2013). While we have dealt with these threats
in the past, as demonstrated in the case of lead, developing threats such as nanometals,
microplastics, and pharmaceuticals lack coherent regulation (Trujillo, 2016; Vasquez,
Lambrianides, Schneider, Kümmerer, & Fatta-Kassinos, 2014).
While the most egregious endocrine-disrupting chemicals, including heavy metals,
DDT, and polychlorinated biphenyls (PCBs) are now largely regulated, they persist in the
environment alongside hundreds of other chemicals that threaten to increase morbidity on
a global scale (Khetan, 2014; Ma et al., 2015; Rasmussen et al., 2015; Tanabe, 2002).
Many of these persistent organic pollutants (POPs), once released into the environment,
can continue to accumulate in animals and sediments for years or even decades after
production has ceased. Because the threat posed by many endocrine disruptors is
persistent, accumulating in body fat and continuing to exert endocrine-disrupting effects,
1
the question of regulation impacts not just the present, but the future health of people and
the environment.
Washington is one of the states leading the charge on endocrine disruptor legislation
within the United States. Several high-profile Washington State laws have, in recent
years, found themselves translated to federal regulations (Food and Drug Administration,
2012; Children’s Safe Products Act, 2008; Consumer Product Safety Improvement Act,
2008; Safe Baby Bottle Act, 2010). If we could understand how the policy in Washington
has developed, and what issues are holding back its further development, that knowledge
could be used to streamline the production of protective policies within Washington and
within the United States as a whole. Furthermore, that information could be used to
provide insight and guidance for the regulation of other emerging threats.
1.1 Lead as an example of the case for regulation
The regulation of lead in modern times represents a basic case study that demonstrates
the value of and potential for future regulation (Khetan, 2014). While endocrine
disruptors are a relatively unknown class of chemicals, knowledge of their impacts seem
to be following a similar path as the knowledge of the dangers of lead. Similarly, lead’s
use in industry and the resulting political issues that occurred in the pursuit of regulation
closely reflect the debate surrounding endocrine disruptors today. The current state of
lead regulation presents a well-regulated end-state for endocrine-disrupting chemicals
(Bridbord & Hanson, 2009; Muennig, 2009).
Lead has been recognized as a toxin for millennia, and has been indisputably proven
as such in modern times. Reference to the dangers of lead date back as far as ancient
2
Rome. As research on lead has become more and more precise, with larger studies and
more accurate measurements, the impact of lead exposure on childhood development has
consistently been shown to be more devastating than previously thought (Grandjean,
2010). Recent tests have shown that there is no minimum threshold for exposure to lead,
and that any exposure leads to decrease in intelligence, and associated behavioral issues
(Pichery et al., 2011; Vorvolakos, Arseniou, & Samakouri, 2016).
While the dangers of lead were being explored by physicians, the benefits of lead were
being explored by engineers. Since Roman times, lead has consistently proved acutely
useful, due to its ubiquity and unique physical properties. Lead’s easy malleability and
relative robustness led to millennia of its use as pipe material in water infrastructure
(Delile, Blichert-Toft, Goiran, Keay, & Albarède, 2014). Since the advent of circuitry,
lead has been a crucial component in electronics manufacturing (Almeida, Madureira,
Bonilla, & Giannetti, 2013). Lead has also found uses in house paint, as a gasoline
additive, and in wheel and fishing weights (Kristensen, 2015; Levallois et al., 2014)
Studies tying the hazard of lead to economic and social costs have led to regulation,
and kick-started the process of identifying reasonable substitutes; however it has not
always been a simple process (Bridbord & Hanson, 2009). The staggering lifetime costs
of contemporary lead exposure have recently been documented, and make a strong
economic argument for immediate and thorough exposure mitigation (Muennig, 2009;
Pichery et al., 2011). Minimizing lead exposure and identifying non-toxic substitutes for
lead have become crucial elements of state, federal, and international policy (Davies et
al., 2009; State of Washington Office of the Governor, 2016).
3
Many of the uses of lead were without substitute upon introduction and, despite our
best efforts, some remain that way. While lead pipes have largely been replaced by
copper and plastic pipes, these materials have their own drawbacks, including the
potential for endocrine disruption (Skjevrak, 2003). While lead in solder has been
restricted in the European Union since 2006, many of the beneficial properties of lead in
solder have not been emulated, and exceptions to the ban still allow for the use of lead in
applications where no suitable substitutes exist (Menon, George, Osterman, & Pecht,
2015).
The case of lead resembles closely that of endocrine disruptors in general. Many
endocrine disruptors were assumed to be entirely benign when first put into production at
the turn of the 20th century, and demonstrate properties that had and still have no direct
substitute. As their impacts were being explored by biologists and physicians, new
permutations with new properties were being produced and put into widespread use
(Khetan, 2014). As their effects have become more apparent, certain endocrine disruptors
produced in the high volumes or causing self-evident impacts have been regulated at the
federal or state level. These chemicals continue to enter the environment even after the
end of their production due to dissemination from legacy sources (Wattigney, IrvinBarnwell, Pavuk, & Ragin-Wilson, 2015). Many endocrine disruptors with and without
robust bodies of research surrounding their effects have been replaced due to consumer
pressure (Baluka & Rumbeiha, 2016). However, many more remain in use due to a lack
of a known substitute and a general lack of consideration.
4
1.2 Issues related to recognizing and regulating endocrine disruptors
Endocrine disruptors are present in many of the products we use on a daily basis,
including those we eat and drink from and those we trust to clean our hands (Khetan,
2014). Perhaps the most widely recognized endocrine disruptor of the twenty-first
century is bisphenol A, or BPA, a manufacturing additive that makes plastics less brittle,
continually leaches from resultant products during normal use, and mimics the hormone
estrogen when ingested (Michałowicz, 2014). BPA remains a common ingredient in
canned food liners, and is frequently found in the foods contained therein, even after
rinsing (Lorber et al., 2015). Triclosan has regularly been added to soaps as an
antimicrobial agent. While it has been found to be inefficient in that role and has been
found to cause endocrine disruption, it continues to be used in that context (Giuliano &
Rybak, 2015; Wang & Tian, 2015).
While some endocrine disruptors are used in critical applications where practical
substitutes do not exist, many are employed to little practical benefit, despite the presence
of viable, non-toxic alternatives, in the name of marginal cost-savings. BPA is present in
almost all thermal receipt paper, in the form of a dust which easily sloughs from the paper
and sticks to fingers (Björnsdotter, de Boer, & Ballesteros-Gómez, 2017). Recent public
and regulatory pressures have led to the removal of BPA from this application. While
known, safer alternatives, such as a vitamin-C-based mixture, exist, the BPA in these
applications is frequently substituted instead with chemical analogues that also
demonstrate endocrine-disrupting properties due to reasons of cost (Björnsdotter, Jonker,
Legradi, Kool, & Ballesteros-Gómez, 2017).
5
Endocrine disruption’s mode of action is well-understood and endocrine disruptors’
effects are predictable. For the past three decades, the science of endocrine disruption has
become more and more sophisticated, as possible pathways for disruption have been
enumerated and explored (Yang, Kim, Weon, & Seo, 2015). Tests in yeasts and higher
animals have provided consistent methods for detection. As more chemicals have been
evaluated for their endocrine-disrupting properties, computer models have become adept
at predicting endocrine-disrupting potential based upon chemical structure (Wilson et al.,
2016).
However, because endocrine disruptors do not act in the same manner as traditional
toxins, and because their effects are so dispersed, their dangers are not easily quantifiable
and are thus difficult to definitively communicate. Because the indirect, time-delayed
mode of action specific to developing organisms was so at odds with the focus of
traditional toxicology, endocrine disruption was not commonly defined or recognized
until the 1990s (Colborn, vom Saal, & Soto, 1993). While the effects of certain endocrine
disruptors have been quantified, because we are exposed to such a variety of endocrine
disruptors simultaneously, and because of the great time delay involved in many of the
most detrimental outcomes, it is difficult to concretely associate real-world impacts with
a single chemical (Woodruff, Carlson, Schwartz, & Giudice, 2008). Furthermore, certain
endocrine disruptors have recently been shown to produce not only delayed effects, but
effects on the offspring of exposed individuals (Manikkam et al., 2013; Skinner, 2014).
While these impacts are observable in controlled laboratory settings and have been
observed in the high-profile case of diethylstilbestrol* (Nilsson & Skinner, 2015), it is
*Diethylstilbestrol, or DES, is a chemical closely related to BPA that was prescribed to pregnant women in
the 1960s as an anti-nausea medication. Exposure to DES in utero was found to cause birth defects.
6
unlikely that these same effects can ever be conclusively demonstrated in humans for
most endocrine disruptors.
In addition to issues faced in demonstrating the effects of endocrine disruptors, one
primary obstacle to comprehensively demonstrating the adverse effects of endocrine
disruptors is the great range of impacts they may have, and the variance therein. The
impacts of a single chemical can vary greatly, due to the non-linear relationship between
dose and effect (Vandenberg & Bowler, 2014), the specific window of exposure
(Burggren & Mueller, 2015; Fudvoye et al., 2014), and the broad range of interaction
effects produced by the presence of other endocrine disruptors (Kortenkamp, 2014).
These wide-ranging issues preclude accurate assessment of the risks borne by exposure to
the vast majority of endocrine disruptors; making regulation that much more difficult in
jurisdictions, such as Washington State, where risk characterization is an expected
element of regulatory debate.
As a result, endocrine disruptor policy in Washington State is an inefficient patchwork
that fails to limit human and environmental exposure to many known and suspected
endocrine disruptors. While BPA has been banned from baby dishware and sports bottles
in Washington State, closely related chemicals that serve the same purpose in
manufacturing and have the same effect in the body have largely taken its place (Liao &
Kannan, 2013; Rochester & Bolden, 2015). While the recent Children’s Safe Products
Act implicitly considers endocrine disruption as an adverse outcome, it only requires the
reporting of chemical concentration, and does not in itself regulate endocrine disruptors
on the basis of that criteria.
7
1.3 Qualitative policy analysis of endocrine disruptor regulation in Washington
State
This thesis attempts to answer the question: What are the current limitations to
Washington State endocrine disruptor policy, and how could state endocrine disruptor
policy be improved through consideration of advances in the science and of the evolving
policy landscape at home and abroad?
Formulating a comprehensive approach to this question required gathering insight
from the scientific and political communities surrounding the issue and reflecting the
broader interests of Washington State and its residents. In order to provide a framework
for local discussion of the issue and to predicate more fine-grained analysis in the future,
I adopted a qualitative policy analysis lens. Using this framework, I performed a casestudy of the history and trajectory of endocrine-disruptor policy in Washington State,
involving a comprehensive policy document review, as well as an international policy
and state-of-the-science review. Significant open questions and potential future directions
highlighted by these reviews were used to produce a series of questions that served as the
basis for a series of open-ended stakeholder interviews that attempted to evaluate all
salient perspectives on the issue. Finally, the interviews were coded based upon the same
identified concerns and others that were suggested by the interview process, and results
were compared and contrasted to highlight both politically feasible paths forward and
more fundamental issues of contention amongst stakeholders.
The results of this research revealed that further regulation of endocrine disruptors is
limited by the high degree of scientific uncertainty, the lack of available data, the
potential of incurring regrettable substitutions, and a general lack of funding and political
8
will. While recent scientific advances have made inroads by reducing uncertainty and
streamlining the data collection and evaluation process, the nature of endocrine disruption
as a mechanism prohibits risk assessment, and thus inhibits the traditional legislative
assessment process. Nonetheless, existing endocrine disruptor regulations pave the way
for both subtle and fundamental improvements in regulation, and efforts by state
agencies, industries and advocacy groups promise to improve chemical management
through non-regulatory means.
To communicate the nature and significance of these findings, I begin by providing
background information related to the characteristic effects that define endocrine
disruptors, their history of production and use, and extant policy approaches employed in
Washington, the United States, and internationally. I then critically review the state of the
science to highlight disagreements and uncertainties that may impact regulation, and
models of science-policy interaction to aid in identifying feasible future policies. I then
describe in detail my methodology and the results of my research before discussing the
implications thereof, and providing recommendations for Washington State policy and
future research.
9
2. Background
Endocrine disruptors represent an emerging hazard that will become more regulated as
time goes on, and for that reason, it is worthwhile to search for presently feasible fixes to
bypass that regulation process. While the precise definition of endocrine disruptor may
remain somewhat unclear, many chemicals in widespread use today fulfill even the most
conservative requirements. Regardless of precisely what definition is used, a wide variety
of chemicals alter the normal, beneficial function of the endocrine system and lead to
negative health outcomes for humans and animals at all stages of development (Bergman
et al., 2013; Groshart & Okkerman, 2000; TEDX, 2017). For these reasons, a systematic
review of state policies relating to this class of chemicals is warranted. I begin by
describing the basic tenets of endocrine disruptors. I then review established approaches
to toxics policy, and evaluate their efficacy based upon established criteria. Finally, I
review external influences on Washington State policy, along with past and current
policies to establish to state of endocrine disruption regulation and to establish a
framework for future policy recommendations.
2.1 Definition
There exist today several different definitions of endocrine disruptor that differ in
subtle but significant ways. Perhaps the broadest definition was proposed by the U.S.
EPA (1997), for which an endocrine disruptor is an “exogenous agent that [interferes]
with the synthesis, secretion, transport, binding, action, or elimination of natural
hormones in the body that are responsible for the maintenance of homeostasis,
reproduction, development, and/or behavior.” The operative word “interferes” here is
open to interpretation. The WHO presented a slightly more concrete definition of
10
endocrine disruptor: “an exogenous substance or mixture that alters function(s) of the
endocrine system and consequently causes adverse health effects in an intact organism, or
its progeny, or (sub)populations” (Damstra et al., 2002). “Adverse health effects”
indicates that the outcome must be negative, rather than simply different, as could be
interpreted in the EPA definition. Further, this definition applies to humans and animals
alike, and puts greater emphasis on transgenerational effects. The WHO also
distinguishes between known and potential endocrine disruptors, clarifying that the latter
“might be expected to lead to endocrine disruption” (Damstra et al., 2002). Most recently,
the European Food Safety Authority distinguished endocrine disruptors as a sub-class of
the larger body of "endocrine active substances," with the latter encompassing "any
chemical that can interact directly or indirectly with the endocrine system, and
subsequently result in an effect on the endocrine system, target organs and tissues," with
the further classification as an endocrine disruptor depending upon "the type of effect, the
dose and the background physiological situation" (Barlow et al., 2010). While there exist
other definitions of endocrine disruptors, the main points of contention seem centered
around the qualitative nature of their impact. This definitional debate is further explored
in the literature review.
It is also important to bear in mind that the delineation of “endocrine disruptor”
characterizes chemicals by their mode of action, rather than their precise endpoint*. What
is clear in the above definitions, and as will become clear with further examples, is that
these chemicals are classed together through their mechanism. It is this characteristic that
makes them more difficult to coherently regulate; changes in developmental stage,
*For example, if one was diagnosed with breast cancer following chronic exposure to BPA, cancer would
be considered the endpoint, and endocrine disruption the mode of action.
11
exposure level, and simultaneous exposures will radically alter the impact of the same
basic mechanical alteration (Burggren & Mueller, 2015; Fudvoye et al., 2014;
Kortenkamp, 2014; Vandenberg & Bowler, 2014). To understand this, it is important to
understand the role of the endocrine system with the body.
2.2 Human effects of endocrine disruptors characterized by life stage
The endocrine system acts as a signaling pathway wherein the movement and varying
concentrations of hormones alter the development and day-to-day operation of other
organ systems. During periods of growth, such as the perinatal period & puberty, the
endocrine system guides the development and refinement of the various organs of the
body (Khetan, 2014). During periods of relative stasis, the endocrine system regulates the
normal functioning of those same organ systems. Disruption of the endocrine system
during different periods of the life cycle manifest in different ways.
Many endocrine-disrupting chemicals have only had their impacts directly
demonstrated in animal studies. While broad predictions of population-level effects can
be made based upon animal studies, and while population-level statistics can be recorded
that support those studies, it remains very difficult to conclusively link a certain chemical
with population-level outcomes, for the reasons cited above and due to legal and moral
objections against direct human testing (Bergman et al., 2013).For example, it is difficult
to conclusively demonstrate an association between fetal endocrine disruptor exposure
and adult male semen quality because there are so few cohorts “with stored blood
samples from mothers during pregnancy and with offspring of sufficient age to perform
follow-up studies. Therefore… one of the core elements of the endocrine disruptor
hypothesis has remained untested for almost 20 years” (Vested et al., 2014). The same
12
applies to any effects that are only demonstrated in offspring or grandchildren of the
exposed.
The endocrine system is most sensitive to disruption during fetal development and
during the first year of life (Palanza et al., 2016). While organs are developing from
previously undifferentiated cells, the exact nature of their development is heavily
dependent on hormones. Alterations to the signals that cells and organs receive at this
time can take several different forms. In the case of the sexual organs, perinatal endocrine
disruption has been associated with physical malformation; incompletely formed
uteruses, and undescended testes and hypospadias, or more generally leading to reduced
sperm count and viability, menstrual irregularity, or simply reduced fertility (Costa et al.
2014; Knez, 2013; Vested et al., 2014). Effects on the thyroid lead to hyperthyroidism
and obesity (Gutleb et al., 2016; Heindel et al., 2015). Neurodevelopment of the brain is
altered, altering brain function and leading to decreased intelligence and behavioral
disorders such as autism (Schug et al., 2015). In addition to these immediately apparent
changes, subtler structural changes to these systems can also occur, that manifest as
cancers and chronic diseases much later in life (Gibson & Saunders, 2014; Hu et al.,
2016; Knower et al., 2014; Rezg et al., 2014; Soto et al., 2013).
Much of the same is true during puberty, although the subtler physical changes lead to
similar, subtler maldevelopments (Fudvoye et al., 2014). This difference persists, and is
described in the concept of “body burden” (Huang et al., 2014). In effect, younger people
are more sensitive to endocrine disruptors because their bodies require proportionally
more food and water, etc. in order to maintain growth. Further, children metabolize faster
than adults and are thus more frequently exposed to acute doses of endocrine disruptors
13
present in the environment. While near puberty the visibly altered outcomes may be more
of degree than of kind (e.g., earlier onset of puberty), exposures at this period often take
the form of delayed effects as described above.
During periods of relative stasis, namely post-puberty, endocrine disruptors primarily
act as agonists (Fudvoye et al., 2014; Rezg et al., 2014). During these periods or life,
hormones typically regulate the body, and ensure the stability and maintenance of
existing systems. While during periods of development, issues are primarily associated
with acute doses of endocrine disruptors that alter development, chronic exposure during
adulthood stresses the target organ, potentially limiting its ability to operate beneficially
and contributing to premature failure.
It has recently been demonstrated that exposure to certain endocrine disruptors can
alter gene expression in subsequent generations, leading to effects of exposure that may
only manifest in the grandchildren of those exposed (Manikkam et al., 2013; Rissman &
Adli, 2014; Skinner, 2014). These transgenerational epigenetic impacts may mean that
the effects of exposure to certain endocrine disruptors may not manifest until the
grandchildren of the exposed individuals are born, perhaps not until decades later.
2.3 Salient characteristics of known endocrine disruptors
Perhaps the most infamous endocrine disruptor, beyond heavy metals such as arsenic
and lead, is the pesticide DDT, the subject of Rachel Carson’s Silent Spring (1962) and a
catalyst for the environmental movement of the 1960s and 1970s. The additive Bisphenol
A, or BPA, is nearly as well known today for its ubiquity in consumer products and its
14
insidious association with breast cancer, among other endpoints (LaKind & Naiman,
2015).
Endocrine disruptors represent a threat to the health of the environment and human
beings alike, and in a large variety of ways, many of which are only now becoming wellunderstood. Many steps go into producing an accurate estimate of the risks of any given
chemical, some of which cannot be taken with our current knowledge (Bergman, 2013).
In contradistinction to traditional chemicals considered in toxicology, the manner in
which the endocrine system operates leads to a nonlinear relationship between dose and
response (Vandenberg & Bowler, 2014). Because much of the exposure to endocrine
disruptors comes from non-point sources (Tijani et al., 2013), their movement through the
environment must be modeled and tested. Even if all of these characteristics were
evaluated in isolation, the interaction effects of multiple endocrine disruptors on the
endocrine system are effectively unpredictable, rendering the real-world evaluation of
risk unattainable (Kortenkamp, 2014).
In order to fully evaluate the risk of an endocrine disruptor in isolation, a series of
functional questions must be answered. Endocrine disruptors have historically been
classed and researched based upon the specific hormone which they disrupt, with much
of the research focusing on estrogen and testosterone (Bergman, 2013). Once the specific
hormone that is impacted is identified, it must next be determined the mode of action of
disruption, be it altering the “synthesis, secretion, transport, binding, action, or
elimination” of the hormone (U.S. EPA, 1997). Once the affected hormone and mode of
action are identified, the dose-response relationship still needs to be determined.
15
Endocrine disruptors, depending on their mode of action, frequently subvert the idea
that “the dose makes the poison.” Hormones do not exert linear effects; in practice, the
lack of a hormone might inhibit a certain response, a small dose might prompt that same
response, and a larger dose might overwhelm the system, and have the same effect as no
dose at all (Vandenberg & Bowler, 2014). This inverted-U relationship, or non-monotonic
dose-response relationship, undermines the established scientific understanding, and was
largely responsible for the delayed acknowledgement of the issue posed by endocrinedisrupting chemicals.
The definition of endocrine disruptor encompasses a wide variety of chemicals, many
of which act in very different ways. A large subset of endocrine disruptors is specifically
designed to move rapidly through our environment and accumulate in body tissues (Faure
& Lefevere, 1998). Certain pesticides, for example, are designed to move through the
water column, integrate into plant and pest tissues, and persist throughout at least a whole
growing season. The same attributes that make them effective in disrupting pests leads
them to accumulate in the environment and in humans and other animals. Persistence in
the environment leads to chronic exposure for organisms in that environment (Faure &
Lefevere, 1998). Persistence in body tissue, or bioaccumulation, may manifest as chronic
exposure or as acute exposure (Wang et al., 2015). Certain other endocrine disruptors
present a threat not due to their persistence, but due to their ubiquity. For example, BPA
degrades quickly in the environment and is quickly metabolized, but because of its
presence in so many products that humans handle and directly ingest every day, the
majority of people are chronically exposed to it (LaKind & Naiman, 2015).
16
Figure 1: Monotonic and non-monotonic dose-response curves: These graphs represent
an organism’s response to the presence of a chemical. The x-axis represents the dose of
the chemical, and the y-axis represents the degree of organism response. Traditional
toxicology assumes a continuous increase until saturation, as illustrated by the black
curve on the left. Non-monotonic curves can take many forms, for instance an inverted U
shape, indicated by gray curves on the right, which may be caused, for example, by two
contradictory monotonic curves. (Adapted from UNEP/WHO, 2013)
Even if the direct mode of action and source and magnitude of exposure are known for
any given chemical, to precisely predict its impact on individuals requires analyzing the
combined effect of multiple simultaneous endocrine disruptors (Kortenkamp, 2014;
Trasande et al., 2016). While one chemical may act to mimic estrogen, another may act to
inhibit estrogen receptors. Their effects may magnify each other or cancel each other out,
and that relationship may itself be non-monotonic, producing different synergistic effects
as the concentration of one or both of the endocrine disruptors changes. To account for
the combined effect of the total range of endocrine disruptors in the environment would
require the production of an increasingly complex model. Considering the uncertainties
present in the preceding steps, this holistic modeling is currently unattainable.
2.4 Historical context of endocrine disruptor proliferation and knowledge
While the endocrine system has always been altered by the environment, the nature
and magnitude of exposure changed upon the advent of the chemical revolution. Many
natural chemicals impact and alter the expression of the endocrine system to some
degree, such as phytoestrogens, plant-based chemicals that mimic estrogen (Sirotkin &
17
Harrath, 2014). However, many manufactured chemicals are more resistant to
degradation, are more potent, and are more liable to bioaccumulate than naturallyoccurring endocrine active substances. The proliferation of hydrocarbons as a resource
for chemical manufacturing led to a chemical revolution that is ongoing, and has
continued to accelerate. Every year, thousands of novel chemicals enter into commerce,
without requirement that they be tested for their unintended effects (U.S. EPA, 2017).
Not only do there exist uncertainties relating to the effects of these chemicals, but also
uncertainties as to precisely which chemicals are in use and at what concentrations. In the
United States, there exists no requirement that the precise use of chemicals be reported,
much less the magnitude; thus, it is difficult to rank chemicals in terms of their ubiquity,
and to make informed decisions as to which are most imminently in need of evaluation.
The increase in the production and distribution of endocrine disruptors, and the recent
recognition of their far-reaching effects, has prompted an increase in the variety and
sophistication of tests for endocrine disruption. Before the mode of action of endocrine
disruption was well-understood, most chemicals were tested for safety via administration
in large doses to adult organisms (Shukla et al., 2010). As our understanding of the
endocrine system developed, it was discovered that the effects of endocrine disruptors
often did not reflect the typical dose-response curve, and thus the limited data points
evaluated in traditional toxicological tests frequently failed to correctly predict effects.
Further, in testing only adult organisms, reproductive and developmental effects were not
tested for and thus not acknowledged. Essentially, toxicology considered cancer as the
result of acute chemical exposure, failing to encompass the most serious impacts of
endocrine disruptors (Buonsante, 2014).
18
Awareness of these other modes of action were publicly acknowledged in the latter
half of the 20th century, and inspired more targeted testing methods. Perhaps the most
critical demonstration of endocrine-disrupting effects was the rash of birth defects caused
by diethylstilbestrol, a drug with endocrine-disrupting properties prescribed to cure
morning sickness in pregnant women (Troisi et al., 2016). Following these revelations,
the sophistication and variety of testing methods increased, leading to the adoption of
studies on pregnant organisms, and multigenerational studies. Recent moral objections to
live animal testing have led to the production of increasingly sophisticated testing
methods involving single-celled organisms or computer models, which allow for precise
demonstration or prediction of the effect that a chemical will have on the endocrine
system, from which the broader impacts can be synthesized (Doke & Dhawale, 2015).
Increasingly sophisticated models of environmental chemical movement have been
developed, paralleling the development of these internal tests and predictive models
(Khetan, 2014). These tests allow for increasingly sophisticated predictions of the impact
of the environmental release of endocrine disruptors, in addition to exploring how
endocrine disruptors enter and move within living beings.
While these tests and models focus on chemicals in their known form, equally
important is an understanding of the degradation pathways of these chemicals.
Frequently, the chemical constituents that endocrine disruptors degrade into in the natural
environment have endocrine-disrupting properties of their own (Makarova et al., 2016).
Thus, effective evaluation of the impacts of the release of an endocrine disruptor into the
environment are incomplete without an understanding of their iterative reduction to
benign components.
19
While degradation models will allow us to better understand the movement and effects
of currently produced chemicals, it also allows us to predict the impact of future
chemicals, which is something that has been incorporated into the recent field of green
chemistry (Schug et al., 2013). “Green chemistry” refers to a process of chemicals
development, wherein the chemicals, before mass manufacture, are tested for both their
beneficial characteristics and their potential to cause unwanted outcomes including
endocrine disruption. This typically involves predictive modeling, wherein a chemical is
tested to see if it shares characteristics with known endocrine-disrupting chemicals.
Green chemistry may also involve early and formalized tests for endocrine disruption,
often in the form of simple yeast assays.
Green chemistry is often employed in a process of “alternatives assessment,” wherein
a product manufacturer who recognizes the potential hazard of a certain chemical
searches for a less hazardous alternative. In this approach, relative hazard is considered
alongside differences in cost and effectiveness.
2.5 Justification for review of Washington State endocrine disruptor policy
Despite and because of the current limitations to the study of endocrine disruptors, a
systematic review of endocrine disruptor policy in Washington state is merited. Assuming
business as usual, even as tests for endocrine disruption become more sophisticated, the
lag between production and policy adoption means that it will be a decade or more before
extant approaches are adopted into law. Furthermore, endocrine-disrupting chemicals, as
a classification, reflect other chemical and material classes that pose similar health and
environmental hazards. The history and potential regulatory responses to the issue of
20
endocrine disruption may be useful in formulating policy responses to these analogous
issues.
The primary issue worth exploring in the context of endocrine disruptor policy is the
temporal disconnect between the time in which a hazard is first established and the time
in which a corresponding risk is concretely established. Because many of the advances in
testing procedures are still in development, it is unlikely that the risk posed by any
endocrine disruptor will be evaluable for many years to come. Further, the sheer
proliferation of endocrine-disrupting chemicals and their synergistic effects make it
unlikely that their true risk can ever be precisely established. Nonetheless, recent
legislation has demonstrated that there is popular support for improved and proactive
regulation, and has begun to evaluate endocrine disruptors from a hazard-based approach.
To understand this phenomenon, it is worth fully exploring the context of current
policy, to form as coherent a picture as possible. To understand the limitations and
opportunities of state-level policy, both federal policy and local policies should be
reviewed. To see how future policies should be broached and couched, this thesis reviews
historical Washington State regulations and proposed regulations, attempting to evaluate
the conditions that led to their success or failure. To bolster this work, this study collects
a variety of stakeholder perspectives, to qualify the results and to address questions of
political feasibility.
This work is designed to find resonance with and application in other geographies and
other disciplines. While every state is different, other states can nonetheless learn from
the example of Washington State, and aide in their own paths forward. Several other
states have already addressed the issue of endocrine disruptors, or have adopted policies
21
analogous to those discussed in this work. Interstate collaboration surrounding this issue
exists amongst a handful of states, but approaches to regulation are infrequently unified.
Different classes of chemicals or materials have and will proceed through similar
science-policy arcs, and a clear examination of the typical characteristic of these arcs
could be rewarded by streamlining the process in future conflicts. Prior to endocrine
disruptors, heavy metals and carcinogens have followed a similar arc and have begun to
reach a regulatory equilibrium that serves as a model for the end-goal for endocrine
disruptor regulation. There are similar narratives developing in regards to microplastics,
nanometals, and pharmaceuticals, material and chemical classes that face unique
scientific challenges to understanding and political challenges to regulation, but which
broadly speaking could be compared to endocrine disruptors. By explaining the historical
struggle for endocrine disruptor regulation and the potential thereof going forward, it is
my hope that the struggles surrounding these other classes can be streamlined, and their
regulation can more quickly reflect our understanding of their dangers.
2.6 Established approaches to toxics policy
Within the sphere of public policy, there are many alternate approaches to chemical
regulation that may produce comparable regulatory ends. As with all policy issues, each
approach faces its own set of barriers to implementation, and each approach favors the
resulting certainty of a certain variable or variables while allowing the others to alter in
order to compensate. For example, while a tax may allow legislators to carefully control
the cost of certain chemicals, it provides no direct control over the volume of chemicals
purchased. Thus, the choice of regulatory approach is carefully tied to questions of
22
political practicality and political vision. Table one, below, illustrates the seven most
common regulatory approaches to chemical management, as identified in the literature.
Table 1: Policy mechanisms applicable to endocrine-disrupting chemicals
Policy
Mechanism
Description
Ban
Cessation of production, distribution, sale, Remove chemical from
etc.
commerce/environment
Environmental
Point-source regulation of soil, water,
Quality Standard airborne chem. concentration
Explicit Goal
Limit pollution rate
(typically in accordance
with external goals)
License
Require chemical producers/importers to Demarcate chemical
seek explicit permission to produce/import producers/importers
chemicals
Permit
Distribute rights to limited amount of
chemical production/import over a
specified time period
Delimit chemical
proliferation (and fund
remediation)
Tax
Add surcharge to chemical
production/import/distribution process
Increase cost of chemical
use (and fund
remediation)
Reporting
Requirement
Require retailers to publicly test for and
indicate chemical presence/concentration
Increase consumer
awareness/empower
decision-making
Labeling
Requirement
Require indication of chemical
presence/concentration on retail products
Increase consumer
awareness/empower
decision-making
While there is debate over which criteria impact the success or failure of policy
(discussed further in the literature review), Eisner (2007) provides a comprehensive list
of criteria that may impact the feasibility of environmental legislation, several of which
23
can be readily assessed for existing policy alternatives: “certainty of results;” “cost,”
either to the public or the government; “corrigibility,” or ease of alteration in the face of
changing knowledge or norms; “timeliness” of expected outcomes; and “compatibility
with normative values.” In addition to these, Eisner delineates three further criteria which
are highly context dependent: “administrative feasibility;” “robustness” in the face of
varying circumstances; “dynamic efficiency,” or the effect of the regulation on innovation
within the affected field; and “public acceptance.” While administrative feasibility can be
assessed for certain policy mechanisms based upon existing regulations, the latter three
are either too context sensitive to be generalized. The table below illustrates how
chemical management policy mechanisms compare in terms of Eisner’s metrics.
Table 2: Summary of Eisner’s (2007) criteria relevant to policy choice for applicable
mechanisms
Policy
Mechanism
Certainty Cost
of results (direct
public)
Administr Corrigibil Timelines Compatib
ative
ity
s
ility w/
feasibility
normative
values
Ban
High
High
Low
Low
High
Low
Quality
Standard
High
High
Moderate
Moderate
High
High
Moderate
Moderate
High
Moderate
License
Moderate
Permit
High
Variable
Moderate
Variable
Low
Moderate
Moderate
Low
Reporting
Requirement
Low
Moderate
High
High
Low
High
Labeling
Requirement
Low
Moderate
Moderate
High
Low
High
Tax
Moderate [unknown] Moderate
[unknown] Moderate
24
2.6.1 Regulatory mechanisms
The most immediate approaches to chemical regulation involve the control of
production, sale, or use of a certain chemical or chemicals (Eisner, 2007; Faure &
Lefevere, 1998; Richards, 1999). If a chemical disrupts the beneficial functioning of the
endocrine system, the logic goes, ensure that it does not reach our endocrine systems.
While this approach may be justified in the face of a particularly deleterious chemical,
the example of BPA* demonstrates the issue that substitutability presents to this approach.
If we ban a single chemical, or even a class of chemicals, users will frequently substitute
the most similar available chemical or class of chemicals, in order to minimize cost.
While these chemicals will be likely to have the same beneficial properties, a ban alone
provides no incentive for reducing the negative impacts of as-yet unregulated chemicals.
From a strictly economic perspective, a ban is also inefficient in the sense that the
marginal benefits of chemical use may exceed the marginal costs up to a certain threshold
(Faure & Lefevere, 1998). In other words, strictly banning certain chemicals may incur a
social cost greater than that of allowing their limited use.
One policy instrument that is more useful as a complement to existing legislation than
as a standalone source of authority is a quality standard, which dictates a maximum
acceptable concentration of a chemical in a given medium, such as soil, air, or wastewater
treatment plant effluent (Faure & Lefevere, 1998). Washington State has water quality
standards, regulating the concentration in public waterways of a variety of chemicals
* BPA was banned from use in sports bottles in Washington State. In its stead, many sports bottles now
contain BPS, BPF, or another chemical closely related to BPA that serves the same purpose in bottle
construction and has similarly deleterious effects in the human body.
25
including endocrine disruptors. Frequently, as in this case, acceptable concentrations are
designed to limit the calculated human health risk to a certain level of adverse outcomes.
However, dictating chemical policy in this indirect way makes it difficult to improve
when the quality standards aren’t met, because environmental pollution is often caused by
a variety of dispersed sources that are difficult to regulate simultaneously, and may not be
applicable direct sources of endocrine disruptor exposure, such as sports bottles or receipt
paper.
For this reason, standards are often established in tandem with a licensing or
permitting system (Faure & Lefevere, 1998). In this system, producers of large
environmental discharges must receive a license to discharge certain chemicals, or a
permit to discharge certain chemicals in certain amounts. While this can lend itself to the
regulation of endocrine disruptors within environmental discharges, it could just as easily
apply to the concentration of endocrine disruptors in consumer products. Polluters found
to be out of compliance with the system are penalized.
If regulators wished to directly alter the demand, thus indirectly reducing the supply,
of endocrine-disrupting chemicals, they could tax them (Mason, 1998). Taxes could be
applied to the production, distribution, or consumption of endocrine-disrupting
chemicals. In raising the cost of using specific chemicals, taxation would make preferable
solutions more cost-competitive, and generally reduce the incentive to use endocrinedisrupting chemicals. However, this method leads to much more inexact outcomes than
direct methods of regulation, as it is difficult to determine the relationship between cost
and demand. In extreme cases, where no substitute exists, a tax may not alter the
production or dissemination of a chemical at all.
26
As is the case with the Children’s Safe Products Act (CSPA), the most relevant current
regulation in Washington State, endocrine disruptors could be subject to reporting
requirements. In contrast to the above-mentioned regulations, this approach requires
fewer resources to implement, and is easily validated with laboratory testing. Further,
because this approach does not restrict the actual concentration of any chemicals, the
requirements to add a chemical to a reporting list are frequently less stringent than for a
ban list. If there is reason to believe that a chemical is hazardous, but its risk cannot be
quantified, collecting concentration information from producers and distributors allows
for a more robust understanding of risk to be produced in the future. In practice, this
approach may act as a sort of “soft ban,” as it indicates to companies that these chemicals
are under consideration for subsequent command-and-control regulation.
Similarly, labeling requirements could see the presence and potential impacts of
endocrine disruptors reported on consumer packaging, akin to the Surgeon General’s
warnings on cigarette packaging. While this approach may be more amenable than
command-and-control, it requires a high degree of certainty to be implemented, at which
point command-and-control regulations may be preferable.
2.7 Broader policy considerations
While not reducible to a formulaic regulatory archetype, there are several more
fundamental conceptual changes that could be made to the regulation of endocrine
disruptors which would alter their production and distribution.
Perhaps the most important conceptual shift would be in the burden of proof, which
would begin with a minimum data set requirement (Khetan, 2014). Currently, at the
27
federal level and by extension at the state level, when a new chemical is manufactured
and enters commerce, there is no requirement that any safety testing be done on that
chemical; in fact, no data about the chemical need be supplied at all. Thus, if anyone has
reason to believe that a product is harmful and should be regulated (assuming that they
can even identify the active chemical in the first place without being stymied by
obfuscating trade secret laws), they must isolate or synthesize the chemical themselves
and perform their own tests to prove harm before they have any chance of enabling
regulation. In contrast to the American system, the European union requires that any
chemicals manufactured in high volumes be thoroughly tested at the expense of the
manufacturers. Adopting a system where the manufacturer must demonstrate the safety of
their product would perhaps lead the chemical industry to adopt principles of green
chemistry.
Barring the implementation of a minimum data set, and a shift in the burden of proof
of safety/risk, regulators can provide other incentives and opportunities to industry
members to test for hazards posed by chemicals prior to their large-scale production
(Eisner, 2007). Washington State has developed programs to promote self-regulation of
industry through the principles of green chemistry and alternatives assessment. These
programs teach the use of existing frameworks that test for adverse outcomes throughout
the process of chemical development and production, with an attempt to minimize the
cost of failure if hazardous endpoints are discovered. While these programs may increase
awareness, in the absence of external pressure in the form of regulation, they provide
little incentive for chemical manufacturers to actually alter their processes.
28
2.8 Influences of agencies and extra-judicial policies on state policies
2.8.1 State Agencies
Before discussing the precise policies, it is worth considering the agencies tasked with
enforcement, the differences in approach between agencies, and the differing authorities
that relevant agencies wield. To that end, the agencies most frequently empowered by
endocrine-disrupting legislation are the Department of Ecology (Ecology), the
Department of Health (Health), and the Department of Enterprise Services (Enterprise
Services).
The majority of proposed EDC legislation grants enforcement authority to Ecology, as
most of the proposed legislation uses approaches that target existing areas of its authority.
Nearly any command-and-control legislation falls under Ecology’s purview, given its role
in regulating air and water quality, and its existing role in regulating business and
industry. Outreach programs sponsored by Ecology more frequently court business and
industry participation than citizen participation. Judgments made within Ecology
frequently accept as evidence demonstrations of hazard, even if they haven’t been
calculated to specific risks, and Ecology is frequently able to act on such a basis
(Steward, 2016). The Department of Health, when empowered by legislation, frequently
plays a complementary role to that of Ecology. While Ecology has capacity for
environmental and materials testing, most biological studies are enacted by Health.
The Department of Enterprise Services, which administers state purchasing, has also
been the subject agency of several laws regarding preferential purchasing of endocrinedisruptor-free products. Given the limited relevant scope of Enterprise Services’
29
authority, these laws are more precisely targeted and perfunctory in nature than those
addressed to Health or Ecology.
Table 3: Division of departmental authority regulating endocrine-disrupting chemicals
Department:
Approach
Ecology
Hazard-based Air/water quality; industry and commerce
Health
Risk-based
Enterprise Services N/A
Areas of Responsibility
Human studies, health advisories
State purchasing
2.8.2 Federal and interstate policies
In order to understand the potential range of policies, and contextualize shifting
policies over time, it is important to understand the shifting federal policy landscape.
While much of the legislation relevant to endocrine disruptors and the media in which
they may be regulated have been established since the wave of environmental legislation
in the seventies or earlier, regulatory changes at the EPA, and a recent amendment to the
Toxic Substances Control Act (which has yet to be enforced) have directly impacted
Washington’s regulatory authority.
The longest-standing federal regulation relevant to endocrine disruptors is the Federal
Food, Drug, and Cosmetic Act (FFDCA), which was passed in 1938. The act regulates
what can be added to and sold as food, drugs, and cosmetics. The section related to food
is fairly robust, and grants the Food and Drug Administration authority to evaluate food
additives. The cosmetic portion of the act, however, does not provide very stringent
protections or limitations on cosmetic additives. While there have been recent pushes to
30
improve cosmetics regulation, including the Safe Cosmetics Act and Safe Personal Care
Products Act, they have not been successful thus far.
The Clean Water Act of 1972, and the Water Quality Act of 1987 (collectively CWA)
dictate and regulate acceptable levels of water pollution in the United States. Under these
acts, states are allowed to set their own water quality standards above and beyond those
mandated federally, under criteria deemed as or more stringent than those in the CWA.
The Toxics Substance Control Act (TSCA) of 1976 granted the EPA authority to
catalogue and evaluate the safety of chemicals in commerce. The act grandfathered in all
chemicals in commerce prior to its passing, granting them immunity from scrutiny via
assumption of safety. While it granted the EPA authority to test the safety of new
chemicals on the market, and ensured the reporting of new chemicals, it made no
provisions for chemical manufacturers to report the amount of the chemical being
produced, or whether it was ever commercially produced at all. Thus, while the list of
chemicals regulated under TSCA grows every year, the EPA has no basis on which to
determine which chemicals are actually present in commerce. While EPA has the
authority to regulate all these chemicals, doing so first requires evaluating them for
specific detrimental endpoints, a process requiring much more funding than the EPA has
historically had access to. Thus, only the most obviously detrimental chemicals end up
being regulated (Markell, 2010).
Recently, TSCA has been amended by the Chemical Safety Improvement Act (CSIA).
This act allows EPA to require chemical manufacturers to perform tests on the impacts of
their chemicals, but only if defensible evidence is produced to support that request. The
act increases the sophistication of tests that can be used to evaluate chemical safety. It
31
also presses the EPA to identify the chemicals presenting the greatest hazard and evaluate
them for acceptable risk level and regulate them appropriately. One of the most
contentious changes to the act involves state preemption; when the EPA determines an
acceptable risk level for a certain chemical going forward, states will now be preempted
from substituting a more restrictive level.
2.8.3 International policies
The Registration, Evaluation, and Authorization of Chemicals (REACh) laws in the
EU, established in 2007, and their subsequent analogues in non-member countries have
provided a contemporary model for chemical regulation. The REACh model connects
manufacturers of chemicals produced in high volumes into clearinghouses, requires those
clearinghouses to produce minimum data sets for those chemicals, and then regulates the
chemicals based upon the produced information. While the production of minimum data
sets for high-volume chemicals hints at a potential source of chemicals data upon which
to base policy in Washington, much of those data are suppressed beyond the EU in the
name of trade secrecy, or available only in digest form, insufficient to provide a basis for
regulation.
REACh requires chemical manufacturers to use the best available science to evaluate
the safety of all widely produced chemicals in commerce. Crucially, REACh relies
strongly on the precautionary principle, accepting demonstration of hazard as sufficient
justification for stringent chemical regulation. Additionally, REACh is critical of animal
testing, prompting the production of non-animal analogues.
32
Because many chemical manufacturers produce chemicals for multiple markets,
REACh is likely already impacting the chemicals manufactured or imported into
Washington State. As more countries adopt a REACh-like approach to chemical safety, so
will the global impact of the approach. Even if chemicals with demonstrable hazard
remain unregulated in Washington, many chemicals that would otherwise enter
Washington waters from overseas will be prevented from doing so.
2.9 Enacted Washington State policies
Perhaps the first policy in Washington State that was written with the concept of
endocrine disruptors in mind is the Chemical Action Plan, which was established by
executive order. This plan prompted Ecology to produce a list of the ten chemicals or
chemical classes of greatest concern, and systematically produce comprehensive
statewide plans to address regulation of those chemicals. The order begins by defining
persistent toxic chemicals through the example of mercury, dioxin and polychlorinated
biphenyl, the latter two of which are classically understood as endocrine disruptors. It
cites concern for these chemicals because they “are toxic in small amounts, remain in the
environment for long periods of time, and build up in humans, fish and animals.” The
first chemical class to be explored under this act were the flame retardant PBDEs.
Following evaluation of PBDEs by Ecology, the 2008 session law 65 was passed greatly
reducing their acceptable use within Washington. At that same time, however, Ecology
transitioned from focusing on Chemical Action Plans to enforcing another law passed that
year. The following table represents significant legislation related to endocrine disruptors.
33
Table 4: Enacted Washington State policies relating to endocrine-disrupting chemicals
2002 Exec. Order No. 02-03
Promotes Sustainable Practices by Agencies
2004 Exec. Order No. 04-01
Orders the regular production of Chemical Action Plans
2008 Wash. Sess. Laws 65
Reduces acceptable uses for polybrominated diethyl
ethers (PBDEs)
2008 Wash. Sess. Laws 288
Children’s Safe Products Act
2009 Wash. Sess. Laws 243
Promotes substitution of lead wheel weights
2010 Wash. Sess. Laws 140
Phases BPA out of children’s food containers and sports
bottles
2010 Wash. Sess. Laws 147
Phases copper out of brake pads
2014 Wash. Sess. Laws 135
Encourages state agencies to avoid purchasing PCBcontaining products
The Children’s Safe Product Act has become the central element of state policy related
to endocrine disruptors. It allows for Ecology to require chemical manufacturers and
importers to test for and report the presence and concentration of a list of chemicals that
Ecology can modify at its discretion. Most crucially, the definition of relevant chemicals
was open-ended both in the source of research accepted as evidence and in the endpoints
considered as cause for concern. Section 70.240.010 of the revised code of Washington
states the following:
‘High priority chemical’ means a chemical identified by a state agency, federal agency,
or accredited research university, or other scientific evidence deemed authoritative by
the department on the basis of credible scientific evidence as known to do one or more
of the following:
(a) Harm the normal development of a fetus or child or cause other developmental
toxicity;
(b) Cause cancer, genetic damage, or reproductive harm;
34
(c) Disrupt the endocrine system;
(d) Damage the nervous system, immune system, or organs or cause other systemic
toxicity;
(e) Be persistent, bioaccumulative, and toxic; or
(f) Be very persistent and very bioaccumulative.
While these considerations leave the door open for broad reporting requirements, the
law also guided Ecology to choose only fifty chemicals for the list, to reflect the initial
cost estimate for the law, which was budgeted based upon an assumption of fifty
chemicals. Since that time, however, Ecology has expanded and continues to expand the
list to reflect other legislative concerns, and to support other agency work.
Beyond creating a reporting requirement, CSPA limited the acceptable concentration
of pthalates and cadmium in children’s products, although the determined values were
later preempted by federal regulations allowing for slightly higher concentrations.
In general, Health works closely with Ecology in determining chemicals to regulate
under CSPA, and in relation to many other policies with implications to both human and
environmental health. A provision in CSPA requested the creation of an information
campaign, which has been pursued by the Department of Health. While not the direct
result of state policy, Health performed a multi-year, federally funded study testing for
the presence and concentration of certain chemicals in various populations around
Washington, including many endocrine disruptors. Unfortunately, funding was not
renewed and this study has not persisted, withering one potential avenue for evaluating
the efficacy of future policies.
The Department of Enterprise Services administers policies related to the chemicals
present in state-funded purchases, and is authorized by the remaining listed laws, as well
35
as the CSPA, to alter their purchasing habits. Unsurprisingly, changes to purchasing are
often accompanied by a requirement to consider relative costs as well as relative safety.
2.10 Proposed state policies
While seven bills relating to endocrine disruptors have passed into law since the
beginning of the 2007-8 biennium, in that same time period about 70 more bills related to
endocrine disruptors have been proposed and failed to pass. Many of these bills are
replicates of the same bill, proposed and refined between legislative sessions. While lead
and mercury have consistently been the subject of perennially proposed legislation, there
has been a slight increase in bills targeting more exotic endocrine disruptors—benzene,
bisphenol A, PCBs, flame retardants, and perfluorinated chemicals, which all impact
humans by means of the endocrine system. While most of the proposed bills are targeted
at negative human endpoints, many of the lead, mercury, and benzene bills are concerned
with the environmental release of the chemical. Recent proposed legislation relating to
flame retardants and electronics has sought to extend the reach of CSPA in a piecemeal
fashion. Highly specialized legislation which would be administered by Ecology, such as
the recently proposed ban of perfluorinated chemicals in food packaging, is being
increasingly frequently brought before the legislature.
36
Table 5: List of proposed state policies related to endocrine-disrupting chemicals
Bienni HB
um
No.
SB
No.
Title
Related
to*
2017 1596
Requiring manufacturers of electronics to report the
presence of high priority chemicals under the children's
safe products act.
CSPA
1744
Concerning the use of perfluorinated chemicals in food
packaging.
CSPA
1842
Taking action to address lead in drinking water at facilities
frequented by children.
Pb
1925
Taking action to address lead in drinking water in schools. Pb
1738
Continuing to protect water quality by aligning state brake
friction material restrictions with the requirements of a
similar nationwide agreement.
Cu
Concerning imposing a surtax on the possession of
5501 hazardous substances.
Addressing contaminated drinking water stemming from
the lead content in drinking water infrastructure, including
5745 pipes, connections, and fixtures.
Pb
15-16
1049 5021 Concerning cadmium in children's jewelry.
CSPA
6042 Concerning cadmium in children's jewelry.
CSPA
Concerning use of chemical action plans to require safer
1472 5406 chemicals in Washington.
Concerning the use of chemical action plans for
5056 recommendations of safer chemicals.
1984
Concerning the use of certain chemicals in food.
1174 5684 Concerning flame retardants.
CSPA
1845 5577 Concerning pharmaceutical waste.
6540 Ensuring safe playgrounds and turf fields.
Conducting remedial actions under the model toxics
5829 control act.
MTCA
6131 Requiring safer chemicals in Washington.
6570 Prioritizing the expenditure of funds associated with the
MTCA
* CSPA = Childrens Safe Products Act; Cu = Copper; DES = Department of Enterprise Services; Hg
= Mercury; MTCA = Model Toxics Control Act; Pb = Lead; PCBs = Polychlorinated Biphenyls;
37
model toxics control act for the cleanup of toxic pollution.
13-14
2779
Concerning the use of certain chemicals in food.
1294 5181 Concerning flame retardants.
CSPA
5933 Concerning flame retardants.
CSPA
Banning certain flame retardants in children's products and
5984 residential upholstered furniture.
CSPA
6048 Concerning flame retardants.
CSPA
Banning tris(1,3-dichloro-2-propyl)phosphate and tris(2chloroethyl)phosphate flame retardants in children's
6540 products and residential upholstered furniture.
CSPA
Directing the department of health to review the impact of
5348 chemicals on public health.
6086 Reducing PCBs in Products Purchased by Agencies
PCBs
Concerning polychlorinated biphenyl(PCB)contamination
6501 in Used Oil Recycling
PCBs
11-12
1319
Regarding the safety of certain children's products.
CSPA
2241
Reducing the introduction of lead into the aquatic
environment.
Pb
2266 6120 Concerning children's safe products.
CSPA
6630 Concerning children's safe products.
CSPA
2821 Concerning children's safe products.
CSPA
6369 Protecting environmental quality and human health.
09-10
1342
Creating a pilot program to screen children for lead
poisoning.
Pb
1345
Creating a pilot program to screen children for lead
poisoning.
Pb
1346
Concerning the labeling of lead-containing products.
Pb
1799
Reducing the release of mercury into the environment.
Hg
1809
Reducing the release of mercury into the environment.
Hg
Limiting the use of copper and other substances in vehicle
3018 6557 brake pads.
Cu
2818
Reducing the environmental health impact of cleaning in
state facilities.
DES
2914
Reducing the release of mercury into the environment.
Hg
6248 Concerning the use of bisphenol A.
1165
BPA
Providing for the safe collection and disposal of unwanted
38
drugs from residential sources through a producer provided
and funded product stewardship program.
1180
07-08
Regarding the use of bisphenol A.
BPA
5813 Reducing the release of mercury into the environment.
Hg
Regarding the testing of the chemical content of products
5977 sold at retail.
CSPA
1355
Incorporating human health analysis into environmental
review under chapter 43.21C RCW.
2166
Enacting the Washington safe cosmetics act of 2007.
CSPA
2185
Reducing the levels of benzene in groundwater and
drinking water.
Benzene
2696
Testing for elevated levels of lead in children.
Pb
4007
Requesting Congress and the Environmental Protection
Agency to further regulate benzene.
Benzene
1464
Reducing the environmental impact of cleaning state
facilities.
DES
1570
Authorizing a biomonitoring program.
1601
Creating the children's environmental health and protection
advisory council.
CSPA
1847
Providing for lead poisoning prevention education and
screening.
Pb
2143
Requiring the use of alternatives to lead wheel weights.
Pb
2613
Reducing the environmental impact of cleaning state
facilities.
DES
2695
Creating a pilot program to screen children for lead
poisoning.
Pb
2800
Regarding the use and disposal of mercury-added products. Hg
2818
Concerning the duties of the Department of Ecology's
office of waste reduction and sustainable production.
2882
Concerning the labeling of lead-containing products.
Pb
3059
Requiring coverage for lead blood level assessments.
Pb
3167
Evaluating environmental health conditions in state office
buildings.
DES
6502 Reducing the release of mercury into the environment.
Hg
39
3. Literature Review
As I am interested in evaluating the practical and political limitations to endocrine
disruptor regulation, this literature review evaluates the state of understanding of the
science of endocrine disruption, theories of science-policy interaction and case studies of
chemicals policy broadly and the evolution of endocrine disruptor policy specifically.
3.1 State of the science of endocrine disruption
While the science of endocrine disruption has consistently sought to establish a
language and understanding that supplements and contrasts with that of traditional
toxicology, many of the earliest points of contention remain unresolved as new
distinctions are explored and established. For example, while there is ample evidence that
certain chemicals exert non-monotonic dose responses in animals, the exact mechanism
producing these odd responses remains unclear. It appears likely that these responses are
attributable to multiple different underlying mechanisms, varying with chemical and
context. Further, while it is clear that some endocrine disruptors exert effects at
concentrations below those considered in traditional toxicology or regulatory
frameworks, there is still no consensus on precisely what constitutes a “low-dose effect.”
Similarly, evidence pointing to the perinatal and pubertal periods as being critical
windows of exposure has been produced at all observational levels, but truly robust
demonstrations of the broad concept are inherently impractical. Perhaps most illustrative
of the difficulty faced in resolving the diverse research of the field are the steady
proliferation of acknowledged modes of action and the fundamental shift in testing
methods required to accurately assess the impact of interaction effects.
40
3.1.1 Non-monotonic and low-dose effects
Uncertainty as to the mechanism and prevalence of non-monotonic dose-responses
(NMDRs) and low-dose effects has been a consistent point of contention since the
development of the field, and are perhaps the features of endocrine disruption research
that most distinguish it from toxicology. Although many persuasive mechanistic
explanations for the observed effects have been proposed, difficulties in holistic causal
demonstration have allowed uncertainty as to the pervasiveness and significance of nonmonotonic and low-dose effects to persist (Barlow et al., 2010; Melnick et al., 2002;
Testai et al., 2013).
A review by Barlow et al. (2010) demonstrates the continuously contentious nature of
these two effects, pointing back to the 1992 wingspread conference and subsequent metaanalyses, all of which continue to grapple with the same issue. Barlow et al. cite two
commonly accepted explanations of NMDRs: that they can be explained by
superimposition of grosser chemical effects onto typical hormone effects that themselves
act at very low doses, and that they can be explained by competing observed agonistic
and antagonistic effects of EDs on hormone receptors. Nevertheless, Barlow et al. note
that a lack of scientific consensus persists due to frequent difficulties in replicating
studies that claim to identify non-monotonic effects and a lack of holistic empirical
demonstration of simultaneous contradictory effects of specific chemicals. Barlow et al.
do note that one large critique of the non-monotonic hypothesis had recently been
resolved: while skepticism had surrounded the theory due to the relatively low affinities
of EDs to the two best-studied nuclear estrogen receptors, recent research demonstrated
that ED’s effects can indeed be larger and more rapid, as illustrated by their action on
41
membrane estrogen receptors. In other words, these effects can be observed in animals
without contradicting our granular understanding of effect pathways because they occur
through alternate, understudied pathways. Thus, future study into membrane estrogen
receptors promises to explain the effects of EDs currently inexplicable by nuclear
receptors alone.
A review by Testai et al. (2013) enumerates elements of uncertainty within the NMDR
hypothesis. Many theoretically plausible explanations for NMDRs can explain their
appearance through mechanisms not specific to the endocrine system, such as
cytotoxicity and loss of ED solubility at high doses. Were such generic causes the norm,
it would undermine the separation of endocrine disruption from toxicology. Further,
Testai et al. argue that, while methods of extrapolation from in vitro to in vivo tests is
forthcoming, such translation remains insufficiently demonstrated. This speaks to the
larger failures of translation: extrapolation from cells to organs to organisms remains
beyond our current abilities, as does distinguishing between adverse and adaptive
responses to exposure when observed only at the cellular level. Finally, they argue that a
procedural requirement of the REACh regulations is increasingly at odds with the
development of the field: while statistical power is best increased by increasing the
number of in vivo replicates, this necessity is at odds with a tenet of REACh that requires
measures to be taken to decrease the reliance on animal testing. This move away from
animal testing in the EU impacts our understanding and ability to assess many of the
uncertainties within the field.
Zoeller et al. (2014) take the evaluations of present practical limitations to their logical
extremes, rightly concluding that observational studies can never provide satisfactory
42
evidence of low-dose effects due to limits to observable concentrations in vivo.
Furthermore, for the purposes of human toxicity, Zoeller et al. argue that differences in
sensitivities between individuals (due to age, gender, sympathetic exposures, etc.) renders
general exposure thresholds indefinable. They explain that the determination of
thresholds of effect relies on assumptions relating to mode of action and effect that
cannot be borne out in whole organisms, thus rendering attempts to evaluate safe
exposure thresholds, and by extension attempts to assess risk in a unified fashion, moot.
Zoeller et al. conclude that “to move this debate forward, we must acknowledge first that
dose thresholds are impossible to prove or disprove experimentally” (Zoeller et al., 2014,
p. 5).
Zoeller et al. further note that “Low-dose” itself is a term used inconsistently, based
upon whether it is couched in the toxicological presumption of effects, the observed
biological active effect range, or the estimated exposure level. Thus, to move the debate
forward, they contend that “low dose” must be employed consistently and “adverse”
effects must be explicitly distinguished from adaptive effects.
Thus, it seems clear that NMDRs and low-dose effects, while understood with
increasing sophistication at the granular level, defy translation to the population level.
While there are certainly incremental improvements to be made in predictive abilities,
and while the discipline would benefit from increased repetition of experimentation to
thoroughly defend the prevailing theories, there is reason to believe that NMDRs and
low-dose effects will stymie efforts at risk profiling at the population level for the
foreseeable future.
43
3.1.2Critical periods of exposure
While evidence from animal studies continues to build the case that exposure to
endocrine disruptors during the perinatal period increases the likelihood of developing
many non-communicable diseases in adulthood, and while there is evidence in both
humans and animals of immediate developmental impacts of endocrine disruptors during
that period, translating early-life effects observed in animal studies to human effects
remains a primary point of contention (Fudvoye et al., 2014; Testai et al., 2013;
Vandenberg et al., 2009; Zoeller et al., 2014).
A review by Vandenberg et al. (2009) provides a substantive distinction between
critical and non-critical periods of exposure. They argue that exposure during non-critical
periods is primarily of an “activational” nature, altering the expression of established
systems solely during the period of exposure. In contrast, exposure during critical periods
causes “organizational” impacts, effecting long-term development and overall health. In
support of this, they point to the hypothesis, provisionally demonstrated, that “different
receptors are likely represented in different cell types at different developmental times
and response stages,” (Vandenberg et al., 2009, p. 81) meaning that EDs interact with
different systems during those critical periods than they would at other times. Vandenberg
et al. note that critical periods of exposure have been observed in human subjects, as in
the case of DES, in which subsequent studies of “DES daughters” and sons have
correlated exposure at different times during pregnancy to different outcomes.
Vandenberg et al. also point out the utility of unintentional chemical exposures for
demonstrating this principle, citing an accidental dioxin release in Seveso, Italy,
44
following a chemical plant explosion, where breast cancer risk was observed to increase
most in in perinatal, pubertal, and pregnant individuals.
Attributing the difference critical and non-critical periods to a different mechanism,
Testai et al. (2013) argue that adult organisms contain mechanisms for “adaptive
endocrine modulation” that developing organisms lack, leaving the latter more
susceptible to adverse reactions following ED exposure. Testai et al. also make the point,
in illustrating the limits of our current testing methods to accurately distinguish the
impact of critical periods, that current procedures for evaluating perinatal exposures fail
to observe all possible endpoints, and have not been satisfactorily proven to be
translatable to human exposures. Thus, we most likely do not have a complete
understanding of the difference in effect between critical and non-critical periods of
exposure.
A meta-analysis of early-life effects of endocrine disruptors concluded that direct
causal relationships between EDs and fetal development are scarce, but the heightened
significance of hormones in fetal development is well-understood (Fudvoye et al., 2014).
Further, paralleling the claims by Vandenberg et al. and Testai et al., Fudvoye et al. note
that fetuses may lack biotransformation enzymes that are known to reduce ED
concentrations in adult organisms. While Fudvoye et al. similarly noted the limitation of
human data availability, they further noted that even in extant studies, estimates of
exposure level remain too stochastic to provide predictive power for effect estimates in
moderately-scaled studies. Further, they note that PBDE exposure in utero has been
associated with immediate adverse effects in humans, but has produced non-analogous
results in animal studies, suggesting further the difficulty of translation from animal to
45
human of even short-term effects. Specifically looking towards the critical period of
puberty, Fudvoye et al. note that impacts of EDs on pubertal development are difficult to
evaluate due to the inconsistent timing of puberty and the uncertain relationship between
perinatal exposure and subsequent effects on and during puberty.
Ultimately, limitations to our understanding of developmental biology impose
themselves on our understanding of ED exposure during periods of development. While
animal testing allows for rapid evaluation of multiple generations-worth of effects within
a short period of time, there are demonstrable difference between effects of certain EDs
in animals and humans. Further, while studies of these effects in humans remain unethical
and impractical, accidental chemical exposures have decisively demonstrated the proof of
concept.
3.1.3 Multiple modes of action
One significant aspect of endocrine disruption, mentioned earlier as a partial
explanation of both non-monotonic dose-responses and critical periods of exposure, is the
fact that a single chemical may exert myriad effects through discrete modes of action
(Barlow et al., 2010; Testai et al., 2013; Vandenberg et al., 2009). From the original
concern with estrogenic nuclear steroid at the 1991 Wingspread conference (Colborn and
Clement, 1992), concern has expanded to include many additional hormones and
hormone systems, for example androgens and thyroid hormones, and many additional
endpoints, for example membrane receptors and steroid inhibitors (Barlow et al., 2010).
Further, additional endpoints caused by disruption of these various mechanisms has been
associated with an increasing number of effects; the proliferation of “diseases,
mechanisms and modes of action” requires an increasing number of sophisticated tests to
46
evaluate, leaving the evaluation of many modes of action in the realm of modeling and
hypothesis for the time being (Barlow et al., 2010). Furthermore, even in isolation, a
single chemical can exert differing effects depending on which tissues it is exposed to;
these effects become increasingly unpredictable when considering exposure to multiple
endocrine disruptors (Barlow et al., 2010).
3.1.4 Mixture and interaction effects
Mixture effects of endocrine disruptors have long been studied within the discipline
(Barlow et al., 2010; Testai et al., 2013). Most studies in vivo have focused on multiple
chemicals known to exert similar effects, such as multiple anti-androgens, or chemically
related, such as multiple PCBs or pthalates, and have found additive effects,
demonstrating the potential for straightforward modeling of combined effects from the
same mode of action (Barlow et al., 2010; Testai et al., 2013). A recent meta-analysis of
studies of pesticide cocktails, in which 90% of reviewed studies focused on endocrine
disruption as a mode of action, found that about half of studied mixtures reported additive
effects, and about one third reported synergistic effects (Rizzati et al., 2016). While the
fact of these effects may be well established, one of the largest barriers to comprehensive
analysis, especially for synergistic effects, is the question of ensuring consideration of all
relevant chemicals in modeling (Kortenkamp, 2014). This issue is exacerbated by the
broad lack of data on the endocrine-disrupting potential of many known and suspected
endocrine disruptors.
47
3.2 Transitioning from toxicology to endocrinology
Roberts (2009) characterizes the issue of low-dose toxicity as posing a challenge to the
production of both science and policy, highlights difficulties in transitioning from
toxicology to endocrinology, and reflects on issues in transitioning to a more proactive
approach to federal chemical policy in the United States. Roberts characterizes
endocrinology as treating the body as “a system in constant communication with its
environment,” (Roberts, 2009, p. 8) with multi-modal and iterative responses that
undermine linear dose-response relationship that characterizes toxicology in
contradistinction. Roberts attributes the new paradigm of endocrinology to improvements
in “analytical and instrumental technologies and experimental methods,” (Roberts, 2009,
p. 8) citing increasing numbers of biomonitoring studies as indicators of this trend.
Roberts then describes a proposed REACh-like system for the US, summarized as a
permission-based model of chemical regulation, and a trespass model, wherein chemical
exposure is interpreted as a trespass on our own human bodies when such exposure
occurs without permission. However, Roberts characterized both of these approaches as
being impracticable due to a lack of robust pre-market methods for toxicity testing.
Roberts concludes that, regardless of the precise policy instrument adopted, any new
regulations should allow for the dynamic adoption of new testing methods, to ensure that
our regulatory “institutions [are] as flexible as the current science” (Roberts, 2009, p. 21).
A more recent consensus statement by Vandenberg et al. (2013) addressed this issue,
and made the case for the employment of endocrine disruption principles to regulate EDs,
largely due to the distinctions addressed in section 3.1, above. Responding to the lack of
certainty highlighted by Roberts, Vandenberg et al. promote a weight of evidence
48
approach to chemical evaluation that aligns with principles of endocrinology. Vandenberg
et al. claim that a weight of evidence approach, wherein evidence supporting one side of
an argument is compared to evidence supporting the other side in order to determine
which is more valid, is the default response to uncertain and contradictory studies. This
granted, Vandenberg et al. further claim that evaluating the weight of evidence requires
expressing one’s own values and using one’s own professional judgment. Vandenberg et
al. cite several common, illogical normative decisions made in evaluating the worth of
contrasting claims, namely that failure to refute a null hypothesis is commonly conflated
with accepting that null hypothesis, and that “Good Laboratory Practice” guidelines are
taken as a proxy for appropriate study design, despite only being an indicator of
procedural quality. Vandenberg et al. then contrast these capricious weighting methods
with methods couched in the current understanding of endocrinology: that reliable studies
must find no evidence of contamination by agonists or antagonists related to the effect
being studied, must be capable of identifying low-dose effects if they occur, must test
species or strains sensitive to the effect in question, and should contain both positive and
negative controls. These four requirements concisely illustrate the core distinctions
between toxicology and endocrinology as applied to risk management or any other
political endpoint, and it is clear that adherence to these principles would lead to more
uniform and defensible decisions in the face of uncertainty.
A review by Molander (2015) focused on providing “insights and methods related to
the risk assessment and risk management of chemicals in consumer products” (Molander,
2015, p. vii) identifies two key tenets for improving the risk assessment process:
developing an evaluation process, akin to that proposed by Vandenberg et al. (2013) that
49
identifies relevant and valid studies more accurately than currently accepted guidelines,
and the use of web-based tools to facilitate in the identification and evaluation of relevant
studies. Reflecting Vandenberg et al.’s complaint, the development of alternative study
selection guidelines was spurred by overemphasis of procedural quality and other noncrucial considerations in commonly accepted guidelines. The SciRAP model developed
by Molander, and first described in Beronius et al. (2014), divides criteria of reliability
into two tiers, fulfillment of the first of which which requires only qualitative analysis
and indicates presence of all information essential for the “evaluation of reliability,”
(Molander, 2015, p. 25) and fulfillment of the second of which, in combination with
external criteria to determine relevance, provides the user with a final evaluation of the
“study’s adequacy for health risk assessment” (Molander, 2015, p. 25). Molander
combined the insight derived from SciRAP with another web-based tool that provides
visualization features for study results. Molander’s initial exploration of these tools seems
to indicate that they are effective in distinguishing the more significant studies from the
less. One major impediment to extensive testing of these tools is a lack of transparency in
study design.
It seems clear that commonly accepted evaluation criteria for endocrine disruptor
studies do not reflect the criteria that endocrine disruptor researchers consider to be most
important. Further it seems that these procedural requirements may be occurring at the
expense of more significant but less understood tenets of endocrine disruption, such as
ensuring the absence of unintended agonists or antagonists within studies and thoroughly
documenting study design. While the dedicated work of a small number of individuals
could easily revolutionize the ability to stratify studies based upon their regulatory utility,
50
as is evidenced by the recent creation of web-based tools, future insistence on onerous
and relatively insignificant procedural laboratory requirements may do more to limit the
availability of useful studies than to promote them.
3.3 Models of science-policy interaction
How does a science with as many remaining uncertainties as endocrine disruption
interact with the policy sphere, and how do stakeholders interpret that interaction? This
interface has been elaborated on in many different forms, all of which have both positive
and normative implications as to how best to promote scientific research and structure
policy based upon incomplete knowledge. Being able to associate stakeholders’
understanding of science-policy interaction would recommend certain approaches to
policy and science improvements above others, and could help to ensure that research
being produced aligns with the requirements of the policy sphere. Millstone et al. (2004)
illustrate a clear hierarchy of increasingly complex models, all of which begin from the
precept of a clear division between science and policy. Funtowicz (2006) describes
similar models in order to contrast them with others which normatively defy the
distinction between science and policy.
Millstone et al. (2004) review three regulatory conflicts with the aim of evaluating
stakeholders’ understanding of the science-policy nexus. They compare regulator’s
perspectives on the issues, highlighting differences between the US and EU and within
the EU. Millstone et al. further evaluate the prescriptive and descriptive value of several
models of science-policy interaction: the technocratic, decisionist, and a novel
“transparent” model, all of which they argue reflect different concomitant elements of
policy production.
51
Prescriptively, a technocratic model implies that every regulatory decision is made
with deference only to robust scientific evidence, in a manner nominally more objective
than regulations based upon political considerations. However, Millstone et al. observe
that the assumption of adherence to a purely technocratic model leads representatives to
assume a priori that regulations within their jurisdiction are based objectively on sound
science, and that any jurisdictions producing contradictory regulations must be influenced
by political considerations. Millstone et al. illustrate this point with an exploration of how
precaution is treated in differing jurisdictions. Noting that, in general usage, precaution is
treated as a response to scientific uncertainty, Millstone et al. note that the technocratic
perspective either assumes precaution as an inherent result of sound science, or dismisses
precaution as an unscientific consideration. Millstone et al. go on to note that absolute
deference to scientific certainty can lead to paralysis in its absence, due to a lack of
evidence sufficient to support debate and decision. Given the increasing
acknowledgement of scientific uncertainty, Millstone et al. conclude that most decisions
cannot be explained on the basis of scientific truth alone, and that the technocratic model
fails to account for social elements of the decision-making process.
Millstone et al. contend that the decisionist model is dominant in all their studied
jurisdictions. The decisionist model contends that science occurs independently of and
prior to political decisions, and that political decisions are founded on scientific
conclusions, but inexorably account for “social, political, cultural and economic”
concerns. In this formulation, the scientific process of risk assessment is explicitly and
entirely distinct from the political process of risk management, the former being objective
and the latter subjective. While this framework more readily explains the differences in
52
regulations between jurisdictions with access to the same scientific knowledge, it fails to
explain differences in presumed-objective risk assessments. In this framework, precaution
is applicable only to the political, second step, and is entirely distinct from the scientific
process.
Addressing the shortcomings in the decisionist model, Millstone et al. propose a
modified version, the “transparent” model, that presents the process of risk assessment
not as being entirely objective, but as being influenced “by legal requirements and by
social, economic and political judgments” (Millstone et al., 2004, p. 26). This model is
labeled “transparent” in the belief that, were those “up-stream” assumptions that
influence risk assessment made transparent, the reasons for differing results of risk
assessments would be self-evident. Belief in this model would prompt risk assessors to be
more transparent about the assumptions underpinning their own work, and would policy
makers to carefully consider the impacts of risk assessment policy on risk management.
In this model, precaution can be considered as one of the up-stream assumptions that risk
assessors must implicitly or explicitly consider in framing the risk assessment process.
Funtowicz (2006) presented a series of models of science-policy interaction developed
within the Joint Research Centre of the European Commission. These models are
distinguished by the importance that positive factual “contributions of experts” and
normative value-laden “contributions of other sectors” have in the policy making process,
with a goal of “assuring the quality of knowledge-inputs to the decision making process.”
(Funtowicz, 2006, p. 139) The practical application of these models to environmental
problems has subsequently been tested in several case studies (Dessai & Van Der Sluijs,
2007; Lemus, 2015; Udovyk, 2014; Van Der Sluijs et al., 2008). Funtowicz describes four
53
models of science-policy interaction: The Initial Modern Model, The Precautionary
Model, The Model of Framing, and The Model of Extended Participation.
The “modern” model of science policy interaction, also called the linear or
knowledge transfer model (Pregernig & Böcher, 2012) and closely paralleling the
“technocratic” model described by Millstone et al. (2004), sees the relationship of science
and policy as “speaking truth to power.” Frequently associated with the concept of
technocracy, this model assumes that all scientific knowledge is purely positive, but
provides sufficient (commonly formulated as complete) information to enable normative
policy decisions to be made. In this formulation, science and policy spheres are
completely separate, with the latter simply drawing from the knowledge of the former;
hence a “linear” model of science-policy interaction.
The precautionary model discussed by Funtowicz reflects a strong formulation of
the precautionary principle. Essentially, any possible threat must be responded to and
avoided, regardless of cost. This model does not take comprehensive scientific
knowledge as a given, and in fact assumes persistent scientific uncertainties. Nonetheless,
in this model policy remains wholly dependent upon scientific understanding, with the
added caveat that policy-makers may frequently call upon scientists to improve the state
of understanding to relieve the state of uncertainty. Thus, policy makers may have large
impacts on the precise avenues of scientific inquiry, but said inquiry is still assumed to
remain wholly positive.
Another formation of this principle, not explored by Funtowicz is the “weak”
formulation of the precautionary principle, commonly associated with the 1992 UN Rio
Declaration on Environment and Development, which requires precaution only in the
54
face of the possibility of “serious or irreversible damage.” This reformulation primarily
changes the requests that policy-makers would make of scientists, and subsequently the
issues that scientists would study, effectively implementing a narrower form of triage
than the “strong” formulation.
The consensus model explored by Udovyk (2014) is in many respects similar to
Millstone et al.’s (2004) “decisionist” model. Rather than there being a single truth
illuminated by science, the consensus model admits of multiple, conflicting truths. The
realm of science is thus tasked with consensus-seeking, prior to presenting finding to the
policy realm. In this formulation, however, the two realms remain distinct and science is
still seen as having a solely positive role. Certainty of evidence, in contrast to the modern
model, is not considered a priori to be absolute, and thus it is only through conscious
efforts to confirm the robustness of data that it can then be found of use in policy-making:
the onus remains on the science realm to demonstrate significance before an issue is
considered in the policy realm.
Funtowicz (2006) approaches this model somewhat differently, in his “Model of
Framing.” This model ascribes differences in truths to differences in the scientific
framework from which the question was approached, adjoining a normative aspect to the
scientific process based upon the underlying assumptions of the varying disciplines. By
introducing normative values to the scientific sphere, this model blurs the lines between
the two, and presents disciplines as stakeholders defending their beliefs. Choosing one
over the other then is inherently a matter of choice, and thus disagreements in this model
are normatively resolved with little respect to the degree of certainty underlying any
given result.
55
The extended participation model denies science an authoritative position as the
sole purveyor of truth, giving equal weight to the informal knowledge of citizens or other
groups (Funtowicz, 2006). By giving equal footing to all knowledge, this model sustains
no distinction between the science and policy spheres. Because scientific findings are
only indirectly equated with knowledge, the question of certainty becomes only indirectly
relevant. The “co-production of knowledge” (Funtowicz, 2006, p. 139) ensures that the
question of uncertainty is dealt with before entering the realm of science-policy
interaction.
The different models of science-policy interaction all differ in many respects. Of
special interest to this thesis is the degree of certainty required to stimulate policy
production. Both the adaptive management and the extended participation model ignore
the question of uncertainty during policy production, and as such cannot be grouped. The
remaining models, however, can be ordered along a spectrum of required certainty (Table
6).
Table 6: Policy models in order from most to least reliance upon certainty
Model
Degree of uncertainty acceptable for policy production, from least
to greatest
“Modern”/Technocratic Model
Burden of proof rests on the shoulders of the science sphere in order to
influence the policy sphere
Consensus/Decisionist Model
Inter-disciplinary certainty required to inform policy
Model of Framing/Transparent
Model
Significance of certainty dwarfed by prior assumptions of disciplines
most relevant to policy
“Weak” Precautionary Model
Uncertainty maintaining the possibility of “serious or irreversible
damage” is taken seriously.
“Strong” Precautionary Model
All uncertainty is acknowledged and actively embraced
56
3.4 Relevant case studies
Wurtz and Sorenson (2011) report on a series of workshops organized by the Nordic
Council of Ministers. Motivated by the idea that robust regulation of endocrine disruptors
would not be enacted for decades, participants in the workshops reviewed soft regulatory
approaches and risk communication strategies intended to reduce chemical exposure in
the interim. Workshop participants described the regulatory landscapes, successes, and
failures in the Nordic countries. The Danish EPA executed a campaign to raise awareness
about ED hazards posed to toddlers, publishes list of medications known to not contain
pthalates, regularly produces lists of “undesirable chemical substances,” and regularly
conducts consumer surveys focusing on chemical exposure. This work is supplemented
by Danish NGOs who regularly pursue chemical bans. The Norwegian Climate and
Pollution agency stressed stressed cooperation and an understanding of the target
audience as being crucial to soft regulatory success. The agency promotes “eco-labeled
products” and “green public procurement,” coordinates with industries to promote
chemical reform, and administers a web page describing and highlighting common
sources of “the 13 most dangerous substances in consumer products.” Similar to the
Danish EPA, Norway’s agency targets its information campaigns towards groups capable
of moderating the exposure of the most vulnerable groups. The Swedish government
developed a national action plan due to increasing consumer concern, largely due to a
documentary illustrating pervasive contamination. The goal of the plan was to improve
the dialogue with industries and to promote voluntary regulation. This work has been
supplemented by consumer outreach materials, primarily targeting the parents of young
children and and daycare centers. In contrast to the other Nordic countries, Finland
57
observed a decrease in NGO interest in chemicals following the passage of REACh. The
government is able to cooperate effectively with business groups, and chemical
regulations are generally understood to be adequate. While consumer awareness
campaigns had also been pursued, the largest shift in chemicals management in Finland
was the unification of regulatory authority within a single agency.
Shamasunder and Morello-Frosch (2015) interviewed chemical regulation
stakeholders throughout the USA to identify points of agreement or contention in regards
to biomonitoring as a scientific and policy tool. Shamasunder and Morello-Frosch
confidentially interviewed advocacy, government, industry, and academic scientists who
had been vocal in the literature and/or in government or industry meetings. Interviews
were in-depth and semi-structured, and were coded using a two-tiered system, wherein
primary codes related to the interviewee’s relationship to the issue, and sub-codes
reflected specific issues, including the intersection of biomonitoring and policy.
Shamasunder and Morello-Frosch observed that while advocacy scientists believe that
biomonitoring influences policy, and that biomonitoring is relevant in promoting policy
change, industry scientists believed neither statement to be true. Regulatory scientists
found the first statement to be true in certain cases, and considered the second statement a
matter of open debate. Academic scientists indicated that the former statement was
partially true, but generally agreed with the latter. The simple delineation of issues
surrounding biomonitoring, valuation of responses as being either positive, negative, or
mixed, and clear prior delineation of scientific spheres lend themselves to concise,
comprehensible results.
58
A study of this kind, combining document study and semi-structured interviews, was
performed by Udovyk (2014). Udovyk attempted to characterize the model of sciencepolicy interaction underlying bisphenol-a management at the international, national, and
municipal level in Gothenburg, Sweden. This thesis adopts many of the models of
science-policy interaction put forth in Udovyk’s study. Udovyk found that most policy
was informed by the modern model of science-policy interaction, and thus had the most
to say about the particular pros and cons of that model. Subsequently, the other models
were only evaluated relative to the de facto standard of the modern model.
Cáceres, Silvetti & Díaz (2016) use a case study of an Argentinian law that
protects native forests in order to evaluate the descriptive power of two alternate
formations of science-policy interaction in the case of legislative failure. They found that
the “information deficit model,” which corresponds most closely to the “modern” model,
failed to adequately describe the origin of enacted legislation. Instead, they found that a
“power-dynamics model,” in which legislation is iteratively influenced by many “actors
and types of knowledge,” (Cáceres et al., 2016, p. 57) best explained why certain
legislation with less stringently scientific underpinnings was enacted in lieu of a more
thoroughly researched policy. They concluded that a “modern” model of science-policy
interaction fails to reflect the roles played by “institutions, subjectivities, values, interests,
power relationships, [and] knowledge” (Cáceres et al., 2016, p. 62) in the creation of
policy.
Lemus (2015) compared the differing approaches of Denmark to BPA regulation,
with a focus on the role of uncertainty. Lemus found that Denmark’s more active
approach to regulating BPA at a national level stemmed from approaching the problem
59
from an “endocrine-perspective,” which encouraged common understanding among
multiple stakeholders, including scientists and regulators. This in turn led to the
exploration of technical questions in a regulatory setting, and subsequent funding support
to adequately answer these questions. The case of Denmark, according to Lemus,
highlights the failure of the “modern” model of science-policy interaction to account for
multiple and divergent truths with which science speaks to policy (Funtowicz and Strand,
2007). In the context of Norway, Lemus observes that BPA has consistently been
regulated as an environmental, rather than as an endocrine-disrupting, chemical of
concern. Lemus explains that this occurs as the result of the greater power afforded to the
technocratic food regulators, who treat BPA from a risk management perspective.
Meanwhile, the environmental authorities of Norway, who promote a precautionary
approach, lack the influence to enact such an approach. Thus, stemming from differing
normative goals, there exists an uncomfortable disagreement amongst Norwegian experts
as to the best approach for regulating BPA.
60
4. Methods and Results
While the science underpinning endocrine disruptors continues to develop, it is clear
that the available data remains insufficient to engender their regulation under the
preexisting regulatory framework: the mode of action of endocrine disruption undermines
many of the primary assumptions of toxicology, necessitating—at a minimum—
translational tools that do not yet exist and cannot be produced without a significant
amount of primary research. Meanwhile, many alternative policy instruments and holistic
methods of measuring hazard have been developed and employed to varying degrees of
success in jurisdictions beyond Washington. From the review of proposed and recently
enacted legislation related to endocrine disruptors in Washington, it is clear that certain
proposed policy alternatives and testing methods have been afforded regulatory
consideration, while others have not. To determine how the choices that have defined and
delimited our current policy landscape were made, and with what degree of autonomy, I
pursued a qualitative policy analysis focused on a document review of primary and
secondary sources relating to proposed regulations and on stakeholder interviews
encompassing the full spectrum of active participants, focusing my efforts on current
policy limitations and contemporary scientific and geopolitical advances.
4.1 Qualitative policy analysis
I chose to approach the issue of endocrine disruptor policy from a qualitative
perspective due to a vast range of perspectives relevant to the study. Because policies in
Washington have rarely addressed the concept of “endocrine disruptor,” it was unclear
initially whether the concept was broadly used, or whether it was translated into other
terms for the purposes of policy. Further, without a broad understanding of historical
61
attempts to regulate endocrine disruptors, any quantitative approach might have
overlooked or overemphasized certain approaches. Further, evaluating models of sciencepolicy interaction involves an understanding of historical intent, and within the body of
literature consistently relies on qualitative analysis. Most generally, addressing the issue
through a qualitative lens allowed for the discovery of salient considerations that would
potentially be overlooked in a purely quantitative study. Approaching the issue from a
qualitative perspective enriches any subsequent research by providing a baseline context
and preliminary interpretations.
I was mindful of my own biases throughout the research process, and strove to
overcome them by making them explicit. Having more experience with toxicology than
endocrinology, I was more comfortable with the uncertainties of the field than most
stakeholders would, and was by extension predisposed to accept a hazard based approach
as being a sound basis for regulation. Bearing these biases in mind as I evaluated my
data, I focused more on the positive logic of my interviewees statements than the
normative content of their statements, attempting to weigh equally all internally sound
arguments.
4.2 Research design
The overall format of this project was primarily influenced by three earlier papers
focused on chemicals management. Lemus’ (2015) case study and comparative analysis
of Bisphenol A regulation in Denmark and Norway was conducted primarily through
document review and stakeholder interviews. The questions asked in that case study, and
the epistemological considerations discussed, were useful in the framing of my own
research. Shamasunder’s (2011) study on biomonitoring data’s influence on US chemical
62
policy was useful in its content and its methodological approach, specifically in terms of
its approach to defining and delineating policy stakeholders. Matus, Clark, Anastas, and
Zimmerman’s (2012) analysis of barriers to green chemistry implementation provided a
further delineation of stakeholders, as well as a framework for evaluating and
distinguishing issues. Trujillo’s (2016) case study of microplastics, demonstrated the
feasibility of looking at a class of materials or chemicals at a broader scale.
I chose to perform a case study because of the lack of existing case studies looking at
chemical policy on the same scale. While the above-mentioned papers reviewed
individual chemicals, scientific methods and policy tools, Trujillo’s focus on
microplastics as a broad class, encompassing many aspects of policy, demonstrated the
feasibility and utility of such an approach. While chemical policy in Washington State
addresses persistent bioaccumulative toxins, this grouping is slightly different from that
of Europe and other jurisdictions introducing REACh-like chemicals policy. Thus, a case
study focusing on endocrine disruptors in the context of Washington State stood to reframe existing policy, allowing for subsequent comparisons to other jurisdiction’s
policies, and for ease of future policies on those grounds within the state.
4.3 Data collection
4.3.1 Policy documents
My primary concern in collecting policy documents for analysis was casting a wideenough net to ensure the capture of the full range relevant documents. Within the
legislature, I reviewed all enacted laws, and house and senate bills dating back as far as
were available. Having identified laws relating to endocrine disruptors, I identified the
63
range of State Departments granted regulatory authority. I then reviewed the regulations
related to those laws as enacted by the various departments, and further searched for
other relevant programs and initiatives sponsored by those agencies. In pursuit of relevant
policies, and given the un-institutionalized nature of the concept “endocrine disruptor,” I
initially took a broad view of what could be considered relevant policy. Additional
documents submitted by non-governmental stakeholders in response to various policies
were considered when available. Finally, I systematically included a question regarding
past and current relevant policies in the interview process, to identify anything of
significance that I may have missed.
4.3.2 Stakeholder interviews
The first step in approaching stakeholders was classifying them in terms of their
relation to endocrine disruptor policy. I began by defining groups of stakeholders based in
part upon the groups described by Shamasunder (2011) and Matus et al. (2012), and
identifying local organizations, agencies, and institutions that fit those categories. This
identification process stemmed from reviewing authorship of the aforementioned policy
documents and observing and reviewing public comment sessions relating to proposed
legislations and regulations. From the broad list of stakeholders, I classified individuals
and organizations as being either policy-makers, policy-analysts, policy-enforcers,
scientists, public interest representatives, or private interest representatives. Additionally,
during the interview process, additional interview participants were identified using the
snowball technique.
Having classified the various groups, I strove to interview a representative portion of
each group, to collect as many perspectives and present as broad a view of each group as
64
possible. Interviews were arranged either by phone call, e-mail, or in person and occurred
between March and May of 2017. Interviews were recorded and subsequently
transcribed, and were conducted under the auspices of confidentiality. In total, 11
interviews were conducted, ranging from 15 to 90 minutes. Unfortunately, I was unable
to interview what I believe to be a representative portion of policy-makers or private
interest representatives.
While the interviews were conducted in an open-ended fashion, I also relied on a set
of questions designed to correspond to the questions of interest identified by the literature
review. This list of questions was stylistically inspired by the questions in Lemus (2016).
See appendix two for the list of questions. However, this list served as more of a baseline
than as a rigid structure for the interview; different sections were more salient to different
stakeholders’ experiences, and certain questions were skipped if earlier questions
indicated a lack of relevance.
4.3.3 Iterative approach
It is important to note that these processes did not occur independently of each other.
While I began with an initial document review, additional documents were subsequently
brought to my attention through the interview process, as were additional points of
interest meriting additional lines of inquiry in subsequent interviews.
4.4 Data analysis
Once the interviews had been transcribed, they were coded using the RQDA library of
the R statistical computing platform. The coding structure was based in part on the format
used by Shamasunder (2011), in conjunction with the discrete areas of consideration
65
identified in the literature review. After initial coding, additional categories presented
themselves, and interviews were re-coded. From the identified policy documents, those
which were considered most relevant were also coded in the same manner.
Following the coding, the basic perspectives of the different stakeholder groups were
compared, both semi-quantitatively, by comparing the presence and frequency of certain
perspectives and arguments, and qualitatively. Responses were used to evaluate
compatibility with various policy approaches and consideration. Additionally, the various
theories of science-policy interaction were compared to the facts and opinions expressed
in the interviews.
4.5 Results
Interviewees largely agreed as to the limitations of Washington State’s current
endocrine disruptor policies. The information in the following section is presented in a
manner intended to aggregate the most common concerns and highlight disagreements
amongst stakeholders where they exist. Results related to scientific advances and
potential policy improvements are presented without reference to the role of their
originator, except to highlight the occasional differences between stakeholder groups in
these matters: while many suggestions were made for future policy approaches, few of
them were directly contradictory.
4.5.1 Current limitations of Washington State endocrine disruptor policy
Undoubtedly the largest impediment to the further adoption of endocrine disruptor
legislation described by interviewees was that of scientific uncertainty. Most interviewees
referenced a lack of information sufficient to expand legislation under the current system.
66
Certain scientists and citizens group representatives, however, referenced bodies of
evidence that they saw as being sufficient cause for increased legislation. However, while
they described this evidence as being sufficient to necessitate policy, it was
simultaneously insufficient to dictate the precise formulation of that policy. The table
below delineates the most frequently cited impediments, and their perceived significance
within stakeholder groups.
Table 7: Stakeholder assessment of current impediments
Impediments
Limitation
Of concern to…?
Regulato Legislati
ry
ve
Industry
Advocac Academi
y
c
Scientific Uncertainty
Yes
Yes
Yes
Yes
Yes
Shared Data Inavailability
Yes
---
Yes
Yes
Yes
Yes
---
Yes
No
Yes
Yes
Yes
No
Yes
---
Yes
No
No
Yes
---
Regrettable Substitution
Yes
---
Yes
Yes
Yes
Funding
Yes
Yes
---
Yes
Yes
Political Will
Yes
Yes
Yes
Yes
Yes
Multiple Definitions of
EDC
Inability to Target
Vulnerable Demographics
Reliance on Individual
Chemical Regulation
While representatives of the business community felt that there was extensive testing
being done on commercial chemicals, they felt that there was still extreme uncertainty
surrounding the sources of exposure. This relationship between source and outcome was
repeatedly mentioned as being under-researched, albeit not as strongly emphasized by
other interviewees. An interview from the Department of Ecology highlighted these
issues stating that the department doesn’t “have methods [to biomonitor] most …
67
chemicals [and] ... only [has] methods for about 500 chemicals… and we’ve got probably
30,000 [chemicals] in commerce, and we’ve probably got a couple thousand that are high
production volume.” In general, the failure to firmly establish cause and effect in an
environmental context was seen as clouding the case for legislation.
One of the most immediate problems, as discussed by scientists, is that at the finest
level of observation, adverse and adaptive changes to the endocrine system are
indistinguishable, necessitating more taxing observations of indirect effect to determine
the ultimate endpoint. Ultimately, the difficulty of establishing a cause and effect
relationship appears to be the greatest remaining point of contention, and the most
immediate impediment to the implementation of any broad or precautionary endocrine
disruptor legislation.
One concern that added to the uncertainty was that of trading data with the EU and
Canada. Many interviewees expressed remorse and dissatisfaction stemming from the
difficulties of sharing data with Canada, despite our countries’ parallel efforts in this
field. This lack of collaboration between Canada and Washington state was attributed to
Washington’s inability to ensure the confidentiality of provided data, due to the state’s
right-to-know laws. This has led to repetition and duplication of existing work, and
contributed to the unnecessary persistence of uncertainty.
Contentions of the definition and scope of the phrase “endocrine disruptor” among
interviewees mimicked those of the overall debate. Interviewees from state agencies
acknowledged the issue and clarified their own stance as essentially coming down to
triage; excluding from consideration, for example, those that don’t have an effect at an
environmentally relevant level. While most were uncomfortable with the fuzziness of the
68
definition, participants generally considered this to be a smaller issue than that of
uncertainty.
Because of practical limitations, policies attempting to target a certain demographic
must employ proxies for that demographic that are frequently imprecise. As described by
an Ecology employee, “the sports containers [clause] was the attempt to get at women of
childbearing age, since half of all pregnancies in the US are unintended.” In the case of
legislation addressing specific materials or products, there was consensus that the choice
of material or product to regulate influenced which demographics were impacted by the
legislation.
Paralleling the consideration of the level of specificity at which to regulate products is
the question of scale at which to regulate chemicals, if a broader policy were to be
implemented. While interviewees who mentioned the topic agreed on the insufficiency of
regulating a singular chemical, several mentioned difficulties of addressing chemical
classes as a whole. In the example of the proposed PFAS in grease-resistant food liners
legislation, it was mentioned that while palpable risk has been demonstrated for longchain PFASs, short-chain PFASs have not been demonstrated to present the same risk.
While manufacturers have largely switched to short chain PFASs in that application, the
combined uncertainty contributed to the bill’s failure. Additionally, an Ecology
interviewee mentioned the issue of generalizing whole classes for the purposes of
endpoints, as chemicals in a given class may be generally dangerous, but may be so by
differing mechanisms. Thus, the question of the specificity of endpoint becomes
entangled with the question of specificity of chemical or class.
69
A corollary to this frustration with the frequent limitations to scope of endocrine
disruptor regulation is the demonstrable fear that regulation of individual chemicals in the
absence of replacement guidelines leads to regrettable substitutions. This was illustrated
by an advocate as having taken place following the PBDE ban: because only a single
flame retardant was regulated, despite contemporary evidence that similar chemicals
within the same family had similarly deleterious effects, PBDEs in plastics were slowly
replaced by a similar chemical with similar beneficial and harmful properties. This
concern was shared by industry groups, who expressed interest in avoiding subsequent
replacement costs that would result from the continued expansion of regulation.
Along these lines, a department of ecology employee highlighted the difficulty of
public education in the context of products that may be doing both good and harm:
There’s chemicals in sunscreen that are probably endocrine disruptors, but you have
to be very careful about that in public health, because we also want people to
protect their skin from the sun, because we know UV light is a mutagen. We know
it causes skin cancer… So, same thing with chemicals in breast milk … we have to
be very careful when we’re putting those messages out, because we want women to
breastfeed.
This sentiment reflects a prime concern shared by the majority of stakeholders,
illustrating a fundamental issue posed by uncertainty. There may not, at least in the short
term, be substitutes for critical chemicals that do not share some of their negative effects.
Thus, the majority of stakeholders acknowledged that evaluating both risks and hazards
requires an understanding of the benefits and harms of the chemical in question as well
as the benefits and harms of potential substitutes.
Alongside purely practical matters of implementation, many stakeholders described
the economic restraints hindering the implementation of wider-reaching policies. In
70
regards to current policy, however, representatives from Ecology noted that the
Department is in a relatively comfortable position, specifically in comparison to other
states. However, Ecology and business representatives mentioned that Ecology’s
capacities for product testing were dwarfed by many companies in the state. Health’s
work in relation to endocrine disruptors is less well funded. While outreach is funded
through the state, the biomonitoring program was the result of a federal grant that went
un-renewed due to competition from other states. This lack of population survey data
impacts the ability of Health to best inform citizens, and for Ecology to determine which
chemicals are of greatest concern.
Interviewees presented a unified view in describing the political considerations
relevant to passing policy. Certain legislators have, in the past, championed specific
causes, and repeatedly brought bills before the house or senate. Now, on the other hand,
legislative committees on environmental issues are relatively cool to any toxics
proposals. Furthermore, by some unspoken tradition, it is unlikely for bills to gain
traction during their first or second legislative session. This is indicated by the multiple
“failed” bills that end up being iterated on for several years before being passed, and
appears to be an accepted aspect of the process. Broadly speaking, changing membership
of relevant house and senate committees, and partisan politics in the house and senate
mean that the actual passage of certain regulations in any given year is due more to
political caprices than to the actual substance of the bill.
4.5.2 Scientific advancements and endocrine disruptor policy
As indicated by the literature review, progress surrounding endocrine disruptors has
expanded the scope of the field while simultaneously coming up against obstacles in
71
analysis that are understood to be insurmountable. Nevertheless, while advances in
understanding may not have materialized, many incidental aspects of the science have
improved and potentially lend themselves to incorporation into future regulation. The
following table delineates scientific advances by the role that they would play in future
regulation, and indicates the specific examples of each role discussed in the interviews.
Table 8: Elements and specific applications of scientific advances
Scientific Advances
Blacklists
EU and California Lists of Chemicals of Concern
Whitelists
EPA Safer Chemical Ingredients List
Knowledge-sharing
Interstate Chemicals Clearinghouse (IC2)
Hazard Assessment
Tools
GreenScreen and TiPED
Mixture Effects
NRC’s Pthalates and Cumulative Risk Assessment (2009)
Data Collection
Efficiency
EPA procedure for high-throughput chemical screening
While various blacklists, or lists of suspected or known endocrine disruptors exist, the
lists administrated by the European Union and by California denote chemicals thoroughly
demonstrated to be endocrine disruptors. These lists were identified as being of use by a
regulatory representative, who explained that they are sufficiently robust to recommend
the addition of chemicals to the CSPA list without much further analysis.
While a robust analysis of the potential harm of a chemical or chemicals may remain
beyond the means of an individual state, the Interstate Chemicals Clearinghouse collects
chemical studies produced by Washington and other member states for the benefit of all.
72
This reduces redundant testing and allows for access to thorough documentation of a
wide range of chemicals that can be interpreted to reflect any regulatory standards.
In contrast to the lists of known bad actors indicated in the above-mentioned
blacklists, the EPA Safer Chemical Ingredients List, provides a whitelist that guarantees
the safety of a wide range of chemicals. This is of use both in the Department of
Enterprise Service’s current procurement role, and in Ecology’s work promoting green
chemistry and alternatives assessment to industry.
Central to Ecology’s advocacy work in the field of green chemistry are GreenScreen
and TiPED, screening protocols for chemical production that are designed to maximize
the efficiency of the testing process by identifying endocrine disruptors through the
simplest possible process. Essentially, both protocols delineate a series of tests ascending
in complexity from computer model-based to in vivo mammalian experiments, balancing
the cost of iterative assays that increase in certainty against the high base cost of in vivo
assays alone.
While the above scientific advances all take the form of redundancy avoidance,
advances in the understanding of mixture effects, the high-throughput chemical screening
assay developed by the EPA in service of the Endocrine Disruptor Screening Program and
the principles described by the National Research Council in Pthalates and Cumulative
Risk Assessment promise to reduce the work needed to estimate dose-response curves
and to evaluate the interaction effects of multiple chemicals from the same class,
respectively. These both reduce the cost and time required to fully evaluate the endocrinedisrupting properties of a given chemical, and allow for extrapolation of that information
into the context of interaction effects.
73
4.5.3 Policy recommendations
The recommendations presented in the table below and further described at length
represent aggregations of recommendations made by multiple interviewees.
Recommendations primarily fell into one of the four categories below. Of
recommendations that directly addressed the existing regulatory framework, some
fundamentally altered some aspect of the framework while others merely adapted it in
attempts to better the nature of the endocrine disruptor issue. Beyond these
recommendations that were intended to alter stakeholders’ relationships to existing
regulation, either by addressing industry or by addressing consumers. The most frequent
and most open-ended recommendations are laid out in the following table and described
below, as are several more specific recommendations.
Table 9: Recommended policy changes, grouped by degree of alteration and target
audience
Fundamental
Alterations:
Minor
Adaptations:
Industry-Focused Consumer-Focused
Recommendations: Recommendation:
Cede deliberative Expand CSPA to Share standardized
authority to
regulate
risk/hazard metrics
agencies
additional
w/ industry
products/
materials
Encourage consumer
exercise of
preferential
purchasing
Adopt a hazard
assessment
paradigm
Group chemicals Promote or require
by class by
alternatives
default
assessments
Increase consumer
education
Mandate
Biomonitoring
---
---
Promote or require
green chemistry
74
While controversial among legislators and industry representatives, the
recommendation to cede agencies, specifically the Departments of Ecology and
Health, the authority to regulate chemicals following their own thorough analysis,
was supported by the remaining stakeholders. This step was seen by many as a
natural extension of the current process, wherein Ecology develops the case for
regulating a chemical, goes before the legislature to receive authority to regulate
said chemical, and proceeds to regulate the chemical.
Similarly, the recommendation that chemical regulation be founded on the basis
of hazard is echoed by regulatory, advocacy, and academic stakeholders. Notably,
however, this view was not expressed by representatives of the Department of
Health, wherein risk analysis is more frequently required and produced. Both
industry and legislative interviewees considered the determination of risk as
significant to future regulation of endocrine disruptors, for the purposes of costbenefits analysis and of evaluative legislative parity, respectively.
While not explicitly mentioned by all stakeholders, the suggestion to mandate
biomonitoring was proposed by regulatory, advocacy, and academic stakeholders.
Biomonitoring was described as a crucial tool in the ultimate development of risk
and exposure estimates, and thus would seem to reflect the interests of industry
and legislative representatives who rely on risk analysis.
Two simple adaptations of the current law, illustrated by the failed electronics
amendment and PFAS bill from the 2017 legislative session, are the expansion of
75
CSPA to mandate reporting on a broader range of products and materials, and the
holistic regulating of chemical classes, rather than individual chemicals. While
both bills were criticized during session for their expanded scope, wide-scale
support for their implementation remains.
While the exact nature of the relationship differed in the telling between
industry and regulatory interviewees, both parties seemed interested in fostering a
closer relationship. The sharing of standard metrics, much like the IC2 data
sharing, would obviate the possibility of potentially disruptive differences in
interpretation of scientific data and of regulatory requirements. For instance,
industry and agency scientists agreed to use the same tests and decision framework
in testing for evidence of endocrine disruption. Further, in line with existing
programs at Ecology, both parties would be happy to see the Department expand
their advocacy for alternatives assessment and green chemistry. Certain regulators
and advocates would like to see these principles enshrined in regulation.
Finally, given that interviewees worked directly or indirectly for the public,
there was near-unanimous interest in improving consumer awareness, and
enthusiasm for consumers’ power to direct market forces away from endocrine
disruptors, if so inclined. While note in opposition to the notion, legislators noted
the wide range of issues that the public is expected to be aware of, and questioned
the emphasis of this issue over others. Additionally, advocates and academics
76
recommended caution and context in the explanation, so as to produce a lasting
understanding of the issue rather than ephemeral panic.
77
5. Discussion, Conclusion and Recommendations
5.1 Discussion
It is very possible that certain seeming differences between stakeholder groups are in
fact merely artifacts of the improvement of my interview style over time. I spoke to
industry representatives near the beginning of the interview process, and while I
subsequently reviewed my questions, I did not have the opportunity to meet with industry
representatives later in the process. Conversely, I met with advocacy representatives near
the close of my research, and perhaps for that reason spoke with them on a wider variety
of subjects. It would no doubt have been helpful to increase the number of interviewees,
or to perform small follow-up interviews to address topics previously overlooked.
By far the most diverse stakeholder group was that of the legislators, representing as
they did different parts of the state with wildly varying interests. It is this stakeholder
group that is simultaneously most significant for the evaluating of policy practicality, and
this group that is more stochastic. No doubt further interviews with legislators would
have yielded the greatest improvements to my results.
Perhaps unsurprisingly, given the relatively steady state of the science as characterized
in the literature review, there was very little contention or discussion about the validity of
the science. Certain interviewees preferred a risk-based approach to a hazard-based
approach, or vice versa, but none seemed concerned with the uncertainties surrounding
endocrinology. Whether this is because they understood the science to be effectively
settled, as reflected in the numerous consensus statements made by endocrinologists, or
didn’t realize that there were fundamental questions unanswered and unanswerable, was
not immediately clear. More to the point, it seems as if the public conversation is simply
78
focused on issues more fundamental than uncertainty or interaction effects for the time
being.
5.2 Conclusion and recommendations
From what I’ve gathered of the legislative process, the simple adaptations of existing
regulation have already been set in motion, and are awaiting a politically favorable
climate, and substantial public scrutiny, on their roundabout journey to legalization. Even
some of the grander regulatory goals are being fomented; Ecology continues to
demonstrate their ability to evaluate the safety of chemicals in a manner satisfactory to
the legislature,
It is clear, however, that the slow pace of scientific advancement combined with the
deliberative legislative system has ensured that this is an issue that will evolve not over
years but over decades. Barring a sea-change in one process or the other, the dominoes
will continue to fall slowly.
While the robustness of these results are limited due to the small number of
interviews, the underlying topics of interest identified and the specific policies suggested
provide a substantial underpinning for follow-up, semi-quantitative research, such as
stakeholder surveys. Such a survey would provide broader evaluation of the potential
popularity of many of the concepts merely identified within this work, allowing for a
much clearer delineation of values.
It is for that reason that the application of this thesis to the legislation of endocrine
disruptors in other states, or other potentially harmful classes of materials in any
jurisdiction is so important: any pitfall identified in the process as being easily skirted
79
could point to the format of future legislation and allow that legislation to approach its
final form that much earlier in the process. A review of this thesis, contrasting the
observed process with that which occurred during the regulation of heavy metals, or that
occurring with the regulation of microplastics, pharmaceuticals, and nanometals, could
highlight even more fundamental stages of policy progression, and allow for future
motions to be moved through with less resistance.
Other states, such as Oregon, have begun to explicitly follow in Washington’s
footsteps in regards to endocrine disruptor policy, producing their own response to
the Children’s Safe Products Act. There is already healthy communication between
Washington and Oregon, but a careful analysis of similarities and differences
would no doubt provide valuable information both about the jurisdictions and the
process.
80
References Cited
Food and Drug Administration. (2012). Indirect food additives: polymers. Fed Reg, 77,
41899-41902.
Almeida, C. M. V. B., Madureira, M. A., Bonilla, S. H., & Giannetti, B. F. (2013).
Assessing the replacement of lead in solders: effects on resource use and human
health. Journal of Cleaner Production, 47, 457-464.
Baluka, S. A., & Rumbeiha, W. K. (2016). Bisphenol A and food safety: Lessons from
developed to developing countries. Food and Chemical Toxicology, 92, 58-63.
Barlow, S., Schlatter, J., Öberg, T., Castoldi, A., Cutting, A., Jacobs, M., ... & Rortais, A.
(2010). Scientific report of the endocrine active substances task force. EFSA
Journal, 8(11), 59.
Baumgartner, F. R., & Jones, B. D. (1993). Agendas and Instability in American Politics.
Chicago, IL: University of Chicago Press.
Beronius A, Molander L, Rudén C and Hanberg A (2014). Facilitating the use of nonstandard in vivo studies in health risk assessment of chemicals: a proposal to
improve evaluation criteria and reporting. Journal of Applied Toxicology 34(6),
607-617.
Beronius, A., & Vandenberg, L. N. (2015). Using systematic reviews for hazard and risk
assessment of endocrine disrupting chemicals. Reviews in Endocrine and
Metabolic Disorders, 16(4), 273-287.
Bergman, Å., Becher, G., Blumberg, B., Bjerregaard, P., Bornman, R., Brandt, I., ... &
Iguchi, T. (2015). Manufacturing doubt about endocrine disrupter science–A
rebuttal of industry-sponsored critical comments on the UNEP/WHO report “State
of the Science of Endocrine Disrupting Chemicals 2012.” Regulatory Toxicology
and Pharmacology, 73(3), 1007-1017.
Bergman, Å., Heindel, J. J., Kasten, T., Kidd, K. A., Jobling, S., Neira, M., ... & Brandt, I.
(2013). The impact of endocrine disruption: a consensus statement on the state of
the science. Environmental Health Perspectives, 121(4), A104-A106.
Björnsdotter, M. K., de Boer, J., & Ballesteros-Gómez, A. (2017). Bisphenol A and
replacements in thermal paper: A review. Chemosphere, 182, 691.
81
Björnsdotter, M. K., Jonker, W., Legradi, J., Kool, J., & Ballesteros-Gómez, A. (2017).
Bisphenol A alternatives in thermal paper from the Netherlands, Spain, Sweden and
Norway. Screening and potential toxicity. Science of The Total Environment, 601,
210-221.
Bridbord, K., & Hanson, D. (2009). A personal perspective on the initial federal healthbased regulation to remove lead from gasoline. Environmental Health Perspectives,
117(8), 1195.
Bryman, A. (2012). Social Research Methods. Oxford, UK: Oxford university press.
Buonsante, V. A., Muilerman, H., Santos, T., Robinson, C., & Tweedale, A. C. (2014).
Risk assessment׳s insensitive toxicity testing may cause it to fail. Environmental
Research, 135, 139-147.
Burggren, W. W., & Mueller, C. A. (2015). Developmental critical windows and sensitive
periods as three-dimensional constructs in time and space. Physiological and
Biochemical Zoology, 88(2), 91-102.
Cáceres, D. M., Silvetti, F., & Díaz, S. (2016). The rocky path from policy-relevant
science to policy implementation—a case study from the South American Chaco.
Current Opinion in Environmental Sustainability, 19, 57-66.
Carson, R. (1962). Silent Spring. Boston, MA: Houghton Mifflin Harcourt.
Colborn, T. and Clement, C. (1992). Chemically-Induced Alterations in Sexual and
Functional Development: The Wildlife/Human Connection. Advances in Modern
Environmental Toxicology, 21, 1-8.
Colborn, T., vom Saal, F. S., & Soto, A. M. (1993). Developmental effects of endocrinedisrupting chemicals in wildlife and humans. Environmental Health Perspectives,
101(5), 378.
Costa, E. M. F., Spritzer, P. M., Hohl, A., & Bachega, T. A. (2014). Effects of endocrine
disruptors in the development of the female reproductive tract. Arquivos Brasileiros
de Endocrinologia & Metabologia, 58(2), 153-161.
Creswell, J. W., & Clark, V. L. P. (2007). Designing and Conducting Mixed Methods
Research. Thousand Oaks, CA: SAGE Publications.
82
Damstra, T., Barlow, S., Bergman, A., Kavlock, R., & Van Der Kraak, G. (2002). Global
Assessment of the State-of-the-Science of Endocrine Disruptors. Geneva: World
Health Organization.
Davies, H., Washington State Department of Ecology, Washington State Department of
Health. (2009). Washington State Lead Chemical Action Plan. Lacey, WA:
Washington State Department of Ecology.
Dekant, W., & Colnot, T. (2013). Endocrine effects of chemicals: aspects of hazard
identification and human health risk assessment. Toxicology Letters, 223(3), 280286.
Delile, H., Blichert-Toft, J., Goiran, J. P., Keay, S., & Albarède, F. (2014). Lead in ancient
Rome’s city waters. Proceedings of the National Academy of Sciences, 111(18),
6594-6599.
Dessai, S., & van der Sluijs, J. P. (2007). Uncertainty and Climate Change Adaptation: A
Scoping Study. Utrecht, The Netherlands: Copernicus Institute for Sustainable
Development and Innovation, Department of Science Technology and Society.
Doke, S. K., & Dhawale, S. C. (2015). Alternatives to animal testing: A review. Saudi
Pharmaceutical Journal, 23(3), 223-229.
Eisner, M. A. (2007). Governing the environment: The Transformation of Environmental
Regulation. Boulder, CO: Lynne Rienner Publishers.
Faure, M. G., & Lefevere, J. G. (1998). An analysis of alternative legal instruments for
the regulation of pesticides. Regulating Chemical Accumulation in the
Environment: The Integration of Toxicology and Economics in Environmental
Policy-making, 249.
Foster, W. G., & Agzarian, J. (2008). Toward less confusing terminology in endocrine
disruptor research. Journal of Toxicology and Environmental Health, Part B, 11(34), 152-161.
Fudvoye, J., Bourguignon, J. P., & Parent, A. S. (2014). Endocrine-disrupting chemicals
and human growth and maturation: a focus on early critical windows of exposure.
Vitam Horm, 94, 1-25.
83
Funtowicz, S.O. (2006) Why knowledge assessment? In Pereira, A., Vaz, S., & Tognetti,
S. (Ed.). Interfaces Between Science and Society. Sheffield, UK: Greenleaf
Publishing.
Funtowicz, S., & Strand, R. (2007). Models of science and policy. Biosafety first:
Holistic approaches to risk and uncertainty in genetic engineering and genetically
modified organisms, 263-278.
Gallagher, M. J. (2000). Proposed Strategy to Continually Reduce Persistent,
Bioaccumulative Toxins (PBTs) in Washington State. Lacey, WA: Washington State
Department of Ecology.
Gibson, D. A., & Saunders, P. T. (2014). Endocrine disruption of oestrogen action and
female reproductive tract cancers. Endocrine-Related Cancer, 21(2), T13-T31.
Giuliano, C. A., & Rybak, M. J. (2015). Efficacy of triclosan as an antimicrobial hand
soap and its potential impact on antimicrobial resistance: a focused review.
Pharmacotherapy: The Journal of Human Pharmacology and Drug Therapy,
35(3), 328-336.
Groshart, C., & Okkerman, P. C. (2000). Towards the Establishment of a Priority List of
Substances for Further Evaluation of Their Role in Endocrine DisruptionPreparation of a Candidate List of Substances as a Basis for Priority Setting. BKH
Consulting Engineers for European Commission Directorate-General for the
Environment: Delft, The Netherlands.
Gunningham, N., Grabosky, P., & Sinclair, D. (1998). Smart Regulation: Designing
Environmental Policy. Oxford, UK: Clarendon Press.
Gutleb, A. C., Cambier, S., & Serchi, T. (2016). Impact of endocrine disruptors on the
thyroid hormone system. Hormone Research in Paediatrics, 86(4), 271-278.
Hass, U. (2005). OECD Conceptual Framework for Testing and Assessment of Endocrine
Disrupters as a Basis for Regulation of Substances with Endocrine Disrupting
Properties. Copenhagen, Denmark: Nordic Council of Ministers.
Heindel, J. J., Newbold, R., & Schug, T. T. (2015). Endocrine disruptors and obesity.
Nature reviews. Endocrinology, 11(11), 653.
84
Hu, D. P., Hu, W. Y., Xie, L., Li, Y., Birch, L., & Prins, G. S. (2016). Actions of
estrogenic endocrine disrupting chemicals on human prostate stem/progenitor cells
and prostate carcinogenesis. The Open Biotechnology Journal, 10(1), 76-97.
Huang, F., Wen, S., Li, J., Zhong, Y., Zhao, Y., & Wu, Y. (2014). The human body burden
of polybrominated diphenyl ethers and their relationships with thyroid hormones in
the general population in Northern China. Science of the Total Environment, 466,
609-615.
Johnson, C. H. (2004). Transsexualism: An Unacknowledged Endpoint of Developmental
Endocrine Disruption? Olympia, WA: The Evergreen State College.
Kabir, E. R., Rahman, M. S., & Rahman, I. (2015). A review on endocrine disruptors and
their possible impacts on human health. Environmental Toxicology and
Pharmacology, 40(1), 241-258.
Kerr, S., & Newell, R. G. (2003). Policy‐Induced Technology Adoption: Evidence from
the US Lead Phasedown. The Journal of Industrial Economics, 51(3), 317-343.
Khetan, S. K. (2014). Endocrine Disruptors in the Environment. Hoboken, NJ: John
Wiley & Sons.
Kingdon, J. W. (1984). Agendas, Alternatives, and Public Policies. New York, NY:
Longman Publishing Group.
Knez, J. (2013). Endocrine-disrupting chemicals and male reproductive health.
Reproductive Biomedicine Online, 26(5), 440-448.
Knower, K. C., To, S. Q., Leung, Y. K., Ho, S. M., & Clyne, C. D. (2014). Endocrine
disruption of the epigenome: a breast cancer link. Endocrine-Related Cancer,
21(2), T33-T55.
Kortenkamp, A. (2007). Ten years of mixing cocktails: a review of combination effects of
endocrine-disrupting chemicals. Environmental Health Perspectives, 115(Suppl 1),
98.
Kortenkamp, A. (2014). Low dose mixture effects of endocrine disrupters and their
implications for regulatory thresholds in chemical risk assessment. Current
Opinion in Pharmacology, 19, 105-111.
85
Kristensen, L. J. (2015). Quantification of atmospheric lead emissions from 70 years of
leaded petrol consumption in Australia. Atmospheric Environment, 111, 195-201.
Kummerer, K. (2010) Emerging Contaminants. In Wilderer, P. A., Frimmel, F. (Ed.).
Treatise on Water Science. Amsterdam: Elsevier.
LaKind, J. S., & Naiman, D. Q. (2015). Temporal trends in bisphenol A exposure in the
United States from 2003–2012 and factors associated with BPA exposure: spot
samples and urine dilution complicate data interpretation. Environmental Research,
142, 84-95.
Lamb, J. C., Boffetta, P., Foster, W. G., Goodman, J. E., Hentz, K. L., Rhomberg, L.
R., ... & Williams, A. L. (2014). Critical comments on the WHO-UNEP State of the
Science of Endocrine Disrupting Chemicals–2012. Regulatory Toxicology and
Pharmacology, 69(1), 22-40.
Lamb, J. C., Boffetta, P., Foster, W. G., Goodman, J. E., Hentz, K. L., Rhomberg, L.
R., ... & Williams, A. L. (2015). Comments on the opinions published by Bergman
et al.(2015) on Critical Comments on the WHO-UNEP State of the Science of
Endocrine Disrupting Chemicals (Lamb et al., 2014). Regulatory Toxicology and
Pharmacology, 73(3), 754-757.
Lemus, D. (2015). The Regulation of Bisphenol A in Denmark and Norway: How the
Problem of Chemical Safety is Framed and Addressed Amidst Scientific
Uncertainty Ås, Norway: Norwegian University of Life Sciences.
Levallois, P., St-Laurent, J., Gauvin, D., Courteau, M., Prévost, M., Campagna, C., ... &
Rasmussen, P. E. (2014). The impact of drinking water, indoor dust and paint on
blood lead levels of children aged 1–5 years in Montréal (Québec, Canada).
Journal of Exposure Science & Environmental Epidemiology, 24(2), 185.
Liao, C., & Kannan, K. (2013). Concentrations and profiles of bisphenol A and other
bisphenol analogues in foodstuffs from the United States and their implications for
human exposure. Journal of Agricultural and Food Chemistry, 61(19), 4655-4662.
Lorber, M., Schecter, A., Paepke, O., Shropshire, W., Christensen, K., & Birnbaum, L.
(2015). Exposure assessment of adult intake of bisphenol A (BPA) with emphasis
on canned food dietary exposures. Environment International, 77, 55-62.
Lynn, S. G. (2012, July). EDSP21: The Incorporation of In Silico Models and In Vitro
High Throughput Assays in the Endocrine Disruptor Screening Program (EDSP)
86
for Prioritization and Screening. In Vitro Cellular & Developmental BiologyAnimal, 48, 13.
Ma, Y., Halsall, C. J., Crosse, J. D., Graf, C., Cai, M., He, J., ... & Jones, K. (2015).
Persistent organic pollutants in ocean sediments from the North Pacific to the
Arctic Ocean. Journal of Geophysical Research: Oceans, 120(4), 2723-2735.
Markell, D. (2010). An overview of TSCA, its history and key underlying assumptions,
and its place in environmental regulation. Washington University Journal of Law
and Policy, 32, 333-375.
Mason, R. (1998). The analysis of market and regulatory failure. In Regulating Chemical
Accumulation in the Environment. Swanson, T., & Vighi, M. (Ed.). Cambridge,
UK: Cambridge University Press.
Makarova, K., Siudem, P., Zawada, K., & Kurkowiak, J. (2016). Screening of toxic
effects of bisphenol A and products of its degradation: zebrafish (Danio rerio)
embryo test and molecular docking. Zebrafish, 13(5), 466-474.
Manikkam, M., Tracey, R., Guerrero-Bosagna, C., & Skinner, M. K. (2013). Plastics
derived endocrine disruptors (BPA, DEHP and DBP) induce epigenetic
transgenerational inheritance of obesity, reproductive disease and sperm
epimutations. PloS One, 8(1), e55387.
Matus, K. J., Clark, W. C., Anastas, P. T., & Zimmerman, J. B. (2012). Barriers to the
implementation of green chemistry in the United States. Environmental Science &
Technology, 46(20), 10892-10899.
McLeod, M. (2005). Washington state Department of Ecology historically speaking: an
oral history of the first 35 years, 1970–2005. Olympia, WA: Washington State
Department of Printing.
Melnick, R., Lucier, G., Wolfe, M., Hall, R., Stancel, G., Prins, G., ... & Moore, J. (2002).
Summary of the National Toxicology Program's report of the endocrine disruptors
low-dose peer review. Environmental Health Perspectives, 110(4), 427.
Menon, S., George, E., Osterman, M., & Pecht, M. (2015). High lead solder (over 85%)
solder in the electronics industry: RoHS exemptions and alternatives. Journal of
Materials Science: Materials in Electronics, 26(6), 4021-4030.
87
Michałowicz, J. (2014). Bisphenol A–sources, toxicity and biotransformation.
Environmental Toxicology and Pharmacology, 37(2), 738-758.
Millstone, E., van Zwanenberg, P., Marris, C., Levidow, L., & Torgersen, H. (2004).
Science in Trade Disputes Related to Potential Risk: Comparative Case Studies.
Luxembourg: European Communities.
Muennig, P. (2009). The social costs of childhood lead exposure in the post–lead
regulation era. Archives of Pediatrics & Adolescent Medicine, 163(9), 844-849.
National Research Council. (2009). Phthalates and Cumulative Risk Assessment: the
Tasks Ahead. Washington, D.C.: National Academies Press.
National Research Council. (2013). Environmental decisions in the face of uncertainty.
Washington, D.C.: National Research Council.
Nilsson, E. E., & Skinner, M. K. (2015). Environmentally induced epigenetic
transgenerational inheritance of disease susceptibility. Translational Research,
165(1), 12-17.
Organisation for Economic Co-operation and Development. (2008). OECD
Environmental Outlook to 2030. Paris, France: Organisation for Economic Cooperation and Development.
Palanza, P., Nagel, S. C., Parmigiani, S., & vom Saal, F. S. (2016). Perinatal exposure to
endocrine disruptors: sex, timing and behavioral endpoints. Current Opinion in
Behavioral Sciences, 7, 69-75.
Pichery, C., Bellanger, M., Zmirou-Navier, D., Glorennec, P., Hartemann, P., &
Grandjean, P. (2011). Childhood lead exposure in France: benefit estimation and
partial cost-benefit analysis of lead hazard control. Environmental Health, 10(1),
44.
Pregernig, M., & Böcher, M. (2012). Normative and analytical perspectives on the role of
science and expertise in environmental governance. Environmental Governance:
The challenge of legitimacy and effectiveness, 199-219.
Rasmussen, J. J., Wiberg-Larsen, P., Baattrup-Pedersen, A., Cedergreen, N., McKnight,
U. S., Kreuger, J., ... & Friberg, N. (2015). The legacy of pesticide pollution: An
overlooked factor in current risk assessments of freshwater systems. Water
research, 84, 25-32.
88
Rezg, R., El-Fazaa, S., Gharbi, N., & Mornagui, B. (2014). Bisphenol A and human
chronic diseases: current evidences, possible mechanisms, and future perspectives.
Environment international, 64, 83-90.
Richards, K. R. (1999). Framing environmental policy instrument choice. Duke
Environmental Law and Policy Forum, 10, 221.
Riley, E.P., Vorhees, C.V. (1986). Handbook of Behavioral Teratology. Berlin, Germany:
Plenum Press.
Rissman, E. F., & Adli, M. (2014). Minireview: transgenerational epigenetic inheritance:
focus on endocrine disrupting compounds. Endocrinology, 155(8), 2770-2780.
Rizzati, V., Briand, O., Guillou, H., & Gamet-Payrastre, L. (2016). Effects of pesticide
mixtures in human and animal models: An update of the recent literature. ChemicoBiological Interactions, 254, 231-246.
Rochester, J. R., & Bolden, A. L. (2015). Bisphenol S and F: a systematic review and
comparison of the hormonal activity of bisphenol A substitutes. Environmental
Health Perspectives, 123(7), 643.
Sabatier, P. A. (1999) Theories of the Policy Process. Boulder, CO: Westview Press.
Schug, T. T., Abagyan, R., Blumberg, B., Collins, T. J., Crews, D., DeFur, P. L., ... &
Hayes, T. (2013). Designing endocrine disruption out of the next generation of
chemicals. Green Chemistry, 15(1), 181-198.
Schug, T. T., Blawas, A. M., Gray, K., Heindel, J. J., & Lawler, C. P. (2015). Elucidating
the links between endocrine disruptors and neurodevelopment. Endocrinology,
156(6), 1941-1951.
Shamasunder, B. (2011). Body Burden Politics: How Biomonitoring Data is Influencing
Chemicals Governance in the US. University of California, Berkeley.
Shukla, S. J., Huang, R., Austin, C. P., & Xia, M. (2010). The future of toxicity testing: a
focus on in vitro methods using a quantitative high-throughput screening platform.
Drug Discovery Today, 15(23), 997-1007.
Sirotkin, A. V., & Harrath, A. H. (2014). Phytoestrogens and their effects. European
Journal of Pharmacology, 741, 230-236.
89
Skinner, M. K. (2014). Endocrine disruptor induction of epigenetic transgenerational
inheritance of disease. Molecular and Cellular Endocrinology, 398(1), 4-12.
Skjevrak, I., Due, A., Gjerstad, K. O., & Herikstad, H. (2003). Volatile organic
components migrating from plastic pipes (HDPE, PEX and PVC) into drinking
water. Water Research, 37(8), 1912-1920.
Slama, R., Bourguignon, J. P., Demeneix, B., Ivell, R., Panzica, G., Kortenkamp, A., &
Zoeller, T. (2016). Scientific Issues Relevant to Setting Regulatory Criteria to
Identify Endocrine Disrupting Substances in the European Union. Environmental
Health Perspectives, 124(10), 1497-1503.
Soto, A. M., Brisken, C., Schaeberle, C., & Sonnenschein, C. (2013). Does cancer start in
the womb? Altered mammary gland development and predisposition to breast
cancer due to in utero exposure to endocrine disruptors. Journal of Mammary
Gland Biology and Neoplasia, 18(2), 199-208.
State of Washington Office of the Governor (2016). Assisting Community and Agency
Responses to Lead in Water Systems(Directive of the Governor 16-06). Olympia,
WA: State of Washington Office of the Governor
Steward, K. (2016) CSPA Rulemaking Workshop. Slideshow presented at the Department
of Ecology, Lacey, WA.
Swanson, T. M., & Vighi, M. (1999). Regulating Chemical Accumulation in the
Environment. Cambridge University Press.
Tanabe, S. (2002). Contamination and toxic effects of persistent endocrine disrupters in
marine mammals and birds. Marine Pollution Bulletin, 45(1), 69-77.
TEDX (2017). TEDX List of Potential Endocrine Disruptors. Retrieved November 8th,
2017 from https://endocrinedisruption.org/interactive-tools/tedx-list-of-potentialendocrine-disruptors/
Testai, E., Galli, C. L., Dekant, W., Marinovich, M., Piersma, A. H., & Sharpe, R. M.
(2013). A plea for risk assessment of endocrine disrupting chemicals. Toxicology,
314(1), 51-59.
Thatcher, M. (1998). The development of policy network analyses from modest origins to
overarching frameworks. Journal of Theoretical Politics, 10(4), 389-416.
90
Tickner, J. A. (2003). Precaution, environmental science, and preventive public policy.
NEW SOLUTIONS: A Journal of Environmental and Occupational Health Policy,
13(3), 275-282.
Tijani, J. O., Fatoba, O. O., & Petrik, L. F. (2013). A review of pharmaceuticals and
endocrine-disrupting compounds: sources, effects, removal, and detections. Water,
Air, & Soil Pollution, 224(11), 1770.
Trasande, L., Zoeller, R. T., Hass, U., Kortenkamp, A., Grandjean, P., Myers, J. P., ... &
Bellanger, M. (2016). Burden of disease and costs of exposure to endocrine
disrupting chemicals in the European Union: an updated analysis. Andrology, 4(4),
565-572.
Troisi, R., Hatch, E. E., & Titus, L. (2016). The Diethylstilbestrol Legacy: A Powerful
Case Against Intervention in Uncomplicated Pregnancy. Pediatrics,
138(Supplement 1), S42-S44.
Trujillo, S. (2016). Zero-valent Metal Nanomaterials in Waterways: Using Microplastics
as a Case Study to Develop the Overdue Policy Platform. Olympia, WA: The
Evergreen State College.
Tyshenko, M. G., Phillips, K. P., Mehta, M., Poirier, R., & Leiss, W. (2008). Risk
communication of endocrine-disrupting chemicals: improving knowledge
translation and transfer. Journal of Toxicology and Environmental Health, Part B,
11(3-4), 345-350.
Udovyk, O. (2014). Models of science–policy interaction: Exploring approaches to
Bisphenol A management in the EU. Science of the Total Environment, 485, 23-30.
UNEP/WHO (2013). World Health Organization, United Nations Environment
Programme (WHO-UNEP). In: Bergman, A., Heindel, J.J., Jobling, S., Kidd, K.A.,
Zoeller, R.T. (Ed.). State of the Science of Endocrine Disrupting Chemicals.
Geneva, Switzerland: World Health Organization.
U.S. EPA (1997). Special report on Environmental endocrine disruption: An effects
assessment and analysis office of research and development. REPA/630/R-96/012.
Washington, D.C.: United States Environmental Protection Agency.
U.S. EPA (2017) TSCA Chemical Substance Inventory. Accessed November 8th, 2017
from https://www.epa.gov/tsca-inventory
91
Van der Sluijs, J. P., Petersen, A. C., Janssen, P. H., Risbey, J. S., & Ravetz, J. R. (2008).
Exploring the quality of evidence for complex and contested policy decisions.
Environmental Research Letters, 3(2), 024008.
Vandenberg, L. N., Ågerstrand, M., Beronius, A., Beausoleil, C., Bergman, Å., Bero, L.
A., Bornehag, C-G., Boyer, C.S., Cooper, G.S., Cotgreave, I., Gee, D, Grandjean,
P., Guyton, K.Z., Hass, U., Heindel, J.J., Jobling, S., Kidd, K.A., Kortenkamp, A.,
Macleod, M.R., Martin, O.V., Norinder, U., Scheringer, M., Thayer, K.A., Toppari,
J., Whaley, P., Woodruff, T.J., & Rudén, C. (2016). A proposed framework for the
systematic review and integrated assessment (SYRINA) of endocrine disrupting
chemicals. Environmental Health, 15(1), 74.
Vandenberg, L. N. (2014). Non-monotonic dose responses in studies of endocrine
disrupting chemicals: bisphenol a as a case study. Dose Response, 12(2), 259-276.
Vandenberg, L. N., & Bowler, A. G. (2014). Non-monotonic dose responses in EDSP Tier
1 guideline assays. Endocrine Disruptors, 2(1), e964530.
Vandenberg, L. N., Colborn, T., Hayes, T. B., Heindel, J. J., Jacobs, D. R., Lee, D. H., ...
& Welshons, W. V. (2013). Regulatory decisions on endocrine disrupting chemicals
should be based on the principles of endocrinology. Reproductive Toxicology, 38, 115.
Vandenberg, L. N., Maffini, M. V., Sonnenschein, C., Rubin, B. S., & Soto, A. M. (2009).
Bisphenol-A and the great divide: a review of controversies in the field of
endocrine disruption. Endocrine Reviews, 30(1), 75-95.
Vasquez, M. I., Lambrianides, A., Schneider, M., Kümmerer, K., & Fatta-Kassinos, D.
(2014). Environmental side effects of pharmaceutical cocktails: what we know and
what we should know. Journal of Hazardous Materials, 279, 169-189.
Vested, A., Giwercman, A., Bonde, J. P., & Toft, G. (2014). Persistent organic pollutants
and male reproductive health. Asian Journal of Andrology, 16(1), 71.
Vorvolakos, T., Arseniou, S., & Samakouri, M. (2016). There is no safe threshold for lead
exposure: alpha literature review. Psychiatriki, 27(3), 204-214.
Walt, G., & Gilson, L. (1994). Reforming the health sector in developing countries: the
central role of policy analysis. Health Policy and Planning, 9(4), 353-370.
92
Walt, G., Shiffman, J., Schneider, H., Murray, S. F., Brugha, R., & Gilson, L. (2008).
‘Doing’health policy analysis: methodological and conceptual reflections and
challenges. Health Policy and Planning, 23(5), 308-317.
Wang, C. F., & Tian, Y. (2015). Reproductive endocrine-disrupting effects of triclosan:
Population exposure, present evidence and potential mechanisms. Environmental
Pollution, 206, 195-201.
Wang, L., Asimakopoulos, A. G., & Kannan, K. (2015). Accumulation of 19
environmental phenolic and xenobiotic heterocyclic aromatic compounds in human
adipose tissue. Environment International, 78, 45-50.
Wattigney, W. A., Irvin-Barnwell, E., Pavuk, M., & Ragin-Wilson, A. (2015). Regional
variation in human exposure to persistent organic pollutants in the United States,
NHANES. Journal of Environmental and Public Health, 2015, 571839.
Wilson, V. S., LeBlanc, G. A., Kullman, S., Crofton, K., Schmieder, P., & Jacobs, M. N.
(2016). Where do we go from here: Challenges and the future of endocrine
disrupting compound screening and testing. PeerJ Preprints, 4, e2605v1.
Woodruff, T. J., Carlson, A., Schwartz, J. M., & Giudice, L. C. (2008). Proceedings of the
summit on environmental challenges to reproductive health and fertility: executive
summary. Fertility and Sterility, 89(2), e1-e20.
Yang, O., Kim, H. L., Weon, J. I., & Seo, Y. R. (2015). Endocrine-disrupting chemicals:
review of toxicological mechanisms using molecular pathway analysis. Journal of
Cancer Prevention, 20(1), 12.
Zoeller, R. T., Bergman, Å., Becher, G., Bjerregaard, P., Bornman, R., Brandt, I., ... &
Skakkebaek, N. E. (2014). A path forward in the debate over health impacts of
endocrine disrupting chemicals. Environmental Health, 13(1), 118.
93
Appendices
Appendix 1: List of interview questions
Opening Questions:
Could you please explain your work as it relates to endocrine disrupting chemicals?
What regulations, policies or practices are relevant for your work with EDCs?
Questions relating to the science of EDCs
Are there debates in your field related to EDCs, or is there largely consensus?
How do the fields of toxicology and endocrinology factor into the decision-making
process?
In your own opinion, what is the best way of dealing with scientific uncertainty? Who
decides?
Where do you get your scientific information from? Is there enough research/competence
on the topic?
Questions relating to state policy
Could you explain how local, state, federal and international policies impact your work
with EDCs?
How much leeway is there for individual states to regulate EDCs under the federal
umbrella?
How economically prepared are we to deal with EDCs in Washington State?
How are the responsibilities relating to EDCs organized in Washington?
Are EDCs high on the political agenda?
94
Questions regarding best policy approach
In your opinion, how successful is current Washington State EDC policy?
Which factors should drive policy-making: scientific, environmental, social, ethical,
economic?
What are some simple changes that could substantially improve WA EDC policy?
What would an ideal EDC policy look like?
Questions regarding local stakeholders
What stakeholder groups are impacted by the regulation of EDCs?
Are there some stakeholders that are more influential than others in dictating policy?
95
Appendix 2: List of codes
Policy Considerations
Economic – References to $/labor/time costs of ED legislation
Environmental
Political
Practical – physical/temporal limitations to/requirements of implementation
Redundancy – related to question of nexus of power – is this already done by companies
or mandated at the federal level?
Social – Which groups do/should care and why?
Uncertainty
Policy Debates
Collaboration – Between regulatory bodies e.g. Health, Ecology, other states, feds, etc.
DefinitionEDC – Which definition is being/should be used?
Funding
Goal-Vision – does the regulation adopt a goal-based or a vision-based approach?
GreenChem – are principles of green chemistry/predictive modeling etc. being
employed?
Institutionalization – who has the authority; how much authority; applied how broadly?
Risk-Hazard – should we maintain a risk-based or a hazard-based approach to
regulation
StakeholderInvolvement
StateVsFed – At what level should/can regulations occur? Policy Considerations
96
97
